ML050120361

Letter from Mark J. Wetterhahn to Diane Curran Enclosing Duke Energy Corporation'S Fourth Supplemental Response to Blue Ridge Environmental Defense League'S First Set of Discovery Requests to Duke Regarding Bredl Security Contention 5

ML050120361
Person / Time
Site:	Catawba
Issue date:	01/04/2005
From:	Wetterhahn M Duke Energy Corp, Winston & Strawn, LLP
To:	Curran D Harmon, Curran, Harmon, Curran, Spielberg & Eisenberg, LLP, Office of Nuclear Reactor Regulation
Byrdsong A T
References
50-413-OLA, 50-414-OLA, ASLBP 03-815-03-OLA, RAS 9141
Download: ML050120361 (20)

Text

S PA . REORRESPONDE WINSTON & STRAWNLLP Electronic Letterhead 1400 L STREET N.W, WASHINGTON DC 20005-3502 202-371-5700 03S r4. 333 80 0 10 CON I * . 43 SC.Ob8 - A cVes - *0 F*

CALL 006064703 " ouw lOld6ISl3 tAOAE. CA 071-11543 c e0 0411 .804 204O O 76 P. , 3 312-458.4600 2.2044700 813416.1700 41401-1000 41.22317-75-75 35-6434*a242 L-K E. .O. C NSHH 44-0Z074A2-0000 January 4, 2005 DOCKETED USNRC VIA ELECTRONIC MAIL January 10, 2005 (3:18pm)

OFFICE OF SECRETARY RULEMAKINGS AND Diane Curran, Esq. ADJUDICATIONS STAFF Harmon, Curran, Spielberg & Eisenberg, LLP 1726 M Street, NW Suite 600 Washington, DC 20036 Re: Duke Energy Corporation Catawba Nuclear Station, Units 1 and 2 (Docket Nos. 50-413-OLA, 50-414-OLA)

Dear Ms. Curran:

Enclosed is "Duke Energy Corporation's Fourth Supplemental Response to Blue Ridge Environmental Defense League's First Set of Discovery Requests to Duke Regarding BREDL Security Contention 5."

Also enclosed is an errata sheet reflecting corrections to the transcript of the oral deposition of Howard B. Williams. With these corrections, Mr. Williams is prepared to sign the original copy of the transcript whenever you make it available to him.

If you have any questions, please contact me.

Sincerely, Mark J. Wetterhahn Counsel for Duke Energy Corporation cc: Service List DC:394279.1 1Fmpklte=.sc s y o03s S6ac %- O.-2d UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of: )

)

DUKE ENERGY CORPORATION )

) Docket Nos. 50-413-OLA (Catawba Nuclear Station, ) 50-414-OLA Units 1 and 2) )

)

)

ERRATA SHEET ORAL DEPOSITION OF HOWARD WILLIAMS TUESDAY, DECEMBER 7,2004 PAGE LINE ACTION 26 4 Insert "in".before operations 44 5 Delete "fought" and insert "qualified" 46 15 Delete "on" and insert "a" 48 7 Delete "PF" and insert "PA" 55 1 Delete "air" and insert "area" 55 10 Delete "parameter" and insert "perimeter" 59 17 Delete "nozzle" and insert "muzzle" 64 17 Delete "MT used" and insert "MTUs" 82 3 Delete "knock bender" and insert "up-ender" 94 7 Insert "cavalry" after "air"

January 4, 2005 UNITED STATES OF AMERICA NUCLEAR REGUI;ATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of: )

)

DUKE ENERGY CORPORATION )

) Docket Nos. 50413-OLA (CatawbaNuclear Station, ) 50414-OLA Units 1 and 2)

)

DUKE ENERGY CORPORATION'S FOURTH SUPPLEMENTAL RESPONSE TO THE BLUE RIDGE ENVIRONMENTAL DEFENSE LEAGUE'S FIRST SET OF DISCOVERY REQUESTS TO DUKE REGARDING BREDL SECURITY CONTENTION 5 Duke Energy Corporation ("Duke") further supplements its July 2, 2004 response to the June 19, 2004 "Blue Ridge Environmental Defense League's First Set of Discovery Requests to Duke Energy Corporation Regarding Security Plan Submittal" as follows:

Supplemental Response to General Interrogatory No. I The information contained in this response was provided by Mr. Steven P. Nesbit of Duke. An affidavit to that effect is included as Attachment 1 to this response.

Supplemental Response to General Document Production Request No. 3 BREDL's General Document Production Request No. 3 seeks production of: "All documents (including experts' opinions, work papers, ajffdavits, and other materialsused to render such opinion) supporting or otherwise relating to testimony or evidence that you intend to use in the hearing on Contention 5. "

Duke hereby further supplements its July 2, 2004 response to this General Document Production Request by providing the following documents, copies of which are included as Attachments 2 and 3, respectively, to this discovery response:

• D. Linzer, "Nuclear Capabilities May Elude Terrorists, Experts Say," The W~ashington Post (Dec. 29, 2004).

• NUREG-1738, Technical Study of Spent Fuel Pool Accident Risk at Decommissioning Nuclear Power Plants (Feb. 1, 2001), Appendix 1.B ("Temperature Criteria for Spent Fuel Pool Analysis").

Respectfully submitted, David A. Repka Mark J. Wetterhahn Anne W. Cottingham WINSTON & STRAWN, LLP 1400 L Street, NW Washington, D.C. 20005-3502 Timika Shafeek-Horton DUKE ENERGY CORPORATION 422 South Church Street Mail Code: PB05E Charlotte, N.C. 28201-1244 ATTORNEYS FOR DUKE ENERGY CORPORATION Dated in Washington, District of Columbia This 4th day of January 2005 2

ATTACHMENT I January 4, 2005 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of: )

)

DUKE ENERGY CORPORATION )

) Docket Nos. 50-413-OLA (Catawba Nuclear Station, ) 50-4 14-OLA Units I and 2) )

)

AFFIDAVIT OF STEVEN P. NESBIT Steven P. Nesbit hereby declares under penalty of perjury that the following statements are true and correct of his own knowledge:

1. I have reviewed the information contained in "Duke Energy Corporation's Fourth Supplemental Response to the Blue Ridge Environmental Defense League's First Set of Discovery Requests to Duke Regarding BREDL Security Contention 5:' dated January 4, 2005.

2. Such information is true and correct to the best of my knowledge, information and belief.

Steven P. Nesbit

washingtonpost.com: Nuclear Capabilities May Elude Terrorists, Experts Say Page I of 4 WahIngtonpostQCOm Ad ATTACHMENT 2 Nuclear Capabilities May Elude Terrorists, Experts I Say By Dafna Linzer Washington Post Staff Writer Wednesday, December 29,2004; Page A01 -becomes

.; financial Of all the clues that Osama bin Laden is after a nuclear weapon, perhaps the most .p~anner~onllne significant came in intelligence reports indicating that he received fresh approval last

-year from a Saudi cleric for the use of a doomsday bomb against the United States. LAN MOREW For bin Laden, the religious ruling was a milestone in a long quest for an atomic weapon. For U.S. officials and others, it was a frightening reminder of what many consider the ultimate mass-casualty threat pos6d by modem terrorists. Even a small nuclear weapon detonated in a major American population center would be among history's most lethal acts of war, potentially rivaling the atomic destruction of Hiroshima and Nagasaki.

Despite the obvious gravity of the threat, however, counterterrorism and nuclear experts in and out of government say they consider the danger more distant than immediate.

They point to enormous technical and logistical obstacles confronting would-be nuclear terrorists, and to the fact that neither al Qaeda nor any other group has come close to demonstrating the means to overcome them.

So difficult are the challenges that senior officials on President Bush's national security team believe al Qaeda has shifted its attention to other efforts, at least for now.

"I would say that from the perspective of terrorism, the overwhelming bulk of the evidence we have is that their efforts are focused on biological and chemical" weapons, said John R. Bolton, undersecretary of state for arms control and international security. "Not to say there aren't any dealings with radiological materials, but the technology for bio and chem is comparatively so much easier that that's where their efforts are concentrating."

Still, the sheer magnitude of the danger posed by a nuclear weapon in terrorist hands -- and classified intelligence assessments that deem such a scenario plausible - has spurred intelligence and military operations to combat a threat once dismissed as all but nonexistent. The effort includes billions of dollars spent on attempts to secure borders, retrain weapons scientists in other countries and lock up dangerous materials and stockpiles.

"The thing to keep in mind is that while it is extremely difficult, we have highly motivated and intelligent people who would like to do it," said Daniel Benjamin, a former National Security Council staff member and senior fellow at the Center for Strategic and International Studies. Each type of weapon of mass destruction --

nuclear, biological and chemical -- presents special challenges for the groups seeking to acquire them, but also opportunities that can be exploited by people determined to unleash their awesome destructive powers. This is the first of three articles aimed at exploring those-risks and challenges.

Difficult Course Without sophisticated laboratories, expensive technology and years of scientific experience, al Qaeda has two primary options for getting a bomb, experts say, both of which rely on theft -- either of an existing weapon or httn://wt^<v.washinsotannnot.cm/ac2vwn-dvn/A3 22R5-2004De.2)RInmrn cep=nrintpr 1M, to)nnec

washingtonpost.com: Nuclear Capabilities May Elude Terrorists, Experts Say Page 2 of 4 one of its key ingredients, plutonium or highly enriched uranium.

Nuclear scientists tend to believe the most plausible route for terrorists would be to build a crude device using stolen uranium from the former Soviet Union. Counterterrorism officials think bin Laden would prefer to buy a ready-made weapon stolen in Russia or Pakistan, and to obtain inside help in detonating it.

Last month, Michael Scheuer, who ran the CIA's bin Laden unit, first disclosed in an interview on CBS's "60 Minutes" that bin Laden's nuclear efforts had been blessed by the Saudi cleric in May 2003, a statement other sources later corroborated. As early as 1998, bin Laden had publicly labeled acquisition of nuclear or chemical

'weapons a "religious duty," and U.S. officials had reports around that time that al Qaeda leaders were discussing attacks they likened to the one on Hiroshima.

A week after his CBS appearance, Scheuer said at breakfast with reporters in Washington that he believed al Qaeda would probably seek to buy a nuclear device from Russian gangsters, rather than build its own.

There were as many as a dozen types of nuclear weapons in the hands of the Soviet Union at the end of the Cold War, but Russian officials have said that several kinds have since been destroyed and that the country has secured the remainder of its arsenal. The nature and scope of nuclear caches are among the most tightly held national security secrets inRussia and Pakistan.

It is unclear how quickly either country could detect a theft, but experts said it would be very difficult for terrorists to figure out on their own how to work a Russian or Pakistani bomb.

Newer Russian weapons, for example, are equipped with heat- and time-sensitive locking systems, known as permissive action links, that experts say would be extremely difficult to defeat without help from insiders.

"You'd have to run it through a specific sequence of events, including changes in temperature, pressure and environmental conditions before the weapon would allow itself to be armed, for the fuses to fall into place and then for it to allow itself to be fired," said Charles D. Ferguson, science and technology fellow at the Council on Foreign Relations. "You don't get it off the shelf, enter a code and have it go off."

The strategy would require help from facility guards, employees with knowledge of the security and arming features of the weapons, not to mention access to a launching system.

Older Russian nuclear weapons have simpler protection mechanisms and could be easier to obtain on the black market. But nuclear experts said even the simplest device has some security features that would have to be defeated before it could be used.

"There is a whole generation of weapons designed for artillery shells, manufactured in the 1950s, that aren't going to have sophisticated locking devices," said Laura Holgate, who ran nonproliferation programs at the Pentagon and the Energy Department from 1995 to 2001. "But it is a tougher task to take a weapon created by a country, even the 1950s version, a tougher job for a group of even highly qualified Chechen terrorists to make it go boom."

Transporting a weapon out of Russia would provide another formidable obstacle for terrorists.

Most of the ready-made bombs that could be stolen would be those made with plutonium, which emits far higher levels of radiation and is therefore more easily detected by passive sensors at ports than is highly enriched uranium, or HEU.

"I wouldn't rule out plutonium altogether, but if one were a terrorist bent upon demonstrating a nuclear explosion, the HEU route is technically much easier," said William C. Potter, director of the Center for Nonproliferation Studies at the Monterey Institute of International Studies in California.

http://www.washingtonpost.com/ac2/wp-dyn/A32285-2004Dec28?language=printer 1/3/2005

washmngtonpost.com: Nuclear Capabilities May Elude Terrorists, Experts Say Page 3 of 4 Building a Bomb Such difficulties have led some nuclear experts to believe bin Laden would be more likely-to try to build an improvised nuclear weapon using a combination of uranium and conventional explosives. That design, known as a gun-type device, was used in the atomic bomb over Hiroshima.

While the technology is relatively simple and has been described in dozens of published scientific studies and policy journals, the path to development is filled with technological and logistical challenges -- the most significant of which is obtaining at least 50 kilograms of bomb-grade uranium. That amount would yield a slightly smaller device than "Little Boy," the code name for the Hiroshima bomb, but would be enough to obliterate any life or structure within a half-mile radius of the blast zone.

"If they got less material than that, it would be really dicey that they could build such a bomb," said Ferguson, at the Council on Foreign Relations.

According to a database maintained by the United Nations' International Atomic Energy Agency, there have been 10 known incidents of HEU theft in the past 10 years, each involving a few grams or less. Added up, the stolen goods total less than eight kilograms and could not be easily combined because of varying levels of enrichment. Most important, the thieves -- none of whom was connected to al Qaeda -- had no buyers lined up, and nearly all were caught while trying to peddle their acquisitions.

"Making the connection between buyer and seller has proved to be one of the most substantial hurdles for terrorists," said Matthew Bunn, a senior researcher at Harvard University's Project on Managing the Atom. Of the few known attempts by al Qaeda to obtain HEU, each allegedly stumbled because there was either no seller or the material on offer was fake. "Each time they tried, they got scammed," said Bruce Hoffman, a counterterrorism expert at the Rand Corporation who has tracked al Qaeda for years.

A September report on terrorism by the Congressional Research Service warned that terrorists could "obtain HEU from the more than 130 research reactors worldwide that use HEU as fuel." The report noted that the nations of "greatest concern as potential sources of weapons or fissile material are widely thought to be Russia and Pakistan."

The largest stocks outside the United States are in Russia and around the former Soviet Union, some in facilities with notoriously weak security and safety procedures.

"Once you have the fissile material, it's a matter of basic chemistry, basic machinery and a truck," said Holgate, now a vice president at the nonprofit Nuclear Threat Initiative. "You have to have some technical capability, but once you have those skills, it's certainly within the grasp of the kind of sophisticated, planning-capable terror organizations out there."

Even so, there are a great many steps between obtaining the material and setting off an explosion. That may account for why such an attack has not materialized, despite intelligence warnings.

The uranium would have to be smuggled out of the facility and then transferred, possibly across several borders, seaports and airports, to a location where the device could be assembled. As described in unclassified literature, the gun-type bomb works when one mass of uranium is shot into another inside a tube. Such a device would be small enough to hide in a corner of a shipping container, but that would mean getting it to a port, onto a container and probably bribing a shipper or cargo crew to transport it.

An oil shipment would be optimal for a ready-made device, according to the congressional report, because the "size of the supertanker and thickness of the steel, especially with the use of double hulls," renders some detection equipment unusable.

http://wwxv.washingtonpost.com/ac2/wp-dyn/A32285-2004Dec28?language=printer 1/3/2005

washingtonpost.com: Nuclear Capabilities May Elude Terrorists, Experts Say IPage 4 of 4 But HEU emits low levels of radioactivity anyway, and that could be masked with lead shielding. A primitive device could be assembled in a small garage using machine tools readily available at an auto shop and concealed in a lead-plated ,delivery truck about the size of a delivery van, experts said.

It is also unclear how a terrorist group would know if its weapons development effort was on the right track.

Nations with nuclear bombs conduct tests, including explosions that can be detected by scientists and governments. Bunn, who has published two studies on nuclear terrorism,-said terrorists would not necessarily need to conduct such tests, but doing without them would increase chances that human error would foil plans or delay progress.

The most elaborate known effort by a terrorist group to develop a nuclear program was undertaken by the Japanese cult Aum Shinrikyo, which instead of stealing enriched uranium planned to mine and enrich the material itself.

Members of Aum Shinrikyo, intent on world destruction when it began its 1993 quest for a nuclear weapon, had all the means to pull it off, on paper at least: money, expertise, a remote haven in which to work, and most important, a private uranium mine.

But the group made dozens of mistakes in judgment, planning and execution. It shifted course, launching its chemical attack on the Tokyo subway in 1995.

"There are valuable lessons in Aum's experience, and there are false lessons," said Benjamin, co-author of "The Age of Sacred Terror." "The valuable lesson is that WMD terrorism is hard to do," he said. "But given that they didn't try what would be the most efficient way to put together a nuclear bomb, we shouldn't overrate their example as a reason why it's not going to happen."

Al Qaeda has been on the run since the United States deprived it of a haven in Afghanistan, making it more difficult for the group to operate on such an ambitious scale.

"At this moment, they are less capable of carrying out an operation like this because it would require so many different experts and operatives," Benjamin said. "But even a depleted group could do it if they got the right breaks."

© 2004 The Washington Post Company Advertising Unks by Google What's his?

Help Tsunami Victims Learn How to Help Tsunami Survivors World Vision - Giving Since 1950 www.WorIldVision.org UNICEF Help the children devested by the tsunami InSoutheast Asia www.unicef.org http://wwv.washingtonpost.com/ac2/wp-dyn'A32285-2004Dec28?language=printer 1/3/2005

The Path to aNuclear Weapon Terroristswould have to overcome enormous technicaland-logisticalobstacles before they could set off a nuclearbomb in a U.S. city. Coun r and nucearexpertssay the danger.ismore distant than immediate.

• C.u':e: _ _ . ,m .n.clearexpe~ts i.s' th damp is Steal sufficient quantities of bomb-grade . 7 Most facilities have security measures and..

uranium or plutonium from nuclear .low quantities of materials. Thieves have facilities, storage sites or reactors around i lty cnnustomers.

the . . .. .

LUranium or plutonium would have to be .. . Numerous people would nieed to be hired removed from the facilities, shielded from or bribed at border crossings, airports and detection and then transported across. seaports. Lead shielding would be needed borders. :to mask uranium, but plutonium ismuch easier for passive sensors to detect.

.The simrplest method would be to build a Proficiency In explosives, weapons crude nuclear device, known as a gun- construction, uranium handling and other type bomb, using at least 50 kilograms of cotexpertise arearequired in andiotrto g

...uranium and conventional explosives. iiii uequipment, materials and a facility for assembling the device.

An oil tanker hull has steel that Isthick Plutonium can be registered by passive enough to protect a plutonium weapon sensors at border controls. A large truck from detection at seaports. Smuggling or ship risks Inspection.

routes used by drug, cigarette and' Illegal Immigrant runners could provide a way In. -

A cruide nuclear device can be set off by Sophisticated locking systems on any bomb conventional explosives,'and a gravity. . other than one made solely by the terrorist bomb can be dropped from a plane. group would have to be dismantled.

Terrorists would not risk testing If they had only srsmall amount of material.

Possible Uraniumn Sources  :

TeVroris cduld obtdin uraniumnforbuilding'abomb from one of more than 130 researchreactorsworldwide that use hidghly

• enriWed uraniumasfuel. While afill list was undvailable, theEnergpDepa en! has targded105researchreactorsworiduide'.

for conversion to a typ offuel that would be diffcult to use in weapo?&A thirdof them have been at leas partly converted.

• CountriesWithatat . J.54 JLjRANY leastoonererch. P reactorinthe ~..3ILLNM~ERLANDSL~A SmeB b s . , 3.UNITEDI.

Department's ,, 211 CZECH B program =-.

than one iT~

rk

/ 'ovr~o I... '. :I , ,'. ' " '...

Denrnua .  ;.~

Roamnla, Sovenla' _____*

• SwbeflandUn Simple Bomb Designs -

A gun-typfsin eievuc st"~Ltl o bomb dropped on Hiroshima,wvIdd be'the most likelydse'uedba n wclwrterrorist.An imPlosion,bomb, ~niAc as the '¶Fat Man bomb drpe nNgskwould bemr ~'cl omake.

T I 777 tinitlatv., Prolect on Managing the Atom. General Accounting Offke. How Stuff Works ..

'.  ;, ..  ; ' '. . ' ,.I .

. . .1-1., _

. I

ATTACHMENT 3 APPENDIX 1.B TEMPERATURE CRITERIA FOR SPENT FUEL POOL ANALYSIS

1. BACKGROUND The engineering analyses performed to address spent fuel pool (SFP) performance during various accidents have, in the past, used a temperature criterion to evaluate the potential for significant fuel damage. This temperature was intended as an acceptance criterion beyond which one would expect the onset of significant, global, fuel damage and substantial release of fission products (e.g., 50-100% of inventory of volatiles) associated with such damage. Further, the temperature criterion cited (generally about 900SC) has been selected on the basis that It represented a threshold for self sustained oxidation (Ref. 1) of cladding in air and on that basis it has been argued that if cooling of the spent fuel could limit fuel temperatures In equilibrium below this threshold then large releases of fission products need not be considered. Self sustaining reaction in this sense means the reaction rate and thus heat generation rate Is sufficient, to roughly balan6e heat losses for given cooling mechanisms, resulting In an Isothermal condition. Once the fuel temperature exceeds this threshold temperature (altematively Identified as an Ignition or autognitlton temperature) It was presumed that subsequent heat up and further Increases In reaction rates would be escalating and rapid and.

that serious fuel damage would ensue. The temperature escalation associated with oxidation in this regime would not be balanced by any reasonable cooling afforded by natural circulation of air. While It was not expected that fission product releases associated with core melt accidents would Immediately emerge at this temperature (based on reactor research In various steam and hydrogen environments) It was recognized that the time window for subsequent fuel heating would be relatively small once oxidation escalated. This also did not preclude gap type releases associated with fuel failures below the threshold temperature but these generally were not considered to be significant compared to the releases associated with higher fuel temperatures and significant fuel damage.

In the report, 'Draft Final Technical Study of Spent Fuel Pool Accident Risk at Decommissioning Nuclear Power Plants," February 2000, the temperature criterion selected, 800 'C, was used in two ways. First, it was used to determine the decay heat level and corresponding time at which heat generation and losses for complete and Instantaneous draining of the pool would lead to heating of the fuel (to 800 *C) after 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />. This time period would allow for the Implementation of effective emergency response without the full compliment of regulatory requirements associated with operating reactors. Secondly, the temperature criterion was also used to evaluate the decay'heat level and time (critical decay time') at which heat generation and losses for a fully drained fuel pool would rbsult in an equilibrium temperature of 800 "C (typically this critical decay time has been on the order of 5 years). On that basis it was reasoned that since serious overheating of the fuel had not occurred, the fission product release associated with core'melt need not be considered.

2. AIR OXIDA1ON AND TEMPERATUR8 ESCALATION The NRC has received a number of comments related to the use of this temperature criterion and has reassessed the appropriateness of such a value for both its intended purposes. At the Appendix .1B A1131 .

outset RES acknowledges that an ignition temperature, or more precisely Inthis case a temperature for incipient temperature escalation Is dependent on heat generation and losses which Inturn Is dependent on system geometry and configuration. In fact, much of the data on oxidation is produced in isothermal tests up to near the melting temperature of zirconium. In examining an appropriate criterion, it is useful to consider the range of available data including core degradation testing In steam environments, since it is likely that many SFP accidents may involve some initial period during which steam oxidation kinetics controls the initial oxidation, heatup, and release of fission products. Invarious experimental programs around the world (e.g., PBF-SFD, ACRR, CORA, NSRR, PHEBUS and QUENCH) repeatable phenomena have been observed for the early phase of core degradation (in steam) which proceeds Initially at temperature increase rates associated with decay heat (at levels characteristic of reactor accidents) until claddingoxidation becomes dominant and a more rapid temperature escalation occurs. The point at which the escalation occurs, which does vary between tests, has been attributed to heat losses (Ref. 2) characteristic of the facility and to phase changes of ZrO2 over a temperature range. The threshold at which temperature escalation occurred has been reported to vary from approximately 100xC to1600 *C. In a CORA test performed witha lower.

initial heatup rate (to simulate reduced decay heat during shutdown conditions) It was reported that uncontrolled temperature escalation did notoccur, raising the prospect that heating rate maybe afactor. (This Is probably because of the formation of a thicker oxidation layer buit up over the protracted timeit lower temperature such that when higher temperatures are attained, the thicker scale resultsin a lower oxidation rate relative to a thinner scale at the same temperature.) In more recent QUENCH tests (Quench 04 and 05) the effect of preoxidation was evaluated for its effect on hydrogen generation and temperature escalation. In Quench 04 temperature escalation was reported to occur at 1300 'C; In Quench 05 with approximately 200pm preoxidation temperature escalation was reported to be delayed until the fuel rod temperature reached 1620 "C.

Because ofinterest in airingression phenomena for reactor accidents, recent severe accident

• research has also examined oxidationin air environments. Publication of results from the DRESSMAN and CODEX test programs (Ref. 3) has provided much of the transient data on fuel rod and rod bundle behavior for air kinetics as well as dataon fuel oxidation and volatility. Early studies of zirconium oxidation Inair (Refs. 4 and 5)were performed by comparing Isothermal oxidation and scaling of fresh samples to determine the influence of drifferent atmospheres and materials as well as to examine potential for fire hazards. The general observation Is that, at least at higher temperatures(>1000 'C), the oxidation rateIs higher In air thanIn steam.

Another observation of the early studies was, under the same conditions, oxidation in an air environment produced an oxide layer or scale less protective than that for steam owing to the possible Instability of a nitride layer beneath the outer oxide layer leading to scale cracking and a breakaway In the oxidation rate. The onset of this breakaway in the oxidation rate occurred at about 800 "C after a time period of 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> in the studies performed by Evans et al (Ref. 4) and after a period of approximately 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> In studies by Leistikow (Ref. 9). The Leistkow studies were performed on fresh cladding, however, and It Is expected that breakaway would occur after a longer time delay with preoxidized cladding. As breakaway oxidation occurs the oxidation behavior observed no longer reflects a parabolic rate dependence but takes on a linear rate dependence. Also, at lower temperatures the kinetics of reaction indicate near cubic rate dependence thus the representation of the oxidation behavior at both high and low temperatures witli a parabolic rate dependence may introduce unnecessary simplification and an Appendix 18 AB-2 I

I.

S understatement of the low temperature behavior. Breakaway scaling In an isothermal test may not translate to similar behavior under transient heatup conditions where Initial oxidation occurs at lower temperatures and may involve steam oxidation. The presence of hydrides In the cladding may also increase the potental for exfoliation and a breakaway in the oxidation; the effect of this has, however, been seen more clearly In testing conducted with steam and high hydrogen concentrations. Also, zirconiurn hydride will be dissolved at 700 'C and above, thus Its contribution to exfoliation and breakaway will be minimal.

Autoigntlon Is known to occur In zirconium alloys and zirconium hydride, especially when clean metal or hydride Is suddenly exposed to air. The temperature of Ignition Is highly dependent on the ratio of surface area to volume and the degree of surface cleanliness. Generally, spent fuel rod cladding Is covered with a'relatively thick oxide layer (20-100im), therefore, unless ballooning and burst occur in the cladding during heatup, clean high-temperature Zircaloy metal will not be exposed to air In an SFP accident. However, if there Is cladding failure by ballooning and burst (expected to occur over a temperature range of 700-850 OC), hot oxide-free clean metal will be abruptly exposed to air. Zirconium hydride is expected to dissolve into the metal matrix during the slow heatup to these temperatures. At the moment of burst, some clean surface area of Zr metal will be exposed to air in the location of the rupture. Although data applicable to this situation is quite limited, considering the relatively small surface-to-volume ratio of the exposed metal, likelihood of ignition and subsequent propagation of the burning front of Zr metal Is believed to be small (Ref. 8).

In the CODEX tests annular cladded fuel von a 9 rod bundle) were heated with an Inner tungsten heater rod to examine fuel degradation, with preoxidized cladding, In an air environment.

Zircaloy oxidation kinetics were evaluated as well as the oxidation of the fuel. In the CODEX AIT-1 test the early phase of the test Involved creating a preoxidation using an argon-oxygen mixture. The Intent was to achieve a controlled preoxidation at a temperature of 900-950 'C, but It was reported (Ref. 3) that preoxidation was started at a slightly higher temperature than planned. What subsequently occurred was an uncontrollable temperature escalation up to approximately 2200 C before it was cooled with cold argon flow. After restabilization of the rods at 900 C air Injection was started, electrical heatup commenced, and a second temperature escalation occurred. In the CODEX AIT-2 test, designed to proceed to a more damaged state,-the preoxidation phase was conducted In an argon/steam mlxturb at 820 C and 950 'C (a malfunction occurred during the preoxidation phase resulting In the admission of a small airflow as well). No temperature escalation was seen during the preoxidaton phase.

Following the restabilization of the fuel rods, a linear power increase was started and a temperature excursion subsequently occurred.

In addition to examining relevant test data RES also looked at determining a temperature based threshold for temperature escalation in an air environment by determining equivalent heat generation from steam transient tests. In this exercise we posited that at equivalent heat generation rates, I.e., accommodating different reaction rates and different heats of reaction for air and steam, we should be able to predict the corresponding temperature for escalation In air based on temperature escalations seen In severe fuel damage tests conducted In steam. Using this approach, the heat generation rate was estimated, assuming parabolic kinetics, and the following equation for a rate constant In air:

Appendix 1B A13-

___ MEMMUMM

. kP=52.67 exp(-17597MT) {kghm' 2-sec [f[rate constanit of 0~mass produced]

It was predicted that based on an escalation temperature of 1200 'C Insteam (observed In many of the steam tests), the equivalent heat generation rate In air would produce a

. temperature escalation at approximately 925 *C. The above equation for air kinetics was identified in Reference 3 as the best fit forthe CODEX AiTtest data, i.e., it provided the best agreement to the temperature transient Inthe peak position. For steam kinetics, the rate equation used In MELCOR was selected for calculating the heat generation rate. The prediction

• . of an escalation temperature in air using this approach seems to conform quite well with the observed behavior in the transient CODEX tests and lends further credence to the relative effect

.* of oxidation Inair with respect temperature escalation. The assumption of parabolic kinetics Is routine in oxidation calculations and has been shown to provide a good match with a wide spectrum of experimental data even though, over select temperature ranges, deviations from

• that formulation have been observed. At temperatures above 900 "C, the reaction rate Inair Is high, regardless of whether parabolic or linear kinetics is assumed at that point and distingUishind between the rates of escalation is unimportant for our purposes. *

• - 3. ACCEPTANCE CRITERIA Inassessing a temperature criterion for escalation of the oxidation process and subsequent temperature escalation it Is necessary to reconsider the intended uses of the criterion: 1)to evaluate the decay time after which the fuel heatup, Inthe case of complete fuel uncovery, leads to reaching that temperature at 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> and 2)to evaluate the decay time after which the fuel

.heatup, in the case of complete uncovery wiYl never exceed the temperature criterion.

On balance It appears that a reasonable criteria for the threshold of temperature escalation inan air environment is a value of approximately 900 -C. This value issupported both by limited experimental data as well as by Inference from the more abundant steam testing data. While certain weight gain data indicate the onset of a break away Inthe oxidation rate at lower temperatures after.a period of 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />, this additional time period then exceeds the Ume interval for which the first use of the criterion is intended. With regard to the second use of the criterion, determination of the point at which severe fuel heatup is precluded, the onset of

. *

• breakaway Indicated In certain tests Indicates that the temperature criterion should be lowered to 800 "C. It Is important to stress that, Inboth instances, the temperature criteria should be used together with a thermal-hydraulic analysis that considers heat generation (i.e., decay heat and zircaloy reactions) and heat losses. For the second use of the criterion, i.e., establishing a threshold for precluding escalation, the analysis must demonstrate that heat losses, through convection, conduction and radiation, are sufficient to stabilize the temperature at the value selected.

' 'In the case of slow, complete draining of the pool, or partial draining of the pool it is appropriate to consider use of a highertemperature criterion for escalation, perhaps as high as 1100 to 1200 'C. This would be appropriate if the primary oxidation reaction was with steam. Such a temperature criterion is relevant for the first Intended use of the criterion, determining the point at which the temperature Is not exceeded for 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />, however it Isnot appropriate for use as a

• long-term equilibrium temperature since over long Intervals at such high temperature, one might

. reasonably expect significant fission product releases.

Appendix IB AlB-4 I

In addition to comments on the selection of an ignition temperature, the staff received comments related to the effect of intermetallic reactions and eutectic reactions. With respect to intermetallic reactions, the melting temperature of aluminum, which is a constituent In BO*RAL poison plates in some types of spent fuel stolage racks, is approximately 640 *C. Molten aluminum can dissolve stainless steel and zirconium In an exothermic reaction forming intermetallic compounds. In the SFP configuration, zircaloy cladding will be covered with an oxide layer and unless significant fresh metal surface Is exposed through exfoliation there will be no opportunity to interact-metallic zircaloy with aluminum (which similarly will be oxidized).

Aluminum and steel will form an Intermetallic compound at a temperature of 1150 *C, (Ref. 5) which is above the temperature criterion selected for fuel damage.

Besides intermetallic compounds, eutectic reactions may take place between pairs of various reactor materials, e.g., Zr-lnconel (937 C), Zr-steel (937"C), Zr-Ag-In-Cd (1200 'C), Zr-B4 C (1627 *C), steel-B 4 C (1150 OC), etc. (Ref.6 ). Consideration of eutectics and intermetallics is important from the standpoint of heat addition as well as assuring the structural integrity of the storage racks and maintaining a coolable configuration. Noting the eutectic and intermetallic reaction temperatures, however it does not appear that formation of these compounds Imposes any additional temperature limit on the degradation of cladding in an air environment.

Since the temperature criterion is also a surrogate of sorts for the subsequent release of fission products it Is useful to consider the temperature threshold versus temperatures at which cladding may fall and fission products be released. Cladding is likely to fail by ballooning and burst In the temperature range of 700-850 'C, resulting In the release of fission products and fuel fines. At burst, clean Zircalloy metal will also be exposed, leading to an Increase In oxidation although the total amount of metal Involved will be limited. Creep failure of the cladding at or above 600 "C Is also a possibility. This temperature limit Is roughly associated with the 10 hour1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> creep rupture time (565 TC) which has been used as a regulatory limit. While failure of the cladding at these lower temperatures will lead to fission product release, such release is considerably smaller than that assumed for the cases where the temperature criterion is exceeded and significant fuel heatup and damage occurs. Low temperature cladding failures might be expected to produce releases similar to those associated with dry cask accident conditions as represented in Interim Staff Guidance (ISG)-5. This NRC guidance document prescribes release fractions for failed fuel (2x104 for cesium and ruthenium and 3x104 for fuel fines). Use of these release fractions would reduce the estimated offaite consequences dramatically from the fuel melt cases, early fatalities would be eliminated and latent cancer fatalities would be reduced by a factor of 100. As the temperature limit is increased from 600 'C to 900 'C there is some evidence that ruthenium releases would be increased based on ORNL test data from unclad pellets. Canadian data indicate though, that in the case of clad fuel the ruthenium release did not commence until virtually all of the cladding had oxidized. By this point It might be surmised that the fuel configuration would more closely resemble a debris bed than Intact fuel rods. Selection of a temperature criterion for fuel pool damage also depends on the intended use, i.e., whether it Is Intended as the criterion for the 10 hour1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> delay before the onset of fission product release or whether It is being used as a threshold for long-term fission product release. If the criteria is being used to judge when 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> are available for evacuation, then it may be argued that a higher temperature could be adopted, one associated with the significant release of fission products,1200 'C, since the release of fission products at lower temperatures will likely be small. However, in air it may be that the oxidation rate above 900 "C is sufficient to Appendix I B AIB-5

@I reduce the additional time gained to reach 1200 *Cto a relatively small amount. Selection of a*

temperature criterion for long-term fuel pool Integrity needs to consider that ruthenium release rates, In air, become significant at approximately 600-800 IC, based on the data of Parker et at.

0 (Ref. 7).

Selection of an acceptance 6ritenion for precluding significant offsite release after roughly 5 years, should also consider that ruthenium with a 1 year half life will be substantially decayed and that at 5 years cesium (and perhaps fuel fines such as plutonium) will dominate the dose calculation. 'For these reasons RES believes that the long-term viability of the pool in a completely drained condition (air environment), if it concerns time periods of approximately 5 years, pool degradation should be assessed for a temperature of approximately 800 "C...

Again, an analysis needs to be performed to demonstrate that at that temperature an equilibrium condition can be established. While this would result Inan offsite release, there would be substantial time available to take corrective action after a 5 year decay time for the most recently loaded fuel. If shorter decay time periods are proposed for achieving the longterm equilibrium temperature criterioh, then the Impactof ruthenium releases would dictate reconsideration of this value.

4. SUMMIARY In summary, we conclude that for assessing the onset of fission product release under transient conditions (to establish the critical decay time for determining avaflabirity of 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> to evacuate) it Is acceptable to use a temperature of 900OC if fuel and cladding oxidition occurs in air. If steam kinetics dominate the transient heatup case, as It would In many boildown and draindown scenarios, then a suitable temperature criterion would be around 1200 'C. For establishing long-term equilibrium conditions for fuel pool integrity during SFP accidents which preclude significant fission product release It is necessary to limit temperatures to values of 600 *C to 800 '*C. If the critical decay time Issufficiently long (>5 yrs) that ruthenium inventories have substantially decayed then It would be appropriate to consider the use of a higher temperature, 800 *C,otherwise fission product releases should be assumed to commence at 600 C. These cases are marked by substantial time for corrective action to restore cooling and prevent smaller gap type releases associated with early cladding failures. A tabulated summary of the suggested criteria Is listed below.

Precluding Large Precluding Large Adequacy of 10 hrs Release Release for Evacuation Fuel c5yrs Fuel'>5yrs Dorminant Air Environment 900 "C 600 'C 800 'C Dominant Steam Environment 1200 *C N/A NWA The degradation7of fuel during SFP accidents Isan area of uncertainty since most research on severe fuel degradation has focused on reactor accidents in steam environments. Because of this uncertainty, we have tended to rely on the selection of conservative criterion for predicting Appendix 1 AIBS-I

the global behavior of the SFP. it is our recommendation that the modeling of SFP accidents be performed with codes capable of calculating the heat generation and losses associated with the range of accidents, Including phenomena associated with both water boiloff and air circulation.

Further, the calculation of critical decay times for establishing both the validity of ad hoc evacuation and precluding fission product release must also Include consideration of the exothermic energy of reactions (I.e., reacions with air and steam) with cladding, or alternatively ;

demonstrate that such energy contribution Is negligible in comparison to decay heat at that point. Severe accident codes, such as MELCOR, developed for modeling the degradation of reactor cores, would seem to be a reasonable approach for analysis of Integral behavior and would possess the general capabilities for modeling liquid levels and vapor generation, air circulation, cladding oxidation and fission product release. Use of a severe accident code also facilitates the use of self consistent modeling and assumptions for the analysis. The proper calculation of fission product releases depends In large part on the prediction of thermal-hydraulic conditions. More detailed CFD modeling would Improve the calculation of boundary conditions for air circulation and could be used In conjunction with integral codes to better evaluate convective cooling. The kinetics of cladding reactions should be confirmed with .

experiments designed to simulate the range bf conditions of interest under steady state and transient heating. The experimental database on ruthenium releases under conditions applicable to SFP accidents is Inadequate and we are currently extrapolating data from conditions which tend to maximize such releases.

While there is uncertainty in the analysis of spent fuel degradation, especially for the conditions of air ingression, it Is also true that elements of the analysis contain conservatism. The assumption of 75-100 percent release of ruthenium initiated at lower temperatures is based In large part on tests with bare fuel pellets, testing of cladded fuel indicates that the cladding acts as a getter of oxygen limiting release of ruthenium until virtually all of the cladding has oxidized.

Further, before significant ruthenium release occurs (in Its more volatile oxide form) the surrounding fuel matrix must be oxidized. During transient-heatup of an SFP with temperature escalation one would expect the ruthenium release to follow the oxidation of the cladding at which point the fuel would more likely resemble a debris bed (the seismic event may also contribute In that regard) limiting the release fraction. The competition between formation of hyperstoichiometric U0 2 and U30, may also limit the release fraction below that seen In the data.

The use of a temperature criterion of 600 IC to preclude significant fission product releases is conservative In that it is based In large part on data that discounts the effect of cladding to limit releases. The cladding failures at low temperatures will still allow substantial retention of fuel fines and the presence of unoxidized zircalloy will prevent formation of volatile forms of ruthenium. More prototypic experimental data on releases under these kinds of conditions may reveal that the onset of significant releases, especially ruthenium, would not occur under SFP accident conditions until fuel rod temperatures reached much higher temperatures associated with complete oxidation of the cladding.

Use of the hottest fuel assemblies to predict global release of fission products from the entire spent fuel Inventory Is a significant conservatism as well. Transient fuel damage testing indicates that at the time of local temperature escalation not all of the rod bundle undergoes rapid heating, cooler regions can avoid the oxidation transient. Prediction of the propagation of the temperature escalation to the cooler regions of the pool needs to be carefully examined to see if significant benefit can be gained, at a minimum it will lengthen the period of fission product Appendix 1S A1B-7

111111111111110 release reducing the concentration of activity in the plume of fission products for offslte consequence analysis.

5. REFERENCES

1. Benjamin, A., et al., Spent Fuel Heatup Following Loss of Water During Storage,'

NUREGICR-0649, SAND77-1371, Sandia National Laboratories, 1979.

2. Haste, T., et al., lin-Vessel Core Degradation In LWR Severe Accidents,' EUR 16695 EN, 1996.

3. Shepherd, I., et al., *Oxidation Phenomena in Severe Accidents- Final Report,'

INV-OPSA(99)-P0008, EUR 19528 EN, 2000.

4. Evans, E., et al., 'Critical Role of Nitrogen During High Temperature Scaling of Zirconium,'

Proc. Symp. Metallurgical Society of AIME, May 1972.

6. Fischer, S. and M. Gruebelich, 'Theoretical Energy Release of Therrnites, Intermetallics, and Combustible Metals,' 24'h International Pyrotechnics Seminar, July 1998..

6. Gauntt, R.; et al, 'MELCOR Computer Code Manuals Version 1.8.4," NUREG/CR- 6119, SAND97-2398, July' 1997.

7. Powers, D. et al., WA Review of the Technical Issues of Air Ingression During Severe Accidents,- NUREG/CR-6218, SAND-0731, September 1994.

8. Letter from H. M. Chung to J. Flack, dated August 24, 2000.

9. Leistikow, S. and H. V. Berg, *Investigation under Nuclear Safety Aspects of Zircaloy-4 Oxidation Kinefics at High Temperatures in Air,' Second Workshop of German and Polish Research on High Temperature Corrosion of Metals, Juelich, Germany, December 2-4, 1987.

Appendix 1B AIB-8 I

UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of: )

)

DUKE ENERGY CORPORATION )

) Docket Nos. 50413-OLA (Catawba Nuclear Station, ) 50414-OLA Units 1 and 2) )

)

)

- r-.CERTIFICATE OF SERVICE _

I hereby certify that copies of "DUKE ENERGY CORPORATION'S FOURTH SUPPLEMENTAL RESPONSE TO BLUE RIDGE ENVIRONMENTAL DEFENSE LEAGUE'S FIRST SET OF DISCOVERY REQUESTS TO DUKE REGARDING BREDL SECURITY CONTENTION 5" in the captioned proceeding have been served on the following by deposit in the U.S. mail, first class, this 4th day of January, 2005. Additional e-mail service, designated by *, has also been made this same day, as shown below.

Ann Marshall Young, Chairman* Anthony J. Baratta*

Administrative Judge Administrative Judge Atomic Safety and Licensing Board Atomic Safety and Licensing Board U.S. Nuclear Regulatory Commission U.S. Nuclear Regulatory Commission Washington, DC 20555-0001 Washington, DC 20555-0001 (e-mail: AMYnrc.gov) (e-mail: AJB5@nrc.gov)

Thomas S. Elleman* Office of the Secretary*

Administrative Judge U.S. Nuclear Regulatory Commission 5207 Creedmoor Road, #101 Washington, DC 20555 Raleigh, NC 27612 Attn: Rulemakings and Adjudications Staff (e-mail: ellemangeos.ncsu.edu) (original + two copies)

(e-mail: HEARINGDOCKET.nrc.gov)

Office of Commission Appellate Adjudicatory File Adjudication Atomic Safety and Licensing Board Panel U.S. Nuclear Regulatory Commission U.S. Nuclear Regulatory Commission Washington, DC 20555 Washington, DC 20555

Susan L. Uttal, Esq.* Diane Curran*

Antonio Fernandez, Esq.* Harmon, Curran, Spielberg &

Shana C. Zipkin, Esq.* Eisenberg, LLP Office of the General Counsel 1726 M Street, N.W.

U.S. Nuclear Regulatory Commission Suite 600 Washington, DC 20555 Washington, DC 20036 (e-mail: slu~nrc.gov) (e-mail: dcurran~haimoncurran.com)

(e-mail: axf2@nrc.gov)

(e-mail: scz~nrc.gov)

Mark J. Wetterhahn Counsel for Duke Energy Corporation 2