Ontario Superior Court of Justice

COURT FILE NO.: CV-10412884

DATE: 2020-03-30

BETWEEN:

MDS INC. and MDS (CANADA) INC., c.o.b. MDS NORDION Plaintiffs

• and -

FACTORY MUTUAL INSURANCE COMPANY, c.o.b. FM GLOBAL Defendant

Counsel: Brian Brock and Camille Walker, for the Plaintiffs David Liblong and James Kitts, for the Defendant

HEARD: March 4, 2019; September 3, 9-13, 16-20, 2019

J. WILSON J.

[1] Atomic Energy of Canada Limited (“AECL”) conducts research and produces radioisotopes at the Nuclear Research Universal Reactor at Chalk River, Ontario (the “NRU”). On May 14, 2009 an unanticipated leak of heavy water containing radioactive Tritium was discovered coming through the calandria wall in the NRU. The shutdown as a result of this leak, anticipated to be thirty-six hours, lasted fifteen months. The Plaintiffs, MDS Inc. and MDS (CANADA) Inc. c.o.b. MDS Nordion (“MDS” or the “Plaintiffs”) purchased the radioisotopes produced at the NRU and in turn processed them for sale for profit worldwide. Radioisotopes are used for cardiac imaging, cancer treatment and sterilization of medical products.

[2] At the time of the shutdown, the defendant insurer, Factory Mutual Insurance Company, FM Global (“FM” or the “Insurer”) had issued a worldwide All-Risks policy to MDS (the “Policy”) covering property as described in the Policy against losses from all risks of physical loss or damage except as excluded.

[3] The Policy has a limit of liability of US$700,000,000, subject to specifications for limits of various aspects of the coverage.

[4] FM concedes that the NRU constitutes property of the type insured located at a Contingent Time Element Location within the meaning of the Policy.[^1] This aspect of the Policy has a limit of US$25,000,000. It is conceded that MDS experienced loss of profits during the period of the shutdown of the NRU from May 2009 to August 2010 in the amount of CA$121,248,000.

[5] MDS submitted a claim for loss of profits under the Contingent Time Element coverage on May 21, 2009. Various communications took place. FM denied coverage on August 4, 2009 claiming that exclusions in the Policy apply. This action was commenced in October 2010.

[6] A number of questions must be answered to decide this case:

• Are the Plaintiffs’ losses recoverable or do the corrosion or nuclear radiation exclusions in the Policy apply?
• If the corrosion exclusion applies, does the resulting damage exemption apply allowing the Plaintiffs’ recovery?
• If there is resulting physical damage, does the idle period exclusion limit the period of time for the recovery of the resulting damage?

[7] In the Statement of Claim the Plaintiffs claim damages and expenses for the Insurer’s breach of contract for failing to pay the amounts due under the Policy. In lieu of the pre-judgment interest rate stipulated by the Courts of Justice Act, the Plaintiffs seek compensation flowing from the Insurer’s breach reflecting the actual cost of borrowing by MDS from the date of loss to the release of this decision.

[8] If the Plaintiffs are successful in their claim, should pre-judgment interest be awarded based upon the Plaintiffs’ actual borrowing costs over the years on a compounded basis?

[9] As this is a lengthy judgment, I outline a Decision Index noting paragraph numbers and page references:

DECISION INDEX

PART I: OVERVIEW

Overview – paras. 10-22, pages 6-7

How a nuclear reactor works – paras. 23-37, pages 7-9

AECL Information Not Available to the Public – paras. 38-44, page 9

Known Leaking from the Reflector to the J-rod Annulus and Corrosion – paras. 45-59, pages 9-11

The Two Kinds of Corrosion – paras. 60-74, pages 11-12

• Generalized Corrosion
• The Second Unanticipated Corrosion Causing the Leak

Trial Procedure and the Appointment of a Court-Appointed Expert – paras. 75-84, pages 12-13

Lay Witnesses – paras. 85-87, page 13

Issues for Determination – paras. 88-90, pages 13-15

The Shutdown and Steps that Followed – paras. 91-122, pages 15-18

PART II: THE ABSENCE OF WATER TESTING

Admissibility of Evidence concerning Water testing – paras. 126-130, pages 18-19

• Ruling on Admissibility – paras. 131-137, pages 19-20
• Dr. Mancey’s evidence on water testing – paras. 138-149, pages 20-22

Conclusions as to water quality – paras. 150-162, pages 22-24

PART III: LEGAL PRINCIPLES AND UNDISPUTED FACTUAL CONTEXT

General Legal Principles: Interpretation of All-Risks Policies – paras. 162-172, pages 24-26

Fortuity in All-Risk Policies – paras. 173-175, page 26

Overview of the Policy: What is covered by this ‘All Risks’ Policy? – paras 176-183, pages 26-27

Factual Context: The Relationship between the Parties prior to the Shutdown – paras. 184-191, pages 26-27

Chronology of communication between MDS and FM after the shutdown to the denial of coverage – paras. 192-216, pages 28-32

The Information Available to FM by August 4, 2009 – paras. 217-230, pages 32-34

PART IV – ISSUE 1: DOES THE CORROSION EXCLUSION APPLY PRECLUDING COVERAGE?

Preliminary Matters – paras. 233-238, pages 34-35

Corrosion Issues – para. 239, page 35

Question 1: Can the Applicability of the Corrosion Exclusion be Determined as Suggested by the Insurer by Definition Alone? – paras. 240-268, pages 36-40

• The Parties’ Positions – paras. 240–251, pages 36-37
• Conclusion: The Meaning of Corrosion in the Policy is Ambiguous – paras. 252-254, page 37
• FM Position Taken in the Denial Letter – paras. 255-259, pages 37-38
• Review of Evidence of Mr. Lodge and Mr. Turcotte – paras. 260-268, pages 38-40

Question 2: If the Term Corrosion is Ambiguous, what is its Meaning in the Context of the Policy? – paras. 273-316, pages 40-46

• Caselaw on Meaning of Corrosion – paras. 273-286, pages 40-42
• Lantheus is Distinguishable – paras. 287-300, pages 43-44
• Fortuity and the US Cases – paras. 301-310, pages 44-45
• The Definition of Corrosion in this Policy – paras. 311-316, pages 45-46

Question 3: What is the cause of the corrosion at J-41? Review of the Expert Evidence paras. 321-401, pages 47-58

• Early Notification Report May 27, 2009 – paras. 321-326, page 47
• Dr. Mancey’s Evidence – paras. 327-329, page 47
• The Interim Mancey Report and the External Review – paras. 330-336, page 48
• The Final Mancey Report – paras. 337-343, page 49
• Further Research and Reports after the Release of the Final Mancey Report – paras. 344-346, page 50
• Dr. Newman’s Evidence – paras. 347-348, page 50
• The Rule 53 Experts and the Court-Appointed Expert – paras. 349-353, pages 50-51
• Overview of the Rule 53 Expert Evidence – paras. 354-357, page 51
• Dr. Kirk’s Explanation of the Pitting Corrosion Process – paras. 358-365, pages 51-52
• Dr. Revie’s Reports and Evidence – paras. 366-375, pages 52-53
• Dr. Frankel’s Reports and Evidence – paras. 376-383, pages 53-54
• Dr. Frankel’s Conclusions – paras. 384-390, pages 54-57
• Further Studies at AECL – paras. 391-394, page 57
• Conclusions on the Expert Evidence – paras 395-401, pages 57-58

Question 4: Considering the factual context, the reasonable expectation of the parties, and my conclusions on the expert evidence, has the Insurer met its onus to prove that the corrosion that caused the leak at J-41 is excluded under the Policy? – paras. 402-430, pages 59-62

• Factual Context of this Policy and the Reasonable Expectation of the Parties: the Insurer has not met its onus – paras. 402-412, pages 59-60
• Rules of Construction – paras. 413-425, pages 60-61
• Contra Proferentem – paras. 426-430, pages 61-62

PART V – ISSUE 2: ALTERNATIVELY, DOES THE RESULTING PHYSICAL DAMAGE EXEMPTION APPLY ALLOWING COVERAGE?

Principles When Considering an Exception to an Exclusion – paras. 434-436, pages 62-63

The Meaning of Resulting Physical Damage in This Policy and Position of the Parties – paras. 437-446, pages 63-64

Relevant Provisions in the Policy – paras. 447-459, pages 64-65

Caselaw Interpreting the Meaning of Physical Damage – paras. 460-475, pages 65-67

Facts Relevant to Resulting Physical Damage – paras. 476-495, pages 67-70

Application of Principles to this case – paras. 496-515, pages 70-73

Conclusions on the Meaning of Resulting Physical Damage – paras. 516-521, page 73

PART VI – ISSUE 3: IDLE PERIOD EXCLUSION

What is meant by the “Idle Period” Exclusion listed in the Time Element Exclusions Clause 5.A? – paras. 523-525, page 74

Has the Insurer met its Onus of Demonstrating that what Occurred in this Case is Excluded by the “Idle Period”? – paras. 526-538, pages 75-76

PART VII – ISSUE 4: THE NUCLEAR RADIATION EXCLUSION

The Evolving Nuclear Radiation Arguments – paras. 539-550, pages 76-78

Question 1: Is the Nuclear Radiation Exclusion ambiguous? – paras. 551-575, pages 78-84

• Context: the Nuclear Radiation Exclusion in the Policy – paras. 560-575, pages 80-84

Question 2: The Role of Nitric Acid in the Leak: Is the wording of this exclusion an anti-concurrent causation exclusion and if so how is it to be interpreted? – paras. 576-586, page 84-85

Question 3: If the Nuclear Radiation Exclusion is Ambiguous, How Should the Exemption be Interpreted? – paras. 587-613, pages 85-89

• Nullity of Contract – paras. 601-610, pages 87-88
• Contra Proferentem – paras. 611-613, page 89

Overall Summary of Conclusions – paras. 614-619, page 89

PART VIII – THE PLAINTIFFS’ CLAIM FOR ACTUAL COST OF BORROWING AS PREJUDGMENT INTEREST

The Statutory Framework: sections 128 to 130 of the Courts of Justice Act – paras. 624-632, pages 90-92

The Pleadings – paras. 633-638, pages 92-93

Evidence of the Plaintiffs’ expert – paras. 639-647, pages 93-94

Relationship Between an Insurer and the Insured – paras. 659-664, pages 97-98

Consideration of the Statutory Factors – paras. 665-666, pages 98-99

Overview of Principles in the Cases – paras. 667-668, pages 99-100

The Arguments of the Parties – paras. 669-675, page 101

Consideration of Bank of America Canada – paras. 676-688, page 102-103

Facts Relevant to the Claim for Enhanced Interest – paras. 689-733, pages 103-109

Conclusions on pre-judgment interest – paras. 734-741, pages 109-110

Costs and post-judgment interest – paras. 742-745, page 110

PART I: OVERVIEW

Overview

[10] On May 14, 2009 there was a generalized power shutdown in parts of Eastern Ontario and Quebec, which automatically tripped a shutdown of the NRU.

[11] During the shutdown, an unanticipated leak of heavy water from the calandria into the J-rod annulus was discovered. The leak was small, with heavy water entering the J-rod annulus at the rate of 5 kilograms of heavy water per hour. As a term of the licence, AECL was required to notify the Canadian Nuclear Safety Commission (“CNSC”) about the leak due to the presence of radioactive Tritium in the ventilation system.

[12] The location and reason for the leak was unknown. After investigation, it was revealed that the leak of heavy water had unexpectedly and without explanation penetrated though the calandria wall towards the base at location J-41.

[13] The leak was localized and unanticipated. Upon investigation it was determined that the leak was caused by an unanticipated corrosion.

[14] The cause of the corrosion resulting in the through-wall penetration at J-41 and leak of heavy water containing Tritium is an important issue in determining whether the Policy engages or whether the Corrosion or Nuclear Radiation Exclusions apply.

[15] Any unexplained leak of heavy water containing Tritium is significant and is potentially dangerous.[^2]

[16] In the June 2009 public hearing, the members of the CNSB were critical that AECL did not know the nature of the unanticipated corrosion causing the leak, and that the problem causing the leak may have been missed on earlier inspections.

[17] In August 2009 after AECL confirmed that at J-41 there was a through-wall penetration of the calandria causing the leak of heavy water to the J-rod annulus, CNSC established a safety protocol before the NRU could be reopened and required a report that the NRU was fit for its purpose.[^3] The requirements to reopen the NRU were amended and enhanced in January 2010.

[18] The NRU was shut down from May 14, 2009 to August 2010 by the CNSC. This time was required to allow AECL to inspect the leak site, conduct the required analysis of the cause of the through wall penetration, to devise what was a complex plan for repair including mock ups and practice sessions, and to implement the repair. This time frame also allowed the root cause report explaining the cause of the leak required by the CNSC to be prepared.

[19] Dr. David Mancey, a longtime employee of AECL, was tasked with overseeing the writing of the root cause report required by CNSC as a condition of restarting the NRU. He wrote an interim report in July 2009 that was sent for peer review (the Interim Mancey Report). The peer review of the Interim Report questioned the theory that nitric acid was the cause of the leak at J-41.

[20] Dr. Mancey was also the author of the report entitled “Corrosion to the J-Rod Annulus of the NRU Reactor” AECL & R&D Report Number 153-150200-440-001 Revision 0, September 2010 referred to in this trial as the “Mancey Report”.

[21] The Interim and the final Mancey Report conclude that nitric acid and carbonic acid were the likely causes of the through-wall leak in the calandria, although the final report recommended further study. There was no external review of the final Mancey Report.

[22] The Plaintiffs’ Rule 53.03 expert, and one of the independent experts who conducted the peer review of the Interim Mancy Report who testified at the trial, challenge the conclusions reached in the Mancey Report and opine that the leak at J-41 was caused by classic pitting corrosion, precipitated by the presence of an unknown aggressive agent, probably chlorine. As will be seen, the court-appointed expert also challenges the conclusions in the Mancey Report. He confirms the likely cause of the through wall leak at J-41 as pitting corrosion, requiring the presence of an aggressive species, most commonly chlorine, corroborating the theory of the Plaintiffs’ Rule 53.03 expert.

How a nuclear reactor works

[23] It is essential understand how a nuclear reactor works to understand the exclusions and exemption issues raised by the parties. I outline a simple explanation of the structure and functioning of the NRU. I attach as “A” an illustration of the NRU.[^4]

[24] The original NRU began operation in 1957. The internal component parts were all replaced in 1974. Except for shutdowns for repairs, the NRU operates 24/7. There are approximately 15 employees directly running the NRU. There are approximately 40 other employees of AECL at Chalk River who conduct unrelated research from nearby facilities.

[25] At the beginning of the trial, Mr. Brock, counsel for the Insured suggested by way of analogy that the NRU is like an enormous thermos. Except this thermos where the radioactive activity takes place is very large and is as wide as it is tall.

[26] The main components of the NRU relevant to this proceeding include the calandria, the J-rod annulus, and the reflector. There are drains in the interior of the NRU at the base of the J-rod annulus, and the reflector.

[27] The interior core of the NRU is called the calandria. It is filled with heavy water and fuel rods. The radiation processes take place inside the calandria at the fuel rods producing Tritium, which in turn produces radioisotopes. The calandria is approximately three meters wide and three meters tall. It is made of 5052 aluminum.

[28] The fuel rods in the interior of the calandria where the nuclear reactions take place reach very high temperatures and for safety reasons it is essential that they be cooled. The heavy water circulating in the calandria assists with this cooling function and also serves as a moderator of the nuclear process. It is also essential that the fuel rods in the calandria remain covered by the heavy water when the NRU is in operation.

[29] Like a thermos, there is then a space. This space is called the J-rod annulus. It is a 10 cm narrow compartmentalized space running the entire height and circumference of the calandria. The J-rod annulus as designed is filled with pressurized carbon dioxide. Its purpose is to maintain a space between the calandria of the reactor, and the reflector.

[30] Again, like a thermos, there is an outer layer. This is known as the reflector. The reflector is filled with 13,000 liters of treated process light water originally sourced from the Ottawa River. It is also made of 5052 aluminum. This outer layer is again three plus meters wide and three meters tall, following the circumference and height of the J-rod annulus. The reflector’s purpose is to cool the extreme heat in the inner core of the calandria generated by the radioactive activity at the fuel rods.

[31] Each of the three layers within the NRU is intended to be discrete. The J-rod annulus is not supposed to contain water, be it light water leaking from the reflector, or heavy water leaking from the calandria. There are drains in case there is leakage.

[32] On top of the circular calandria there are circular guide tubes topped by an upper deck plate. The guide tubes and the deck plate measure some 6 meters high. The total circular structure including the 3 meter circular calandria, and the 6 meter guide tubes is therefore 9 meters high.

[33] Accessing the leak site at the base of the calandria nearly 9 meters below the deck plate was challenging as access was only possible from the upper deck through the holes in the guide tubes measuring a mere 12 cm in diameter.

[34] There is finally an external concrete structural shell of unknown dimensions containing the reactor components allowing sufficient space for workers to perform their various tasks.

[35] The radioactive isotopes the Plaintiffs acquire from the AECL are produced when heavy water is exposed to neutron radiation. The large volume of the NRU reactor allows for the bulk production of these isotopes.

[36] There is routine daily monitoring of any water draining out of the J-rod annulus to ensure that heavy water is not leaking from the calandria. The presence of heavy water leaking into the J-rod annulus is identified by the presence of radioactive Tritium in the ventilation systems.

[37] If there is an unintended leak of heavy water, the design of the NRU is intended to allow for heavy water containing Tritium entering the J-rod annulus from the calandria to be drained by a sump pump system. Due to the nuclear reactive Tritium, this drained heavy water is required to be stored in sealed underground drums indefinitely.

AECL Information Not Available to the Public

[38] The internal functioning of the AECL and all AECL internal documents and reports are protected and not available to the public without bringing a Freedom of Information application.[^5]

[39] AECL was obliged by its licence to report the leak to the CNSC due to actionable levels of Tritium. Public information became available only through bulletins prepared and posted by AECL on their website.

[40] After this lawsuit commenced, Mr. Liblong, counsel for the Insurer, brought a Freedom of Information application to obtain documents from AECL.

[41] AECL produced to Mr. Liblong the redacted 347-page Mancey Report, which he in turn provided to counsel for the Plaintiffs.

[42] In 2019, shortly before the trial resuming, Mr. Brock obtained three other reports relevant to the issue of corrosion that had been produced pursuant to another Freedom of Information application in another proceeding relating to the shutdown of the NRU.[^6]

[43] Clearly all documentation in the possession of AECL relevant to the issue of the cause of the corrosion has not been produced. As Mr. Liblong counsel for the Insurer confirmed during the cross-examination of Dr. Kirk discussing another issue, “God only knows what was not produced”.[^7]

[44] Of particular concern is the absence of comprehensive water testing information at the NRU.

Known Leaking from the Reflector to the J-rod Annulus and Corrosion

[45] Notwithstanding the NRU design, contemplating no leakage from either the reflector, or the calandria into the J-rod annulus, it was known that there was leaking of the light water from the reflector going into the J-rod annulus beginning only three months after the NRU went into operation in 1974 causing known tolerated generalized corrosion to the outer walls of the calandria and reflector.

[46] The generalized corrosion was caused by nitric acid, which was produced by the radiation of air in the J-rod annulus.

[47] This leakage and production of nitric acid was tolerated, as the generalized corrosion was not perceived by AECL staff to be a safety issue.

[48] I accept the evidence of Thomas Jackson, a former supervisor at the NRU, that the light process water coming from the Ottawa River that ultimately goes into the reflector is tested and treated with chlorine. This treated process water is used for other services at the NRU, but not for drinking water.

[49] The 13,000 liters of treated light process water in the reflector at the NRU originally sourced from the Ottawa River is heated to become steam, and when the stream cools it becomes a condensate that is collected in the liquidex tank. The water going into the reflector is drawn from the liquidex tank located below the calandria.

[50] The light water in the liquidex tank going to the reflector is designed to be free from contaminants by using an ion system.

[51] The quantity of water that leaked from the reflector to the J-rod annulus was monitored.[^8]

[52] The quantity of water leaking from the reflector increased over time. There is very little concrete evidence in the Mancey Report about the reasons for this significant leak of water other than charts produced at pages 167 and 221.[^9]

[53] Shortly before the NRU shut down in May 2009, in January 2009, the leakage of water from the reflector to the J-rod annulus increased dramatically to 8,000 liters per day. The capacity of the reflector was 13,000 liters. This significant leaking went on until February 10, 2009 when a special seal plug was installed.[^10]

[54] After the repairs, the rate of leaking from the reflector to the J-rod annulus was reduced to approximately 1,000 liters per day.[^11]

[55] Contrary to the intended design, it is undisputed that the drains from the J-rod annulus were largely blocked with significant bulky general corrosion product preventing the water that leaked from the reflector into the J-rod annulus from properly draining.

[56] It is also undisputed that there was corrosion of the lip of the gutter at the base of the J-rod annulus which was intended to keep water from contacting the wall of the J-rod annulus.

[57] The combination of leaking water from the reflector to the J-rod annulus, the blocked drains and the corrosion on the gutter lip created an environment in the J-rod annulus that retained the light water leaking from the reflector in contact with the calandria wall. The nature of this environment relevant to the leak will be canvassed when I consider the expert evidence.

[58] An issue in this proceeding is, notwithstanding the design, whether there were microscopic traces of agents such as chlorine in the light water that had leaked from the reflector to the J-rod annulus that were the aggressive agents precipitating the pitting corrosion process that caused the leak at J-41.

[59] As will be discussed, comprehensive water testing results, although requested by the Plaintiffs’ expert, were not produced by AECL and were not in the evidence at trial.

The Two Kinds of Corrosion

Generalized Corrosion

[60] It was known by AECL since 1974 that as a result of the light water leaking from the reflector, the air in the J-rod annulus was subject to radiation producing nitric acid in the J-rod annulus.

[61] It was known as well by AECL that this nitric acid caused, over time, a generalized corrosion on the outer walls of the calandria and the reflector.

[62] This known, non-fortuitous corrosion was monitored. This generalized corrosion over the 35-year period of activity in the NRU from 1974 until 2009 reduced the wall thickness of the calandria from 8 mm to 7.5 mm.

[63] This gradual reduction in the wall thickness of the calandria caused by nitric acid creating generalized corrosion was known by AECL and accepted by AECL staff as in their view it did not affect the safe operation of the NRU.

[64] The Plaintiffs acknowledge that had this known generalized corrosion caused the leak at J-41 that the corrosion exclusion would apply precluding recovery.

The Second Unanticipated Corrosion at J-41 Causing the Leak

[65] During the shutdown, a second kind of unanticipated localized corrosion of a different character was identified on the common wall between the calandria and the J-rod annulus. The location of the through wall leak at the base of the calandria was at J-41. This unanticipated corrosion at J-41 caused the penetration through the 8 mm wall of the calandria.

[66] The evidence before me is uncontradicted that it was never anticipated that the external wall of the calandria could be penetrated by general corrosion causing the leak of heavy water. In the words of Dr. Mancey the localized corrosion was “surprising”, “not expected”, “not known” and “unique”.

[67] It is the Plaintiffs’ position that the corrosion exclusion in the Policy does not apply to the unanticipated, fortuitous corrosion at J-41 where the through-wall leak took place.

[68] The cause of the second kind of corrosion at J-41 is a key issue in this lawsuit as to whether the corrosion and the nuclear radiation exclusions as characterized by the Insurer apply.

[69] Was there some form of contaminant or aggressive agent such as chlorine present to cause the corrosion at J-41 in the specific environment in the J-rod annulus, as suggested by the Plaintiffs’ expert, or was the corroding agent the known nitric acid that was in the J-rod annulus as suggested by Dr. Mancey and the Insurer’s rule 53 expert reports?

[70] Dr. Mancey, research scientist with AECL and author of the Mancey Report, provided an affidavit prepared by counsel for the Insurer and testified as a participant expert as a fact witness.

[71] The Mancey Report dated September 2010 opines that the localized corrosion at J-41 was caused by nitric acid, and carbolic acid, although that conclusion is not unequivocal and he suggests further research is needed. [^12]

[72] Dr. Mancey conducted further studies and research to test his conclusions. He was of the view that the subsequent research confirmed the conclusions in the Mancey Report that nitric acid caused the localized corrosion resulting in the through-wall leak at J-41.

[73] The studies confirmed that carbonic acid was not responsible for the corrosion at J-41. All experts confirm that carbonic acid was not a contributing causal factor to the through-wall penetration. Hence the trial focused on the role of nitric acid in the leak at J-41.

[74] The Plaintiffs retained a Rule 53.02 expert to review the conclusions reached in the Mancey Report. Dr. Donald Kirk disagreed with the conclusions reached in the Mancey Report and opined that pitting corrosion at J-41 caused the unexpected leak. The Insurer in turn retained Dr. Winston Revie as a responding Rule 53.02 expert. He supported Dr. Mancey’s conclusions.

Trial Procedure and the Appointment of a Court-Appointed Expert

[75] When this matter came before me for trial in March 2019, I concluded that it was not ready. The usual trial management had not taken place, which was particularly problematic in a case of this complexity.

[76] There were disclosure issues raised by counsel on the first day of trial, as AECL had not produced un-redacted copies of the Mancey Report to the Rule 53.02 experts. A cooperative order was worked out on consent with counsel from AECL on March 4, 2019, the first day of the trial, for the sealed un-redacted Mancey report to be made available to the experts, but not to be filed publicly.

[77] In light of the technical complexity and importance of the cause of the corrosion in this litigation, pursuant to Rule 52.03 of the Rules of Civil Procedure, I directed that a court-appointed expert be appointed by the parties, in consultation with their experts, to assist the court with the theories of causation of the corrosion as outlined in the Mancey Report and the opinions of the parties’ Rule 53 experts.

[78] Rule 52.03 of the Rules of Civil Procedure, provides that on motion by a party or on his or her own initiative, a judge may, at any time, appoint one or more independent experts to inquire into and report on any question of fact or opinion relevant to an issue in the action.

[79] The parties, in consultation with their experts, appointed Dr. Gerald Frankel, a renowned corrosion expert from Ohio to assist the court in this case.

[80] He has no connection with either party, any of the experts or with counsel.

[81] The original trial date was March 4, 2019. Various trial management procedures took place on that date, and the Plaintiffs raised a preliminary motion to exclude the admissibility of the Insurer’s Rule 53 expert. In separate reasons I dismissed that motion.

[82] As it turns out on the March 4, 2019 date, unbeknownst to the parties, there were several other AECL reports relevant to the corrosion issue that had not been provided. These three further reports were provided by AECL pursuant to court orders on a cooperative basis between the original trial date in March 2019 and the ultimate trial date continuing on September 4, 2019.

[83] The Rule 53 experts prepared further reports back and forth commenting on this new disclosure. There were a total of 9 reports produced. The parties agreed that the various reports would be filed as part of their evidence-in-chief.

[84] The assistance of Dr. Frankel as a court-appointed expert streamlining the technical factual issues in this proceeding is gratefully appreciated. Without his assistance, I would probably have been caught up in the minutiae of the 9 reports exchanged between the Rule 53 experts. The trial and this judgment would inevitably have been much longer. Dr. Frankel’s participation narrowed and focused the factual issues in dispute.

Lay Witnesses

[85] The Plaintiff called Thomas Jackson, an engineer, who worked as the senior shift supervisor at AECL until 2001. He testified on the first day of trial on March 4, 2019 before the matter was adjourned for further production and for the appointment of the court-appointed expert. His transcript forms part of the evidence. Mr. Jackson described the operation of the NRU until he left in 2001, so had no first hand knowledge of the shutdown in May 2009. His evidence was not disputed. He is now employed with MDS controlling its isotope production.

[86] As part of the trial management, on March 4, 2019, the first day of the trial, I requested that the parties provide affidavit evidence of the evidence of all the witnesses who were testifying to form part of their evidence-in-chief when the trial resumed in September.

[87] The Plaintiffs produced an affidavit of Dr. Newman, one of the peer review experts of the Interim Mancey Report. The defence lay witnesses included William Pilkington, chief nuclear officer at Chalk River at the time of this incident, Michael Lodge, the adjuster handling this claim, as well as a supervisor, John Turcotte. Dr. Mancey also produced an affidavit.

Issues for Determination

[88] When the matter first came before me in March, I asked counsel to define both the factual and legal issues that the court was being asked to determine.

[89] Counsel for the Plaintiff prepared a list of legal issues. Counsel for the Insurer prepared a list of the factual issues framed in such a way as the questions posed foreshadow the arguments that the Insurer would advance that the Policy does not engage to cover the Plaintiff’s losses due to the application of the corrosion, nuclear radiation and idle period exclusions.

[90] These two documents assist in providing a roadmap for this decision. The following are the issues for determination:

The Insurer argues that the corrosion exclusion applies precluding coverage

• Can the applicability of the corrosion exclusion be determined as suggested by the Insurer by definition alone or is the term ambiguous?
• If the term corrosion is ambiguous, what is its meaning?
• What is the cause of corrosion at J-41 having regard to the expert evidence?
• Considering the factual context, the reasonable expectation of the parties, and the factual findings on the expert evidence, has the Insurer met its onus to prove that the corrosion that caused the leak at J-41 is excluded under the Policy?

If the corrosion exclusion applies, the Plaintiffs argue that the resulting physical damage exception to the exclusion applies

• What is the meaning of resulting physical damage?
• Is the meaning of resulting physical damage ambiguous?
• If there is ambiguity, what is meant by resulting physical damage in this Policy’ taking into account the factual context of this case and the reasonable expectations of the parties?
• Applying the meaning of the term, have the Plaintiffs met the onus to prove that the resulting physical damage exception applies?

If the resulting damage exception applies, the Insurer raises the idle period exclusion.

• What is meant by the “Idle Period” exclusion listed in the Time Element Exclusions clause s. 5.A.?
• Has the Insurer met its onus of demonstrating that what occurred in this case is excluded by the “idle period”?

Alternatively, the Insurer argues that the radiation exclusion applies precluding coverage

• What is the meaning of the radiation exclusion?
• Is the meaning of the radiation exclusion ambiguous?
• If the radiation exclusion is ambiguous, how should the ambiguity be interpreted given the Policy, the factual context in this case, and the reasonable expectation of the parties?
• Is the exclusion an anti-concurrent exclusion, and if so, how should it be interpreted?
• Consideration of principles of nullity of contract, and contra proferendum
• Has the Insurer met its onus of proof that the radiation exclusion applies?

Prejudgment Interest

• If the Plaintiffs recover their losses in this proceeding, is it just in this case to exercise statutory or common law discretion to award interest based upon the actual cost of borrowing by the Plaintiffs, rather than simple interest as provided for in the Courts of Justice Act?

The Shutdown and Steps that Followed

[91] Although any heavy water discharged into the J-rod annulus is collected and removed by a sump pump system, a small amount of heavy water evaporates before it can be collected resulting in the release of Tritium into the ventilation system.

[92] On Thursday May 14, 2009 a power outage resulted in the automatic trip and safe shutdown of the NRU. During this shutdown period, increased levels of Tritium were observed through routine monitoring of the ventilation streams indicating that heavy water was leaking from the calandria into the J-rod annulus. The Tritium was at actionable levels, in the sense that its presence triggered mandatory reporting to the CNSC, though not dangerous levels.

[93] AECL determined that the leak was occurring at the rate of about 5 kg per hour and likely had been occurring for several hours before the power outage. The leak of heavy water from the calandria to the J-rod annulus resulted in the loss of approximately 3,368 kg of heavy water in 28 days until the heavy water was drained from the calandria.[^13]

[94] The shutdown of the NRU in 2009 was a newsworthy event causing public concern about safety at the NRU and a short supply of medical isotopes. All information was protected and not available to the public.

[95] The reporting to the public was limited to the publication bulletins authored by AECL. Understandably, to minimize public concerns about potential health hazards for the public and the NRU employees, the senior staff at the AECL took a reassuring, non-alarmist tone in any public reporting as to the reason for the continued shutdown of the NRU in the AECL bulletins on the AECL website, and in any CNSC hearings.

[96] For example, in all of their bulletins between May 15, 2009 and August 12, 2009, AECL confirmed “there is no threat to workers, the public, the environment or nuclear safety related to this event”.[^14] AECL indicated that a “small”, “contained” heavy water leak was found, which had also resulted in small amounts of tritium-producing heavy water evaporation[^15] “well below regulatory limits”.[^16] Subsequent bulletins outlined the anticipated length of the NRU shutdown and the steps that AECL was taking to investigate and fix the leak.[^17]

Evidence of Mr. Pilkington

[97] Mr. Pilkington, a mechanical engineer, was employed by AECL from 2005 to September 2010. Initially he worked in the Candu services unit in Mississauga then in 2008 became the senior vice president of research and technology and the Chief Nuclear Officer for AECL at the Chalk River site.

[98] Mr. Pilkington was new in his task of Chief Nuclear Officer and licence holder for the NRU. He was appointed in 2009 and left AECL in September 2010 before the Mancey Report was released.

[99] As Chief Nuclear Officer and holder of the site licence, Mr. Pilkington was obliged to report to the CNSC when nuclear reactor activity exceeded reportable levels as defined by the governing regulations.[^18]

[100] I pause here to comment on his evidence.

[101] The witnesses called by the Insurer from AECL including Mr. Pilkington and Dr. Mancey were clearly aligned with the defence position. It is not surprising that Mr. Pilkington, Chief Nuclear Officer, may tend to underestimate the gaps in the protocol for inspecting the NRU, and overestimate what inquiry may have taken place by AECL prior to the leak. He acknowledged that CNSC was critical of the safety culture at AECL. He was in charge of that culture at least at the time of the leak.

[102] I accept the evidence of Mr. Pilkington in his cross-examination in preference to the carefully crafted selective text in his affidavit and aspects of his evidence-in-chief.

[103] Each shift tests for the presence of Tritium in the J-rod annulus. The presence of Tritium indicates a leak from the calandria to the J-rod annulus.

[104] Following the shutdown on May 14, 2009 Tritium was discovered in the ventilation streams, indicated there was a leak of heavy water that met the reportable threshold levels triggering the obligation by AECL to report to the CNSC.

[105] Mr. Pilkington confirmed that there were two concerns: “One certainly was the elevated Tritium release and the second was the unknown leakage of heavy water in an area where it would not be expected during operation.”[^19]

[106] In the affidavit, and in his evidence-in-chief Mr. Pilkington minimized the importance of a leak of heavy water from the J-rod annulus that began on May 14, 2009 implying that this was minor and suggesting that the NRU could continue to operate notwithstanding the leak. This suggestion is misleading and not true.

[107] Once it was determined that the leak was through the wall of the calandria and not merely located at a reparable mechanical fitting where leaking may be anticipated to occur, the NRU could not return to service until the protocol requirements of the CNSC were satisfied.

[108] Mr. Pilkington confirmed that a leak of heavy water due to penetration of the wall of the calandria in a nuclear reactor was a “serious event” and “major event”[^20] resulting in a protocol imposed by the CNSC that had to be met for design and safety requirements before the NRU could reopen.

[109] Any attempt by Mr. Pilkington or the Insurer to soft pedal the importance of this unanticipated, unexplained through-wall leak of heavy water in the wall between the J-rod annulus and the calandria and to suggest that the NRU could have continued operations in light of the heavy water leak in the context of the facts of this case is not true.

Steps taken by AECL

[110] Understandably, the AECL staff were concerned about taking any steps that could have made the leak bigger, and which could escalate to a potentially dangerous situation exposing the fuel rods in the calandria.

[111] AECL therefore decided remove the fuel rods in the calandria to be able to identify the location of the leak and to be able to inspect it. This process took 28 days.

[112] The AECL staff were then able to drain the heavy water from the caldaria. The drainage of the heavy water took 24 hours.

[113] These steps were necessary to enable AECL to safely inspect the leak site by non-destructive means without the risk of increasing the size of the leak. The goal was to determine if the leak could be safely repaired, and if so to develop repair options.

[114] It must be remembered that the leak at J-41 was located nearly 9 meters below the deck plate making inspection of the leak site very challenging, as the openings on the deck plate were only 12 cm in diameter. Special tools were designed and tested in a mock up protocol for the repairs.

[115] Once the calandria was drained the inspection revealed an unanticipated through-wall hole in the wall between the J-rod annulus and the calandria at location J-41.

[116] The leak was at a single isolated location. However, the damage identified was in a one-centimeter band on the exterior wall of the calandria and was more extensive. There were ten sites of localized corrosion that had not yet penetrated the calandria wall that needed repair located in that band. As well there were vulnerable scalloped areas in that band that Dr. Frankel opined were caused by other aggressive agents in the water including potentially copper or iron that I will discuss later in this decision.

[117] This vulnerable one-centimeter band was located at the base of the J-rod annulus where the leaked water from the reflector pooled, below the interface where the leaked water from the reflector met the air and carbon dioxide, and above the stagnant floor of the J-rod annulus.

[118] The leak of heavy water at J-41 was caused by the unanticipated localized corrosion. Once the through-wall leak was discovered, Dr. Mancey was asked to be part of the team charged with determining the cause of the corrosion causing the leak at J-41, as corrosion was his area of expertise.

[119] After the discovery of the unanticipated through-wall leak, CNSC established a protocol on August 14, 2009 [Exhibit 11] that was revised and enhanced in January 2010 that established conditions that must be met by AECL before the NRU could restart.

[120] The process to identify the cause of the leak, to conduct the repairs and to meet the CNSC protocol took 15 months. As counsel for the Plaintiffs stated it was a “long journey” with a lot of steps along the way to fix the problem. In total, ten sites were repaired. The steps taken are relevant to the resulting damage and idle period issues. These facts will be explored in more detail when these issues are discussed.

[121] The CNSC allowed the reactor to reopen in August 2010, on the condition that a final root cause report to the CNSC was prepared within 90 days identifying the cause of the leak, providing assurances that the NRU design was fit for the purpose and confirming that the NRU met safety applicable standards.

[122] The Mancey Report was issued to the CNSC on September 22, 2010 to meet these requirements as a condition to reopen the NRU. The Mancey Report was not available to the public and was not subject to any external review.

PART II: ABSENCE OF WATER TESTING

[123] After the first date of trial in March 2019 and prior to the trial resuming, counsel for the Plaintiffs conveyed Dr. Kirk’s request to AECL for further documents including requests for the water testing results from the water going into and draining from the J-rod annulus. All counsel appeared before me prior to the trial resuming to canvass the issue. Counsel for AECL objected to the Plaintiffs’ request as it was too vague and voluminous. Counsel for the defence did not join in the request for the water testing documentation.

[124] Counsel for the Plaintiffs did not pursue the issue by way of a formal motion prior to the trial presumably to avoid prejudicing the September trial date.

[125] Counsel for the Insured took no steps to obtain the water testing results.

Admissibility of Evidence Concerning Water Testing

[126] Before reviewing the corrosion issue, and the expert evidence, I pause to consider a matter that came up during Dr. Mancey’s evidence during the trial concerning water quality and water testing, and to outline my findings with respect to the water leaking from the reflector to the J-rod annulus.

[127] The water testing issue was a hotspot for both counsel as it relates to the presence of aggressive agents or contaminants in the water leaking into the J-rod annulus and pitting corrosion.

[128] During the trial at different points in time, each counsel was concerned about the absence of water testing data. I was not prepared to simply adjourn this complex trial to pursue the water testing, which could take months or longer given the need to bring a Freedom of Information application to obtain the information.

[129] Both counsel were well aware of the significance of the water testing issue months before the trial resumed and chose to proceed.

[130] Initially counsel for the Plaintiffs considered requesting an adjournment, to obtain the water tests, or suggesting a potential mistrial. Mr. Brock reconsidered his position and decided to soldier on.

[131] Later Mr. Liblong raised concerns during Dr. Mancey’s evidence wanting to adjourn the trial to locate the water testing information. I was not prepared to simply adjourn this complex trial mid-stream. I gave Mr. Liblong the opportunity for me to declare a mistrial, and to allow him to pursue the Freedom of Information application for the water testing results, on the condition that he pay all costs of the Plaintiffs for the trial to date. After consulting with US in-house counsel, who was present in the courtroom throughout the trial, counsel for the Insurer declined this option. The trial continued based upon the available evidence.

Ruling on Admissibility of Dr. Mancey’s Evidence on Water Quality

[132] At my request, counsel prepared an affidavit for Dr. Mancey, along with all the other fact witnesses in this proceeding, to narrow his evidence from his 347-page report, and to identify what evidence he would be giving relevant to issues in this trial. Counsel for the Insurer prepared this affidavit.

[133] There is no mention in the Mancey Report or in Dr. Mancey’s affidavit sworn July 24, 2019 about water testing procedures for the water going into the reflector, or testing of the quality of the water leaking from the reflector to the J-rod annulus. As noted above, prior to Dr. Mancey’s affidavit being sworn, requests had been made for the water testing results by the Plaintiffs’ expert and had been communicated to counsel for FM.

[134] In his evidence in chief, Dr. Mancey gave evidence about the quality of the water in the reflector and testing procedures in place that was not included in the Mancey Report or in his affidavit prepared for the trial. This evidence was objected to by counsel for MDS.

[135] As pointed out by the defence counsel, Dr. Mancey was a participating expert testifying as a fact witness.

[136] Strict limits are placed upon Rule 53 experts who stray from the contents of their report, subject to directions from the trial judge.[^21] The same limits do not apply to the evidence of a participating expert witness, compared to a Rule 53 expert. In Westerhof v. Gee Estate, the Court of Appeal held that “participant experts” may give opinion evidence without complying with Rule 53.03 where they have not been engaged by or on behalf of a party to the litigation and: (1) the opinion to be given is based on the witness’s observation of or participation in the events at issue; and (2) the witness formed the opinion to be given as part of the ordinary exercise of his or her skill, knowledge, training and experience while observing or participating in such events.[^22]

[137] Notwithstanding his objection in chief, during cross-examination of Dr. Mancey, Plaintiffs’ counsel asked questions directly relevant to the issue of the process water and steps that took place before the water goes into the reflector relevant to whether a contaminant was present.

[138] In my gatekeeper function, I allowed Dr. Mancey to testify in chief on this issue as a fact witness and as a participating expert relying on Westerhof v. Gee. I ruled that I would determine what weight to attribute to this evidence at the conclusion of the trial.

[139] As counsel for the Plaintiffs asked questions during cross-examination of Dr. Mancey relevant to water testing, I also allowed further evidence to be given by Dr. Mancey, during cross-examination and re-examination, notwithstanding Plaintiffs’ counsel’s attempt to limit this evidence.

Dr. Mancey’s evidence on water testing

[140] The evidence given by Dr. Mancey in-chief that was objected to by Plantiffs’ counsel relevant to water testing was the following:

• The light water going into the reflector was made into distilled water by boiling the process water and collecting the condensation. The distilled water system is subject to further purification by an ion system, which Dr. Mancey confirmed is designed to remove any dissolved copper, or chloride.
• When asked “And if that system’s working as it’s designed to do … any water leaking into the J-rod annulus, would you expect it to contain any copper or chloride ions?” to which he responded “The – the concentrations of those sort of contaminants would be extremely low.”. According to Dr. Mancey, further analysis of the water leaking into the J-rod annulus showed that “those solutions were now acidic solutions of aluminum nitrate”.

[141] In cross-examination Dr. Mancey testified:

• Process water is a lower grade of water than that found in the distilled water system, which was supplied via steam condensate. Dr. Mancey believed, but was not sure, that the “Liquidex” – referred to in the Alcoa report – was a tank in the distilled water system. [Alcoa was one of the independent experts reviewing the Interim Mancey Report]
• When asked whether process water is used to cool the heavy water in the reactor, Dr. Mancey did not know whether process water was used for this function.
• There were cracks in several welds in the reflector that would permit water to leak out of the reflector and into the J-rod annulus. Dr. Mancey testified that the reflector typically holds 13,000 liters of water, and may leak up to 1,000 liters of water per day, most of which would pass through the J-rod annulus. These cracks, and other issues, would usually be addressed during periodic maintenance shutdowns of the NRU.
• Dr. Mancey confirmed that the reflector was leaking up to 8,000 liters of water per day in January 2009, which was approximately two-thirds of the capacity of the reflector. When asked whether the leaking of two-thirds per day of the reflector water system taxed the liquidex tank, and whether they had to use process water to keep the reflector filled he answered, that he was “not certain of that design aspect.”[^23]
• The J-rod annulus was designed with two separate drain systems, each intended to capture light water from the reflector or heavy water from the calandria, respectively.[^24]
• Dr. Mancey’s cross-examination indicated that charts contained on pages 167-169 of the Mancey Report “demonstrate the corrosion product exiting the annulus in the water”. Dr. Mancey confirmed that these charts were not the results of any search for a corrosion product such as chloride, copper, or iron, and that no such search results were undertaken, nor included in the final report.[^25]
• Dr. Mancey testified that water going into the Liquidex was drawn from the Ottawa River, in all likelihood treated in some fashion, and boiled into steam condensate.
• Dr. Mancey agreed that between January 2008 and January 2009, the reflector was leaking water at a rate of 2,000 liters per day, which increased to 8,000 in January 2009. Dr. Mancey denied that process water was used in the reflector, as he suggested that only distilled water was ever used in the reflector. He reiterated his belief, but not certainty, that the Liquidex is a part of the distilled water system.
• In re-examination, Dr. Mancey confirmed that he took “into consideration whether or not anodes and cathodes and contaminates were present inside the annulus when [he was] doing [his] investigation”. Dr. Mancey stated that he did not detail this in his report, however, because the report was prepared for a very specific reason to satisfy the CNSC protocol and because “The CNSC had access to all kinds of information about the reactor. So, in terms of, you know, the – the question about the quality of the water in the reflector was really a non-issue because both parties accepted that the water was a very high quality and did not contain contaminants at levels that would constitute a threat and might cause pitting corrosion.”[^26] [Emphasis added]

[142] I conclude that the evidence of Dr. Mancey, though admissible, is far from complete or conclusive as to the quality of the water leaking into the J-rod annulus from the reflector. Many of his answers noted above were speculative, and not based upon first-hand knowledge as a participating fact expert.

[143] Water testing results were not considered in the Mancey Report analysis. The Mancey Report relied on general results and appears to assume that the NRU would function in accordance with its design without the presence of aggressive agents/corrodents in the leaking water from the reflector.

[144] Respectfully, I do not accept some of the viva voce evidence given by Dr. Mancey on the water testing issue.

[145] His suggestion that the CNSC had “access to all kinds of information” and that the parties “accepted the water was a very high quality” and that therefore water testing did not need to be considered in the Mancey Report strains credulity. The Mancey Report was a 347-page report, intended to be a rigorous, scientific root cause analysis. In light of the significant leaking not part of the NRU design, the lacuna of failing to consider and include basic information of water testing undermines the rigor of the analysis as to the cause of the corrosion at J-41.

[146] The review of the transcripts of the June 11, 2009 CNSC public hearing confirms members were searching for answers, that they were frustrated by the lack of concrete information. I conclude that CNSC members did not have special knowledge about water quality as suggested by Dr. Mancey.

[147] The lacuna and absence of water testing in the Mancey Report is difficult to understand in light of the concerns raised by Dr. Newman in August 2009 in his peer review of the Interim Mancey Report. He stated: “More detailed water chemistry information, including trace anions other than nitrate and bicarbonate, would be helpful.”[^27]

[148] Similarly, Alcoa, the second independent expert reviewing the Interim Mancey Report also recommended water testing.[^28]

[149] Dr. Kirk testified that it was very frustrating that the water testing results were not contained in the Mancey Report and that they were not made available to the experts in spite of his requests that they be produced.

[150] Dr. Frankel also confirmed the importance of the water testing results.[^29]

[151] Dr. Revie initially testified that he “did not think a water analysis would be that helpful”. I do not accept this evidence. In cross-examination Dr. Revie begrudgingly acknowledged that if he had been doing the analysis he would have considered water testing data.

Conclusions on Water

[152] I accept the evidence of Thomas Jackson that the process water from the Ottawa River was treated with chlorine. It was tested regularly and shocked as needed with chlorine to ensure that the process water did not contain microorganisms. I accept Mr. Jackson’s evidence that the plumbing fittings dividing the process water were old, leaking and required frequent repairs.

[153] I accept Dr. Mancey’s evidence that the ion system was designed to take out contaminants, or reduce them to “an extremely low” level.[^30] Dr. Mancey also acknowledged in his evidence that “he was not particularly familiar with the operation or the design of the NRU.”[^31]

[154] I accept the evidence of Dr. Mancey that in a perfect world, based upon the intended design, the chlorine would be removed, or would be in very low levels in the light water going into the reflector as a result of the process water becoming condensation and the further cleansing ion system.

[155] Although Dr. Kirk expressed the view that the water testing results were not adequately canvassed in the Mancey Report, it was Dr. Kirk’s opinion that charts that were contained in the Mancey Report confirm inferentially the presence of contaminants or aggressive agents.

[156] Specifically, Dr. Kirk testified that Table 21, at page 169 of the Mancey Report, contains conductivity results of water drained from the J-rod annulus measuring ions, which indicates by inference a substantial presence of other ions in that water that were not the expected levels of nitrate or ammonia ions.[^32] This evidence was not challenged by any expert evidence of the Insurer.

[157] The evidence is undisputed that the design of the NRU and the day-to-day functioning did not coincide. The NRU was old and leaking. It must be remembered that the design of the NRU did not contemplate continual leaks from the reflector to the J-rod annulus that were permitted to continue for years. There was not supposed to be any water leaking into the J-rod annulus, be it light water from the reflector, or heavy water from the calandria.

[158] In January 2009, shortly before the leak at J-41, the leaking from the reflector with a capacity of 13,000 liters to the J-Rod annulus reached a torrent of 8,000 liters per day.

[159] It was also confirmed that the design of the NRU contemplated that any leaking water into the J-rod annulus would be fully drained. It is not disputed that significant and voluminous general corrosive product blocked those drains retaining water in the J-rod annulus.

[160] As a result of the leaking, combined with the blocked drains, the presence of light water leaking from the reflector to the J-rod annulus became part of the environment at the base of the J-rod annulus.

[161] I accept the evidence of Dr. Kirk that the level of corrodent/aggressive agent required to attack or weaken the oxide layer on the aluminum to precipitate the aggressive pitting corrosion below the water line at J-41 is “Not just microscopic but nanoscopic. It's on the atomic level.”[^33] This evidence was not contested by any of the experts. I also accept the evidence that once the pitting corrosion begins it is self-perpetuating.[^34]

[162] I have no hesitation in concluding the most reasonable inference available on the evidence, and certainly proof on a balance of probabilities, including a consideration of Dr. Mancey’s testimony as to water testing, is that that there were aggressive agents (the term used by Dr. Frankel) or corrodents (the term used by Dr. Kirk) in the reflector light water leaking into the J-rod annulus.

[163] In all probability the aggressive agent was chlorine, perhaps at very low levels, but sufficient to precipitate corrosion. According to the expert evidence, particularly the evidence of Dr. Frankel there were probably other aggressive agents/corrodents present including copper and iron relevant to the issue of the scalloping observed in the affected one centimeter band in the J-rod annulus.

[164] This finding foreshadows my conclusions that I will outline after a review of the expert evidence that pitting corrosion precipitated by chlorine caused the unanticipated, unexpected, and unique corrosion that penetrated the calandria wall at J-41.

PART III: LEGAL PRINCIPLES AND UNDISPUTED FACTS

General Principles: Interpretation of All-Risks Policies

[165] Before considering the specific exclusion arguments raised by the Insurer, I pause to outline the general principles that apply to interpret an all-risks insurance policy.

[166] All-risks policies, such as the Policy in this case, are generally understood to protect against fortuitous losses, unless such losses are otherwise excluded.[^35]

[167] In Progressive Homes v. Lombard General Insurance Co. of Canada, the Supreme Court of Canada defined fortuity: “When an event is unlooked for, unexpected or not intended by the insured, it is fortuitous. This is a requirement of coverage.”[^36]

[168] The guiding principles of insurance policy contractual interpretation were reviewed by the Supreme Court of Canada in Progressive Homes Ltd. (at paras. 22-24):

• The primary interpretive principle is that when the language of the policy is unambiguous, the court should give effect to clear language, reading the contract as a whole (Scalera, at para. 71).
• Where the language of the insurance policy is ambiguous, the courts rely on general rules of contract construction (Consolidated‑Bathurst, at pp. 900-902).
• Courts should prefer interpretations that are consistent with the reasonable expectations of the parties (Gibbens, at para. 26; Scalera, at para. 71; Consolidated-Bathurst, at p. 901), so long as such an interpretation can be supported by the text of the policy. Courts should avoid interpretations that would give rise to an unrealistic result or that would not have been in the contemplation of the parties at the time the policy was concluded (Scalera, at para. 71; Consolidated-Bathurst, at p. 901).
• Courts should also strive to ensure that similar insurance policies are construed consistently (Gibbens, at para. 27). These rules of construction are applied to resolve ambiguity. They do not operate to create ambiguity where there is none in the first place.
• When these rules of construction fail to resolve the ambiguity, courts will construe the policy contra proferentem — against the insurer (Gibbens, at para. 25; Scalera, at para. 70; Consolidated-Bathurst, at pp. 899-901). One corollary of the contra proferentem rule is that coverage provisions are interpreted broadly, and exclusion clauses narrowly (Jesuit Fathers, at para. 28).

[169] The onus lies on an insurer to prove on a balance of probabilities that an exclusion clause applies.[^37]

[170] Where an insurance policy is clear and unambiguous on its face, it is unnecessary to resort to external evidence to interpret its terms. However, the terms of a seemingly clear insurance policy must be examined in light of the surrounding circumstances, sometimes called the “factual matrix”, in order to determine the intent of the parties and the scope of their understanding.[^38]

[171] The recent Supreme Court of Canada decision in Ledcor Construction Ltd. v. Northbridge Indemnity Insurance Co. interprets a standard exclusion clause for faulty workmanship, and the exception to that exclusion for physical damage in a common form all-risk builders’ policy.[^39] The Court reaffirmed the principles enunciated in Progressive Homes Ltd. If the policy language is unambiguous, effect should be given to the clear language of the policy. If the language is ambiguous, the general rules of contract construction apply to resolve any ambiguity. Of importance is consideration of the reasonable objective expectation of the parties:

These rules include that the interpretation should be consistent with the reasonable expectations of the parties, so long as that interpretation is supported by the language of the policy; it should not give rise to results that are unrealistic or that the parties would not have contemplated in the commercial atmosphere in which the insurance policy was contracted, and it should be consistent with the interpretations of other similar insurance policies.[^40]

[172] The Plaintiffs argue in addition that the rules of construction ejusdem generis and noscitur a sociis should be employed when interpreting the Policy. Finally, if the ambiguity cannot be resolved, contra proferentem applies.

[173] Beyond general interpretive principles and rules of construction, it has also been recognized that where there is little or no Canadian authority on a point of insurance law, Canadian courts may consider American law for assistance.[^41]

[174] The all-risks Policy prepared by the Insurer in this case is 83 pages long with numerous schedules. It is a complex document. Its purpose is to protect MDS from various fortuitous losses, and in particular relevant to this case, the separate coverage for Contingent Time Element covers loss of profits if a supplier is unable to furnish product, unless any of the exclusions apply.

Fortuity in All-Risk Policies

[175] I pause to consider the question of fortuity in all-risks insurance policies. As confirmed in Progressive Homes, “When an event is unlooked for, unexpected or not intended by the insured, it is fortuitous. This is a requirement of coverage.”[^42]

[176] All risk insurance is not all loss insurance, and covers only fortuitous or accidental losses subject to express conditions or exclusions. This was recognized in Essex House v. St. Paul Fire & Marine Insurance Co:

The label, “all risk” as applied to policies of the type exemplified in this case has been called a “misnomer.” Friendly, J., sitting by designation, Aetna Casualty & Surety Co. v. Yates, 344 F.2d 939 (5 Cir. 1965). It has been said, however, that the purpose of such policies is to extend coverage to risks that are not ordinarily included in other types of insurance coverage; and that under all risk policies, recovery is allowed for all fortuitous losses unless the policy contains an express provision excluding the loss from coverage. 13 Couch on Insurance 2d, s. 48:138, p. 596. Therefore, plaintiffs’ argument in this connection that “all risk” is a deceptive term because plaintiff thought that by purchasing it he was thereby insured against literally all risks, while perhaps appealing from a layman’s point of view, must be rejected. The term is one of insurance art and has been judicially recognized and dealt with as such.[^43]

[177] Brown & Donnelly’s text, Insurance Law in Canada, discusses the concept of fortuity in the context of all-risks insurance policies. The authors indicate that

Many property policies are written on an "all risks" basis rather than naming specified perils such as fire, wind, or falling aircraft. The typical insuring agreement covers "all risks of direct physical loss of or damage ... from any external cause". Such policies provide extremely broad coverage which is narrowed primarily by the exclusions contained in the policy and “"All risks" encompasses any fortuitous loss.[^44] [Emphasis added]

Overview of the Policy: What is covered by this ‘All Risks’ Policy?

[178] Before I consider the various specific exclusions and arguments raised by FM about coverage, it is important to review the structure of the Policy as a whole.

[179] The Policy is governed by the Insurance Act, R.S.O. 1990, c.I.8 and is to be interpreted in accordance with Ontario law. It covers “property, as defined in the Policy, against ALL RISKS OF PHYSICAL LOSS OR DAMAGE, except as hereinafter excluded, while located as described in this Policy”. The Policy has a limit of liability of US$700,000,000, subject to specifications for limits of various aspects of the coverage.

[180] Sections B and C outline the totality of property that is covered under the Policy. Section B of the Policy covers Property Damage. According to Section B(1),[^45] the Policy insures Real Property, Personal Property (to the extent of the Insured’s interest in such property), and the interests of contractors and subcontractors in insured property, unless otherwise excluded, located at an Insured Location or within 1,000 feet thereof. Insured Location is defined as “a location 1) listed on a Schedule of Locations attached to this Policy. 2) covered as a Miscellaneous Unnamed Location. 3) covered under the terms and conditions of the Automatic Coverage or Errors and Omissions Provisions.”[^46]

[181] According to Section B(3),[^47] the Policy also provides Additional Coverage “for physical loss or damage insured by this Policy”, subject to the applicable exclusions and deductions. These Additional Coverage sections cover a wide range of subject matters, including shortages in accounts receivable, damage to brands/labels, damage to electronic data, and more.

[182] Section C of the Policy, which is acknowledged to engage in this case, subject to exclusions, covers Time Element losses including loss of profits to MDS flowing from physical damage to a supplier “directly resulting from physical loss or damage of the type insured by this Policy” to a maximum of US$25,000,000.00.[^48] Time Element losses are limited generally by conditions found in section C(1), e.g., the loss has to occur during the Periods of Liability described in section C(1).[^49] Section C(2) lists specific forms of Time Element losses,[^50] such as gross earnings loss, extra expenses during certain conditions loss, and lease interest loss.

[183] Section C(3)[^51] extends Time Element coverage to specified situations, such as losses incurred when a civil authority has prohibited access to property, service interruption loss under certain conditions, and losses incurred for reasonable delays to the start up of business operations.

[184] The Policy coverage is broad, subject to the exclusions, and covers MDS for various losses in 95 different locations worldwide.

[185] FM confirms there is coverage for loss of profits of MDS for interruption of supply of isotopes from AECL under section C of the Policy to a maximum of US$25,000,000.00 unless the Insurer can prove that either the corrosion or the nuclear radiation exclusions applies.

Factual Context: The Relationship between the Parties prior to the Shutdown

[186] FM insured MDS since 1993 to the date of loss in 2009.

[187] Mr. Michael Lodge, the FM adjuster assigned to the file after the loss, and Mr. John Joseph Turcotte, the supervisor from the Toronto office provided affidavits and testified to provide background information about the Policy, factual background relevant to the context in this case and what transpired between MDS and FM after the shutdown of the NRU. Both have enjoyed lifetime careers with FM. Mr. Lodge is now Operations Vice President and Senior General Adjuster with FM. Mr. Turcotte is the senior supervisor in the Toronto office.

[188] Mr. Turcotte confirmed that the wording of the exclusion clauses in this 83-page Policy, plus schedules were not negotiated. He further confirmed that the exclusion clauses in question have been in use since 1985. He provided no case law interpreting the meaning of the exclusion.

[189] Mr. Lodge testified that the Contingent Time Element business interruption insurance in the Policy was additional coverage purchased by MDS for the type of loss in this case, when a supplier experienced a physical loss affecting the supply of product and hence profits of the Plaintiffs. The Policy insures MDS for loss of profits flowing from locations controlled by MDS or from suppliers to MDS. Mr. Lodge testified that the Policy treats the loss at the supplier location in the same way as a loss at the insured location. Hence, if a supplier has an interruption of production caused by physical loss or damage, the loss of profits of MDS flowing from that damage would be covered by the Contingent Time Element Coverage in the Policy, unless a Policy exclusion applies.

[190] FM knew that the supply of isotopes from AECL at the NRU was a major component of MDS’s worldwide activities. Mr. Turcotte and Mr. Lodge confirm that they knew prior to entering into the Policy, and prior to the shutdown that the revenues from the supply of isotopes from AECL constituted 45% of the world income of MDS.

[191] Mr. Turcotte also confirmed that FM was aware that MDS was seeking to increase the Contingent Time Element coverage to $80,000,000 targeted to a potential loss of the supply of isotopes at the NRU, as the US$25,000,000 was not adequate protection. Discussions had been underway between MDS and FM for a year before the shutdown to increase the CTE coverage.

[192] It is a unique undisputed fact that FM was also the insurer for AECL at the Chalk River location.

[193] The NRU itself was no longer covered under the policy between AECL and FM and that change was in the last renewal prior to the shutdown.[^52] There is no evidence before me as to whether MDS was aware of this change.

Chronology of communication between MDS and FM after the shutdown to the denial of coverage

[194] On May 21, 2009, one week after the shutdown, the insurance broker for MDS, Aon Reed Stenhouse Inc reported to FM that MDS suffered a “contingent time element loss” arising from the shutdown and the interruption of the supply of medical isotopes.

[195] FM understood that MDS could have a large loss.

[196] On May 26, 2009, Bob Swinton of FM prepared an Advanced Notice of New Large Loss with a reserve of $6,000,000 for a repair outage of 4 weeks based upon CAN $1.5 million per week (although it is noted that “Client estimates that the financial impact will be a loss of US$ 1.8 million per week). This document was copied to Mr. Turcotte.

[197] In the Advance Notice it is of importance to note that “The Peril Code and Name” is “07-Escaped Liquids”: referring to the leak of heavy water outlined in the document. That document stipulates “LTBD”: liability to be determined.

[198] On May 27, 2009 there is an email sent dated from Bob Najdovski to Jeanne Li which appears to include transmission of the Advance Notice. That email confirms that as FM excluded the reactor from the AECL program at the last renewal the FM loss would be limited to Contingent Time Element for MDS. This internal memo refers to the incident as an accident, and discusses extending the waiting period to “take the deductible up”. It suggests that the “leak is due to corrosion which would otherwise be excluded. Make sure you play that up at the next Stewardship”:

From: Najdovski, Bob

Sent: Wednesday, May 27, 2009 9:00

To: Li, Jeanne

Subject: FW: Advance Notice of New Large Loss: MDS Inc. D.O.L./ 15-May-2009 TE: $0-LTBO

Jeanne, when the news broke about Chalk River going down there was not much of a surprise that we would see a CTE claim.

We have excluded the reactor from the AECL program at the last renewal, so our loss should be limited to CTE for MDS exclusively. This brings up the issue of coverages and limits.

At the very least we need to re-rate this exposures if in fact the only source of Radio isotopes will continue to be Chalk River. In addition, you should consider a waiting period or ADV to take the deductible up. This is the second major interruption to production. The first was a regulatory delay that was not insurable but from both of these it is clear that there is no contingency plan to address an alternate supply for the Radio Isotopes.

Too bad that CTE is accidental occurrence. Looks like the tube leak is due to corrosion which would otherwise be excluded. Make sure you play that up at the next Stewardship. Please let me know when you plan to meet next with Mike and what the timing looks like for the Stewardship. [Emphasis added]

[199] This comment by Bob Najdovski seems to support the view that the Insurer understood that fortuitous, unanticipated corrosion characterized as “accidental” was not intended to be covered by the exclusion.

[200] On May 30, 2009 Michael Lodge wrote an email letter to Michael Butler, the Mississauga manager of MDS summarizing the situation from the perspective of FM at that date, and asking questions relevant to the question of the cause of the heavy water leak relevant to coverage. Mr. Lodge did not simply deny coverage although the Insurer knew corrosion was involved with the leak. The following are the questions asked by Lodge in that email letter:

As discussed, it will be necessary to develop a good understanding of what happened at AECL’s facility in order to be able to determine how the terms and conditions of FM Global Policy SU400 will respond in this instance. Specifically, we would need:

• The date and time when electrical power was interrupted at AECL and when it was restored
• A detailed description of the impact of the power outage on the NRU’s operation (ie why it had to be shut down, were there any problems that occurred as a result of the sudden power outage or during the shutdown procedures)
• What were the circumstances that lead to the discovery of the heavy water leak?
• What caused the heavy water leak? How did the leak develop?
• A detailed description of the leaking equipment – What does this equipment do? What size/capacity is it? What material is it composed of?
• What method will be used to repair the leak?
• What is the impact of the corrosion mentioned in the AECL report? Will it impact the leak repair? Are other repairs needed at this time to address other components that are not leaking?
• What other factors are preventing or delaying the return of the NRU to operation?

Of course, these requests are preliminary in nature – as information becomes available it may lead to other questions or need for further information.

[201] The May 30, 2009 letter confirms that “Once we have sufficient information we will be able to review your policy information and confirm any applicable coverages, loss payables and deductibles in effect.” The letter also confirms that it is neither an admission or denial of liability.

[202] Mr. Lodge testified that he knew from the outset that “possibly exclusions would apply” including the corrosion, faulty workmanship or the nuclear reaction exclusions, but that FM did not know conclusively that the exclusions applied. Mr. Lodge testified that early on he “knew there were issues and knew they had to be investigated”.

[203] Mr. Lodge confirmed that FM had not taken any position as of May 30, 2009.

[204] On June 10, 2009 Michael Butler, responded on behalf of MDS confirming that they had very little information about the cause of the leak, and also confirming that MDS was attempting to mitigate by using alternate isotopes such as Thallium and sourcing reactor isotopes from other suppliers, but stated that “Unfortunately these actions will not materially reduce the impact on the supply disruption”. It is not disputed that there was no alternative source for the Molybdenum 99 isotopes, which composed the lion’s share of supply from AECL.

[205] The June 10, 2009 letter also confirms the importance of these isotopes to the revenue of MDS:

List of isotopes produced by the NRU and the relative financial importance of these isotopes to Nordion operations?

MDS Nordion relies on the NRU reactor to supply the majority of its needs for the following radioisotopoes:

• Molybdenum 99
• Iodine 131
• Iodine 125
• Xenon 133
• Iridium 192

Reactor isotopes comprise approximately 45% of Nordion’s revenue. Molybdenum 99, is of particular significance – comprising approximately 40% of Nordion’s revenue. [Emphasis added]

[206] On June 14, 2009 Mr. Lodge prepared a Loss Memo. That memo confirms the various exclusion clauses that may be applicable including the corrosion, faulty workmanship and the nuclear exclusion. A consultant was retained to view the proceedings at the CNSC held on June 10 and 11, 2009. Further the memo confirms that “outside counsel Liblong Digambar PC retained via Johnston Legal to assist in Policy coverage”.

[207] The Loss Memo confirms that AECL was working on a root cause assessment of the leak incident to identify the probable cause(s) of the corrosion creating a pin hole leak, as well as 5 or 6 areas of concern around the perimeter at the base of the vessel. The Loss Memo confirms that the root cause analysis was to be ready before the next public hearing of the CNSC in August 2009 to determine whether repairs to the NRU should be considered permanent contributing to the life expectancy of the NRU to 2024.

[208] The Loss Memo confirms that “at this time we are reviewing Policy terms and conditions based on information presently available in an effort to determine what coverage, if any, may be afforded for the losses incurred by MDS as a result of the incident. In the event that additional information is needed to determine the Policy application we will review what steps may be appropriate to obtain the required information”. The Loss Memo confirms the loss of sales information provided by Butler to FM.

[209] Mr. Lodge confirmed that the no coverage decision was taken by the FM team sometime in early or mid-July 2009, some 6 to 8 weeks after the shutdown, and prior to the release of the interim root cause report.

[210] The written record before me and the evidence of Mr. Lodge and Mr. Turcotte, confirm that the information considered by FM before the Denial Letter dated August 4, 2009 is the AECL public postings until June 10, 2009 as well as the transcripts from the hearings of the CNSC on June 10 and 11, 2009 and finally a Technical Briefing at the Toronto Board of Trade on July 8, 2009.

[211] The subsequent hearings before the CNSC in August 2009, and the Interim Mancey Report were not available and hence not considered by FM in their decision to deny coverage.

[212] On July 30 2009 a meeting occurred between Mr. Lodge from FM and Mr. Butler from MDS. FM confirmed to MDS that they were denying coverage for the losses. There is no specific evidence before me about what was discussed at that meeting.

[213] On August 4, 2009 Mr. Lodge sent an email letter to Mr. Butler formally denying coverage for the MDS losses (the “Denial Letter”).

[214] On August 4, 2009 Mr. Lodge also signed a Loss Memo showing Mark-off as FM concluded that “based upon available information we reviewed the conditions of the Policy and have concluded that the Policy Exclusions apply in this instance. As such there would be no liability under the Policy for any business interruption incurred by the insured as a result of this incident.”

[215] The Loss Memo states that “following an in depth review of available information we reviewed Policy terms and conditions and concluded no liability for this incident exists based upon our current understanding of the event”. Further the Loss Memo states “If new information becomes available we may revisit but this appears unlikely at this time-as such we are presently Marking-off this loss.”[Emphasis added]

[216] The evidence of Mr. Lodge confirms that the decision to deny coverage was not his decision, but was made by the team of seven people who had been discussing the issue for a period of weeks before the Denial Letter was sent. The FM team included three senior FM employees in Canada, two in the US, US corporate counsel, and Mr. Liblong, counsel for the Insurer at this trial who was retained as a consultant to provide advice about coverage.

[217] There was extensive consultation as to the contents of the Denial Letter with various drafts and comments being exchanged so in the words of Mr. Lodge the Denial Letter “matured”. Mr Turcotte confirmed that FM team contributed ideas to the various drafts.

[218] Mr. Turcotte denied that there was any rush to deny coverage, and if there were “additional facts FM would be happy to consider them”. It appears from the loss memo, and the history, that these words of a willingness to review the claim when new evidence is available ring hollow.

The Information Available to FM by August 4, 2009

[219] It is useful to pause to review the information available to FM when the Denial Letter was written.

[220] The most comprehensive information available to FM about the cause of the corrosion was the transcript of the public hearing held by the CNSC on June 11, 2009.

[221] In the CNSC Public Hearing on June 11, 2009 a presentation was given by various AECL employees and then there was a question and answer opportunity for CNSC members to clarify the status of matters. By June 11, 2009 AECL had identified by ultrasonic inspection the location of the leak as well as six other areas of concern. A panel of experts was assembling to review the safe repair options. A root cause report was to be prepared to determine why the corrosion was not identified in other inspections, and to confirm whether the NRU was fit for service before it restarts.

[222] Peter Elder, a senior employee of AECL, confirmed that AECL was preparing a root cause report as to the cause of the leak and conducting non-intrusive examinations. Since there was an unknown state of the wall AECL decided that it was appropriate to de-fuel and drain the calandria [page 17] which also stopped the leak and stopped the emissions of Tritium which were above actionable levels.

[223] I note that the possibility of pitting corrosion causing the leak was discussed during the public hearing, with the deferral of the question to obtaining meaningful answers as to the cause of the leak in the root cause report.

[224] The Director of Operational Engineering Assessment answered the question of “do they have any sense of degradation in properties over 35 years of the 5052” by saying “usually the degradation mechanism is related to the environment of the metal and as long as the environment is not aggressive to the metal, the degradation mechanism will not appear”.

[225] The Chairman of the CNSC asked a very good question:

“What I don’t understand is, we know there has been water leakage from the reflector. I mean there’s the reflector water in there and I am told there was CO2 put in place. I’m still stuck on the fact that we don’t worry about the water and even aluminum and CO2: Its not the right mixture. Why weren’t we able to detect—or shouldn’t we be able to detect a possibility of corrosion?” [Emphasis added][^53]

[226] Mr. Elder went on to clarify the corrosion that was observed in 2004 was the known generalized corrosion. AECL had not presented any understanding of why there would be local corrosion of the lip and the wall that we’re seeing in a number of locations. “So it is not the overall corrosion that was known and expected to continue; it’s the fact that there was some sort of accelerated or localized corrosion in certain regions.” [page 30 to 31]

[227] A review of the transcript of the June 11, 2009 CNSC hearing makes it clear that the members of the CNSC were searching for information, and that they were very frustrated with the absence of clear concrete evidence about why the unexpected leak had occurred. The panel members expressed clear frustration with the absence of transparency and in the difficulty in getting accurate, understandable information in writing.

[228] A great deal of emphasis was placed by the members of AECL speaking to the members of the CNSC about the importance of the root cause report. Member Barnes at page 62 raised concerns about the composition of the expert panel [conducting the root cause analysis] “We have no information of the nature of that panel. Is it internal? Is it a combination of internal and external, that I would hope for? But no information.” We know now that the Mancey Report was prepared internally by AECL, and not subject to any outside neutral peer review experts.

[229] Mr. Elder confirmed that as of the date of the public hearing “We’re observing the technical meetings, we’re sitting in, but we don’t have any written-very little written technical information to review to date” [page 64].

[230] Member Barriault asked an important question at page 68: “So this—actually this failure of this aging program which I understand is what happened was really not predicted in any way”; “What I am saying NRU has an aging management program in place that oversees degradation of the reactor and does corrective action to repair this. I understand now that CSNC staff is supposed to oversee to make sure that is being done. And I guess what I’m wondering is, you know, where did … the train get off the rails here”. [Emphasis added]

[231] The response to the various probing questions by the Chairman and the committee members was that “we have to await the root cause report”.

[232] The answers to these important questions were to be canvassed in the Mancey Report. Neither the Interim Mancey Report nor the final Mancey Report had been released at the time FM denied coverage on August 4, 2009 in the Denial Letter.

PART 1V-ISSUE 1: DOES THE CORROSION EXCLUSION APPLY PRECLUDING COVERAGE

[233] The Insurer relies on several exclusions. The first is the corrosion exclusion.

[234] The Policy states:

C. This Policy excludes the following, but, if physical damage not excluded by this Policy results, then only that resulting damage is insured: …

3. deterioration, depletion, rust, corrosion or erosion, wear and tear, inherent vice or latent defect.

Preliminary Matters

[235] The word “corrosion” is not defined in the Policy.

[236] It is agreed by the parties that the word corrosion must be interpreted in its normal ordinary meaning understood by a lay person guided by legal precedent as well as general principles applicable to all-risks policies. The meaning of corrosion must be interpreted within the context of the Policy.[^54] A scientific definition does not apply.

[237] Dr. Kirk, Dr. Revie, and Dr. Frankel made suggestions about the common person’s understanding of the meaning of corrosion. It was not appropriate for any of the experts to stray into the arena of the common person’s understanding of corrosion. That is not their area of expertise.

[238] They of course can opine on the scientific definition of corrosion, which would be within their area of expertise. The parties agree, however, that is not the definition to apply in interpreting the Policy in this case.

[239] Therefore the parts of the expert reports dealing with the everyday definition of corrosion are not appropriately admitted into evidence and are to be effectively deleted from the various reports, including the report of Dr. Frankel.

[240] The expert reports and their evidence are of assistance in providing opinion evidence as to the cause of the corrosion at J-41 causing the through-wall leak, not the definition of corrosion.

The Corrosion Issues

[241] The issues to determine the applicability of the corrosion exclusion are:

1. Can the applicability of the corrosion exclusion be determined as suggested by the Insurer by definition alone or is the term ambiguous?
2. If the term corrosion is ambiguous, what is its meaning?
3. What is the cause of corrosion at J-41 having regard to the expert evidence?
4. Considering the factual context, the reasonable expectation of the parties, and the factual findings on the expert evidence, has the Insurer met its onus to prove that the corrosion that caused the leak at J-41 is excluded under the Policy?

Question1. Can the Applicability of the Corrosion Exclusion be Determined as Suggested by the Insurer by Definition Alone?

[242] There were two types of corrosion that occurred on the wall of the calandria as discussed in paras 59 to 73.

[243] There was the known generalized corrosion of the outer calandria wall caused by the creation of nitric acid in the J-rod annulus as a result of leaking water from the reflector. This corrosion that occurred over a long period of time was monitored. This generalized corrosion over the entire wall of the calandria over the life of the reactor from 1974 to 2009 caused the 8 mm calandria wall to thin by approximately 0.4 to 0.5 mm.

[244] This known and anticipated generalized corrosion did not cause the leak. If it had, it is agreed by the Plaintiffs that this non-fortuitous corrosion involving the predictable wearing away of the calandria wall over time would be subject to the corrosion exclusion. It is the Plaintiffs’ argument that this is the very corrosion that is contemplated in the corrosion exclusion in the Policy precluding coverage.

[245] It is the second kind of unanticipated corrosion at J-41 that caused the leak of heavy water that is in issue. Causation of this corrosion is relevant to whether or not the corrosion exclusion applies.

The Parties’ Positions

The Insurer’s position on corrosion

[246] A primary interpretive principle is that when the language of the policy is unambiguous, the court should give effect to clear language, reading the contract as a whole (Scalera, at para. 71).

[247] The Insurer argues that the meaning of corrosion is clear, and not ambiguous. Counsel suggests various very broad definitions of corrosion applied in decisions in the United States. The Insurer urges this court to accept these definitions as conclusive of the applicability of the corrosion exclusion, without further inquiry into the evidence as to the cause of the second kind of localized, unanticipated, corrosion.

[248] The Insurer lists several lay and scientific definitions of corrosion in its materials, the majority of which define corrosion as the wearing away of a material by chemical or electrochemical means.[^55]

[249] According to the Insurer, “the definition of ‘corrosion’ most commonly applied by the Canadian and American courts is ‘the wasting or change in the state of material by chemical action”.[^56] The Insurer also cites definitions from The Free Dictionary and Collins English Dictionary, the primary definition being “a process in which a solid, esp. a metal, is eaten away and changed by a chemical action, as in the oxidation of iron in the presence of water by an electrolytic process”.[^57]

[250] The Insurer’s position is that it is unambiguous that the corrosion exclusion is meant to exclude all forms of corrosion: both non-fortuitous as well as fortuitous corrosion. The Insurer argues therefore that all of the expert evidence is not needed to determine whether the corrosion exclusion applies to preclude coverage under the Policy. It does not matter whether the corrosion that caused the leak was from unanticipated, localized tunneling corrosion from nitric acid (the Mancey theory) or pitting corrosion caused by the unanticipated presence of an aggressive agent/contaminant, such as chlorine, at the site of the leak (the Kirk and Frankel theory).

[251] The Insurer also argues that as the meaning of corrosion is clear and unambiguous it is unnecessary to consider the context of the contract, and the objective reasonable expectations of the parties.

The Plaintiffs’ position on corrosion

[252] The Plaintiffs argue that the second corrosion causing the leak was not covered by the corrosion exclusion in the Policy as it was fortuitous: an unanticipated event and the very kind contemplated to be covered by all-risks insurance.

[253] The law is clear that the terms of even a seemingly clear insurance policy must be examined in light of the surrounding circumstances, sometimes called the “factual matrix”, in order to determine the intent of the parties and the scope of their understanding.[^58]

Conclusion: The meaning of corrosion in the Policy is ambiguous

[254] Although the Insurer’s arguments at first blush may seem persuasive, it is clear from the evidence that the applicability of the corrosion exclusion cannot be decided based upon a definition alone, as suggested by the Insurer without consideration of causation of the water leak at J-41 and the evidence.

[255] The witnesses called on behalf of the Insurer conceded that the corrosion at J-41 is fortuitous, and that not all corrosion is excluded by the Policy. This evidence ends the argument that this exclusion can be determined based upon a definition alone, whatever that definition may be. These admissions confirm that the meaning of corrosion in the context of this Policy is ambiguous.

[256] I turn to consider the Denial Letter of August 4, 2009 outlining the reasons why FM took a no coverage position, followed by a summary of the evidence of Mr. Lodge and Mr. Turcotte outlining these important admissions.

FM Position Taken in the Denial Letter

[257] Both Mr. Lodge and Mr. Turcotte were part of the seven person FM team preparing the Denial Letter of August 4, 2009. Counsel also participated in the careful crafting of the Denial Letter, including US in-house counsel and Mr. Liblong, FM’s counsel in this case.

[258] In the Denial Letter, the Insurer confirms that both the generalized and localized corrosion are excluded under the Policy:

Based upon information available to date, it appears that there may be two different types of corrosion damage and/or corrosion mechanisms implicated. The first has been described as a generalized thinning of the walls of the calandria. AECL has been aware for years that this type of “corrosion” has been occurring. The second type is referred to as localized corrosion which AECL appears not to have known about prior to the leak.

Insurance policies only provide coverage for “risks” – they do not provide coverage for certainties. To allow insurance coverage on a certainty has been held to violate public policy and encourage fraud. Accordingly, only “fortuitous” losses are insurable. As AECL was aware that “general corrosion”, or “thinning”, of the walls of the calandria was taking place for at least several years prior to the leak occurring, and expected this “thinning” to continue during the Policy period, it is our view that this “thinning” corrosion was not fortuitous. Rather, it was certainty and, therefore, not insurable.[^59]

[259] The Statement of Defence of the Insurer pleads:

19. FIMC (sic) further pleads that the only physical damage detected by AECL after thorough investigation of the NRU consisted of the known and ongoing general corrosion of the calandria, and the localized corrosion of the calandria wall and light water reflector lip. FMIC pleads that such “corrosion” is expressly excluded from coverage under the Policy pursuant to exclusion 5.C.3) in Section B of the Policy.[^60]

[260] The Denial Letter refers to the two types of corrosion, then appears to conflate both kinds of corrosion as being non-fortuitous corrosion that for public policy reasons cannot be covered in an all-risks policy. The pleading similarly conflates both the fortuitous and the non-fortuitous corrosion in alleging the corrosion exclusion applies.

[261] The Plaintiffs acknowledge that the non-fortuitous, generalized corrosion caused by nitric acid, and causing the generalized thinning of the calandria wall over time would be captured by the exclusion. The issue is whether the unexpected fortuitous corrosion at J-41 causing the through-wall penetration and leak is excluded.

Review of evidence of Mr. Lodge and Mr. Turcotte

[262] I review the admissions made by Mr. Lodge and Mr. Turcotte in their viva voce evidence at trial.

[263] First they acknowledged that the corrosion that caused the leak at J-41 was fortuitous as it was unique, unexplained and unexpected.

[264] Second, they both acknowledged that not all corrosion under the Policy is excluded from coverage.

[265] Mr. Lodge confirmed in cross-examination that the leak at J-41 was fortuitous:

Q. There was another corrosion going on that you did not know about, AECL did not know about, FM Global did not know about, and that was not general thinning, it was a different type of corrosion. Is that correct?

A. Correct.

Q. And this corrosion, this particular corrosion, a localized corrosion, it was not known about by AECL and you say that, right?

A. Indeed, right then. Right. Localized was not known by AECL.

Q. It was a surprise to them and not picked up at any prior investigation, correct?

A. That’s my understanding, yes.

Q. The issues we’ve got with corrosion is also unique.

A. I’m sorry. Okay.

Q. All right. So, in talking about whether something is fortuitous or not fortuitous, it being unique would be on the side of fortuity, right?

A. I, I think we, we agree that the second corrosion mechanism was fortuitous. [Emphasis added.][^61]

[266] Mr. Turcotte, the supervisor to Mr. Lodge, also confirmed the evidence of Mr. Lodge that the corrosion at J-41 was fortuitous:

Q. And Mr. Lodge told us that as far as he understood, this was a fortuitous loss.

A. The localized corrosion was.

Q. Yes. You agree with that?

A. Yes.[^62]

[267] Mr. Turcotte’s examination-in-chief also confirmed that corrosion is not always excluded from coverage.:

Q. Okay. Did it matter to you as to what the precise cause of the corrosion was?

A. I think – it does matter.

Q. Can you explain that?

A. Like – so some causes are explicitly excluded. Some causes can trigger coverage for some types of corrosion.

Q. And are you able to assist with respect to a cause that trigger [sic] coverage for corrosion?

A. Sure. Something like a sprinkler leakage, that’s an insured peril. A sprinkler leakage can cause flash corrosion and we would – and that sprinkler leakage is covered and corrosion is a direct result of that covered peril. We would pay for that type of corrosion.[^63]

[268] Mr. Turcotte also made a somewhat convoluted admission in his discovery transcript that not all corrosion is excluded by the wording in the Policy. The answer was modified five years later to include corrosion which occurs as a result of some covered event, or in other words, resulting damage.[^64]

[269] Based upon the evidence called by the Insurer that the corrosion in this case is fortuitous, and that not all corrosion is covered by the exclusion it is clear that this case cannot be decided based upon a simple definition of corrosion. The circumstances for coverage where the exclusion does not apply may well be broader than the concessions made in the Insurer’s evidence.

[270] Therefore, based upon this cogent evidence I find the meaning of the exclusion is ambiguous. Further analysis is required.

Question 2. If the term corrosion is ambiguous, what is its meaning in the context of the Policy?

[271] The Denial Letter confirms that the Insurer denied the Plaintiffs’ claim as it alleges the corrosion leak at J-41 was non-fortuitous. However, the representatives of the Insurer admitted in their evidence at trial that the corrosion leak was fortuitous.

[272] The Plaintiffs submit that it was open to the Insurer, as drafter of the Policy, to define corrosion broadly to explicitly exclude both fortuitous and non-fortuitous corrosion. The Insurer did not do so.

[273] Simply because the exclusion clause says “corrosion” as confirmed by the evidence of the Insurer, it does not mean all forms of corrosion, however caused, are excluded. As the Supreme Court of Canada said in Consolidated-Bathurst v. Mutual Boiler:

[L]iteral meaning should not be applied where to do so would bring about an unrealistic result or a result which would not be contemplated in the commercial atmosphere in which the insurance was contracted. Where words may bear two constructions, the more reasonable one, that which produces a fair result, must certainly be taken as the interpretation which would promote the intention of the parties.[^65]

[274] Given the evidence of the representatives of the Insurer, there can be no doubt that the meaning of the term corrosion in the Policy is ambiguous.

[275] The applicability of the corrosion exclusion can only be determined in light of the caselaw, the terms of the Policy considered as a whole, the factual context, and the parties’ reasonable expectations considered objectively.

[276] The Plaintiffs also rely upon the interpretive doctrines of ejusdem generis and noscitur a sociis and finally, if required, by applying contra proferentem.[^66]

Caselaw on Meaning of Corrosion

[277] As the term “corrosion” is not defined in the Policy, I must apply its normal and ordinary meaning and be guided by legal precedent, as well as the general principles applicable to all-risks policies to interpret the meaning of this term within the context of the Policy.

[278] None of the Canadian insurance textbooks consider the meaning of corrosion in all-risks policies.

[279] The Plaintiffs rely on obiter comments in a case from this court, PCL Constructors Canada Inc. v. Allianz Global Risks US Insurance Co.[^67] This appears to be the only Canadian decision considering the meaning of corrosion in the context of an all-risks policy.

[280] In PCL Constructors, Myers J. commented in obiter that exclusions for rust and corrosion in all-risks policies are “generally understood in law to relate to the normal risk of wear and tear of property left exposed to the elements. Such natural processes are not considered to be ‘fortuitous’ events of the type encompassed by an all risks policy”: at para. 16. He also considered whether the ejusdem generis rule of construction may “limit the meaning of ‘corrosion’ to only ‘naturally occurring corrosion’ rather than corrosion brought on by untimely exposure to corrosive liquid as occurred in this case”: at para. 17.

[281] The Plaintiffs urge this court to adopt Myers J.’s definition of corrosion. The Insurer argues that all these comments are obiter and not binding, and refers to the American jurisprudence for the appropriate definition of corrosion.

[282] The Insurer argues the exclusion clause considered in PCL Constructors is different from the exclusion clause in this case. The court considered the following corrosion exclusion: “5(j) Loss or damage caused directly or indirectly by rust or corrosion, frost or freezing, pollution or contamination unless caused directly by a peril not otherwise excluded herein.”

[283] The Insurer argues that the corrosion exclusion in the Policy can be distinguished from the clause in PCL Constructors as there is no exception for corrosion caused by an otherwise insured peril; the exception to the corrosion exclusion in the Policy only applies to resulting physical damage not otherwise excluded caused by the corrosion.

[284] The Insurer also argues that the statement that the corrosion exclusion applies only to “non-fortuitous” events of damage is nonsensical. All-risks policies only insure “fortuitous” events of physical loss or damage. If physical loss or damage is not fortuitous, it does not fall within the basic coverage grant. There is, therefore, no need to exclude it and the exclusion would be redundant. The Supreme Court of Canada has stated clearly in Progressive Homes that any interpretation which would render the exclusion redundant is to be avoided.[^68]

[285] I do not accept these arguments.

[286] In the affidavit of John Turcotte he expressed his views about the interpretation of the Policy in paragraphs 4 and 5. Mr. Brock appropriately objected to this evidence. The law is clear that the subjective intention of the parties is irrelevant: it is the reasonable expectations of the parties assessed objectively in the context of the commercial environment when the Policy was entered into that must be ascertained.[^69]

[287] As there is a paucity of caselaw in Canada considering the meaning of corrosion, the Insurer urges that, in these circumstances, I may look to caselaw in the US for guidance.[^70] This is particularly so where the same provision under consideration is in common use by the insurance industry in Canada and the United States and where the American authorities have applied rules of construction not materially different from those applied by Canadian courts.[^71]

[288] The American rules of construction are not materially different from those found in Canada. Couch on Insurance indicates that plain language is to be given its ordinary meaning, but where there is ambiguity, the reasonable intentions of the parties should be sought. Coverage is to be interpreted broadly, exclusions narrowly. The onus of proof for exclusions and exceptions are the same as those in Canada.[^72]

[289] Although the guiding principles appear consistent with Canadian caselaw, a review of the American cases illustrates a more literal, definitional approach in interpretation of insurance contracts, compared to Canadian caselaw. Our highest court make it clear that the meaning of the words of an insurance policy must be assessed not only with regard to a dictionary definition of the word in question, but in the context of the entire contract, the facts and reasonable expectation of the parties at the time the contract is entered into.[^73] The Ontario Court of Appeal confirms in Dunn v. Chubb Insurance Co. of Canada that the terms of a seemingly clear insurance policy must be examined in light of the surrounding circumstances, sometimes called the “factual matrix” at the time of the contract, in order to determine the intent of the parties and the scope of their understanding.[^74]

[290] The Insurer urges that I apply the strict definition of corrosion found in American jurisprudence, without consideration of the factual context and the reasonable expectations of the parties.[^75] Counsel argues that I should apply the cases that confirm that corrosion exclusions exclude all forms of corrosion damage regardless of the cause.[^76] In particular, the Insurer relies upon the US decision of Lantheus Medical Imaging Inc. v. Zurich American Ins. Co., which arose out of the same shutdown of the NRU in the context of a motion for summary judgment with another plaintiff and another insurer interpreting another policy of insurance.[^77]

Lantheus is Distinguishable

[291] The Insurer places great emphasis on the importance of Lantheus, and urges that it is dispositive of this case.

[292] One key difference between the two cases is the acknowledgments made in this case by witnesses from the Insurer confirming that the corrosion in this case is fortuitous and that not all corrosion is excluded under the Policy. These admissions are clear evidence of ambiguity as to the intended meaning of the corrosion exclusion, and preclude an approach by this Court based on a strict definition of the term corrosion, without a consideration of the evidence.

[293] There are other differences of note.

[294] The wording of the two exclusion clauses is quite different and the exclusion clause in Lantheus appears to be more specific, confirming all corrosion is excluded.

[295] In this case the wording of the exclusion is

This Policy excludes the following, but, if physical damage not excluded by this Policy results, then only that resulting damage is insured: …

3. deterioration, depletion, rust, corrosion or erosion, wear and tear, inherent vice or latent defect.

[296] The relevant exclusion clause in the policy in Lantheus provides:

We will not pay for loss or damage resulting from any of the following; such loss or damages is excluded regardless of any cause or event that contributes concurrently or in any sequence to the loss or damage, except as specifically provided. …

5. Developing, Latent and Other Causes The effects or cause of: … b. Deterioration, depletion, rust, corrosion, erosion, loss of weight, evaporation [,] or wear and tear[.] …

But if any of these results in a covered cause of loss, this exclusion does not apply to the loss or damage caused by the covered cause of loss. [Square brackets in original.]

[297] Finally, the arguments advanced in the two cases and the evidence before the court in Lantheus and this court are very different. Given the wording in the policy in Lantheus, and the restrictive US caselaw, the plaintiff in that case argued that the through-wall penetration of the calandria occurred too quickly to be corrosion for the purposes of the corrosion exclusion clause.[^78] The insured argued that the corrosion covered in the exclusion was that corrosion which occurs gradually and inevitably over the lifetime of a machine.

[298] The court rejected that argument, relying on other American decisions interpreting corrosion broadly that held that all corrosion, however caused, is excluded from coverage.[^79]

[299] The court in Lantheus also rejected that the cause of the through-wall penetration was the formation of an electrochemical cell, given its speed and the absence of an acid base, and not “corrosion”. The court did not accept this submission: “The Court finds nothing in the dictionary definition that requires the presence of an acid; to the contrary, the dictionary includes ‘wearing away ... by a[n] ... electrochemical ... process.’ This definition fully embraces Lantheus's assertion that an ‘electrochemical cell’ caused the [redacted] Penetration.”

[300] The court also rejected attempts to apply noscitur a sociis and ejusdem generis.

[301] The Plaintiffs in this case argue that the corrosion, however long it takes to develop, was fortuitous and unexpected, due to the presence of an aggressive agent or corrodent that precipitated the corrosion. This unanticipated circumstance is the exact kind of fortuitous event usually covered in all-risks policies.

[302] There was no expert evidence called in Lantheus in support of the position that the corrosion causing the leak in the calandria was unforeseen, fortuitous corrosion caused by an unanticipated aggressive agent as opposed to the non-fortuitous, known and anticipated general corrosion. The court in Lantheus did not consider the distinction between non-fortuitous and fortuitous corrosion, and did not consider evidence of the presence of an aggressive agent.

[303] The Lantheus decision was a summary judgment motion with no viva voce evidence.

[304] Notwithstanding that the Lantheus case involved the breakdown of the NRU and a corrosion exclusion, that case can be distinguished based upon the different wording of the exclusion clause itself, the very different evidence before this court and the arguments advanced, and most importantly the admissions made by the Insurer in this case that confirms that the meaning of the term corrosion in the Policy in this case is ambiguous.

Fortuity and the US cases

[305] There is no American case that discusses the cause of corrosion being the presence of an unanticipated aggressive agent or corrodent. As well, a review of the American caselaw referred to by counsel does not focus on the question of fortuity.

[306] I note that Adams-Arapahoe Joint School Dist. No. 28-J v. Continental Ins. Co. is somewhat of an exception and distinguished between fortuitous and non-fortuitous corrosion based upon the exclusion clause in that particular policy.[^80]

[307] In Adams-Arapahoe, part of a high school roof collapsed due to corrosion damage. The corrosion was caused by a covered risk, defective design and/or construction, as the construction company had used a form of concrete known to cause corrosion. Adams-Arapahoe’s claim was denied by its insurer. A partial summary judgment decision held that the corrosion exclusion did not preclude coverage. A jury trial was held on the remaining issues, and Adams-Arapahoe was granted judgment. The decision was appealed to the 10th Circuit Appeals Court.

[308] The appeal court in Adams-Arapahoe concluded that the corrosion exclusion clause at issue excluded only naturally occurring corrosion, and that fortuitous corrosion would be covered.

[309] The 10th Circuit Court of Appeals first confirmed that generally speaking, “under Colorado law the word ‘corrosion’ unambiguously refers to all corrosion, however brought about.” However, because a latter portion of the corrosion exclusion clause in that policy stated “[the Policy excludes loss caused by] … corrosion … unless such loss results from a peril not excluded in this policy”, on appeal the court held that “In effect, the corrosion exclusion applies only to naturally occurring corrosion.”

[310] The 6th Circuit Court of Appeal in Libbey-Owens-Ford Co. v. Insurance Co. of North America, applied Adams-Arapahoe on the basis of the “unless” clause (i.e., a covered peril caused damage that would have otherwise been excluded).[^81]

[311] In the case of Leprino Foods Company v. Factory Mut. Ins. Co., the Colorado District Court recognized that “[the court in Adams-Arapahoe] interpreted an exclusion for corrosion in an all-risks insurance policy with an exception for loss resulting from a ‘peril not excluded in this policy’ to limit the exclusion to naturally occurring corrosion.”[^82]

[312] Although the exclusion clauses in the respective policies are different, the principles in Adams-Arapahoe, Leprino Foods Company and Libbey-Owens-Ford Co. are useful by analogy in interpreting the unique facts of this case.

[313] The other cases relied upon by the Insurer, Sports Arena Management, Lantheus, City Brewing, Alex. R. Thomas, Gilbane and Bettigole do not consider fortuity when applying the corrosion exclusion. The word fortuity may be mentioned, usually simply reiterating that the insurer has alleged the loss caused by corrosion was non-fortuitous, but there is no substantive discussion of the concept of fortuitous verses non-fortuitous corrosion.

[314] For these reasons, although I have carefully considered the US cases presented by the Insurer, I am not persuaded that the US caselaw is helpful in determining the issues in this case. I am guided by the clear principles of interpretation outlined by the Supreme Court of Canada in Progressive Homes and Ledcor.[^83] I also must consider the evidence in this case, which is quite different from the evidence and arguments advanced in Lantheus and the other cases, and apply the established principles confirmed by our highest court to the facts of this case.

The Definition of Corrosion for this Policy

[315] I have considered all of the various definitions of corrosion submitted by counsel and considered the caselaw.

[316] Applying the general principles in Progressive Homes Ltd. to the evidentiary record before the court, I reject the Insurer’s position that all corrosion, non-fortuitous and fortuitous alike, is covered by the corrosion exclusion in this Policy. I reject the all-inclusive definitions of corrosion advanced by the Insurer as enunciated in the American caselaw.[^84]

[317] In the context of this Policy, the evidence in this case and the reasonable expectation of the parties, I find that the exclusion applies only to non-fortuitous anticipated corrosion.

[318] The Canadian Oxford Dictionary defines “corrosion” in simple lay person’s terms as “the process of corroding, esp. of a rusting metal.” “Corrode” is defined as “wear away, esp. by chemical action.”

[319] I adopt this Canadian lay person’s definition of corrosion for the purpose of interpreting the corrosion exclusion in the Policy but add the words “anticipated and predicable” to make in clear that the exclusion is intended to cover non-fortuitous corrosion, not fortuitous corrosion.

[320] I conclude that the definition of corrosion for the Policy in this case having regard to the Policy considered as a whole, the factual context and the reasonable expectation of the parties is:

“The anticipated and predictable process of corroding, esp. of a rusting metal.” “Corrode” is defined as “wear away, esp. by chemical action.”

Question 3. What is the cause of the corrosion at J-41: a review of the expert evidence?

[321] To assess the applicability of the corrosion exclusion in light of the definition above, it is necessary to canvas the expert evidence to understand what caused the leak at J-41 in the calandria wall.

[322] Was the leak at J-41 caused by the known presence of nitric acid in accordance with the theory outlined in the Mancey Report and adopted by the Insurer? Or was the leak caused by pitting corrosion precipitated by an unanticipated aggressive corrodent or agent at J-41 in the particular environment of the J-rod annulus in accordance with the theory of the Plaintiffs?

[323] For the reasons I will outline in some detail, I accept the evidence of Dr. Kirk as confirmed and amplified by Dr. Frankel as to the exact mechanism of the localized pitting corrosion. I find that nitric acid was at most part of the environment in the J-rod annulus, but was not the direct or indirect causal factor precipitating the pitting corrosion at J-41.

[324] By way of foreshadowing, taking a narrow view of the corrosion exclusion clause,[^85] and applying the definition of corrosion outlined above, I conclude that it would be within the parties’ reasonable expectations, considered objectively, that the fortuitous, unanticipated unpredicted corrosion at J-41 causing the leak of heavy water into the J-rod annulus would be covered by the all-risks Policy unless another exclusion applies.

Early Notification Report May 27, 2009

[325] Peter Elder of AECL prepared the Early Notification Report[^86] reporting the leak to CNSC pursuant to its licence obligation on May 27, 2009. Neither counsel referred to this document.

[326] The Early Notification Report confirms in the initial findings “As the inspections performed in 2004 for the Life Assessment of the calandria had not detected any generalized wall thinning, the leak is now believed to be due to localized pitting. The mechanism for this has not been confirmed yet: AECL is in the process of carrying out inspections and assessments to determine the extent of corrosion near the bottom of the vessel”.

[327] Under Planned/ to be Taken includes ultra sonic examinations to be witnessed by CNSC staff. The Early Notification Report confirms “The presence of widespread pitting in a small area would be the worst situation, as it could result in lost mechanical resistance of the vessel in the region.” [Emphasis added.]

[328] At the June 11, 2009 public hearing of the CNSC the possibility of localized pitting corrosion being the cause of the leak was discussed.[^87] Much emphasis was placed at that hearing on the root cause report being prepared to review the cause of the corrosion, and whether its presence had been missed in prior inspections.

[329] Curiously, the concern that localized pitting corrosion may have been responsible for the leak were not considered by Dr. Mancey in either his interim or final report.

[330] I turn now to review the Mancey Report and the expert evidence.

Dr. Mancey’s Evidence

[331] Dr. Mancey is a chemist with a doctorate in electrochemistry. In 2009 Dr. Mancey was a respected, twenty-five-year career Research Scientist at AECL working at the Chalk River location doing research on corrosion in other contexts.

[332] Prior to this incident, Dr. Mancey had no direct dealings with NRU until the shutdown in 2009. Dr. Mancey had one interaction in his long career with AECL with the NRU assessing corrosion of the nuclear fuel discharged from the core of the reactor in the rod base water filled ponds.

[333] Without being critical, it is fair to point out that Dr. Mancey as chemist had a limited experience assessing physical aspects of corrosion at the NRU. He was a member of the electrochemical society, but not a member of the National Association of Corrosion Engineers, which is an organization with a focus on corrosion not limiting its membership to engineers.

The Interim Mancey Report and the External Review

[334] Dr. Mancey, in consultation with others, produced an interim report (the “Interim Mancey Report”), issued for review and comment in July 2009 and issued as approved for use in October 2009. The Interim Mancey Report suggested that nitric acid was the probable cause of the “localized tunneling” corrosion:

Notwithstanding the current lack of direct evidence relating to the specific wall thinning sites, all available evidence suggests that the degradation is a manifestation of the nitric acid corrosion phenomenon that has been a known degradation mechanism over the lifetime of the NRU reactor

At present it is not possible to reconcile the highly localized final wall penetration causing the vessel leak with the general corrosion expected of nitric acid corrosion of uncompromised material.[^88]

[335] The Interim and Final Reports did not consider the concerns raised by Peter Elder, senior AECL employee, before the CNSC on June 11, 2009 that pitting corrosion may be the cause of the leak.

[336] The Interim Mancey Report had been submitted to two external corrosion experts prior to its release for review and comment.

[337] Both independent experts questioned the tentative conclusions reached in the Interim Mancey Report that nitric acid was responsible for the leak at J-41, and suggested other potential corrosion mechanisms to be considered.

[338] The first independent expert, Alcoa, suggested potential microbiologically induced corrosion (“MIC”) or intergranular corrosion. Microbiological corrosion is a type of corrosion caused by microorganisms. Intergranular corrosion is a selective attack of the alloy at grain boundaries of the alloy crystals and can produce perforation at those specific sites much faster than the uniform corrosion process. The report also raised the importance of water testing.

[339] Dr. Newman suggested in his report that either intergranular or open pitting corrosion were more probable causes of the through-wall penetration challenging the hypothesis that nitric acid was the cause of the leak as suggested in the Interim Mancey Report.

[340] Dr. Newman outlined in the summary of his 2009 report that the likely cause of the through-wall penetration is not nitric acid that he describes as a “self corrosion with a purely local cathode”. He states:

“I believe that the intense pit-like events, whether open or intergranular, did not occur with a purely local cathode [nitric acid]….I believe they occurred following a switch from a fairly slow, uniform acid corrosion mechanism….[caused by nitric acid] to a pit-like or “underdeposit” mechanism where the original acidity in the water is no longer relevant- the acidity in the pit now came from cation hydrolosis, as in a classical localized corrosion”. [Emphasis added.]

The Final Mancey Report

[341] The Mancey Report dated September 2010 opines that the localized corrosion is caused by nitric acid, and carbonic acid, although that conclusion is not unequivocal and he suggests further research is needed.

“The understanding of the conditions in the J-rod annulus, when combined with the insights gained from reviewing the scientific literature, leads to the belief that corrosion of the calandria wall can be attributed to the presence of nitric acid and carbonic acid. The acidic conditions are the underlying cause of all the corrosion in the annulus. However, a number of different corrosion mechanisms are active, affecting the distribution and rate of corrosion.”[^89] [Emphasis added.]

[342] The subsequent experiments ruled out carbonic acid as causing the localized corrosion. At trial, Dr. Mancey confirmed his opinion that nitric acid was the cause of the localized corrosion, supported by Dr. Revie, the Insurer’s Rule 53 expert.

[343] In the final Mancey Report, Dr. Mancey dealt with and rejected the hypothesis of Alcoa suggesting MIC as a potential cause of the corrosion at J-41, as the level of radiation in the J-rod annulus was too high to sustain MIC. He also rejected the theory of intergranular corrosion as a possible cause as the coupon sample cut from the calandria wall near J-41 did not show intergranular corrosion. These two conclusions are not challenged by the Plaintiffs.

[344] Curiously, the final Mancey Report did not consider Dr. Newman’s independent peer review opinion that the corrosion at J-41 is not consistent with the AECL theory of self-corrosion with the purely locale cathode of nitric acid, suggesting classic pitting corrosion caused the leak at J-41.

[345] The final Mancey Report was internally reviewed by 8 employees of AECL. It appears that no one except Dr. Mancey had specialized expertise in corrosion. The Mancey Report was not subject to any external peer review to assess the reliability of the conclusions reached. This seems to be an odd lacuna in the protocol, given the importance of the report, the concerns raised about the conclusions of the Interim Mancey Report by the two independent experts, and the concerns expressed by the CNSC members at the June 2009 hearing.

[346] The failure to have the Mancey Report externally reviewed was criticized by Dr. Frankel, particularly as the external experts reviewing the Interim Mancey Report questioned the preliminary causation theory advanced by Dr. Mancey.

[347] Dr. Mancey testified that the decision not to have the final Mancey Report subject to external peer review was to meet the deadline established by the protocol of the CNSC to reopen the NRU.

Further Research and Reports After the Release of the final Mancey Report

[348] After the Mancey Report was submitted in September 2010, the CNSC required further research to be conducted to verify the conclusions in the Mancey Report as a condition of the NRU resuming operations.

[349] After March 4, 2019, and shortly before this trial resumed in September 2019, AECL disclosed three further reports prepared by Dr. Mancey, including an experiment that Dr. Mancey concluded confirmed his findings that nitric acid was the cause of the leak at J-41.[^90]

[350] The experiment involved concentrated levels of nitric acid and aluminum, in air but without the critical presence of water.

Dr. Newman’s Evidence

[351] Dr. Newman, who was one of two external experts who conducted the independent peer review of the Interim Mancey Report testified at this trial. He filed his original peer review report, and provided an affidavit and viva voce evidence. His evidence supports Dr. Kirk’s theory of causation.

[352] In Dr. Newman’s viva voce evidence he confirmed that the pitting corrosion described in his 2009 report was caused by a switch mechanism from fairly uniform nitric acid corrosion, to a highly localized pit-like corrosion, required the presence of a contaminant, probably chlorine, but potentially aided by other contaminants such as copper or peroxide.

The Rule 53 Experts and the Court-Appointed Expert

[353] The two Rule 53 experts, Dr. Kirk and Dr. Revie, exchanged a total of 9 reports with some bewildering detail and unusual tit for tat chart formats considering the various reports. There were numerous rule 53 reports due to the disclosure of the three additional reports obtained by Mr. Brock after March 2019. These additional AECL reports, relevant to the matters in issue, required further commentary by the experts.

[354] The presence of Dr. Frankel in sorting out issues was extremely helpful.

[355] Thanks to Dr. Frankel, I am able in this decision when considering the expert evidence to narrow the focus to the matters in issue. Many of the peripheral issues in the nine reports canvassed by Dr. Kirk and Revie were identified by Dr. Frankel as being insignificant to the question of the causation of the leak at J-41.

[356] Counsel accepted Dr. Frankel’s approach. Neither did the experts address these peripheral issues in their viva voce evidence, nor were they addressed by counsel during argument. Dr. Frankel’s involvement as a court appointed expert significantly streamlined the complex expert evidence.

[357] As I will outline, I accept the evidence of Dr. Frankel, who adopts the opinions of Dr. Newman and Dr. Kirk, that pitting corrosion caused the leak at J-41.

Overview of the Rule 53 Expert Evidence

[358] Dr. Kirk, on behalf of the Plaintiffs, opined in his report dated October 14, 2014 that the unusual pitting corrosion at site J-41 in the J-rod annulus was enhanced by the oxidizing environment due to radiation, the use of carbon dioxide and most importantly, the leakage of light water into the J-rod annulus from the reflector. He did not agree with the conclusion in the Mancey Report that the cause of the corrosion at J-41 was nitric acid.

[359] Dr. Kirk concluded that the leak at J-41 in the calandria wall was pitting corrosion precipitated by the presence of some contaminant in the water leaking into the J-rod annulus, most probably chlorine, and he opined that nitric acid played no role in the creation of the pitting corrosion that penetrated the calandria wall. He requested the disclosure of any water testing results conducted of the water leaking into the J-rod annulus.

[360] Dr. Revie was retained by the Insurer in 2017 to review the Mancey Report and the Kirk Rule 53 report dated October 14, 2014. He was asked by counsel for the Insurer to “assess the relative strengths and weaknesses of the reports and “if possible my opinion on which of the two experts in correct/closer to the truth”.

[361] At the opening of the trial in March 2019, Mr. Brock sought an order to disallow Dr. Revie’s evidence for a variety of reasons. I have written separate reasons in an endorsement on that motion and need not repeat them here. I allowed Dr. Revie’s reports and evidence to be admitted into evidence, with the question of the weight to be attributed to Dr. Revie’s evidence to be determined at trial.

Dr. Kirk’s Explanation of the Pitting Corrosion Process

[362] I accept the evidence of Dr. Kirk as credible, and logical.

[363] Dr. Kirk was fair in the answers he gave. Even though he was clearly testifying for the Plaintiffs, his evidence makes sense, and accords with the undisputed facts and he conceded the obvious in cross-examination. He testified as a scientist, and was a fair and credible witness. He did not exaggerate and answered questions fairly in accordance with the obligations of an expert to be impartial.

[364] I found Dr. Kirk’s viva voce evidence to be helpful in particular in explaining how the microscopic pitting corrosion process begins and continues on its own as “auto corrosion”. As Dr. Kirk explained:

• Pitting corrosion occurs when a corrodent interacts with the protective layer on the aluminum and weakens it at the atomic level. With the layer weakened, corrosion occurs faster and creates a hole or “pit” in the surface. This occurs at a microscopic level. Pitting corrosion does not require nitrates; it only requires what he calls a corrodent.[^91]
• If a corrodent is present, it keeps the oxygen barrier weak, which allows for the corrosion process to continue unaided. This process is “auto-catalysis” or “auto-corrosion”.[^92]
• “Nitric acid does not cause selective corrosion or pitting, as it’s better called.”[^93]
• Nitric acid attack is quite different in that it is a surface-only attack and is not a pitting agent.[^94]
• “Nitric acid is irrelevant for the pitting process.”[^95]

[365] I accept the evidence of Dr. Kirk explaining the pitting corrosion mechanism at J-41 in the context of the facts and circumstances in this case.

[366] It is not necessary to review all of the points of divergence between Dr. Kirk and Dr. Revie, as I accept the evidence of the court-appointed expert, who confirms the opinions shared by Dr. Kirk and Dr. Newman, and amplifies the explanation for what occurred at J-41. All the points of divergence were not canvassed by counsel either during the trial or in argument.

[367] Dr. Frankel confirms Dr. Kirk’s explanation of the process of pitting corrosion.[^96]

[368] Dr. Kirk was bluntly critical of the Mancey Report in his evidence at trial, suggesting that it was more of an argument advancing the nitric acid theory, rather than a root cause analysis to accurately using the scientific method to establish the cause of the leak at J-41 as required by the CNSC.[^97]

[369] Dr. Frankel politely agreed that the Mancey Report did not meet the rigors of a root cause report. I agree with this criticism.

Dr. Revie’s Reports and Evidence

[370] I agree with the Plaintiffs’ submission that Dr. Revie’s report is more of a summary of positions than an independent analysis. He confirms the opinions of Dr. Mancey.

[371] I conclude that his written reports do not add substance to the debate as to the cause of the leak at J-41.

[372] The same concern applies and is amplified as to the weight to attach to his viva voce evidence.

[373] Dr. Revie’s oral evidence was not helpful to the Insurer’s case. As Mr. Brock politely put it, Dr. Revie was a “difficult” witness. Mr. Liblong referred sparingly, if at all, to the evidence of Dr. Revie in his closing argument. Dr. Revie was dogmatic, argumentative, and would not answer questions asked directly.

[374] Dr. Revie was quite aggressive when Mr. Brock pointed out to him that he did not have all the relevant material when he prepared his first report, and that additional information may be relevant or helpful in forming his opinion.

[375] Dr. Revie would not concede the obvious, and would not change his position tenaciously held even if the facts underpinning that opinion changed. It was difficult to get Dr. Revie to agree to anything that did not accord with his opinion. Getting him to agree that having the water testing results would be important in a corrosion analysis was like pulling teeth.

[376] Even when a statement from his report was put to him in cross-examination Dr. Revie would not agree to it unless counsel pointed out to him that it was in fact his statement, and where it was in his Report.

[377] What was somewhat extraordinary was the emotional commitment Dr. Revie had to his position. When challenged in cross-examination on dry academic questions, Dr. Revie became quite emotional and aggressive. He was clearly an advocate supporting the Insurer position and did not seem to understand his duty to the court to be impartial.

[378] I conclude that, quite apart from his lack of objectivity, the contents of Dr. Revie’s reports and evidence were not convincing and did not include any independent analysis. He did not have the complete record available to him as compared with the other experts, and importantly had not read the independent peer review reports of Dr. Mancey’s Interim Report when he authored his first report. He did not have the benefit of the complete file before committing to his opinion, very tenaciously held. I conclude his evidence has no weight as it is not independent and is partisan.

[379] For these reasons, I do not accept Dr. Revie’s conclusions supporting the findings in the Mancey Report.

Dr. Frankel’s Reports and Evidence

[380] Dr. Frankel’s reports are detailed, comprehensive and technical. I accept his unbiased, independent evidence, both in his reports and in his viva voce evidence.

[381] Dr. Frankel was very complimentary of the detailed scope in the Mancey Report about the historical explanation of the functioning of the NRU up until the time of the unexpected leak.

[382] However, Dr. Frankel pointed out that the Mancey Report was not subject to peer review, as had taken place for the Interim Report. Dr. Frankel was of the view that the Mancey Report was not a root cause report. He was of the opinion that there were flaws in the procedure used to determine the cause of the leak, and that these flaws may have been avoided had the report been subject to external peer review.

[383] After the calandria was drained of heavy water, AECL cut a 3.4 cm circular coupon in the wall of the calandria near J-41 and photos were taken of the leak site. No coupon was taken at the leak site. Dr. Frankel was concerned that the failure to cut through the wall and examine the leak site itself was a failure in the analytical process.

[384] He stated in his evidence that the Mancey Report focused on the nitric acid attack and not on the cause of the localized corrosion. As such, the Mancey Report failed to provide a root cause analysis, a failure which may have been caught had the final report been subjected to peer review.[^98]

[385] I accept Dr. Frankel’s criticism of the Mancey Report that was confirmed in the evidence before me.

[386] It appears that there was a great deal of pressure to re-open the NRU. The background history in the Mancey Report was detailed and complete, but the Mancey Report itself as to the cause of the leak was not a root cause analysis looking at various potential theories for the leak and testing them using the scientific method. The CNSC and the public were waiting for a root cause report.

[387] I wish to be clear that I am not blaming Dr. Mancey. He was part of a team. It appears that Dr. Mancey may have been placed in a difficult position under significant pressure to finalize a report without adequate support, and without a peer review.

Dr. Frankel’s Conclusions

[388] Dr. Frankel’s evidence is not disputed by either counsel. During argument, to my recollection, neither counsel challenged any aspect of Dr. Frankel’s reports or his viva voce evidence.

[389] Dr. Frankel produced two reports and testified via Skype from Ohio. I accept the evidence of Dr. Frankel in his reports and his viva voce evidence.

[390] Dr. Frankel supports the theory of causation advanced by Dr. Kirk.[^99] As the coupon was not cut from the leak site and analyzed, any conclusions reached by any of the experts are necessarily inferential, although I note that the coupon was taken close to J-41, and the inferences drawn are strongly supported by the evidence, though not proved to scientific certainty of 100%. Based upon the available evidence, Dr. Frankel concludes that in all probability pitting corrosion precipitated by an aggressive agent, such as chlorine, created the pit that grew by a standard pitting corrosion mechanism. This pit caused the perforation and leak in the calandria wall at J-41.

[391] Dr. Frankel also confirms that his views align with the amended report of Dr. Newman dated August 29, 2009 commenting on the Interim Mancey Report prepared in 2009.

[392] In his reports and in his evidence Dr. Frankel explains the environment in the J-rod annulus that he opines caused the pitting corrosion and the leak in the calandria wall at J-41. I have summarized his explanation as simply as I can so those who are not corrosion experts or familiar with nuclear reactors may understand what took place:

• The drains from the J-rod annulus were blocked with voluminous corrosive residue, aluminum hydroxide, so that the water leaking from the reflector to the annulus was not being properly drained from the J-rod annulus. This blockage contributed to the environment in the J-rod annulus.
• It is possible that there were nitrate ions in that corrosion product that was blocking the drains as well as the voluminous aluminum hydroxide.[^100]
• The aggressive environment in the J-rod annulus was in the region of where the water that had leaked from the reflector interfaced with the carbon dioxide and air in the J-rod annulus. “Ready access of the top of the water to these species resulted in a form of waterline attack, which promoted corrosion below the surface.”[^101]
• The oxidizing agents in the aggressive environment were nitric acid and carbonic acid. Nitrogen oxides formed as a result of irradiation of the air in the J-rod annulus, which in turn dissolved in water to form nitric acid. Carbon dioxide similarly dissolved in the water to form carbonic acid.
• The bottom of the J-rod annulus was filled with leaked water from the reflector and was a stagnant environment. There was very little corrosive attack at the lowest regions of the calandria wall as there were no oxidizing agents and acidity at that location.
• Below the aggressive environment where the water was in contact with air and carbon dioxide nitric and carbonic acid formed. Below this interface there was the one-centimetre band circling the J-rod annulus and the outer wall of the calandria. This band toward the base of the calandria was where J-41 and the other pitting corrosion sites were located as well as the damaged scalloped regions.
• Dr. Frankel confirmed the opinion of Dr. Kirk that there were different corrosion mechanisms at play creating the scalloped regions by the formation of electrochemical cells where the aluminum of the calandria wall was in contact with other aggressive metal agents including probably iron or copper, as well as the pitting corrosion precipitated typically by chloride.
• Dr. Frankel was of the opinion that the scalloped regions were fragile, and would have resulted in a leak through the calandria wall at some point in time had the pitting corrosion not precipitated the first leak.
• The pitting corrosion occurred just below the scalloped regions. Dr. Frankel opines that “pitting corrosion was promoted where the oxidizing driving force of the waterline [aggressive environment] overlapped with the passivity promoted by the environment at the bottom of the gutter”[^102] “at the interface where the oxidizing power of the surface overlapped with the high pH in the restricted crevice, pitting corrosion was promoted.”[^103]
• Dr. Frankel confirms based upon his observations that the location of the leak and other damaged areas where repairs took place “were pits formed by localized corrosion or the standard pitting corrosion.”[^104]
• He further confirms that for pitting corrosion there needs to be an “aggressive agent, a pitting agent you might say, and it's -- typically, that would be chloride.”[^105]. The aggressive agent initiates the breakdown of the aluminum.
• Once pitting corrosion begins it is a very aggressive “auto catalytic” environment: once the pitting corrosion begins it grow and continues to grow.[^106]
• He confirmed that the actual pitting mechanism does not require nitric acid.[^107]
• The pit at J-41 that caused the leak was not alone. It is not disputed that 10 sites of pitting corrosion needed to be repaired: nine of the sites had not yet penetrated the calandria wall.
• When asked whether this type of pitting corrosion could have taken place in the absence of nitric acid, Dr. Frankel responded:

A. That nitric acid was not a causal factor, right, and what role did nitric acid play. So was nitric acid needed? And your question then, Mr. Liblong, was, would this have happened in the absence of nitric acid?

Q. Yes.

A. And I believe that's an interesting question. I think it's possible that this type of pitting could have happened in the absence of nitric acid. I don't know. I can't state with 100 percent certainty that it would, but I do believe that the nitric acid played a role in the process that took place in this reactor. Maybe it could have happened without the nitric acid, but it was promoted in fact in this by the reactions that took place, you know, above it, let's say, that were driven by nitric acid corrosion. So I think -- does that answer the question that you asked?

Q. I think so. I wasn't so much asking could this happen if there was no nitric acid in there. I guess my question was more: Do you believe the nitric acid played a role in the formation of what you call the scalloping and the pitting corrosion?

A. Yes, I believe the nitric acid played a role in what happens there.

Q. Thank you.

A. Because of this interaction of the different forms of corrosion. I don't believe that you can view them totally independently.[^108]

Q. And the actual pitting mechanism itself does not require nitric acid; is that correct?

A. That is correct.[^109]

[393] I accept the reports and viva voce evidence of Dr. Frankel as credible and reliable evidence in this case. He supports the conclusions of Dr. Kirk.

[394] It appears from the cross-examinations and submissions made, that neither counsel contests the theories and opinions advanced by Dr. Frankel.

Further Studies at AECL

[395] Dr. Kirk was critical of the AECL experiments after the release of the Mancey Report, as it did not replicate the environment in the J-rod annulus where the pitting corrosion and through wall penetration took place. He was also critical of submitting bent aluminum to the nitric acid, as in his view that added an additional stress not present in the environment at J-41. Dr. Kirk was of the view that the additional disclosure did not change his opinion that classic pitting corrosion precipitated probably by chlorine caused the leak at J-41.

[396] Dr. Revie was of the view that these additional disclosures, and particularly the experiment was “great” evidence that nitric acid caused the corrosion at J-41.

[397] Dr. Frankel was in agreement with Dr. Kirk that the environment tested in the subsequent experiments did not replicate the environment at J-41 with the presence of water where the leak took place, and therefore the results of the experiment did not confirm that nitric acid caused the corrosion at J-41. However, Dr. Frankel was not of the opinion that the bending of the aluminum being tested in the experiment caused a different stress undermining the conclusions reached. The bending of the aluminum was in his view irrelevant.

[398] I accept the criticisms of Dr. Kirk and Dr. Frankel about the studies and the conclusions reached. The new disclosure did not change the opinions of either Dr. Kirk or Dr. Frankel that the cause of the corrosion at J-41 was pitting corrosion precipitated by an aggressive agent, probably chloride.

Conclusions on the Expert Evidence

[399] I accept the evidence of Dr. Kirk and Dr. Frankel and Dr. Newman that pitting corrosion caused the leak at J-41.

[400] There was no reliable, tested evidence before me about how long the process would take for the pitting corrosion to take place. Once pitting corrosion commences it is very aggressive in an “auto catalytic” environment. Any reference to the length of time of the pitting corrosion in the transcripts of the CNSC hearing on June 11, 2009 is not reliable tested evidence.[^110]

[401] I do not accept any evidence of Mr. Pilkington suggesting that the localized corrosion had begun as early as 1999.[^111] There was no credible evidence as to how long it took for the second type of corrosion to penetrate the calandria wall at J-41. Mr. Pilkington’s suggestion that he had seen a photo dated 1999 confirming a second kind of corrosion is not reliable evidence as the evidence is clear that it was not possible to photograph the calandria wall. That photograph was never produced. It is not disputed that general corrosion was known to be taking place on the outer calandria wall for years.

[402] It is not contested that when the leak occurred at J-41 and thereafter, the AECL staff did not know the cause of the leak of heavy water from the calandria to the J-rod annulus, did not understand how and why it had happened, and certainly had not expected the disturbing finding of a through-wall penetration of the calandria wall resulting in heavy water containing radioactive Tritium flowing from the calandria into the J-rod annulus.

[403] As I have fully outlined in my finding of fact and my review of the expert evidence, I conclude that the corrosion at J-41 in the J-rod annulus was caused by classic pitting corrosion precipitated by the presence of a contaminant/aggressive agent, probably chlorine interacting with and attacking the aluminum at the microscopic level. Contrary to the conclusions of Dr. Mancey, I find that the pitting corrosion was not caused by nitric acid.

[404] I accept the evidence of Dr. Frankel, Dr. Kirk and Dr. Newman that pitting corrosion caused the leak at J-41. I do not accept the evidence of Dr. Mancey or Dr. Revie that nitric acid was the cause of the localized corrosion. I accept the evidence of Drs. Kirk, Frankel and Newman as to the mechanism of the fortuitous pitting corrosion. Nitric acid is not required for pitting corrosion and did not cause the corrosion at J-41. The pitting corrosion was precipitated by the presence of an aggressive agent, probably chlorine in the water in the J-rod annulus that leaked from the reflector.

[405] I accept the evidence of Dr. Frankel that nitric acid was part of the environment in the J-rod annulus where the pitting corrosion took place and played a role as part of that environment in what took place, although I accept the evidence of Dr. Frankel the pitting corrosion may or may not have taken place had nitric acid not been present.

Question 4. Considering the factual context and the reasonable expectation of the parties, and my conclusions on the expert evidence has the Insurer met its onus to prove that the corrosion that caused the leak at J-41 is excluded under the Policy?

Factual Context and the Reasonable Expectation of the Parties: the Insurer Has Not Met its Onus

[406] I reiterate the test of the reasonable expectation of the parties enunciated in Ledcor.[^112] The interpretation of the Policy should be consistent with the reasonable expectation of the parties, guided by the words of the Policy. Applying this test should not give rise to unrealistic results and should accord with the contemplated commercial atmosphere when the contract was entered into, and should be consistent with other similar insurance policies.

[407] The Denial Letter confirms that “the process [of corrosion] was one which AECL ought to have expected” and took the position that non-fortuitous events were not recoverable in an all-risk policy.

[408] Both Mr. Lodge and Mr. Turcotte acknowledged that the localized corrosion was fortuitous. They also acknowledged that not all corrosion is covered by the exclusion.

[409] Mr. Turcotte confirmed that at the time of the Denial Letter, FM did not know the cause of the corrosion. He acknowledged that the cause of corrosion does matter as some causes could trigger coverage notwithstanding the exclusion clause.

[410] Mr. Lodge confirmed that the Insurer did not retain any expert on corrosion to assess the cause of the localized corrosion

[411] With respect to the precise cause of the localized corrosion the representatives of FM confirmed for coverage to be available that they were looking for:

• physical damage that resulted in the corrosion, such as a flood to trigger coverage.
• “a physical event that triggered the corrosion”.
• “looking for metaphysical damage that would cause the corrosion” or
• “looking for an event or physical damage leading to the corrosion”.

[412] Prior to the litigation Mr. Turcotte testified that FM had no evidence that the localized corrosion was an insured peril triggering coverage. Mr. Turcotte testified that had they received such information it “could potentially change their coverage position.”

[413] For the reasons outlined in detail in this decision canvassing the expert evidence, I conclude that the unanticipated pitting corrosion was caused by the presence of an aggressive agent, probably chlorine, that was not supposed to be in the water that leaked from the reflector to the J-rod annulus.

[414] It appears using the words of the representatives of the Insurer, that the presence of an unanticipated and unpredicted aggressive agent such as chlorine in the leaking water causing the corrosion to the wall of the J-rod annulus would qualify using their logic and approach as a “physical trigger” or “metaphysical damage”, although on the microscopic level. Using their reasoning, the unanticipated presence of chlorine constituting the aggressive agent causing the self-perpetuating pitting would constitute an “event or physical damage leading to the corrosion”.

[415] I conclude considering the reasonable objective expectation of the parties in light of the factual context of this case, that coverage should be available for losses caused by the leak of heavy water at J-41 resulting in a shutdown of the NRU. The cause of the fortuitous corrosion was triggered by an unanticipated and unpredicted microscopic attack of the calandria wall due to the presence of an aggressive agent, or contaminant, probably chlorine, that caused a through wall leak. I refer to my findings of fact and conclusions as to the quality of the water leaking from the reflector to the J-rod annulus confirming the probable presence of chlorine as well as other contaminants [para 150 to 162].

[416] I have defined corrosion in the Policy as “the anticipated and predictable process of corroding, esp. of a rusting metal.” “Corrode” is defined as “wear away, esp. by chemical action.”

[417] Applying this definition to the facts of this case, I conclude that the Insurer has not met the burden of proof that the corrosion exclusion applies to this fortuitous, unanticipated and unpredictable corrosion found at J-41. Hence, I conclude that the corrosion exclusion does not apply.

Rules of Construction

[418] If the conclusion that the meaning of the corrosion exclusion is still in issue after considering the expert and other evidence and the reasonable expectation of the parties, the Plaintiffs rely on certain rules of construction to assist in determining its meaning.

[419] Given my conclusions, the rules of construction are not necessary to my analysis. As they do support the Plaintiffs’ arguments, I will outline the principles that in my view apply.

[420] The Plaintiffs rely on ejusdem generis, noscitur a sociis, and contra preferendum in support of their arguments.

[421] The corrosion exclusion provides:

C. This Policy excludes the following, but, if physical damage not excluded by this Policy results, then only that resulting damage is insured: …

3. deterioration, depletion, rust, corrosion or erosion, wear and tear, inherent vice or latent defect.

[422] Ejusdem generis is “a general rule of construction … that, where a particular class is spoken of, and general words follow, the class first mentioned is to be taken as the most comprehensive and the general words treated as referring to matters ejusdem generis with such class,”: Lunenburg Industrial Foundry v. Commercial Union Assurance Co. of Canada, at para. 51.[^113]

[423] Ejusdem generis can only be employed if “(i) the specific words have some common characteristic which constitute them as a genus, (ii) the specific words do not exhaust a whole genus; or (iii) there is no clear indication that the opposite was intended by the parties”: Lunenburg Industrial, at para. 51. It appears that this rule of construction does not apply as a review of the Policy confirms that “corrosion” is not “general words” that follow a more specific “class of words”.

[424] Noscitur a sociis is defined in Black’s Law Dictionary as “a canon of construction holding that the meaning of an unclear word or phrase, especially one in a list, should be determined by the words immediately surrounding it” (see also Opitz v. Wrzesnewskyj, at para. 40[^114]).

[425] Myers J., in PCL Constructors considered the meaning of corrosion in obiter comments whether the ejusdem generis rule of construction might “limit the meaning of ‘corrosion’ to only ‘naturally occurring corrosion’ rather than corrosion brought on by untimely exposure to corrosive liquid as occurred in this case” given the wording in that policy which excluded “Loss or damage caused directly or indirectly by rust or corrosion, frost or freezing, pollution or contamination unless caused directly by a peril not otherwise excluded herein.”[^115]

[426] The Insurer argues that these rules of construction do not apply. Counsel points out that the clause in question includes reference to inherent vice or latent defect, which have nothing to do with gradual anticipated wearing away.

[427] The Plaintiffs argue that rules of construction interpreting insurance policies add flesh to the objective reasonable expectations of the parties as to the meaning of the corrosion exclusion. They say it supports the view that the corrosion exclusion as defined in the Policy is intended to mean the non-fortuitous corrosion that is anticipated, normal wearing away over time when this word is considered in the context of the Policy and its immediate neighbors in the exclusion clause.

[428] The last two items in the list, inherent vice or latent defect, are outsiders. In the list of exclusions, the words immediately surrounding the word corrosion, that is rust and erosion, fall in the category of normal expected deterioration over time.

[429] The presence of “corrosion” in a list of commonly occurring perils – such perils either being recognized as being inherent to the property itself or generally expected to occur as a by-product of the property’s normal function lends weight to the Plaintiffs’ contention that the word corrosion is meant to exclude only non-fortuitous perils.

[430] Although not necessary for my analysis, it appears that the rule of construction of noscitur a sociis supports the position of the Plaintiffs’ in their argument that the corrosion exclusion applies to non-fortuitous wear and tear, not fortuitous, unanticipated corrosion. It appears that the rule of construction ejusdem generis does not apply.

Contra Proferentem

[431] Finally, contra proferentem confirms “whoever holds the pen creates the ambiguity and must live with the consequences.”: Gibbens v. Co-Operators Life Insurance Co. at para. 25.[^116]

[432] Contra proferentem was succinctly described by the Supreme Court of Canada in Consolidated-Bathurst v. Mutual Boiler: “If there is any doubt as to the meaning and scope of the excluding or limiting term, the ambiguity will be resolved against the party who has inserted it and who is now relying on it.”[^117]

[433] In Indemnity Insurance Co. of North America v. Excel Cleaning Service, the Supreme Court of Canada described contra proferentem as:

a general rule to construe the language used in a manner favourable to the insured. The basis for such being that the insurer, by such clauses, seeks to impose exceptions and limitations to the coverage he has already described and, therefore, should use language that clearly expresses the extent and scope of these exceptions and limitations and, in so far as he fails to do so, the language of the coverage should obtain.[^118]

[434] If after considering the evidence of context, the reasonable expectation of the parties, and the other rules of construction outlined above leaves doubt as to meaning of the corrosion exclusion [which is not my finding] then the principle of contra proferentem applies, such that any ambiguity in the meaning of the corrosion exclusion should be interpreted contrary to the Insurer who drafted the Policy, and in favour of the Insured.

[435] For all these reasons I conclude that the Insurer has not met its onus of proof that the corrosion exclusion applies.

PART IV: ISSUE 2: ALTERNATIVELY, DOES THE RESULTING PHYSICAL DAMAGE EXEMPTION APPLY ALLOWING COVERAGE??

[436] If another court concludes that the corrosion exclusion does apply [which is not my conclusion] I will consider whether the exception to the exclusion for resulting physical damage applies, allowing the Plaintiffs to recover losses notwithstanding the application of the corrosion exclusion.

[437] The Policy states:

C. This Policy excludes the following, but, if physical damage not excluded by this Policy results, then only that resulting damage is insured: …

3. deterioration, depletion, rust, corrosion or erosion, wear and tear, inherent vice or latent defect. [Emphasis added.]

[438] What is the proper interpretation of resulting physical damage in the context of this case?

Principles When Considering an Exception to an Exclusion

[439] The insured bears the onus of proving that an exception to an exclusion clause applies: Ledcor, at para. 52; Progressive Homes, at paras. 26-29, 51.

[440] Exceptions do not serve to create coverage; they only bring an otherwise excluded claim back within coverage where the claim initially fell within coverage: Progressive Homes, at para. 28.

[441] Exceptions to exclusion clauses should be interpreted broadly: Monk v. Farmers and Muskoka Ins., at para. 132.[^119]

The Meaning of Resulting Physical Damage in this Policy and Position of the Parties

[442] Physical damage is not defined in the Policy.

[443] There are conflicting lines of cases on how to interpret the meaning of resulting physical damage. There is no definitive decision defining the meaning of resulting physical damage in all-risks policies in Canada. The meaning of the term is considered in light of the specific wording in each policy as well as legal and factual context.[^120]

[444] Should resulting physical damage be defined narrowly to require actual physical damage, or should it be defined broadly to include loss of use?

[445] The question is:

To qualify as resulting physical damage under the Policy does there have to be actual tangible damage to the J-rod annulus caused by the leak of heavy water? [the Insurer’s position] or

Does resulting physical damage in this Policy include loss of use of the NRU as a consequence of the leak of heavy water? [the Plaintiffs’ position]

[446] Neither Black’s Law Dictionary nor Halsbury’s Laws of Canada define “resulting damage” or “physical damage”. Nor do they define “resulting physical damage” or “resultant physical damage”. There are no Canadian textbooks that define “physical damage”, “resultant physical damage” or “resulting physical damage”.

[447] The Insurer takes a narrow approach in applying this exception. FM argues that the presence of the leaking heavy water in the J-rod annulus did not cause actual tangible damage in the interior of the J-rod annulus. This fact is not disputed. The Insurer argues in these circumstances that the presence of the leaking heavy water is not resulting physical damage, and hence the exception to the corrosion exclusion does not apply.

[448] The Insurer further suggests that the NRU could have continued to operate had there not been a through-wall leak. For reasons I will outline, I do not accept this argument.

[449] The Insurer argues that loss of use is not covered in resulting physical damage.

[450] The Plaintiffs argue that the presence of heavy water containing Tritium in the J-rod annulus constitutes resulting physical damage under the Policy. The character of the J-rod annulus was changed in an important way by the heavy water leak, requiring a shutdown and rendering the entire NRU unusable. The Plaintiffs argue that it is not necessary to have tangible damage to the J-rod annulus itself to qualify for the exception to the corrosion exclusion.

[451] The through-wall leak of heavy water at J-41 caused the NRU to be shut down until the problem of the leaking water and safety concerns relating to that leak were addressed to the satisfaction of the CNSC. A lengthy, convoluted chain of events followed the discovery of the leak of heavy water to safely rectify the leak, and these various steps are included under the umbrella of resulting physical damage.

Relevant Provisions in the Policy

[452] It is important to look at the other provisions in the Policy to assist in interpreting the intended meaning of the resulting physical damage exception.

[453] A review of the Loss of Use Exclusion, as well as the terms of coverage of the Contingent Time Element and the wording of the Period of Liability appear to confirm the Plaintiffs’ position that resulting physical damage is intended to include loss of use of the NRU in the context of the Contingent Time Element Coverage.

[454] I note there is a general loss of use exclusion in the Policy which appears in the same section as all other exclusions. It specifically excludes interruption of business or loss of use except to the extent provided under the Policy. The section provides:

5. Exclusions

A. The following exclusions apply unless specifically stated elsewhere in this Policy:

1. indirect or remote loss or damage.

2. interruption of business, except to the extent provided by this Policy.

3. loss of market or loss of use except to the extent provided by this Policy.

[Emphasis added]

[455] MDS purchased the additional Time Element Coverage to provide coverage if there was an interruption in the business of a supplier affecting the profits of MDS. This coverage applied to the supply of isotopes from AECL produced at the NRU.

[456] Section C of the Policy covers Time Element losses “directly resulting from physical loss or damage of the type insured by this Policy”.

[457] Time Element losses are limited generally by conditions found in section C(1), e.g., the loss has to occur during the Periods of Liability described in section C(1). Section C(2) lists specific forms of Time Element losses, such as gross earnings loss, extra expenses during certain conditions loss, and lease interest loss.

[458] The parties agreed that if any section of the Policy applied, it would be C(3)(B). That section of the Policy states:

This Policy covers the Actual Loss Sustained and EXTRA EXPENSE incurred by the Insured during the PERIOD OF LIABILITY directly resulting from physical loss or damage of the type insured to property of the type insured at Contingent Time Element Locations located within the TERRITORY of this Policy.

[459] The wording of the period of liability for the Time Element Coverage Extension as defined in the Policy is helpful in clarifying whether physical damage at the NRU was intended to cover loss of use of the NRU facility.

4. The PERIOD OF LIABILITY for this TIME ELEMENT COVERAGE EXTENSION will be:

The period of time:

Starting at the time of physical loss or damage causing the disruption of the Normal movement of goods or materials between the Insured Locations; or directly between the Insured Location and the Location of the direct customer, supplier, contract manufacturer or contract service provider to the insured, …

[460] The wording of the Period of Liability for the Time Element Coverage “starting at the time of physical loss or damage causing the disruption of the normal movement of materials” supports the argument of the Plaintiffs that the disruption in the flow of isotopes from loss of use of the NRU is intended to be included in the resulting physical loss.

[461] When the leak of heavy water with Tritium emissions was detected, the NRU was shut down until the source of the leak was identified and repaired. As the location of the leak was unexpected, the complex process to repair the leak and meet the safety protocol of the CNSC took 15 months. The physical loss or damage of the leak of heavy water triggered the shutdown, and caused “the disruption of the normal movement” of the supply of isotopes to MDS until the NRU was reopened in August 2010 as approved by the CNSC.

[462] When the resulting physical damage exception is considered in light of other provisions in the Policy, including Period of Liability for the Time Element Coverage, it appears clear that the meaning of resulting physical damage contemplates loss of use of the NRU. Therefore the Policy is not ambiguous. This would end the analysis on the applicability of the resulting physical damage exemption. The Plaintiffs’ loss flowing from the disruption of the normal supply of isotopes would be covered under the Policy.

[463] I will however consider the other arguments raised by the Insurer.

Caselaw interpreting the meaning of resulting physical damage

[464] There are two very different interpretations of the meaning of physical damage in the various cases, reflecting two approaches which suggests that the meaning of the term “resulting physical damage” is ambiguous.

• Does resulting physical damage require actual, tangible corporal damage to the J-rod annulus where the leaking water was present? (Insurer’s position)
• Alternatively, does the leak of the heavy water causing the shutdown of the NRU constitute resulting physical damage as the leaking water impaired the use or function of the NRU as a whole? (Plaintiffs’ position)

[465] There are few Canadian cases that specifically discuss the interpretation of physical loss or physical damage in the context of an all-risks policy.

[466] From these cases, and a review of caselaw in other jurisdictions, two approaches to the interpretation of physical damage emerge: one the narrow view that limits physical damage to corporeal, tangible damage;[^121] and the other a broader interpretation that encompasses not only tangible damage but also impairment of use or function.[^122]

[467] The Insurer relies upon two cases in support of the argument that there is a requirement that there must be tangible physical damage to qualify for the resulting physical damage exception.[^123]

[468] The Plaintiffs rely upon other cases that counsel argues supports a broader interpretation of resulting physical damage involving loss of use or function.[^124]

[469] A review of the cases confirms that there is no bright line or a single case that is dispositive of the meaning of resulting physical damage for this case. There is a range of factual circumstances and a variety of policy language present in the various cases submitted by counsel that limit the utility of a specific case in considering the unique facts of this case.

[470] The Insurer argues that some of the cases relied upon by the Plaintiffs are liability insurance cases, and are therefore not applicable to the interpretation of all-risks policies as this approach neglects a fundamental difference between the purpose of liability insurance and all-risks property insurance.

[471] I note that only one of these cases relied upon by the Plaintiffs, Canadian Equipment, involves the interpretation of a liability insurance policy, defining “property damage”, as opposed to an all-risks property insurance policy, defining “physical damage”. Notwithstanding the arguable differences between liability insurance and all-risks property insurance, the property damage cases can provide assistance to the extent that the analysis focuses on the interpretation of “physical damage”, as the operative phrase that must be interpreted in this case.

[472] The only case that squarely considers the meaning of physical damage in the context of an all-risks policy in a similar factual context to this case is Jessy’s Pizza: a Small Claims Court decision in Nova Scotia.

[473] In Jessy’s Pizza, a vandalized oil line caused an oil spill under the plaintiffs’ business. This in turn caused an overpowering smell of oil in the store. The plaintiffs tried to mitigate this problem by leaving open windows and doors, to no avail. They were forced to close. The floor of the restaurant was removed and the contaminated soil beneath was dug out.

[474] Adjudicator Johnston looked at the facts in a pragmatic fashion confirming that the physical damage coverage applied to both the tangible direct damage to the floor, as well as to the fumes in the restaurant as “the fumes physically damaged the contents of the business” as the continued operation of the restaurant business was untenable.

Facts Relevant to Resulting Physical Damage

[475] I outline my findings of fact relevant to the context and the meaning of resulting physical damage. There is inevitably some overlap in facts already canvassed relevant to other arguments raised by the Insurer.

[476] The Insurer relies upon the affidavit of Mr. Pilkington and his evidence that “The presence of heavy water in the J-Rod annulus did not present any risk of physical damage to any of the components of the NRU situated within the J-rod annulus as the J-Rod annulus was designed to accommodate any heavy and/or light water that might leak or be discharged into the J-Rod annulus.” Mr. Pilkington’s affidavit also indicated that “the presence of heavy water in the J-Rod annulus did not itself prevent the NRU from returning to operation; it was merely an indication that an abnormal condition existed.”[^125]

[477] I note that Mr. Lodge, the FM adjuster who signed the Denial Letter testified that he did not understand that the NRU remained in a shutdown because of the leak of heavy water. He thought the reason for the shutdown was to address the corrosion issues, and that the “leak was minor” and “it was not an operational issue”.

[478] This evidence of both Mr. Pilkington and Mr. Lodge illustrates a fundamental misapprehension of the facts.

[479] The Insurer argues that the Plaintiffs have not presented any evidence that the NRU had to remain shut down to remove the heavy water leaking into the J-rod annulus which was automatically being collected by the sump system and stored in containers. Again, this is an inaccurate assertion of the facts.

[480] As Mr. Pilkington confirmed “Any leakage of heavy water is significant”.[^126]

[481] The unexpected leak and the Tritium levels triggered the mandatory reporting of this event to the CNSC as a safety issue. The CNSC required the NRU to be shut down until the source of the leak was identified. Once it was determined that the unexplained leak was through the wall of the calandria, the CNSC imposed a safety protocol that had to be met before the NRU could reopen. The through-wall leak of heavy water from the calandria into the J-rod annulus affected the operation of the NRU as a whole, requiring it to be shut down until the conditions imposed by CNSC were met, and permission was granted to renew operation.

[482] The leak was at the rate of 5 kg per hour. The heavy water leaking from the calandria to the J-rod annulus was drained and stored in tanks in perpetuity due to the presence of radioactive Tritium. Until the calandria was drained, a total of 3,368 kilograms of heavy water had leaked from the calandria within 28 days.[^127] The members of the CNSC expressed concern that radioactive emissions of Tritium at twice the actionable/reportable level were being emitted into the environment. Their concerns made it clear that this situation was serious.[^128]

[483] The suggestion of Mr. Pilkington in his affidavit or evidence that the NRU could have continued operation with the through wall leak of heavy water into the J-rod annulus is untrue.

[484] In the CNSC public hearings on June 11, 2009 Mr. Pilkington was specifically asked the question about whether the NRU could continue in operation, given the leak was small and there was a pressing public need for the medical isotopes for patient treatment. Mr. Pilkington confirmed that it would be unsafe to continue operations as they did not know the location of the leak:

“the issue is the knowledge of the nature of the leak. And so if we had knowledge that the leak was from a component that would not further degrade, if we had knowledge that it was from a component that we could isolate from the system, if we could collect the water efficiently at a high isotopic then it would be reasonable to at least consider continued operation. However where the leak is from the primary reactor vessel where we do not know the exact extent of the corrosion and therefore we do not know the potential for the leak to increase, in this case the safe and prudent thing to do is to shut down and do a complete inspection to identify the extent of that condition.”[^129] [Emphasis added]

[485] The leak site at J-41 was through the calandria wall, not at a component part. The AECL staff were understandably very concerned that the leak site at J-41 could become larger, again creating a potentially dangerous situation. It was for this reason the AECL staff decided the appropriate course of action was to safely drain the calandria and remove the fuel rods before investigating the leak site.

[486] The J-rod annulus is a space intended to be filled with carbon dioxide. There was some air introduced into the J-rod annulus, estimated to be 10%. It was not clear to me in the evidence whether this 10% air leaked into the J-rod annulus or whether it was part of the design. The design of the NRU and the J-rod annulus in particular did not include leakage of water, be it light water leaking from the reflector, or more importantly, heavy water containing radioactive Tritium leaking from the calandria.

[487] Had the NRU continued operating, the leaking of light water from the reflector at the rate of 1,000-plus liters a day would have continued. The light water would have been mixed with the leaking heavy water. There is no evidence before me that the sump pump design could handle this quantity of water to be stored in the drums in perpetuity to prevent Tritium contaminating the environment. As well, Tritium would continue to be released into the air in the environment at twice the actionable concentration that triggered the action rate to shut down the NRU.

[488] The relatively small leak of heavy water at J-41 in the damaged one-centimeter band was the tip of the iceberg, but its significant impact is not to be underestimated.

[489] It must be remembered that the purpose of the heavy water in the calandria is to perform a critical safety function: to cool the radioactive rods. These reach very high temperatures when the nuclear activity takes place. It is essential that the rods in the calandria remain covered with heavy water.

[490] Mr. Pilkington testified that there was a concern that, if the leak was coming from the calandria, it could grow to a level that was impossible to maintain, exposing the fuel rods.[^130] The cause for concern is apparent: as Mr. Pilkington testified, if the fuel rods overheat, they may melt and cause a nuclear meltdown.[^131]

[491] If the rate of leaking of the heavy water increased, there was the risk that the rods would not be adequately covered, with the potential for a serious nuclear event. Thankfully, this chain of events did not happen as the NRU was shut down by the AECL and confirmed by order of the CNSC.

[492] Mr. Pilkington’s affidavit and evidence about the absence of physical damage in the J-rod annulus reflects the position of the Insurer that as there was no actual tangible damage to the J-rod annulus, there was no resulting physical damage. I accept that the J-rod annulus, with its sump system, was not physically damaged by the leak of heavy water in the literal sense. That is not the end of the inquiry.

[493] A complex and lengthy process took place to rectify the problem of the leaking of heavy water to satisfy the CNSC that the NRU could be reopened safely.

[494] The CNSC gave permission for the NRU to begin operations in August 2010 after the repairs outlined above had been conducted, on the condition that the root cause report would be provided by AECL to the CNSC within ninety days. The Mancey Report was released on September 22, 2010. Dr. Mancey confirmed that the Mancey Report was not peer reviewed due to the time constraints imposed by the CNSC.

Application of Principles to this case

[495] The most important decision to assist in understanding the intended meaning of the resultant physical damage exclusion is to consider the approach outlined in the Supreme Court of Canada decision in Ledcor. The decision confirms that to determine appropriate meaning in light of ambiguity, the interpretation of the scope of a resulting damage exception will be impacted by the specific language of the policy at issue, as well as the relevant factual matrix including the reasonable expectation of the parties.[^132]

[496] Following the steps in Ledcor, it should first be noted that there is no direct evidence before the court as to the facts surrounding contract formation which would assist in discerning the parties’ reasonable expectations regarding the definition of resulting “physical damage”. The 83 page Policy was written by the Insurer and was not subject to any negotiations about wording.

[497] To inform the reasonable expectation of the parties I must consider the purpose of all-risks coverage, and in particular the purpose of all-risks coverage in the context of coverage for loss of profits in light of the facts of this case.

[498] The Supreme Court of Canada’s discussion of the intended purpose of all-risks policies in Ledcor is directly on point. The court considered a particular type of all-risks policy, “builders’ risk” insurance. However, the court cited a general passage from E.A. Dolden as to the purpose of all-risks policies applicable to this case:

“Urbanization and industrialization in the past 100 years have made the concept of an insurance policy covering all conceivable risks advantageous to both insureds and their insurers. The insured benefits from the extensive nature and scope of the coverage, and insurers benefit from the economies of managing and marketing a policy which, in terms of its scope, has certainty. For these reasons, the “all risk policy,” which creates a special type of coverage extending to many risks not customarily covered under other types of insurance policies, is attractive to both the insurance industry and consumers.”[^133] [Emphasis added.]

[499] The purpose of the Policy in this case is to insure the Plaintiffs for loss of profits in their business operations. It was known, and FM confirmed that the parties were actively discussing an increase of the Contingent Time Element coverage to $80,000,000 given the importance of the isotopes supplied by the NRU in the business operations of the Plaintiffs, constituting some 45% of MDS world income. The Plaintiffs were dependent on the continued operation of the NRU to conduct its business and to generate profits. The Plaintiffs could not mitigate by simply seeking another supplier of radioactive isotopes, as there was no other alternative supplier worldwide. FM was aware of this fact.

[500] As confirmed in Ledcor, all-risks insurance is purchased to provide peace of mind in case the unforeseen occurs.[^134] In this case the purpose of the all-risks Policy in terms of the operation of the NRU is to compensate the Plaintiffs for loss of profits if a fortuitous event interrupted the supply of the radioactive isotopes.

[501] The leak of heavy water at J-41 was the tip of the iceberg that led to the shutdown of the NRU until the CNSC approved the repairs and design of the NRU as fit for its purpose. The process was long and convoluted. Mr. Pilkington’s and Dr. Mancey’s evidence confirms the sequence of steps that had to be taken to fix the leak of heavy water from the calandria to the J-rod annulus. After investigating the leak and identifying the problem area in the calandria, the fuel rods had to be removed (taking one month), and then the heavy water was drained from the calandria. There were then non-destructive examinations performed to assess the extent of the damage. A coupon measuring 3.4 cm was cut in the calandria wall near J-41 and analyzed. In light of the findings, it had to be determined whether the NRU could be safely repaired. AECL considered three separate potential methods of repair for the calandria. AECL had special tools made for the repairs and AECL trained people to use the special tools. Mockups of the calandria and J-rod annulus were created so that the individuals could practice with the special tools before actually doing the repairs. The leak site was located some 30 feet from the top of the J-rod annulus through a sort of a narrow grate. Finally, the repairs were conducted with plates welded onto the calandria wall.[^135]

[502] Apart from J-41 there were nine other areas of pitting corrosion in the one-centimeter band circling the exterior wall of the calandria that needed to be repaired. As well in that band there was the presence of scalloping, precipitated probably by the presence of the aggressive agents of iron or copper in the water. Dr. Frankel confirmed that the scalloped areas were vulnerable to a through wall leak. Although the leak at J-41 was small, it was not minor and could have expanded if the NRU continued in operation. Other areas where pitting corrosion had begun could have resulted in other leaks.

[503] The leak of heavy water has a context. The possibility of meltdown if the leak increased is a unique and important fact. The NRU was prudently shut down to avoid the possibility of a nuclear accident.

[504] The definition of resulting physical damage in this factual matrix should not be limited to the occurrence of an actual nuclear event, which would qualify for the Insurer’s suggested definition of resulting physical damage.

[505] The interpretation suggested by the Insurer falls into what has been described in the Texas Supreme Court in U.S. Metals, Incorporated v. Liberty Mutual Group, Incorporated as a “perverse” result of insurance interpretation:

The result in this case has a perverse aspect to it. Had ExxonMobil been negligent or reckless—had it not tested the flanges, or had it found the defect but decided to risk the danger of leaks—and an explosion had resulted, U.S. Metals would not be denied coverage for the damages to persons and property for want of physical injury. But because ExxonMobil was careful and cautious, U.S. Metals is not entitled to indemnity for the costs of remedying the installation of the faulty flanges. Nevertheless, we think the text of the policy is clear.[^136]

[506] The Texas Supreme Court denied coverage notwithstanding its observation of perversity. This case is distinguishable. AECL was ordered by CNSC to shut down the NRU upon discovering that the integrity of the calandria was compromised in the context of a tightly regulated safety rules for nuclear reactors.

[507] A review of the US caselaw confirms generally a narrow view of physical damage. However, three US cases are of assistance in confirming a broader interpretation of the meaning of resulting physical damage in this Policy to include loss of use. The leak of heavy water from the calandria in this case rendered the NRU unusable, as reflected in the shutdown order by the public safety body of the CNSC.

[508] The first two cases deny coverage as, contrary to the clear facts in this case, the physical damage in those cases did not make the buildings in question uninhabitable, and there was no imminent risk of contamination.

[509] In Port Authority of New York and New Jersey v. Affiliated FM Ins. Co.[^137], the court confirmed “Physical damage” to property means a distinct, demonstrable, and physical alteration of its structure; therefore, under New York and New Jersey law, buildings with asbestos in components did not sustain “physical loss or damage” covered by property insurance policies, the asbestos was not in such form or quantity as to make the buildings unusable or uninhabitable, and nothing indicated an imminent threat of asbestos contamination. [Emphasis added]

[510] In Universal Image Productions, Inc. v. Chubb Corp.,[^138] the court confirmed that the insured claiming that water seepage caused a pervasive odor, mold, and bacterial contamination did not suffer a “direct physical loss” within the meaning of an all-risks policy. The court similarly noted that the insured did not suffer any structural or any other tangible damage to the insured property, and the odor was not so pervasive as to render the premises uninhabitable. [Emphasis added]

[511] These two cases are easily distinguishable from the facts of this case. In this case, the leak of heavy water required the shutdown of the NRU rendering it unusable. With continued operation there was a very real imminent risk of harm.

[512] The final case confirms a broad interpretation of direct physical loss, allowing coverage in a situation where a public body declared that the premises in question were uninhabitable.

[513] In Western Fire Ins. Co. v. First Presbyterian Church,[^139] the property was rendered uninhabitable due to the infiltration and saturation of gasoline vapours from an adjacent property. These circumstances, combined with a government declaration of uninhabitability, amounted to a direct physical loss allowing coverage. This case is helpful in applying the principles to the unique facts of this case.

[514] The order by the CNSC in this case required the continued shutdown of the NRU until its safety protocol was met. By analogy, this is the equivalent to a declaration of uninhabitability amounting to a direct physical loss in Western Fire Ins. Co. v. First Presbyterian Church until the safety conditions imposed by the public body for the building had been satisfied.

Conclusions on the Meaning of Resulting Physical Damage

[515] A broad interpretation of resulting physical damage to include impairment of use confirms the principle that exceptions to exclusions should be interpreted broadly.[^140]

[516] The Policy must be considered as a whole. The other provisions of the Policy including the loss of use exceptions, and the Contingent Time Coverage provisions, with specific regard to the period of liability when read with the resulting physical damage exception, all appear to confirm that loss of use of the NRU caused by the leak of heavy water would constitute resulting physical damage. The effect of the leak of heavy water rendered the NRU inoperable until the safety concerns and protocol imposed by the CNSC had been met.

[517] In assessing the objective reasonable expectation of the parties as to the meaning of physical damage, it makes common sense that if the unanticipated leak of heavy water from the calandria precipitates the shutdown of the NRU ordered by the CNSC to study and rectify the problem causing the leak, that this circumstance rendering the NRU inoperable would constitute resulting physical damage.

[518] Applying the principles of Ledcor to interpret the meaning of resulting physical damage, I conclude that a broad definition of resulting physical damage is appropriate in the factual context of this case to interpret the words in the Policy to include impairment of function or use of tangible property caused by the unexpected leak of heavy water.

[519] This interpretation is in accordance with the purpose of all-risks property insurance, which is to provide broad coverage. To interpret physical damage as suggested by the Insurer would deprive the Insured of a significant aspect of the coverage for which they contracted, leading to an unfair result contrary to the commercial purpose of broad all-risks coverage.

[520] For these reasons, I conclude in the facts of this case that the Plaintiffs have met their onus of proving that the leak of heavy water from the calandria to the J-rod annulus was resulting physical damage, and that if the corrosion exclusion applies, then the exception provisions of resulting physical damage apply allowing coverage for the Plaintiffs losses.

PART VI: ISSUE 3: DOES THE IDLE PERIOD EXCLUSION REDUCE THE PERIOD OF COVERAGE IF THE RESULTING PHYSICAL DAMAGE EXCEPTION APPLIES ?

[521] Further, and in the alternative, the Insurer argues that if the resulting physical damage exception applies, the period of the loss is limited by the idle period exclusion reducing the time frame available for the Plaintiffs’ claim for loss of profits.

What is meant by the “Idle Period” exclusion listed in the Time Element Exclusions clause 5.A?

[522] The Insurer argues that the resulting damage claim is subject to the idle period exclusion in the Policy limiting the time period for the resulting damage. Counsel makes two arguments:

• The Policy only covers the business loss damages for 24 hours, which was the time required to actually drain the calandria of heavy water, once the radioactive rods had been removed from the calandria, or alternatively,
• The Policy only covers the business loss damages for the period of time the heavy water causing resulting damage was present in the J-rod annulus: that is until July 22, 2009. At that point in time the cause of the leak had been identified, the rods had been removed from the calandria, and the heavy water had been drained from the calandria. Any period beyond this date was to repair the corrosion causing the damage, not the resulting physical damage of the water leak. The damage accrued during this period was agreed to be US$10,732,347.00.

[523] The idle period exclusion provides:

TIME ELEMENT EXCLUSIONS

In addition to the exclusions elsewhere in this Policy, the following exclusions apply to TIME ELEMENT loss:

This Policy does not insure against:

A. Any loss during any idle period, including but not limited to when production, operation, service or delivery or receipt of goods would cease, or would not have taken place or would have been prevented due to:

1. physical loss or damage not insured by this Policy on or off of the Insured Location.

2. planned or rescheduled shutdown.

3. strikes or work stoppage.

4. any other reason other than physical loss or damage insured by this Policy.

[524] I have concluded that the corrosion exclusion does not apply. If another court takes a different view of the corrosion, I have concluded that the leak of heavy water causing the shutdown of the NRU constituted resulting physical damage in accordance with the exception found in section 5(c) of the Policy.

Has the Insurer met its onus of demonstrating that what occurred in this case is excluded by the “idle period”?

[525] The Insurer relies upon the Court of Appeal for Ontario case of Algoma Steel Corp. v. Allendale Mutual Insurance Co. (1990), in support of the argument.[^141]

[526] In Algoma, the insured suffered losses when a steel furnace on its property leaked molten material which caused a series of explosions in its facility. The resulting damage related to the inability of an adjacent steel furnace to operate as the accident had damaged a crane runway that serviced the adjacent furnace. There were multiple steel furnaces. One of them leaked molten material damaging the crane runway which serviced another. The insured requested indemnification for economic damages under the business interruption loss provision in their insurance policy. At the time of the loss, the adjacent furnace was being relined and was not operating. This repair was independent and unrelated to the accident, and took place during the period of time it took the insured to fix the crane runway.[^142]

[527] Allowing the appeal, Blair J.A. held that the economic losses due to the damaged crane system were not recoverable due to the idle period exclusion. Even if the crane had been operational, the furnace could still not have produced steel as it was being relined concurrently in an unrelated repair during the relevant period of time: Algoma, at para. 19. Justice Blair concluded that the business losses suffered by the insured’s inability to produce steel from the furnace “[fit] squarely into the ‘idle period’ exclusion because production was prevented for a reason ‘other than direct damage to the said premises from a peril insured against.’”: Algoma, at para. 19.[^143]

[528] In this case it is not possible to separate the corrosion damage causing the leak of heavy water from the leak of heavy water containing Tritium. The two are inextricably intertwined. The leak was the immediate result of the unanticipated corrosion penetrating the wall between the calandria and the J-rod annulus. The leak started when the corrosion reached the point of through-wall penetration. The NRU was unusable because of the leak of heavy water contained Tritium at actionable levels. As well, allowing the leak to continue risked the leak site enlarging with potentially serious consequences. Finally, the CNSC had to give the green light that the NRU could be safely reopened.

[529] Algoma is distinguishable from this case. In Algoma the idle period exclusion applied as the relining of the runway unrelated to the loss was already underway and the time frame for the repairs was concurrent to the damage to the furnace and the time period required for those repairs.

[530] If the corrosion exclusion applies [which was not my conclusion], the proximity in time and place between the excluded corrosion damage, and the covered resultant physical damage of the leak of water precludes the Insurer from denying coverage for business losses based upon the idle period exclusion.

[531] Applying the but for test confirms the Plaintiffs’ entitlement. But for the leak of heavy water, the NRU would have been operational, and the Plaintiff would have been able to receive the isotopes it required and would not have suffered business losses.

[532] For these reasons I conclude there is no merit to the arguments raised that the resulting damage should be limited to 24 hours, the time it took to actually drain the water from the calandria, or for the period of time to diagnose the problem and to empty the calandria of the fuel rods and drain the water. The resulting physical damage continued until the NRU was repaired and cleared by the CNSC to be reopened.

[533] The Plaintiffs also argue that the timeline for the resulting damage must be interpreted considering the Period of Liability as defined in the Policy. It submits that the Period of Liability starts from the time of the physical loss or damage and extends for the entire period of time that the insured’s business is directly affected by such damage. Hence the idle period exclusion does not engage.[^144]

[534] Exclusion clauses are to be interpreted narrowly and coverage provisions are to be interpreted broadly.[^145]

[535] The Time Element Exclusions do not apply as there was no idle period in the steps between the leak of heavy water until the NRU was restarted in August 2010. The leak of heavy water triggered the shutdown, and until the leak site, and other vulnerable areas were repaired to the satisfaction of the CNSC. Until that time, the NRU was inoperable.

[536] I conclude that the period of loss for the resulting physical damage is from the date of the shutdown to the date the NRU was restarted.

[537] For these reasons, I conclude that the Idle Period Exclusion does not apply to limit the period of loss of the resulting physical damage. Therefore, the Plaintiffs are entitled to be indemnified for the business losses they suffered throughout the entire period from the date of the shutdown until the leak was repaired, subject to the maximum coverage stipulated in the Policy.

PART VII: ISSUE 4: THE NUCLEAR RADIATION EXCLUSION

The Evolving Nuclear Radiation arguments

[538] The next exclusion the Insurer relies on is Exclusion 5.B.1 of the Policy which I will refer to as the “Nuclear Radiation Exclusion”.

[539] The Insurer’s argument relying on this exclusion has gone through several transformations.

[540] The Insurer originally took the position that the presence of Tritium in the J-Rod annulus engaged the nuclear exclusion relying on “nuclear reaction” rather than “nuclear radiation” in Exclusion 5.B.1 of the Policy. At discovery, the Insurer confirmed that it was no longer relying on the nuclear reaction exclusion.

[541] Later, after the disclosure of the Mancey Report, the Insurer changed its position, relying on the “nuclear radiation” exclusion due to the presence of nitric acid and carbonic acid in the J-rod annulus. According to the Mancey Report, nitric acid and carbonic acid probably caused the through wall leak at J-41. Carbonic acid was later ruled out as a contributing cause of the corrosion.

[542] In light of the evidence of Dr. Frankel and Dr. Kirk questioning the conclusions in the Mancey Report, the Insurer has reshaped its nuclear radiation argument again relying on the presence of nitric acid, caused by irradiation of air in the J-rod annulus as being part of the “environment” causing the leak at J-41 that directly or indirectly caused the corrosion at J-41. The Insurer argues that the anti-concurrent causation exclusion principles apply precluding coverage.

[543] The Nuclear Radiation Exclusion provides as follows:

5. Exclusions

The following exclusions apply unless specifically stated elsewhere in this Policy:

B. This Policy excludes loss or damage directly or indirectly caused by or resulting from any of the following regardless of any other cause or event, whether or not insured under this Policy, contributing concurrently or in any other sequence to the loss:

1. nuclear reaction or nuclear radiation or radioactive contamination. However:

a) if physical damage by fire or sprinkler leakage results, then only that resulting damage is insured; but not including any loss or damage due to nuclear reaction, radiation or radioactive contamination.

b) this Policy does insure physical damage directly caused by sudden and accidental radioactive contamination, including resultant radiation damage, from material used or stored or from processes conducted on the Insured Location, provided that on the date of loss, there is neither a nuclear reactor nor any new or used nuclear fuel on the Insured Location. This coverage does not apply to any act, loss or damage excluded in item B2f of this EXCLUSIONS clause.

This exclusion B1 and the exceptions in B1a and B1b do not apply to any act, loss or damage which also comes within the terms of exclusions B2b of this EXCLUSIONS clause. [Emphasis added][^146]

[544] Nitrogen oxides formed as a result of irradiation of the air in the J-rod annulus, which in turn dissolved in the water that had leaked from the reflector to from nitric acid. As nitric acid formed in the J-rod annulus by a process of nuclear radiation, the Insurer argues that the nitric acid was the cause, or a concurrent cause of the through-wall leak[^147] and therefore the nuclear exclusion applies.

[545] The first position of the Insurer relies on the Mancey Report, supported by the evidence of Dr. Revie, as to the role played by nitric acid causing the through wall penetration of the calandria. For reasons I have fully outlined in the corrosion exclusion analysis I do not accept the Insurer’s arguments that nitric acid caused the leak at J-41. The localized pitting corrosion was caused by the presence of an aggressive agent in the water leaking from the reflector.

[546] Now, the Insurer submits an alternative theory, that nitric acid was part of the environment causing the corrosion and therefore indirectly contributed to the through wall leak at J-41 engaging the nuclear radiation exclusion.

[547] The Insurer argues that the language of the Nuclear Radiation Exclusion is clear and unambiguous.[^148] If nuclear radiation directly or indirectly caused or contributed to the loss or damage, all coverage for the loss or damage is excluded: see Derksen v. 539938 Ontario Ltd.[^149]

[548] The Plaintiffs argue that if the meaning of the nuclear exclusion is ambiguous, the exclusion in this Policy intends to refer to a nuclear event, not as a result of the anticipated and known radiation that was taking place in the J-rod annulus over the years analyzed on a chemical, microscopic level.

[549] To address the Nuclear Radiation Exclusion, I will address the following questions:

1. Is the nuclear radiation exclusion ambiguous?
2. If the nuclear radiation exclusion is ambiguous, how should it be interpreted?
3. Is the wording of this exclusion an anti-concurrent causation exclusion, and if so how is it to be interpreted?

1. Is the Nuclear Radiation Exclusion ambiguous?

[550] The Policy does not define any of “nuclear reaction,” “nuclear radiation,” or “nuclear contamination”.[^150]

[551] There are no cases or insurance law textbooks considering the meaning of the nuclear radiation exclusion. Black’s Law Dictionary does not define any of these terms.

[552] The Canadian Oxford Dictionary does not define “nuclear radiation”. The relevant definition of “nuclear”, alone, is “using, producing, or resulting from nuclear energy”. It defines “radiation” as “1. The act or an instance of radiating. 2. The emission of energy as electromagnetic waves or as moving particles.”

[553] The Plaintiffs submits that the Nuclear Radiation Exclusion clause is ambiguous and such ambiguity should be resolved in favor of the Plaintiffs. They have provided interpretive principles of contract and insurance law to resolve ambiguity.[^151] While the Plaintiff has provided numerous authorities for how to resolve ambiguity, counsel has not provided any caselaw to determine the preliminary question: whether an ambiguity exists.

[554] The Insurer argues that the nuclear exclusion should apply due to the presence of nitric acid in the J-rod annulus. The Plaintiffs argue that the nuclear radiation clause is intended to refer to a nuclear event, not the known, predictable and accepted processes of radiation taking place in the day-to-day operation of the NRU.

[555] It is my view, although not argued by the Plaintiffs, looking simply at the wording of the Nuclear Radiation Exclusion, its placement in the Policy and the undisputed facts there may not be any ambiguity.

[556] Therefore, before resolving any potential ambiguity, the threshold question is to consider the jurisprudence to determine whether or not the Nuclear Radiation Exclusion is ambiguous.

[557] The caselaw confirms the following principles in assessing ambiguity:

• Mere difficulty in interpreting a contractual provision is not synonymous with ambiguity: St. Lawrence Petroleum v. Bailey Selburn Oil & Gas Ltd.[^152]
• A contractual provision will be ambiguous when it is reasonably capable of more than one meaning: Hi-Tech Group Inc. v. Sears Canada Inc. (2001), at para. 18, and Survey Aircraft Ltd v. Stevenson, at para. 33.[^153]
• A contractual provision will also be ambiguous if one word, one expression or a series of expressions are capable on its face or in its application of two or more meanings: Leitch Gold Mines Ltd. v. Texas Gulf Sulphur Co., at para. 241.[^154]
• The court must be mindful to interpret the words of the contract by giving them the ordinary meaning of the language: Melito Estate v. Melanson, at para. 62.[^155] “The court should not strain to create an ambiguity where none exists.”: Chilton v. Co-operators General Insurance Co. (1997), at para. 26.[^156]
• Determining whether there is an ambiguity requires “the full text of the contract [be] considered, in light of the surrounding circumstances at the time of its execution, if necessary.” Geoffrey L. Moore Realty Inc. v. Manitoba Motor League, at para. 25.[^157]
• Before finding an ambiguity, a court must first turn to the surrounding circumstances of the contract to see if the intention of the parties can be determined on that basis.[^158]

[558] For the reasons I will outline, I find that it is possible to understand the intended meaning of the nuclear radiation exclusion, by reading the exclusion within the full text of the Policy considered as a whole, in light of the surrounding circumstances and considering the objective reasonable expectation of the parties[^159] without further analysis.

Context: The Nuclear Radiation Exclusion in the Policy

[559] The Nuclear Radiation Exclusion clause appears on page 37 of the Policy and is immediately followed by other exclusion clauses that identify almost every conceivable instance of societal disruption or disintegration. These exclusions include[^160]:

5. Exclusions

The following exclusions apply unless specifically stated elsewhere in this Policy:

This Policy excludes loss or damage directly or indirectly caused by or resulting from any of the following regardless of any other cause or event, whether or not insured under this Policy, contributing concurrently or in any other sequence to the loss:

1. nuclear reaction or nuclear radiation or radioactive contamination. However ….

a) if physical damage by fire or sprinkler leakage results, then only that resulting damage is insured; but not including any loss or damage due to nuclear reaction, radiation or radioactive contamination.

b) this Policy does insure physical damage directly caused by sudden and accidental radioactive contamination, including resultant radiation damage, from material used or stored or from processes conducted on the Insured Location, provided that on the date of loss, there is neither a nuclear reactor nor any new or used nuclear fuel on the Insured Location. This coverage does not apply to any act, loss or damage excluded in item B2f of this EXCLUSIONS clause.

This exclusion B1 and the exceptions in B1a and B1b do not apply to any act, loss or damage which also comes within the terms of exclusions B2b of this EXCLUSIONS clause.[^161]

2. a) hostile or warlike action in time of peace or war including action in hindering, combating or defending against an actual, impending or expected attack by any:

(i) government or sovereign power (de jure or de facto);

(ii) military, naval or air force; or

(iii) agent or authority of any party specified in i or ii above.

b) discharge, explosion or use of any nuclear device, weapon or material employing or involving nuclear fission, fusion or radioactive force, whether in time of peace or war and regardless of who commits the act.

c) insurrection, rebellion, revolution, civil war, usurped power, or action taken by governmental authority in hindering, combating or defending against such an event.

d) seizure or destruction under quarantine or custom regulation, or confiscation by order of any governmental or public authority.

e) risks of contraband, or illegal transportation or trade.

f) Terrorism, including action taken to prevent, defend against, respond to or retaliate against Terrorism or suspected Terrorism, except to the extent provided in the TERRORISM coverage in this section of the Policy. However, if direct loss or damage by fire results from any of these acts (unless committed by or on behalf of the Insured), then this Policy covers only to the extent of the Actual Cash Value of the resulting direct loss or damage by fire to property insured. This coverage exception for such resulting fire loss or damage does not apply to:

(i) direct loss or damage by fire which results from any other applicable exclusion in the Policy, including the discharge, explosion or use of any nuclear device, weapon or material employing or involving nuclear fission, fusion or radioactive force, whether in time of peace or war and regardless of who commits the act.

(ii) any coverage provided in the TIME ELEMENT section of this Policy or any other coverages provided in this Policy.

Any act which satisfies the definition of Terrorism as provided herein shall not be considered to be vandalism, malicious mischief, riot, civil commotion, or any other risk of physical loss or damage covered elsewhere in this Policy.

If any act which satisfies the definition of Terrorism as provided herein also comes within the terms of item B2a of this EXCLUSIONS clause then item B2a applies in place of this item B2f exclusion.

If any act which satisfies the definition of Terrorism as provided