ML20059E499

Transcript of 931026 Briefing in Rockville,Md Re Severe Accident Research Program.Pp 1-78.Supporting Documentation Encl

ML20059E499
Person / Time
Issue date:	10/26/1993
From:	NRC COMMISSION (OCM)
To:
References
REF-10CFR9.7 NUDOCS 9311030241
Download: ML20059E499 (121)

Text

{{#Wiki_filter:_ MM%%%'W6WWWW6W6W9VWW6%W6%%WrV;VtV6%%@6%%iflfggggg h 3

AMSMIT'A1. TO:

M Occument Control Dest. 016 Phillips G C f s h iDVANCED COPY TO: The Public Occument Rocm p /0/5F/h3 h CATE: h [ FROM: SECY Correspondence & Records Brancn g k k Attacneo are copies of a Comission meeting transcript and relatec meeting f-l dccument(s). They are being forwarced for entry on the Daily Accession List and g placement in the Public Document Room. No other cistribution is recuested or g g recu1reo. g g 4. . reb [ Meeting

Title:

/bxA[ d-u _. /

e. ex. e__.

h keNue _, a Meeting Date: / /,7 6/9.3 Open X Closed 8 s M U 3 f ! tem Cescriptien n Copies Advanced DCS C

Copy,

{

• s E

to POR o ,5 t k h c

1. TRANSCRIPT 1

1 g u)bsNe w u.e 6 ) d E, i. 'e 3 k 2. s5 i l k. 3- = =c E =2 i: 4 c = w w m c 5 s4 290002 5. ~* B r e 9

• DCR is advanced one copy of eacn document, two of each SECY paper.

[ 8 k C1R Brancn files the original transcript, witn attacnments, withcut SECY 7 l* p', (1 jh [ =g

a ers.

9311030241 931026 PDR 10CFR PT9.7 PDR gyng 8'm===a'OiiiiiW##A%%%' 'ggf^kggh'd"h"N'fi. ^

l UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMIS SION i t a l Title ~ BRIEFING ON SEVERE ACCIDENT RESEARCH PROGRAM 3 LOCatiOL: ROCKVILLE, MARYLAND l 1 h3I6*. OCTOBER 26, 1993 i Pages: 78 rAGts i I i NEALR.GROSSANDCO.,INC. COURT REPORTER $ AWD TRANSCRIDERS 1323 Rhode-Island Avenue, Northwest Washington, D.C. 20005 (202) 234-4433-i

DISCLAIMER This is an unofficial transcript of a meeting of the United States Nuclear Regulatory Commission held on October 26

1993, in the Commission's office at one White Flint North, Rockville, Maryland.

The meeding was open to public attendance and observation. This transcript has not been reviewed, corrected or edited, and it may contain inaccuracies. The transcript is intended solely for general informational purposes. As provided by 10 CFR 9.103, it is not part of the formal or informal record of decision of the matters dise'ssed. Expressions of opinion in this transcript do not necessarily reflect final determination or beliefs. No pleading or other paper may be filed with the commission in any proceeding 'as the result of, or addressed to, any statement or argument contained herein, except as the Commission may authorize. HEAL R. GROS $ CoutT tipotTER$ AHO TRANSCRIttt$ 1313 kHo06 ISLAMO AY9MUS. N.W. 4 (202) 234-4433 WA3mMGToN. 04 20003 (202) 232 4 800 l 1 )

1 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION-b BRIEFING ON SEVERE ACCIDENT RESEARCH PROGRAM a PUBLIC MEETING Nuclear Regulatory Commission One White Flint North Rockville, Maryland Tuesday, October 26, 1993 The Commistion met in open

session, pursuant to

notice, at 10:00 a.m.,

Ivan

Selin, Chairman, presiding.

COMMISSIONERS PRESENT: IVAN SELIN, Chairman of the Commission KENNETH C. ROGERS, Commissioner FORREST J. REMICK, Commissioner E. GAIL de PLANQUE, Commissioner NEAL R. GROSS COURT REPORTERS AND TRANSCAIDERS 1323 RHODE ISLAND AVENUE, N W. (202) 234 4433 WASHINGTON. O C. 20005 (202) 234-4433

j i 2-s STAFF SEATED AT THE COMMISSION TABLE: SAMUEL J. CHILK, Secretary KAREN CYR, Office of the General Counsel' JAMES TAYLOR, Executive Director for Operations t ERIC BECKJORD, Director, Office of Research THOMAS MURLEY, Director, NRR BRIAN SHERON, Director, Division of Systems Research, e RES ASHOK THADANI, Director, Division of System Safety and Analysis, NRR i FAROUK ELTAWILA, Chief, Accident Evaluation Branch, RES i t i f j i l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234-4433 WASHINGTON D.C. 20005 (202) 234 4433 ~

3 1 P-R-0-C-E-E-D-I-N-G-S 2 10:00 a.m. 3 CHAIRMAN SELIN: Good morning. 4 We're pleased to welcome members of the-5 staff to brief the Commission on the Severe Accident 6 Research Program. The goal of this research is to 7 achieve an adequate level of understanding of severe 8 accident phenomena in order to allow regulatory 9 decisions to be made on these severe accident issues.. 10 Understanding severe accident phenomena is 11 necessary for evaluating the extent to which a plant 12 has design features which will prevent these severe 13 accidents or, if necessary, to mitigate their 14 consequences. 15 As most of you know, the NRC has sponsored 16 an active research program on severe accidents in 17 light wate-reactors for quite a few years. 18 Substantial benefits have also been derived from our 19 cooperative agreements on severe accident research 20 with other countries. The focus of the research has 21 been on specific severe accident phenomena that could 22 result in early containment failure and also on code 23 development. 24 Today's briefing will-focus on staff 25 accomplishments since the issuance of its last program NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N.W (202) 234 4433 WASHINGTON, D.C. 20005 (202) 234-4433

4 1 plan update in September 1992. The briefing will 2 cover the current status of the severe accident 3 issues, as well as closure plans for issues that have 4 not yet been closed. The research results are also 5 being used in support of the certification reviews for 6 the evolutionary and for the advanced light water 7 reactors. So, the staff will discuss not only the 8 research program itself and its results, but also the 9 use of those results in our regulatory programs. 10 Copies of the viewgraphs are said to be 11 available at the entrances of the room. 12 Commissioners? 13 Mr. Taylor, would you proceed? 14 MR. TAYLOR: Good morning. With me at the 15 table from the Office of Research, Eric Beckjord, 16 Brian Sheron, Farouk Eltawila. From NRR, Doctor Tom 17 Murley and Ashok Thadani. 18 Mr.

Chairman, I won't repeat.

You've 19 outlined the basis upon which we do most of our work 20 in the severe accident arena and its contributions to 21 the review of the evolutionary and passive reactors. 22 But as you'll hear from the staff this morning, this 23 program has undergone a significant revision in its 24 scope and priorities since being set up in 1987. A 25 number of key issues associated with our understanding NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE N W. (202) 234 4433 W ASHINGTON. D.C. 20005 (202) 2344433

S 1 of severe accidents have been resolved over the last 2 few years and current plans are to resolve the 3 remaining issues by 1996. 4 I'm pleased with the scope of the program 5 that is currently ongoing and also the progress. 6 With that, I'll turn.the meeting over to 7 Eric, who will continue. 8 DOCTOR BECKJORD: Thank you. 9 (Slide) Mr.

Chairman, Commissioners,

.0 slide 2 shows the presentation outline. I'm going to 11 give a background in brief. Doctor Sheron will review 12 the revised Severe Accident Research Program and the 13 update since September 1992. Doctor Eltawila will 14 give a presentation on fuel coolant interactions and 15 debris coolability. Doctor Sheron will then review 16 the utilization of the research results, plans for l 37 future work in the international programs, and if we -l 18 have some time Doctor Eltawila will come back to the 19 additional information. 20 (Slide) If I could have the third slide, 21 please. 22 This is the background. Severe accidents 23 are important safety considerations for several I 24 reasons. First, we're able to calculate in many cases 25 severe accident loads from first engineering i NEAL R. GROSS COUA' -EPORTERS AND TRANSCAIBERS 132; MODE ISLAND AVENUE. N W. (202) 2344433 WASHINGTON. O C. 20005 (202) 2344433

6 1 principles, and they can exceed design limits in some 2 cases. Also, we are able to estimate the frequency of 3 severe accidents by means of probabilistic analysis 4 and equipment failure rates that are derived from 5 operating experience. The combination of these two 6 flows of information by means of what we call the=PRA 7 shows that severe accidents are major contributions to 8 risk. We know this from WASH-1400, from the update a 9 few years ago of NUREG-1150. We know it from PRAs 10 that have been done on individual plants over a period 11 of 15 years and we also know it from the experience of 12 Three Mile Island Unit 2. 13 The margins, as to margins in existing 14 plants, we have studied the f ailure of primary systems 15 and subsequent possible containment failures and we 16 are able to evaluate the margins in plant designs to 17 accommodate severe accident loads, the frequencies of 18 the challenge and the probability given the challenge 19 that the containment would fail. 20 The containment, the importance of 21 containment has been recognized for a long time. I 22 would say it goes back to certainly the Brookhaven 23 Report, WASH-740 which came out many, many years ago. I think it was 1957 and this was clearly demonstrated 24 25 at Three Mile Island. NEAL R. GROSS COURT REPORTERS AND TRANSCRtBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234-4433 WASHINGTON, D C 20005 (202) 234-4433

7 1 The source term, which is listed here, has 2 been investigated very' thoroughly. After Three Mile 3 Island, questions about the size of the source term 4 came up and spurred an extensive review which was 5 carried out by an office, the source term office 6 established by the Commission. The conclusion was 7 that the source term magnitude and the composition of 8 radionuclides depends on accident sequences and on 9 severe accident phenomena. It is not a single value 10 for a reactor type and it had to be determined 11 sequence by sequence. 12 As a consequence on that conclusion, the 13 Severe Accident Research Program was undertaken in the 14 early 1980s and its purpose was to develop an 15 understanding of severe accident phenomenon and to 16 provide a technical basis for regulatory decisions. 17 In 1985, the Commission approved a severe 18 accident policy statement. That statement, in brief, said that based on then current knowledge of operating 19 O 20 plants that there was no undue risk posed by the 21 operation of those plants. However, the Commission l 22 called in that statement for a systematic evaluation l 23 of all operating plants to identify the 24 vulnerabilities to severe accidents that could be 25 corrected, all vulnerabilities and with focus on those NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 2344433 WASHINGTON, D.C. 20005 (202) 2344433

8 1 that could be corrected in a cost effective way. 2 The next item is NUREG-1150. That was an 3 assessment of five nuclear power plants which, in 4

effect, updated the technology of doing risk l

5 assessment from the WASH-1400, making use of the 1 i 6 available results from the Severe Accident Research l 7 Program, which were beginning to come forth in the 8 1986 to '88 period. In 1988, there was issued a SECY 9 paper, 88-147, on the subject of the individua: plant 10 examination, the Containment Performance Improvement 11

Program, accident management, the Severe Accident 12 Research Program and the use of operating experience.

13 In offect, this document described how the Severe ) 14 Accident Research Program was being focused on the 15 vulnerabilities of severe accidents that had important 16 risk significance. In particular, the core damage 17 frequencies and the containment f ailure mechanisms and 18 risks. 19 There have been several revisions of the 20 Severe Accident Program since that time. The latest 21 is NUREG-1365 and that is the December 1992 update and 22 Doctor Sheron will take over to describe that to you. 23 DOCTOR SHERON: Thank you. i 24 (Slide) The next slide, please. i 25 As Jim and Eric said, in 1988 we undertook i ) i NEAL R. GROSS { COURT REPORTERS AND TRANSCRIBERS d 1323 RHODE ISt.AND AVENUE, N W (202) 2344433 WASHINGTON. D.C. 20005 (202) 234-4433

9 1 a complete reexamination of the Severe Accident 2 Research Program. Prior to that it had focused 3 priricipally on just generating general information 4 about severe accidents. What we wanted to do with 5 this reexamination was to refocus it on those areas 6 where there were larger remaining uncertainties and on 7 the phenomena which seemed to be posing the highest a 8 risk in ter of severe accidents. 9 We did this reexamination and went through 10 a rather extensive review process with the ACRS and 11 the NSRRC. It was documented in NUREG-1365 and that 12 was issued in August of '89. So, you can see this 13 took us about a year to really refocus the program. 14 (Slide) Next slide, please. 15 The refocused program tried to focus in on i 16 four major areas. One was we tried to emphasize 17 resolving uncertainties associated with scenarios that 18 produced early containment failure because these are 19 the scenarios which lead to the highest risk. The two 20 major issues there were the Mark I -- the BWRs with 21 the Mark I containment, the liner failure issue. What 22 this was was basically that in a core melt accident 23 you would postulate the melt, after it is released 24 from the vessel would spread through the pedestal 25 region into the dry well and contact the liner of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234 4 33 WASHINGTON, D C. 20005 (202) 234 4 33 -

10 1 containment. It was postulated that this would then l 2 burn through or fail the liner, which then gave you a j i 3 direct pathway to the environment. The question was 4 would this liner fail under these conditions. 5 The other main issue we were focusing on 6 is the direct containment heating. This basically 7 pertains to PWRs and a few postulated core melt 8 accident with the primary system remaining at a 9 relatively high pressure. When the vessel fails the 10 high pressure steam that is in the vessel would drive 11 the core material out into the key way, up into 12 intermediate and perhaps even into the main 13 compartments of the containment. This high 14 temperature melt would interact very quickly with the 15 containment atmosphere, causing an over pressurization 16 and a failure. 17 Other areas were the source term research 18 we were using to update the source term, core concrete 19 interactions which generate the non-condensible gases 20 basically that would cause a slow over pressure 21 failure of the containment, hydrogen combustion which 22 also contributes to the containment loads. 23 The other thing we focused on. was our 24 computer codes. At that time when we took a look at 25 our suite of codes there were numerous codes which NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W (202) 234 4 33 WASHINGTON. D.C. 2000ii (202) 234 4433

11 1 seemed to duplicate one another. They were actually 2 calculating the same thing or nearly.the same thing, 3 they were just being done at.dif ferent laboratories or 4 dif ferent groups. Our objective was to eliminate any 5 vaplication and just have one code to do a single job. 6 The other thing we also discovered was -i 7 that many of our severe accident' experiments really 8 had not put together a very rational scaling analysis. 9 By this I mean that we wanted to be able to take the 10 results from smaller scale experiments and extrapolate 11 them up to a large plant. The detailed scaling 12 analysis, the dimensionless groups, whatever, that 13 would provide that rationale had really not been 14 developed to the extent that we would like. So, we [ t 15 focused in on developing a scaling rationale. 16 Then also, some of our other research was 17 refocused. One area is the melt progression, the core 18 melt progression, which we'll talk more about, fuel 19 coolant interactions, which can provide a source of a 20 energy to possibly fail certain parts of the primary 21 system and so forth, debris coolability which is an 22 issue not only from the standpoint of ' accident 23 me 3gement, also for the advanced passive reactors, 24 and then hydrogen combustion. 25 (Slide) The next slide. NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 AMODE ISLAND AVEt*JE, N W (202) 234-4433 WASHINGTON. D C 20005 (202) 234 4433

l 12 1 This just is a little cartoon which tries 2 to illustrate where the loads on a containment come 3 from during a severe accident. As you can see, 4 pressure is the driving force. What we're worried 5 about is really over pressurizing containment beyond 6 its ultimate strength and failing it. The things that 7 can contribute to increasing the pressure obviously is 8 if you put a break in the primary system you'll 9 release steam and possibly hydrogen from the primary 10 system. Melt, as it relocates to the lower part of 11 the vessel and perhaps out in -- if it fails the 12 vessel into the containment, will interact with the 13 concrete base mat. It will generate non-condensible 14 gases, hydrogen. Also, the melt itself can interact 15 with the containment atmosphere producing heating and 16 a pressurization. All of these contribute to 17 pressurizing tisc containment and ultimately to its 18 possible failure. 19 (Slide) Next slide, please. 20 In the intervening period from 1988 to 21 '92, after we had revised the program and refocused 22 it, we in f act completed research on a number of these 23 issues that I just mentioned. The first one being the 24 Mark I liner failure issue. The results of that 25 research demonstrated that the Mark I containment NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 2344433 WASHINGTON, D C. 20005 (202) 2344433

? 13 1 integrity would be maintained provided water was added 2 to cover the core debris. This was documented in 3 NUREG/CR-5423 and it underwent a peer review, a very i 4 extensive peer review. As a result of that peer 5 review, we identified four areas that required some 6 confirmatory research to confirm the conclusions of 7 the report. We set forth and did that research, 8 carried it out, completed it. The results of that 9 research were incorporated back into the report. It 10 was then peer reviewed a second time to make sure that 11 we had, in fact, addressed the issues of the first 12 peer review. It was and the final report, NUREG/CR-13 6025, was issued in October of '93. 14 And we have taken the results of. that 15 research and we are using it as we review the IPEs 16 right now. So, as we review IPEs, in particular the 17 plants with the Mark I containments, we are looking at 18 how they were dealing with the question of having 19 reliable water supplies to provide water to the dry 20 well. 1 21 (Slide) On the severe accident scaling l 1 22 methodology, we had a task group that was put together 23 consisting of a number of experts in the country which 24 met over a period of -- I guess it was well over a 25 year, to put together this methodology. They applied NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W, (202) 2344433 WASHINGTON. D C. 20005 (202) 2344433

4 14 1 it as an example to the direct containment heating 2 issue and it was done successfully and we now have a 3 NUREG report, NUREG/CR-5809, which is sort of a 4 guidance document for all of our contractors to 5 follow. For any new experiments that are being 6 considered, they have to do a thorough scaling 7 analysis to show why the results of that experiment 8 can in fact be extrapolated up. 9 On source term

issues, our research 10 program is essentially complete with regard to the 11 quantity of the source term, the timing of it, its 12 release and the chemistry or its composition.

One 13 thing we are doing right now is participating in the 14 Phebus project in France which you may know something 15 about. That's a large scale, an integral experiment 16 designed to look at the source term behavior not only 17 in the primary system but as its transported basically 18 from the core to the containment. 19 In terms of our source term research, we 20 are planning on updating the source term. There's a 21 SECY-92-127, which i believe has already ' been sent 22 down to the Commission last year, and we have a final 23 report, NUREG-1465, which we plan to publish at the 24 end of '93. 25 COMMISSIONER REMICK: Brian, before we go NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 2344433 WASHINGTON. D C. 20005 (202) 234-4433 j

15' 1 on, I have a couple questions. 1 2 DOCTOR SHERON: Yes, sir. 3 COMMISSIONER REMICK: Do we know if other i 4 groups are using our scaling methodology that we've 5 developed here? ?' o 6 DOCTOR SHERON: Other groups, you mean -- ? 7 COMMISSIONER REMICK: International' 4 8 groups. 9 DOCTOR SHERON: International groups. Let 10 me ask Farouk. I 11 Are you aware of any? l 12 DOCTOR ELTAWILA: I'm not aware of any, 13 but we received a request for the report from 14 different people in the -- member of the CSR, but we 15 have not got any clear indication if they have used it 16 or not, i 17 COMMISSIONER REMICK: You indicate that 18 the research program is complete in source term. 19 What's the final resolution? I think there was a 20 difference between EPRI and the staff on the -low 21 volatility radionuclides. Has there been anything 22 more recent? 23 DOCTOR ELTAWILA: I think.that difference i 24 still. exists between us and the reason' for the 25 database that we have is not that complete to_ cover NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234-4433 WASHINGTON. D C. 20005 (202) 234-4433

16 1 all the expected range of severe accident. So, the 2 staff still believe that the source tern that we 3 propose in that NUREG document is still adequate. So, 4 I don't think that difference has been resolved yet. 5 COMMISSIONER REMICK: Nothing on that? 6 At the meeting in Seoul on advanced 7 nuclear power system, Pierre Bashear from, I think, 8 EDF, and some others made some strong statements about 9 the need to get various international regulatory 10 groups to come to closure on source term and there 11 were statements that there are differences of several 12 orders of magnitude between source terms in different 13 countries. Would that be related to results from the 14 Phebus project or would you have any idea what would 15 be the basis of the concerns like that that different 16 regulatory bodies were using source terns that 17 differed by several orders of magnitude? 18 DOCTOR ELTAWILA: I think the main reason 19 I think some of the other countries that they have not 20 developed their database as much as we developed ours 21 in the United States here and they'are relying in a 22 much more conservative estimate in their evaluation 23 until they have their own information. I don't 24 believe that the Phebus project is going to change our j 25 opinion about the source term. I believe it's going i l i NEAL R. GROSS COURT REPO9TERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVENUE, N W. (202) 2.4 4433 WASHINGTON. D C. 2000S (202) 234-4433

.m _.m r 17 1 to confirm that we are choose the source term. The i e 2 source term that we are developed in this country is 3 based on a lot of experimental data and the Phebus as + 4 an integral experiment is not going to assess each 5 individual issue, but rather is going to give us the. 6 overall picture and our ability.in calculating the j 1 7 source term can't confirm that rather than to produce v 8 any new information in that area. 9 COMMISSIONER ROGERS: Could I just -- 10 COMMISSIONER-REMICK: Excuse me. Go 11 ahead, please. 12 COMMISSIONER ROGERS: -- piggyback on that 13 a little bit? 1 14 Earlier you mentioned that the source 15 term, in fact, depends very much on the particular 16 reactor in the scenario that you're considering. The 17 question is the variability in those results for U.S. 18 reactors in different scenarios, is that outside of i 19 this gap between that Bashear is referring to.of I 20 orders of magnitude dif ference between U.S. results 1 21 and estimates done elsewhere. 22 COMMISSIONER REMICK: Yes. I have to say j 23 he did not say U.S. and elsewhere, he just. said 24 between different regulatory bodies, yes. 25 MR. TAYLOR: Maybe when we finish NUREG-NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 234 4433 WASHINGTON, D C. 20005 (202) 2344 433 ~ _,. -

18 1 1465 we.ought to be sure it gets distributed widely 2 and then perhaps have a meeting of people on the 3 subject in an international forum since this 4 represents the best of the efforts that we have had. 5 COMMISSIONER ROGERS: That would sound 6 like a very useful -- 7 MR. TAYLOR: And I think that would be w B very useful among the people who practice severe 9 accident -- 10 COMMISSIONER REMICK: Yes. I did not read 11 his comments with any difference of U.S. necessarily. 12 It was kind of an appeal internationally to the need 13 for regulatory bodies to come to closure and be 14 consistent because otherwise it will lead to.ack of 15 public confidence. 16 MR. TAYLOR: Maybe if we stimulate 17 that description. 18 COMMISSIONER REMICK: Yes. 19 MR. TAYLOR: That sounds to me if we're 20 about to wrap up our review, it would be appropriate 21 to do that. 22 Eric, how do you feel about that? I think 23 that we could make those types of arrangements. 24 DOCTOR BECKJORD: Yes, I think so. Did he 25 give a paper? Is it available? Could we -- NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE.,N W (202) 2344433 WASHINGTON, D C. 20005 4202) 2344433

~. 19' 1 COMMISSIONER REMICK: These were ad hoc 2 comments -- 3 DOCTOR BECKJORD: I see. 4 COMMISSIONER REMICK: -- of a discussion. 5 I'm trying to think if Pierre gave a paper. I'm not 6 sure. I'll check on that. But the comments came as' 7 part of a discussion of an international regulatory e 8 body, a harmonization basically. 9 DOCTOR BECKJORD: The director and some of - l his staff from the CEA's Institute of Safety are here 10 this week and we can talk to them and see if they have 11 12 some insight on this. 13 COMMISSIONER REMICK: Yes. 14 DOCTOR MURLEY: Could I add a point here? 15 I don't think the situation is as divergent as its 16 been portrayed. We've, in the context of the EPRI 17 requirements document, of which EDF.is a member of 18 this, we've resolved the source term basically for 19 evolutionary plants and also we've issued our FSER for e 20 the passive plant. So, we've at least.got a 21 regulatory resolution of how to proceed. 22 There will always'be some loose ends,.of. 23 course, but I ~ don't see it as being a. major issue 24 that's going to affect the designs for the passive or 25 the evolutionary. l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234 4433 WASHINGTON, D.C.'20005 . j (202) 23444M -r y

20 1 COMMISSIONER REMICK: No, that certainly 2 was not the point. In no way do I want to infer that 3 there was a difference with U.S. source term. It was 4 an appeal to international bodies to come together and 5 on what should be the source terms used to have some 6 kind of consistency. He claimed there were several 7 orders of magnitude in different parts of the world. He said nothing specifically about U.S. and indicating 8 9 that this can result in lack of public confidence if 10 one nation are using this source term and another 11 nation are using another and so forth. 12 DOCTOR MURLEY: I do agree that we need to 13 understand any differences. It may be the difference 14 between a source term for design basis accidents and 15 a source term for severe accidents. It may be used 16 differently. 17 COMMISSIONER REMICK: Yes. 18 DOCTOR THADANI: In fact, I think -- as 19 Tom was saying, I think that's where the difference 20 is. What source term do you apply for design basis 21 accident? Do you include in-vessel as well as ex-22 vessel part? As Commissioner Rogers said, that's 23 definitely driven by sequences and the source term 24 will depend on specific sequences. The concept, 25 however, in terms of design basis accident would be a NEAL R. GROSS COURT REPORTERS AND TRANSCRtBERS 1323 RHODE ISLAND AVENUE. N W. (202) 234-4433 WASHINGTON, D C. 20005 (202) 2344433

21 1 recoverable accident. So, perhaps one would only 2 consider the in-vessel aspects. But for severe 3 accidents, where you're looking at containment 4 performance, you do have to look at the sequences and 5 the challenges to the containment and actually what t 6 you would expect in terms of ex-vessel interactions as 7 well. That's the approach we've been using in e 8 evolutionary and passive designs. 5 9 DOCTOR ELTAWILA: Can I add one more 10 thing, just not to belabor the points here? There is 11 an inconsistency in the definition of the source term, t 12 In the United States we define the source term as the 13 amount of fission product in the containment itself. 14 Some other country defines the source term about 15 what's released outside the containment. So, if you 16 take the effect of contaminution due to the 17 containment, that can be the reason for the i 18 differences. So, what we might need is a consistent 19 definition of the source

term, rather than the e

20 quantity of it. 21 DOCTOR SHERON: (Slide) Slide 9, please. 22 In the area of core concrete interactions, 23 we've completed all of our dry core concrete 24 interaction experimental research. By dry I-mean that 25 there is no overlying pool of water. We have NEAL R. GROSS COUAT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N w. (202) 2344433 WASHINGTON D.C. 200d5. (202) 234 4 433

22 1 validated models that are currently available for 2 assessing a plant's response and we are continuing to do additional validation on experiments that are being 3 4 conducted worldwide, both in the U.S. as well as in 5 the Russian Republic. 6 In hydrogen transport and combustion, a 7 major effort has been completed on a deflagration to-8 detonation transition criteria under a variety of 9 conditions. This is one that transitions from burning 10 to a detonation. 11 We have research going on right now on 12 some residual issues related to high temperature 13 hydrogen combustion research. This is going on at 14 Brookhaven National Laboratory and it's a joint 15 program between the NRC as well as NUPEC in Japan. 16 (Slide) Next slide, please. 17 We have just recently completed our TMI-2 18 vessel investigation project. This was -- I guess, as 19 you know, we had a meeting in Boston just a week ago 20 which was the final meeting in which we presented the 21 results. What this program did is it provided a 22 substantial amount of information to help us benchmark 23 some aspects of our core melt progression models. We 24 gained a lot of information about reactor vessel 25 integrity. It gave us significant insights on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS t 1323 RHODE ISMND AVENUE, N W. (202) 234-4433 WASHINGTON, D C. 20005 (202) 2344433

23 1 accident management and, most importantly, it allowed 2 us to evaluate the margin of failure of the TMI 3 vessel. 4 Some of the highlights we had uncovered .i 5 from this program is that we saw a hot spot on the l 6 lower head of the vessel, about a' median diameter. It. 7 reached a temperature of 1100 degrees C for about 30 e 1 8 minutes. But away from the hot spot the vessel 2 9 temperature was below the 727 degree C transition t 10 temperature for the steel and, as a consequence, we -l 11 were able to conclude that from that the vessel was 12 not that close to failing under the conditions that 13 were seen at T'!I. 14 There was a number of other highlights 15 there, I think, from the program. Instrument tube 16 failures, for example, for the TMI type scenario are 17 not considered to be a dominant failure mode. Creep 18 rupture is the more controlling failure mode. The l 19 other thing we inferred from the results is that there 20 is greater heat transfer between the core-debris and 21 the overlying pool of water. There may have been ] \\ 22 water ingress into the melt and that removed much more ) 23 heat from the melt than what would be expected if we 24 did our calculations with our current codes. 25 All of this indicated that there was NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234 4433 WASHINGTON. D C. 20005 (202) 234-4433

24 1 substantially more margin in that vessel than we would 2 have predicted had we just done.an outright 3 calculation with our analysis codes. 4 COMMISSIONER REMICK: Have we attempted to 5 calculate the progression of the melt with our codes, 4 6 what we believed to be the case in TMI and if we did 7 so how accurately were we able to do that? Have we w 8 attempted to use MELCOR or something? 9 DOCTOR BECKJORD:

Yes, we did 10 calculations.

11 Farouk? 12 DOCTOR ELTAWILA: There was an effort 13 under the sponsorship of OECD, a comparison exercise 14 using the different code and we used the SCDAP/RELAP 15 code and the MELCOR code. But that was done about 16 five years ago, an older version of this code. We are 17 planning now after we add all the updates to our code 18 to go over the scenario again and see how well we 19 predicted. 20 COMMISSIONER ROGERS: Just before you 21 leave that

topic, as you know I

was at that 22 presentation last week and I just want to say that I 23 think that whole project was an enormous success. I 24 think it was very well managed and I think the NRC did 25 a superb job in bringing the international effort NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 2344433 WASHINGTON. D C. 20005 (202) 2344433

25 ) I 1 together, getting the support-from overseas that was 2 vital to be able to do the job as well. as it was done, 3 and the quality of the work. I couldn't stay for the 4 .whole meeting, but certainly what I heard was very-5 impressive indeed. I think that it really was a 6 superb piece of work. I think that a great deal of 7 credit should go to NRC's management of that effort in e 8 bringing it all together. Fine effort. 9 DOCTOR BECKJORD: Thank you very much. 10 (Slide) Slide 11, please. 11 Last year, which was around four years 12 since we started to look at our revision to the Severe l 13 Accident Research Program -- as you remember that was ) 14 done in 1988 with the 1365 being issued in 1989 -- we 1 i 15 decided that we ought to take another look and now 16 update and see where we are again. What.we did is we 17 updated it through revision 1 of NUREG-1365 and what 18 this revision done is it identifies the issues that 19 have now been completed or are very near completion. o 20 It describes the progress and our understanding of 21 important severe accident phenomena. It defines the 22 research that would lead to closure of core melt 23 progression, direct containment' heating, which as I 24 said before was an early containment f ailure mode, and i 25 fuel coolant interactions and it includes research NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 2344433 WASHINGTON, D.C. 200$ (202) 2344433

26 1 that would be applicable to the advanced passive light 2 water reactors, especially in the area of containment 3 performance criteria during severe accidents. 4 (Slide) Next slide, please. 5 In the revised Severe Accident Research 6 Plan we're focusing our researcli on three 7 phenomenological issues and on code development and 8 assessment. 9 One thing I want to point out is, and I 10 think I said this a couple weeks ago at our -- I may 11 have mentioned this at the thermal hydraulic briefing, 12 and important aspect of our code development program 13 now is peer review. Each one of our major codes has 14 or is undergoing a peer review. These are with 15 outside experts, not within the agency, _but people 16 that are knowledgeable in the area with a wide variety 17 of backgrounds, both in large codes as well as in 18 phenomena _and so forth. And the objective is to identify deficiencies in the code and to identify any 19 20 necessary further work in order to say these codes 21 have reached a good state of maturity and which we can 22 have confidence in their application to severe 23 accidents. 24 COMMISSIONER ROGERS: How was that being 25 conducted? How was that peer review actually carried NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE., N W. (202) 234-4433 WASHINGTON, D C. 20005 (202) 2344433

37 1 out? Some of these codes are enormous' things that are 2 developed.over many, many years. How are they 3 actually doing a peer review of-the codes? 4 DOCTOR BECKJORD: What we do is -- 5 COMMISSIONER ROGERS: Are they piece by 6 piece? 7 DOCTOR BECKJORD: No. No. It's for the a 8 whole code. Let me give you an example for MELCOR. 9 What we did is we asked Los Alamos National Laboratory 10 to head up the peer review effort. They had one. 11 individual who was acting as sort of the project i 12 manager or the coordinator. We wanted to do this just - 13 so that the NRC was still at arm's length from the f 14 peer review process. The individual then selected 15 peer reviewers and I can't remember how many. 16 Do you remember how many there were.for 17 MELCOR? 18 DOCTOR ELTAWILA: Maybe eight or so. 19 DOCTOR BECKJORD: It was around eight or' s 20 so. They had a wide variety of backgrounds. Some: 21 knew large codes. Some knew certain phenomena better 22 than they knew others. So, we had a real spectrum. 23 They would then go of f and they would conduct a number 24 of -- they would do their review. They would come 25 back, they would periodically meet. During those NEAL R. GROSS court REPORTERS AND TRANSCRIBERS 1323 AHODE ISLAND AVENUE, N W. (202) 234-4433 WASHINGTON. D.C. 20005 (202) 234 4 33

E 28 1 meetings the contractors, the developers of the code 2 would give them presentations on.the

models, 3

verification, whatever information that they would 4 need. Sometimes the peer reviewers would need some 5 additional calculations done in order that they could 6 look at, say, sensitivities. So, our contractor would 7 go back and do those calculations. Then this was all = 8 pulled together again by Los Alamos in the form of a 9 final peer review report, which was then issued. 10 We, the NRC, then took the peer review 11 report. We looked at the recommendations and the 12 conclusions. We had to decide what things were 13 important for the NRC, for regulatory. In other 14 words, you would get a lot of recommendations, for 15 example if some model may not have enough validation _. 16 But the question is was that important for the NRC's 17 work? If it wasn't, then we may not want to go off 18 and spend a lot of time or money working on that one 19 item. 9 20 We then would take the peer review report-21 and we would put together an implementation plan on-22 how we would address the peer review comments. Then 23 we would go forth and do that. 24 Farouk, do you want to add anything to 1 25 that? NEAL R. GROSS COURT REPORTERS AND TRANSCR'BERS 1323 RHODE ISLAND AVENUE, N W. (202) 234-4433 WASHINGTON. D.C. 20005 (202) 2344433 i

~. _ 29 1 COMMISSIONER ROGERS: What were the most 2 significant modifications or deficiencies that you-3 felt should be addressed? 4 DOCTOR ELTAWILA: The most deficient 5 things in our code right now is the late phase melt 6 progression because the database is inadequate. About 7 two years ago when we started that program the only_ J 8 database that we had is the TMI accident itself and we 9 did not have any information about that. Some'of the 10 other important problem is the numerics with the code, 11 lack of diversions and it's difficult to run. 12 Documentation was an important issue and assessment of - i 13 the codes. So, all of these we had an expedite, very 14 aggressive program to address all of them. We still 15 have additional database on late phase melt 16 progression, but that has not been incorporated in the 17 codes yet. 18 COMMISSIONER ROGERS: And the 19 documentation, where does that stand > G 20 DOCTOR ELTAWILA: The documentation right 21

now, I

think we improved the documentation 22 tremendously and most of 'our codes right now, the 23 documentation is very adequate. I think we ' have l 24 learned one lesson from the peer review process, that i 25 we should not let -- in the future, anybody develop a l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 2344 433 WASHINGTON, D C. 20005 (202) 2344433 l

30 1 code-should not let the code developer start 2 developing a code and to go along developing very fast 3 without having a peer review process along the way. 4 So, what we have instituted in the 5 division and in the branch in particular, that any 6 time we have a modification to the code, before it is 7 incorporated in the code we form smaller group of L 8 expert to review the model, to see if the database 9 support the implementation of that model. Once we get 10 the approval of the peer reviewer we give the code 11 developer the go-ahead to add this model into the 12 code. So, this way we'll not fall into trap that we 13 have to review a very complex code like we did in the 14 case of the MELCOR code. 15 COMMISSIONER ROGERS: Thank you. 16 COMMISSIONER REMICK: Has RELAP/SCDAP also 17 had a peer review and have any issues there been 18 resolved? 19 DOCTOR ELTAWILA: The SCDAP/RELAP has gone a 20 peer reviewed. The main issue that were resolved is 21 the user friendliness, the documentation and ' the 22 ability to run the code for full plant analysis. We 23 have resolved all this problem. The peer review was 24 just finished about six or eight months ago. So, we 25 have a plan in fiscal year '94 to implement the high NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 234 4433 WASHINGTON. D C. 20005 (202) 2344433

~. 31 1 priority issue of the peer reviewer comments. 2 DOCTOR BECKJORD: Okay.. On direct 3 containment heating closure, we've done a lot of work 4 in this area. We've finished our scaling analysis, as 5 I mentioned. We started running experiments again at 6 the SURTSEY facility at Sandia. We also then found a 7 way to utilize the containment technology test-3 8 facility which is a one-sixth scale concrete 9 containment model at Sandia. So, we had now 10 experiments at Argonne which we had a small one-t 11 fortieth scale. We had the SURTSEY facility at' i 12 Sandia, which was one-tenth scale. We then went to 13 one-sixth scale. So, I think we have a real good base 14 to extrapolate to the prototype plant'on a DCH. 15 We have developed a resolution for DCH for 16 the Zion plant. That report has been issued and is 17 undergoing peer review. We also have a resolution 18 plan now that has been issued for the Surry plant. 19 These were the two designs that we looked at O 20 experimentally because of their difference.in their 21 lower compartment geometries. 22 The report on Surry came out about.a month 23 or month and a half after the Zion report. It also 24 has a generic -- it's called the integrated report. 25 It also looks at the general methodology of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVEN A, N W. (202) 234-4433 WASHINGTON O C. 20005 (202) 234-4433 w

32 1 extrapolating to other designs. Both these reports 2 are undergoing peer review and will follow basically 3 the same process we did on a Mark I issue. 4 COMMISSIONER REMICK: In the SECY document 5 there was no mention of B&W plants. What is being 6 done for the B&W plants. 7 DOCTOR BECKJORD: On direct containment s 8 heating? 9 COMMISSIONER REMICK: Yes. 10 DOCTOR BECKJORD: Direct containment 11 heating, B&W plants for direct containment heating 12 look very similar to a Westinghouse design. The plant 13 that does not in this case is the combustion. The 14 reason is because they do not have lower penetrations i 15 on their vessel head and they do not have that keyway j 16 to bring out the instrument tubes. 17 COMMISSIONEP. REMICK: I see. l 18 DOCTOR BECKJORD: Right now we're looking 19 at whether we have to run one or two more experiments .[ 20 to possibly get some information on the CE design. 21 (Slide) On core melt progression, early 22 phase melt progression we think is fairly well 23 understood. We've run a number of experiments, both 24 in the U.S. as well as abroad to understand the early 25 phases of core melt, the candling down of the claddi ':-g NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W (202) 234 4433 WASHINGTON. D C,20005 (202) 234 4 33

33 1 and the early relocation. One remaining uncertainty 2 is the question of blockage versus drainage, which 3 applies primarily to BWRs. Because you have an ADS 4 system you blow most of the coolant out of the core, 5 out of the primary system during the blowdown. So, 4 6 when the core melts, it is not melting where the lower 7 part of the core is underwater still. So, there's a J 8 question as to whether when the melt comes down it 9 will freeze and the cooler lower portiores are whether 10 it will run right down and relocate immediately to the' 11 lower head. 12 On fuel coolant interactions and debris 13 coolability, we've completed most of our research on 14 the fundamentals of fuel coolant interaction. We have 15 work going on right now cooperatively at the FARO 16 facility in Ispra, Italy involving a large scale 17 integral test. But our ex-vessel debris coolability 18 research results are still inconclusive. 19 These -- e 20 COMMISSIONER REMICK: How do you see that 21 closure on that? 22 DOCTOR BECKJORD: On the fuel coolant 23 interaction? 24 COMMISS70NER REMICK: Yes. 25 DOCTOR BECKJORD: Doctor Eltawila is going NEAL R. GROSS LOURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W (20?) 2344433 WASHINGTON. D C. 20005 (202) 2344433

34 1 to discuss that in his presentation. 2 COMMISSIONER REMICK: Okay. 3 DOCTOR BECKJORD: (Slide) On our computer 4 codes we just talked about those. I'll just touch on 5 these very quickly. 6 Our major severe accident codes are the 7 MELCOR code, which is an integral risk assessment L 8 code. When I say risk assessment it's used primarily 9 for risk assessment. This has been peer reviewed. 10 The comments of the peer reviewers, as I said before, 11 have been taken into account and we have work going on 12 to incorporate these. I think most of it's already 13 been done. 14 On the SCDAP/RELAP code, this is our 15 detailed core melt progression code. The peer review 16 is complete and we have a program underway right now 17 to address the peer review comments. 18 The CONTAIN code, which is our containment 19 analysis code, a detailed peer review was started in 20 April of '93 and we expect it to be completed in about 21 four months or five months, which would be in March 22 '94. 23 We do have other specialized codes which 24 we use for specific type phenomena analyses. These 25 are HMS, which looks at hydrogen transport, COMMIX, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W (202) 234 4433 WASHINGTON, D C. 20005 (202) 2344433

35 1 which we are looking at. That's a three dimensional 2 computer code which looks at the containment behavior 3 for the AP-600 where you're in .a condensing 4 environment. 5 (Slide) Next slide, please. S 6 COMMISSIONER REMICK: For

MELCOR, 7

SCDAP/RELAP and CONTAIN, I'm really pleased with the i 8 f act that you are having the peer review and resolving 9 issues. How about long-term maintenance? Do you have 10 plans to make sure that once you've brought them up to 11 date that we maintain the codes and the expertise 12 knowledgeable in the codes over a longer period of i 13 time? 14 DOCTOR BECKJORD: I address that in my 15 last slide. 16 COMMISSIONER REMICK: Okay. 17 DOCTOR BECKJORD: I just want to touch 18 very quickly on some of the research facilities that 19 the NRC sponsors. We have funded significant 20 experimental research in facilities both in the U.S. 21 as well as around the world. We 've done.in-vessel 22 crust behavior studies in the Annular Core Research a 23 Reactor at Sandia. We have ex-reactor experiments to i 24 study 1.nis question of blockage versus drainage for i 25 the BWRs, also at Sandia. When we.put these slides NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 2344433 WASHINGTON, D C. 20005 (202) 2344433

36 1 together we also had a full-length core melt 2 experiment scheduled at the Canadian NRU reactor. 3 Unfortunately, af ter several delays on the part of the 4 Canadians, we went to test our test bundle and found 5 we had a leak in it. So, the test has been canceled. 6 As I said before, we have fuel coolant S 7 interaction experiments going on at the FARO facility L t 8 in Ispra, Italy and I think you're probably aware we 9 are now looking at a lower head coolability program 10 using a RASPLAV facility in Moscow. 11 COMMISSIONER ROGERS: Could you say just 12 a quick word or two about that RASPLAV program? 13 DOCTOR BECKJORD: Basically what it is is 14 we're looking at a scale model of a cylindrical vessel 15 lower head and what we want to understand are two 16 things. One is the behavior of the melt in the lower [ 17 head itself, the circulatory patterns and therefore 6 18 the heat transfer. Then hopefully later on the 19 question of having -- if there is water on the outside 20 of the vessel, what kind of cooling that would t 21 provide. 22 The Russians right now are still designing I 23 the test facility itself. There's questions about 24 whether we go to a complete hemispheric lower head or 25 whether we use a slab geometry because the heating NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 234-4433 WASHINGTON. D.C. 200(Fs (202) 2344433

37 1 obviously is by induction heating and the Lorenz force 2 becomes very significant when you. have a large 3 hemispherical head and it may distort the phenomena 4 that you're interested in. 5 DOCTOR BECKJORD: That will become another p 6 OECD Nuclear Energy Agency program. It's been 7 approved and I think CS&I will put its blessing on it b 8 at the December meeting. As I think you've seen, 9 there's a very strong logical follow-on to the TMI lo vessel program. 11 COMMISSIONER de PLANQUE: Brian, just a 12 general question and a qualitative answer will do. If 13 you look at all the experimental research being done 14 internationally, what proportion of that research is 15 being done in U.S. facilities and do you see a trend 16 in the change of that proportion? 17 DOCTOR SHERON: In terms of number of 18 experiments or in dollars per experiments? 19 COMMISSIONER de PLANQUE: Well, I don't 20 know what unit I want it in. What will best describe 21 the situation? 22 DOCTOR SHERON: Well, I mean, I would, and 23 this is just my judgment, maybe Farouk -- I would say 24 maybe about, what, two-thirds in the U.S., one-third - 25 - maybe one-half, one-half. Okay? NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234-4433 WASHINGTON, D C. 20005 (202) 234-4433

38 1 In terms of dollars, there's a couple of { 2 major experiments. For example, the FARO f acility is 3 fairly high cost. We have very good leverage on that 4 because we're only paying,

what, 15

percent, I

l 5 believe, of the cost? 6 DOCTOR ELTAWILA: We have the fixed cost 7 of close to S1.75 million that we're sharing with .6 8 EPRI, but the FARO or CEC and the European Community 9 is paying the salaries and the costs of the facility 10 itself, so we're going to run five experiments over 11 there at that fixed cost regardless of the outcome of 12 how long it takes or anything like that. 13 COMMISSIONER de PLANQUE: Do you see any 14 indication that more and more of this work as we go on 15 or get into the advanced reactors will have to be done i 16 at facilities outside the U.S.? 17 DOCTOR ELTAWILA: In the area of severe j 18 accident, we have not identified any new area that can 19 affect the advanced light water reactor. The reason f 20 that some of this work now is starting, for example in 21 Europe, and we don't have any of that work, they start 22 very late compared to us, 23 We have our experimental program going on 24 since the '80s, so we developed a wealth of database 1 25 and now they are trying to develop their own database .l l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVENUE, N W. f (202) 2344433 WASHINGTON, D.C. 20005 (202) 234-4433

l 39 1 and we are taking advantage of that by becoming a i j 2 member of this international program to get the 3 information at very cheap cost and we keep our 4 expertise maintained and involved in this program at 5 the same time. So although they have started, most of t 9 6 the work in severe accident will be in European 7 countries. I think we have our own database that they A 8 are-trying to catch up with. 9 COMMISSIONER de PLANQUE: Okay. 10 DOCTOR BECKJORD: I think I'd like to add I 11 something to that. I think it's very clear that there 12 is a benefit to the international cooperation and I 13 think that's recognized throughout the world and that 14 has meant that there are now more cooperative programs ? 15 outside of the U.S. whereas before ten years ago it 16 was primarily inside the United States. l 17 I think the direction here is going to 18 depend very much on the direction of the nuclear 19 industry in this country. If there's a strong nuclear 20 industry, I think you will find that there will be a 21 lot of projects that are carried out with ~ 22 international cooperation here. I 23 COMMISSIONER de PLANQUE: Okay. Thank 24

you, t

25 DOCTOR SHERON: For the sake of time, NEAL R. GROSS t COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 2344433 WASHINGTON, D C 20005 (202) 2344433

40 1 slide 16 really doesn't say anything other than list 2 some facilities we have. 3 I'd like to just -- if you'd look at the 4 monitors, we've got some much better pictures of some 5 of these facilities. 6 COMMISSIONER REMICK: Brian, before going 7 to that -- 4 8 DOCTOR SHERON: I'm sorry. 9 COMMISSIONER REMICK: I've recently 10 heard that the Japanese have a program of severe 11 accidents called Alpha. It 's not listed here. Do we 12 have any involvement with an Alpha program on severe 13 accidents? 14 DOCTOR ELTAWILA: Yes. As part of our 15 severe accident research program, the Japanese are 16 entitled to our data and we are entitled to their 17 data. I apologize for not including it here'in the 18 viewgraphs. 19 COMMISSIONER REMICK: So we will get the 20 data from their programs? 21 DOCTOR ELTAWILA: We will get'the data. 22 DOCTOR SHERON: (Slide) Okay. If you 23 could put the pictures on the monitor, the first one 24 is a picture of the SURTSEY vessel. As you can see, 25 this is one-tenth linear scale. This is at the Sandia NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 234 4433 WASHINGTON, D C 20005 (202) 2344433

41 1 Laboratories. At the bottom, it 's not very clear, but 2 right inside the vessel in the lower part is where we 3 have the melt generator which drives the melt into the 4 containment and we measure the parameters like the 5 hydrogen a:.d the pressure and so forth. Y 6 COMMISSIONER REMICK: What is the melt 7 material that they're using? ? 8 DOCTOR SHERON: This is usually Thermite, 9 which is an iron-alumina compound. 10 (Slide) The next picture is the CTT 11 facility, the Containment Technology Test facility 12 This is a one-sixth linear scale model of the 13 Westinghouse containment, I believe, and this was 14 actually tested previously where it was pressurized to 15 see its failure mode, and so it was pressurized up to 16 its ultimate strength and then it showed signs of i 17 failure. I think what happened was that the liner 18 cracked and the pump that was pressurizing it just i 19 could not keep up with the leakage. 20 DOCTOR BECKJORD: Reinforced concrete. 21 DOCTOR SHERON: Yes. 22 DOCTOR BECKJORD: It got up to a little 23 bit over 130 pounds per square inch. 24 DOCTOR SHERON: So, we decided that we 25 could use that for our DCH testing as well. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W-(202) 2344 433 WASHINGTON. D C 2000'$ (202) 234 4433 i

43 1 (Slide) The next picture, please. 2 This is a picture of our high-temperature 3 high-speed combustion facility. I think this picture 4 was taken at the construction site where it was built. 5 It has since been taken apart and rciscated to 6 Brookhaven and, if anyone is traveling to Brookhaven, I 7 I think in about another couple weeks you would be 8 able to see this. It's put in a long tunnel down 9 there and we will be looking at high-speed high-10 temperature combustion phenomena on hydrogen. 11 (Slide) Next slide, please. 12 This is just a schematic of the Phebus 13 tacility which will show you the mock-up of the 14 primary system and the containment and the the '5 steam generator is in there too and the pressurizer 16

and, again, it's a matter of looking how fission 17 products distribute in the core, the primary system 18 and in the containment.

19 With

this, I'm going to let Doctor 20 Eltawila talk to you about the next two items on the 21 agenda.

22 DOCTOR MURLEY: Brian, could I just add a 23 point here? 24 DOCTOR SHERON: Yes. 25 DOCTOR MURLEY: From our perspective, this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234 4433 WASHINGTON. O C 20005 (202) 2344433

43 1 is very valuable research and we do use it and I'll 2 give one example. 3 Brian was talking about the TMI-2 vessel 4 results and I had as we were wrapping up our 5 reviews of the evolutionary plants and writing our 9 6 FSERs, I wanted to know whether there had been any 7 kind of integral insights, licensing insights, safety 8 insights that have changed as a result of this 9 research, and so I asked Ashok's staf f what would have 10 been would the containment have failed if the Il vessel had failed at TMI-2?

Now, there was a 12 preliminary study done by the Rogovin group right 13 after the accident, but, in light of all this new 14 information, I asked them to look into that and maybe 15 he can say a few words.

16 The bottom line is that probably it would 17 not have failed using the codes and the best methods 18 that Research has developed as long as the containment 19 sprays were available. If they were available, it 20 would not have failed at all. 21 DOCTOR THADANI: Yes, that's basically 22 what we concluded and in f act some of the calculations 23 have helped us think about some helpful. accident 24 management strategies that one can develop to protect ) 25 containment for long-term. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W-(202) 234 4433 WASHINGTON. O C. 20005 (202) 2344433

44 1 DOCTOR MURLEY: I think our FSER will 2 include a detailed discussion of severe accident 3 behavior of these evolutionary plants and it will of 4 course use -- depend greatly, almost totally on the 5 methods and research that has been developed in the 6 last decade or so. t 7 COMMISSIONER REMICK: That's a good e 8 introduction to a question I was going to ask later 9 on, a question I was recently asked. How would you 10 characterize how we are doing severe accident analysis 11 in the reviews? It's not really a design basis 12 approach, but how would you express how we are going 13 about looking at severe accidents in the designs? 14 DOCTOR MURLEY: I can express it broadly 15 and then perhaps Ashok can add. What we've said from 16 the beginning is that a core melt accident, severe 17

accident, is not a design basis accident in the 18 classical sense because there what one does is insist 19 on diversity and redundancy and basically gold-plating 20 all the equipment that needs to deal with it and we 21 said that's not necessary.

But one must consider core 22 melt accidents in the design and do what you can to 23 mitigate their consequences and where it makes sense, 24 where it seems to be cost-effective, put features in 25 the design to cope with those things, so we're quite NEAL R. GROSS CoVAT AEPORTERS ANJ TRANS7BERS 1323 RHODE ISLAND AVENUE, N Y (202; 2W433 WASHINGTON D C 20005 (202) 23W3J

45 1 confident that we have reduced the conditional 2 containment failure probability given a core melt 3

accident, but we've stopped short of all the 4

paraphernalia that goes with making it a design basis 5 accident. q 6 COMMISSIONER REMICK: I think I didn't do 7 too bad in providing an answer. One thing I did add 8 is rather than conservative calculations too we used' 9 best estimate and realistic-- 10 DOCTOR MURLEY: Yes. 11 COMMISSIONEh REMICK: That's correct? 12 DOCTOR THADANI: That's right. And where 13 there are questions I'll use an example. 14 Initially, if you recall, when we came up with SECY-i 15 90-016 we said quench the core debris was our first 16 thinking and then we realized the experimental basis 17 couldn't really support that. Well, what was our 18 goal? Our ultimate goal was to make sure that the 19 containment integrity would be maintained for those 20 accidents for some time period. You recall we talked 21 about 24 hours, and that then required that we do some 22 analysis and use some best estimate analysis to see 23 what coolability considerations would be, what heat 24 transfer mechanisms would be involved and if in fact 25 one would impact some of the support structures and so NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W (202) 234 4433 WASHif4GTON, D C 20005 (202) 2344433

46 1 on. And it was these codes that we used to analyze to 2 get an understanding of what could happen and how much 3 time there would be before the containment would be 4 pressurized be"ond the ultimate point, so I would say 5 that this has been very helpful and timely for us. 6 COMMISSIONER REMICK: Are the new designs 7 using concrete aggregate that produces less of the 8 noncondensibles or not? 9 DOCTOR T H A D A!1 1 : 11 o. They are using 10 aggregate that uses less gas production, but then 11 there are some other downsides to that also. So 12 that's the way they're going right now and that's 13 what's in our, in fact, ITAACS. 14 COMMISSIONER REMICK: I see. 15 DOCTOR ELTAWILA: I would like to talk 16 about two phenomenological areas of research that we 17 are doing right now in the Accident Evaluation Branch. 18 The first one is the fuel-coolant interaction and the 19 second one related to the debris coolabi]ity, and they. 20 are related somehow. That's adding water or molten 21 material and getting in contact with water. 22 But before I start the detailed 23 discussion, I would like to make a point that whenever 24 we talk about fuel-coolant interaction people always 25 think of the alpha mode steam explosion that was NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W (202) 2344433 WASHINGTON. D C. 20005 {202) 234 4433

47 .l 1 defined in the WASH-1400. There are other aspects of ) i 2 fuel-coolant interaction that we are. interested in. l 3 I will. repeat myself later, but we have concluded the I 4 alpha mode containment failure is a low probability 5 event and we are not looking at it at.all. .t 6 For example, during ti.L TMI event when the 7 material relocated into the lower head there was a B 8 mild fuel-coolant interaction which had -some 9 beneficial effect which resulted in the quenching of 10 the material in the lower head and that's why.we did 11 not see the challenge in the lower head that we used 12 to assume in our analyses before. So, because fuel-13 coolant interaction can occur in the vessel itself as. 14 a result of accident management through flooding the 'l i 15 core, it can happen in the lower plenum as a result of i 16 molten material falling into the water coolant and it ) 17 can happen ex-vessel as a result of reactor pressure I 18 vessel failure and pouring of the material into 19 flooded cavity. ) 20 We have interest in pursuing the issue of 21 fuel-coolant interaction and our main focus is'try to i i 22 identify what are the relocation scenario likely to -] 23 lead into a coolable configuration, what are the 24 effects of fuel coolant interaction on lower head 25 integrity and on the cavity and pedestal integrity, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISt.AND AVENUE. N W, (202) 234 4 33 WASHINGTON. O C. PCOOS (20h 234M33 1

48 I and what is the likely energy release and resulting 2 load from containment? Again, we're not talking about J 3 the alpha mode failure. Here we're talking about the 1 4 potential of producing large quantities of steam and ) 5 hydrogen and see how can that charge the containment 6 hydrostatically or quasi-static pressurization. I i 7 COMMISSIONER ROGERS: Excuse me. Just 8 before you move off that, this question of the 9 relocation scenarios, do you have in mind that it's 10 the possibility of influencing the development of 11 scenarios? In other words, accident management 12 through this work or is that just beyond what you're 13 looking at? 14 DOCTOR ELTAWILA: Absolutely not. I think f i 15 we -- our position that any time you have a water -- 16 you add the water as much as you can. It's not the 17 idea of to influence the scenario itself. It's try to 18 understand the consequences of adding the water, what 19 a person observing severe accidents should one occur 20 expect to happen as a result of adding the water at 21 the different stages of the accident. So, he will 22 have indication about -- should have an idea about the 23 amount of steam that would produce, the amount of 24 hydrogen that can be produced, where that hydrogen 25 will go, how the water addition itself can influence NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 234 4 433 WASHINGTON. O C. 20005 (202) 234 4433

49 1 the source term, can influence the result the rest of 2 the core melt progression. ( 3 So, it's not to alter anything, but to 4 understand the consequences of the action that we'll 5 be taking. t 6 COMMISSIONER ROGERS: Well, that troubles 7 me a little bit, frankly. This is not some abstract 8 kind of scientific study. You don't want to just know 9 what happens. If there's some possibility of a 10 management decision being made to influence the 11 evolution of the scenario, you want to be able to be 12 able to pass that kind of information on. So, it 13 sounded to me as if either there isn't any work going 14 on in that or that that's -- I'm just uncomfortable 15 about your answer on that one because maybe I 16 misunderstood it, but it didn't seem to me that you 17 can pull out of this what would be ultimately the 18 ultimate most usetul result, namely not what has 19 happened, but how do you see that the worst thing 20 doesn't happen when you're adding water. The rate at 21 which you add it and when you add it. 22 DOCTOR THADANI: If I may comment on that. 23 I think this in very valuable information as part of 24 accident management considerations. In fact, this is 25 the kind of information that the industry would be NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W (202) 2344433 WASHINGTON, D C. 20005 (702) 7344 433

50 1 looking at and in fact they are looking at today as 2 part of their accident management program to see what 3 kind of strategies they need to develop and what kind 4 of information would be needed in the development of 5 that strategy and the information that one needs is 6 what comes out of an activity like this. There is a i 7 program in place that the industry has. They expect 8 to implement the accident management program. They 9 expect to implement by the middle of 1997. I mean 10 there's a lot more to it, including the completion of 11 IPEs and so on, but this is very useful information 12 that goes into that program and assessment of what 13 actions should be taken given one understands the 14 state of conditions. 15 COMMISSIONER ROGERS: Yes, but I would 16 hope that it wouldn't be just an open look feedback 17 system, that somehow you'd close that loop so that 18 whatever studies are being done also consider what the 19 management actions might possibly be and then the 20 consequences of those. 21 DOCTOR THADANI: I was being very 22 simplistic, I admit. 23 COMMISSIONER ROGERS: Okay. All right. 24 DOCTOR THADANI: But yes, indeed. In 25

fact, including what kind of instrumentation and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W.

j <m> 2 awn u swNaToN.o c 2 a \\ <m> 2 awn

i 51 i I capability you need l to - be able-to feed back and 2 assess. 3 COMMISSIONER ROGERS: All right. .Okay. 4 That is part of your objective here. 5 DOCTOR THADANI:

Yes, and also the 9

6 decision making process. This is -- I t 7 COMMISSIONER ROGERS:- Now I hear NRR I 8 talking, but my problem was with an RES response. So, 9 I would hope that you would be closing the loop 10 internally here. 11 DOCTOR BECKJORD: The conclusion of my 12 paper dealt with this at the Boston meeting last week 13 and I think there is a very clear lesson from TMI-2 as 14 to how you handle the scenario and the pressurized 15 water reactor that's following that way. 16 There are still some questions that have 17 to be answered and Doctor Eltawila is talking now a 18 about the fuel coolant interaction and that's one of 19 the questions that needs to be dealt with. 'In 20 particular, it's clear that you want to put water in ~ 21 the vessel. It's almost as important for the vessel i 22 as it is for cooling the core. If'the vessel was 23 overheated, and I think we can conclude now that the 24 vessel will be overheated if the core melts and falls 25 to the bottom, you also want to limit the pressure and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W (202} 234-4433 WASHINGTON. D C. 2000$ (202) 2344433 a e e ,e .c-- r- -r-

52 1 bring the pressure down. The question is how far to 2 bring the pressure down and that's still an ) 3 outstanding question that he is dealing with. 4 COMMISSIONER ROGERS: Okay. 5 DOCTOR BECKJORD: And then when you know 6 that the core has been quenched, then you're free and 7 clear. You just keep the water in there, keep the 8 cooling going. 9 COMMISSIONER ROGERS: Sure. Okay. 10 DOCTOR ELTAWILA: (Slide) If I may jump 11 to a viewgraph on page 23 just to try to describe the 12 phenomena of fuel coolant interaction and I just would 13 like to emphasize that this schematic here is a very 14 old and does not represent our understanding right 15 now, but I put it in totality here to see how far we 16 came in understanding the issue of fuel coolant 17 interaction as a result of the research. 18 When the molten core material is 19 accumulated on the lower core plate, it was assumed in 20 WASH-1400 that would be a catastrophic failure of the 21 lower core plate and the material will enter into the 22 water pool in the lower plenum. It will fragment and 23 interact with water produced large quantities of 24 steam. If an explosion occur as a result, for 25 example, of the material impacting the lower plenum or NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234 4 33 WASH!NGTON, D C. 20005 (202) 234 4433

S3 1 any of the plates located in the lower plenum and that 2 shock wave propagates through the system, additional 3 fragmentation will happen which provides more surface 4 area which will interact with the steam and accelerate 5 a slug of liquid material upward and impacting on the q 6 upper head of the vessel, dislodging it and impacting 7 the containment and penetrating the containment. That 8 was the WASH-1400 assessment. 9 Since then there was some ofher research 10 here that indicated if we have enough energy to impact 11 the upper lead and dislodging, that same energy would 12 be able to fill the lower head and that will reduce 13 th9 amount of energy available to impact on the upper 14 head as.d that was the reason for concluding that the 15 alpha Fode failure is a low probability. 16 Again, right now, our research result, 17 which I will discuss it in detail, indicate now there 18 is not enough energy in the system to fill either the 19 lower head or the upper head. So, instead of relying 20 on a failure in the system to eliminate another 21 failure, we believe that the failure mode of the 22 vessel has been under control right now. 23 (Slide) Just to put our approach -- I'll 24 go back to page 22 -- our approach in perspective, in 25 1985 the NRC formed the steam explosion review group NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234J433 C WASHINGTON, D C. 20005 (202) 234 4433

54 1 and that group concluded that the occurrence of a 2 steam explosion was sufficient energy to fail containment has a very low probability. At that time 3 4 the steam explosion review group used engineering 5

judgment, used probabilistic approach and used 6

subjective assessment about uncertain phenomena. They 7 did not have enough database to help them put this 8 issue to rest and they recommended additional research 9 to try to explore the certain key energetic generic 10 phenomenological aspect of fuel coolant interaction to 11 address the complex variation in core melt 12 progression, whether we have a large coherent pour, as 13 they assume in the WASH-14 00, or we can have a slender 14 slow pour as happened at TMI and in accident 15 management whether you add water early, in the early 16 stage of core melt progression or at advanced stage of 17 core melt progression. 18 So, we devised our research program based 19 on the recommendation of the steam explosion review 20 group and we performed a small scale experiment to 21 look at the f undamental of the phenomena itself, which 22 mainly the premixing and this is a hypothesis that 23 when the material located into the lower head, the 24 largest steam that occurred as a result of that 25 relocation would pull the water out of the mixing zone NEAL R. GROSS COURT REPORTERS AND TRANSCRIDERS 1323 RHODE ISLAND AVENUE, N W. (202) 234-4433 WASHINGTON, D C. 20005 (202) 234-4433 .m.

55 1 and will result in depletion of water in that region. 2 So, you have a physical limit on the amount of 3 material that can participate in fuel coolant 4 interaction. 5 That was a hypothesis in 1989. We ran i 6 experimental program and we proved that the issue of 7 water depletion is a real issue and now we have 8 confidence in our assessment of these loads. 9 The second issue about additional 10 fragmentation that can happen as a result of being in 11 an explosion zone and we have completed our research 12 in that area too. We have used the research result 13 and developed analytical tools which have been-14 validated using this experimental result and other 15 results fro'm'ot'ner countries ~too. ~ 16 (Slide) May I have page 24, please? 17 As Doctor Sheron indicated, that we are 18 performing a large scale experiment ~with prototypic 19

material, UO;,

ZrO, and Zirconium. And under 2 20 prototypic conditions representative of accident in 21 the in-vessel at the FARO f acility, we hopefully will 22 run a test later on this year to investigate the melt 23 jet fragmentation, the quenching, the vapor 24 generation, the hydrogen generation, any thermal load 25 on the lower head and any erosion in the lower head. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W (202) 2344433 WA$HINGTON. D C. 20005 (202) 2344433

56 1 By using our analytical tool that we 2 developed based on separate effect tests -to prepredict 3 that experiment and then try to compare the result-4 between preprediction and after the experimental 5 result, we'll be able to have additional confidence in 6 our analytical tool. 7 Our approach to resolve fuel coolant 8 interaction, like any other issue in severe accident, 9 is using a probabilistic framework. Since there are 10 too many variations of the accident scenario, we have 11 to rely on developing a probabilistic framework like 12 the one we did for DCH or the Mark I liner issue and 13 we used the same approach, fuel coolant interaction 14 issue. 15 The result from our fuel coolant 16 interaction research so far indicated that the 17 assumption, for example, that was made in WASH-1400 18 about a massive catastrophic failure of the coil 19 support plate is unrealistic and that came from our 20 core melt progression research. We've seen in the TMI 21 that you get very minor break and relocation process. 22 The fact of the water depletion phenomena again 23 reduced the amount of material that can participate in 24 fuel coolant interaction and reduce the energy that 25 will be dissipated in the system either on the lower NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 2344433 WASH!NGTON, D C. 2(XOS (202) 2344 433

57 1 head and the upper head. And as a result of.that-2 work, we reassess the alpha mode containment failure 3 probability and will confirm the. conclusion that we 4 reached in 1985. That's indeed a low probability, but 5 in addition to that we have concluded too that lower ) 6 head failure is not expected to occur as a result of -g 7 fuel coolant interaction. 8 We feel that our analytical code are 3 9 adequate right now for doing fuel coolant interaction 10 loads for in-vessel as well as ex-vessel. The issue 11 of premixing and the fragmentation phenomena, which-12 are the key for fuel coolant interaction phenomena, 13 are really well understood right now. For ex vessel-14 phenomena we might need -- we're waiting for the FARO 15 test program to see if any variation in the hole size 16 or in the fall high, the water depth and so on might 17 affect our perception about ex-vessel. But right now 18 we have confidence that it might not change that much. 19 Our plan right now is to complete the 20 ongoing research program which is mainly at FARO and 21 at the University of Wisconsin. At.the conclusion of. 22 that work we will convene a group of experts to 23 identify the need for and the value of any additional 24 resecrch in that area. 25 COMMISSIONER.REMICK: What do you foresee NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 FIHODE ISLAND AVENUE N W. (202) 234-4433 WASHINGTON, D C. 20005 (202) 234 4433

58 1 as a schedule for convening that group? 2 DOCTOR ELTAWILA: Of course we're not 3 going to wait until the end of the FARO program 4 because that will take maybe about two or three years. 5 What we expect to see after running one FARO 6 experiment this year and finish the University of 7 Wisconsin work, sometimes maybe by March, April of a 3 next year we can convene the group immediately after 9 that. 10 COMMISSIONER REMICK: '94? 11 DOCTOR ELTAWILA: '94, yes. 12 (Slide) The second issue that I would 13 like to discuss here is the issue of the debris 14 coolability and what our interest is try to understand 15 what is the effect on containment performance from 16 debris-structure-water interaction. If we have core 17 concrete interaction and add water as a result of 18 accident management or as a result of design 19 specification, like an advanced light water reactor, 20 we want to see the effect on how much can the debris 21 be cooled enough to arrest the core concrete 4 s 22 interaction and further erosion and reduce the amount 23 of non-condensible that enter in the containment. And 24 at the same time we'll-try to understand the mechanism 25 that can lead into sustained coolable configuration. NEAL R. GROSS COURT REPORTERS AND TRANSCRIP,ERS 1323 RHODE ISLAND AVENUE, N W. (202) 2344433 WASHINGTON. D C. 20005 (202) 2344433

a r 59 1 We have run several experiments-in the 2 past at Sandia National Laboratory and we are a member 3 of the MACE project which is run by EPRI and include, 4 in addition to NRC,

DOE, EPRI and other foreign i

5 countries. ) 6 The results from the past testing 7 performed at the small scale showed that there are 8 some potential for coolability. Immediately after 9 adding water there are significar.: amount of energy 10 released or removed from the melt. However, because { 11 of the configuration that we test run this 12 experiment which are very small experiment, usually 4 13 about 20 by 20 centimeter or 50 by 50 centimeter, the 14 maximum that we ran, sustained debris coolability 15 appears to be encumbered by the insulating crust i 16 formed on the debris surface that attached to the l 17 wall. 18 (Slide) If I may go to the figure on page 19 28 and try to explain that slide in details here. j i 20 In the lef t-hand side you see the crucible which you have the concrete at the bottom and on top 21 22 of it we have a charge of uranium oxide, zirconium 23 oxide and some concrete material. It's heated 24 electrically and when the melt is complete the charge i 25 is melted and the core concrete interactions start. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234-4433 WASHINGTON. O C. 20005 (202) 234 4433

60 1 Water is added on top of it to cool the debris. What 2 happened in the experiment, if you look at the right-3 hand side, that you see that the dark line on the top, 4 a crust formed and as the core concrete interaction continuing, that the surf ace of the melt keep receding 5 6 down, which caused creating a void between the actual I 7 melt and the water on top of it. So, we were not able 8 to remove heat from that. 9 COMMISSIONER REMICK: Just intuitively, it 10 would seem like when you had the melt concrete 11 interaction you'd be producing a lot of gases which 12 would agitate the melt and break up the crust, but 13 that does not happen? 14 DOCTOR ELTAWILA: Absolutely. That's ' "" ~1 6 exa'ct'l'y'wfiat should have happened. But unfortunately 16 the crust is formed and because of the size of the 17 crucible is so small it attached itself to that wall i 18 and will not be able to penetrate it. ) 1 19 COMMISSIONER REMICK: So, in the actual i 20 case, one might expect that the gases coming off from I 21 the concrete interaction would agitate and allow water 22 to get into the melt, I would think. i 23 DOCTOR ELTAWILA: In fact, in some of { l 24 these tests we see some burst of material coming out, j 25 gases pumping out of that void, breaking the crust and NEAL R. GROSS COUPT AEPOHTERS AND TRAN$CR$ER$ 1323 RHODE ISLAND AVENUC, N W. (202) 234 4433 WA$HINGTON, D C. 20005 (?O2)2344433

61 1 disputing material on top of the water and causing it 2 to quench. 3 (Slide) Now I'll go back to page 27 and 4 again that there are some believe among the 5 researchers, as Commissioner Remick indicated, that t 6 the crust may be fractured by thermal mechanical loads 7 and might be fractured by just the aspersion of the 8 gases from the melt. 9 So, what the MACE project is planning for 10 the next integral test, M3, is to run it as large a 11 scale as possible and we are planning right now to run 12 it at 1.2 meters by 1.2 meters. It's going to have a ~ 13 charge of 2000 kilogram of reactor melt material and 14 20 centimeter of debris bed. 15 We're currently assessing the usefulness 16 of any additional testing versus accepting the current 17 uncertainty if we can live with the fact that 18 containment will not be pressurized beyond its 19 capability within certain time as a result of core 20 concrete interaction. That might be an acceptable 21 answer to us. 22 (Slide) I'd like now to return back to 23 Doctor Sheron. 24 DOCTOR SHERON: Okay. I would just point 25 out that Ashok had told you -- you asked a question NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W (202) 234-4433 WASHINGTON. O C. 20005 (202) 2344433

62 1 about what kind of concrete they were using in 2 advanced designs and you could argue that in one 3 scenario you want to have a low gas content concrete 4 that doesn't put a lot of non-condensible gas into the 5 containment. But from the standpoint of breaking up 6 across and allowing water ingress, you would like a 7 lot of gas. So, it's not clear what's the best way. 8 I'd like to finish up here by talking 9 about utilization of our research results and then 10 talk a little bit about our future plans. 11 With regard to the results from all this 12 research we've been doing, when we talked about the 13 source term research we intend to modify 10 CFR Part 14 100 based on the new information. We've updated the 15 source term for the advanced light water reactor which 16 was documented in NUREG-1465. It's improved all of 17 this research we've done and the area of severe 18 accidents has, I think, significantly improved our 19 understanding of containment performance. This has 20 been reflected in our evaluation of the IPEs, the 21 individual plant examinations, containment performance 22 improvements as well as our reviews of the accident 23 management programs being developed by the industry. 24 We've also used these results to close, as 25 I said, on some of the severe accident issues that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W (202) 234-4433 WASHINGTON. D C 2000$ (202) 234 4433

63 1 were dominating the risk uncertainty, namely those 2 that resulted in early containment failure. As I said 3 before, we've already closed on the Mark I liner issue 4 and put that one to bed and hopefully within a year we 5 will have direct containment-heating resolved for all' s 6 of the PWRs. 7 COMMISSIONER REMICK: What's the 8 anticipated schedule for Part 100? Not surprisingly, in Taiwan and Korea recently there was a 9 when I was l 10 lot of interest in knowing the status of that. What-11 is the schedule for sending - that back up to the i 12 Commission? ? 13 MR. TAYLOR: I'm supposed to see b 14 something, an options paper, by November 19th, I j 15

believe, and that will then be coming up to - the 16 Commission before we receive -- this will provide the 17 options.

I think that's right. 18 DOCTOR SilERON: (Slide) Slide 30, please. 19 The other areas we utilized research e 20 results. One is obviously, I think as Tom said 21 before, we're providing NRR with a database that they 22 can use for evaluating the advanced light water 23 reactor's performance against severe accidents. The 24 areas that are being evaluated were documented. in 25 SECY-90-016 and also, as you know, we had an advanced NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234-4433 WASHINGTON, D.C. 20005 ' (202) 234-4433 ) ~, , =.

64 1 notice of proposed rulemaking on Part 50 to deal with 2 severe accidents for future reactors. 3 Both of these specifically require 4 technical basis for assessing the capability of the 5 containment to withstand the severe accident loads and 6 these loads, as I said before, they're just listed F 7 right here, the DCH, the FCI, direct melt attack on 8 the

concrete, the hydrogen and then the slow 9

pressurization due to steam and non-condensibles. 10 And then

lastly, understanding severe 11 accident phenomena and their range of behavior is 12 important for taking advantage of existing t eactor and 13 containment capabilities.

This is primarily in 14 exploring accident management strategies for the 15 different containment types and developing plant 16 performance criteria for future light water reactors 17 against severe accident challenges. 18 So, I think in summary all of this 19 research is being utilized rather extensively in 20 various facets of the whole regulatory process. 21 (Slide) This will get to Commissioner 22 Remick's concern from before with our future plans. 23 First is we would like to -- if all goes 24

well, we'll complete our research on these major 25 issues that we're still working on by 1996.

That's NEAL R. GROSS CoVAT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 2344433 WASHINGTON, D.C. 20005 (202) 234 4433

65 1 sort of a goal that we've set for ourselves.

However, 2

1996 is not when we turn out the lights, close the-3 door and walk away. We intend to maintain a level, 4-albeit reduced, capability for the agency, for the 5 staff to be able to evaluate and predict severe 1 6 accident behavior. In order to do this we're going to 7 need to identify the necessary disciplines for which 8 expertise should be maintained. We need to identify 9 the minimum suite of analysis codes that we will need 10 to maintain. Once we've done that, then we will 11 identify future efforts needed to maintain 'this 12 expertise. 13 I think I said at the thermal-hydraulic 14 briefing you can't maintain the expertise by'just 15 keeping your researchers sitting around on their hands i 16 waiting to run a code or something. The way we would 17 plan on doing this would be to continue our 18 involvement in international programs, participate in 19 international standard

problems, participate in 20 cooperative research with international partners and 21 so forth.

We would maintain a code development and il 22 assessment effort. I think this will provide the 23 impetus for the researchers to want to improve the 24 codes the same way as I said in the thermal hydraulic 25 area. There may be a need to continue to run small NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 2344433 WASHINGTON, D C. 20005 (202) 2344 433

66 1 scale experiments, although we don't really under - - 2 we don't see a need right now for any large major 3 experimental facilities for severe accidents. 4 COMMISSIONER ROGERS: Before you leave 5

that, just on this whole code question, I know 6

Commissioner Remick is very-interested in it. I'm 7 l 7 interested in it too. But I'11 tell you, I would be 8 a little uncomfortable about developing codes to 9 pursue code development unless there is a really 10 important reason to do that other than to maintain the 11 expertise of the people. I think that's a separate 12 question of how to maintain expertise and whether 13 codes need further development or not. 14 I think I wouldn't mix those two-15 together automatically just to pursue code 16 development, for fine tuning it a little bit more, I 17 know, to keep people alive, their interest alive. I'd 18 be very concerned about that because it seems to me at 19 some point these things ought to stabilize and'not 20 change. Unless there is a good solid technical reason 21 to pursue additional code development, I wouldn't -- 22 I mean there are some systems that once you get 23 once they're complete, they have to be complete. If 24 you keep tweaking

them, you're going to drive 25 everybody crazy.

NEAL R. GROSS COURT REPORTERS AND TRANSCRIBER $ 1323 RHODE ISLAND AVENUE, N W. (20Q 2344433 WASHINGTON, D.C 20005 (202) 2344433

67 j l 1 So, it's just an issue there, a bit'of a- -{ i 2 concern that I have with respect to continued code j 3 development. I'm not opposed to it if there's a l 4 really important technical reason why codes need to be l 5 further developed, but just linking -that with the 3 6 necessity of maintaining expertise is a connection 4 7 that to me is a doubtful one. l 8 DOCTOR SHERON:

Yes, I apologize.

I \\ 9 didn't mean to leave the impression that we were just 10 going to do this just to keep expertise. 11 COMMISSIONER ROGERS: It's sort of a 12 gratuitous comment, but -- 13 DOCTOR SHERON: Yes, I think on.the areas 14 where we feel very comfortable we're certainly not 15 going to pursue any major code development just for i 16 the sake of it. But I think that in the area of 17 severe accidents there's always going to be some areas 18 where we're just not really totally comfortable and we 19 could always feel we could do a little better. I 20 think I said in the thermal hydraulic briefing that 21 the area I see is going to be in the numerics. With 22 the computer environment changing the way it is with 23 these new machines coming on, with much more 24 capabilities, as you saw the other day we can now run 25 these large codes right here from NRC, whereas five NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 2344433 WASHINGTON. D C. 20005 (202) 234 4433

q 68 1 years ago that was a dream. 2 COMMISSIONER REMICK: I think that will 3 help maintain the codes too since we have now 4 capability in Research and in NRR to run some of these 5 on work stations. Our own people will realize where 6 there is a need for improving. 7 CHAIRMAN SELIN: Underground code testing. 8 DOCTOR MURLEY: Mr. Chairman, could I be 9 excused? 10 CHAIRMAN SELIN: Yes, of course. Thank 11 you, Doctor Murley. 12 Before you go on, Commissioner Remick, did 13 you get your questions answered? 14 COMMISSIONER REMICK: Yes. I'm very 15 pleased with that answer. 16 DOCTOR SHERON: And our long-term planning 17 is needed right now to determine the requisite funding 18 level to maintain this capability and that's something 19 we're going to take an IOU because that's something we 20 have to do. 21 (Slide) Just a last slide. I just wanted l 22 to touch on very quickly some of the international 23 programs that we are involved in. f 24 We have bilateral agreements with, as you 25 see, that list the countries who have joined our NEAL R. GROSS j COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 2344433 WASHINGTON. D C. 20005 (202) 234 4433

69 1 Cooperative Severe Accident Research Program. What 2 joining our Cooperative Research Program means is that 3 the participants will receive the NRC severe accident 4 research

results, our computer codes that we've 5

developed and any other information that comes out of 6 our program. In return, what we usually ask is that 7 we receive some cash and, depending upon the size of q 8 the country, perhaps some in-kind contribution from 9 that country. If they're doing research, then we 10 would like to see their research results. 11 Just as our thermal hydraulic' codes, our 12 severe accident codes, I think, set the standard for-7 13 the world. Most of the Europeans and even in Asia are 14 relying extensively on our severe accident codes to 15 analyze their designs. 16 That concludes our presentation. I see 17 it's 11:30, which was the scheduled completion time, 18 so I don't -- unless you would like, we won't propose 19 to discuss anymore issues. 20 CHAIRMAN SELIN: I do have a couple of 21 questions. i 22 I was quite impressed with the i 23 presentation and in particular how you were able to 24 tie the research work to needs in NRR and in the 25 regulatory program. That's very positive and deserves NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE N W. f j (202) 2344 433 WASHINGTON. O C 20005 (202) i'34-4433

70. 1 commendation. P 2 I want to make sure I understand a remark 3 that Doctor Eltawila made, your discussion of our 4 databases being more effected by experimental work and 4 5 therefore not having to have such large margins of 6 computational

well, just large computational 1

7 margins. Is the implication that we use experimental 4 8 results more than some of our more -- the other J l 9 advanced colleagues overseas in doing the codes and in ) 10 doing the regulatory work? f i 11 DOCTOR ELTAWILA: We're using the 12 experimental result in assessing our code and try to 13 expand the database to a large full-scale facility and 14 we are using it more than our colleague overseas 15 because we had that database available to us as a 16 result of that research program that we have over the 17 past ten years. 18 Our colleague overseas, they just start 19 the process itself. They are developing their own 20 code. Even the most advanced one, you know, they are 21 in the beginning of developing their own code or i 22 performing their own experiments, so that's mainly the i 23 reason for the big difference. 24 CHAIRMAN SELIN: I must be missing 25 something. I understand there's quite a bit of 3 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W l (202) 234 4 33 WASHINGTON, D C. 20005 (202) 234 4 33 ,, _ _ ~... -.

= 71 1 communication about the experimental results and that a 2 our programs are available. Are people slow to pick- ~ 3 them up or do they disagree with the way we're using 4 them? 5 DOCTOR ELTAWILA: I don't see disagreement 6 about the way we use them, but I think it's a-matter-7 of timing that they start developing their own 8 capability over there. They have been participating 9 in this program. They see the data. They have been 10 using our

codes, but they mainly in the

past, 11 especially in smaller countries, relied on the NRC for 12 using the code, developing the database and try to 13 just develop the code to a state of maturity that they 14 can use them themselves.

Nowadays they are using the 15 code themselves. In addition to that, they are trying 16 to assess it against either their own developed data 17 or data that develop internationally. I don't know if 18 I answered your question. 19 CHAIRMAN SELIN: You answered the 20 question. I have to tell you that your answer is not 5 21 consistent with my general view of the sophistication 22 of some of these regulatory areas. I do believe that 23 we invest more and run a larger regulatory research 24 program than our colleagues, but I don't have the' 25 sense that the Germans or the French or even the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE tSLAND AVENUE, N W (202) 234M33 WASHINGTON. D C. 20005 (202) 2344433

72 1 Swedes are slow to use the results and to integrate 2 them. 3 COMMISSIONER de PLANQUE: Maybe what we 're l 4 looking here for as an answer, are they repeating some 5 of these experiments, duplicating them, or are they 6 doing additional or variants on the experimental work 7 that's been done here? 8 DOCTOR ELTAWILA: For example, the alpha 9 program that Commissioner Remick mentioned is 10 duplicating everything we have done in the past. They 11 have done only one they have only one unique 12 program about reactivity insertion accident. That's I 13 the only difference between their program and our 14 previous program. 15 The German, we relied a lot on the German 16 data, and there was, for example -- their research 17 program is not as comprehensive as ours. They focus 18 on the particular issue and they address it, i 19 For example, if you look at the French, l 20 they have a major program ongoing right now. Most of l 21 that program, we have information about this result 22 and it's sort of duplication of ours. 23 CHAIRMAN SELIN: We can leave this for . 24 future discussion. 25 I'm concerned about the image of a lot of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE fSLAND AVENUE, N W (202) 2344433 WASHING 77L 0 C. 20005 (202) 2344433

73 1 1 chauvinistic duplication of results rather than..just 2 accepting this as an international database and each l I 3 one building on the other's results, but that's a-big 4 topic and I'd be interested in your insights at some i i 5 future point as to what you think is happening in this 6 international program. 7 I've had the image of a rather. seamless 8 international research program where we do some things l 9 and other people do some things but the answers are 10 used sort of regardless of source and the image that 11 you present is a little more nationalistic and 12 chauvinistic than that. 13 DOCTOR BECKJORD: My sense, Mr. Chairman, 14 is that they take some time to incorporate results, l 15 but I don't think they're duplicating in every case. i 16 Maybe the Japanese -- 17 CHAIRMAN SELIN: This is not a good topic 18 to enter in at 11:35 without some concrete' example, 19 but perhaps in the future we could have a little a 20 discussion about that. (i I 21 MR. TAYLOR: Let's take a look'at it. 'l 22 CHAIRMAN SELIN: Commissioner Rogers? ~ l i 23 COMMISSIONER ROGERS: Yes. We 've -had - l 24 quite an extensive presentation and I've enjoyed it 25 very much. I think that it was very helpful and there NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 2344433 WASHINGTON, D C. 20005 (202) 2344433

74 1 are a lot of things we could talk about and go on all 2 day, so I'll just touch on two things that I thought 3 I might like to hear just a little bit about. 4 The scaling methodology, to what extent 5 have those results been published or are going to be 6 published in peer review journals that would be of ~ 7 interest to a broader community? I'm kind of 8 interested perhaps along the way that Commissioner 9 Remick started to touch on. It seems to me these are r 10 very fundamental kinds of problems, scaling of these 11 phenamena experiments up to larger applications. And 12 I don't know where they might be useful aside from the 13 nuclear industry, but they might very well be in 14 certain other applications and it would seem to me 15 that it would be a good idea to try to see that that 16 methodology gets published in a peer review journal 17 that is widely read by interested people of different 18 technologies, not just nuclear. 19 Excuse me? 20 DOCTOR SHERON: The scaling report, the 21 NUREG report that was done, had not been published in 22. any peer review journals. I think one of the reasons 23 is that we kind of considered the whole development of 24 it was a peer review process because we had 17 experts 25 and -- NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W (202) 2344433 WASHINGTON, O C. 20005 (202) 234 4433

75 1 COMMISSIONER ROGERS: I'm not thinking of 2 it from the point of view of questioning the quality 3 of it. I'm just saying that I think the dissemination 4 of it will only be meaningful in peer review journals, 5 not in internal documents, and for that reason I'm g 6 talking about peer review journals because those are 7 the ones that everybody reads that looks at i? from e 8 outside of this industry and that's the reason I'm 9 stressing that. 10 DOCTOR SHERON: Well, let us take a look 11 and see if there's a way we can get this publiched. 12 COMMISSIONER ROGERS: Because I think 13 these scaling questions are very sticky questions and 14 when people are grappling with them I'm sure that they 15 would love to be able to use something that has the 16 attention of a large collection of experts that our 17 work has had. 18 DOCTOR SHERON: The information is 19 available. It's published in NUREG reports, NUREG CR 20 reports, but I don't think it's been really -- yes, 21 I'm saying it's not available in peer and we're going 22 to take a look and see if we can -- 23 COMMISSIONER ROGERS: I'm

saying, if 24 somebody has got a problem and they're doing a 25 literature search, they're not going to be looking in NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIDERS 1323 RHODE ISLAND AVENUE. N W.

(202) 234-4433 WASHINGTON, D C. 20005 (202) 234-4433

76 1 NUREG reports if they're outside the nuclear industry. 2 DOCTOR SHERON: We will see if we can find 3 a way to get these published in peer review journals. 4 COMMISSIONER ROGERS: I'd like to see 5 that. I think that's very important. 6 The other thing is just on the codes, the P 7 international aspects of codes. Are we getting codes? ~ 8 I mean, our codes are being used and we're providing 9 them to others internationally. What is the 10 reciprocity on that? Are we getting codes from other 11 cooperating bodies? 12 DOCTOR ELTAWILA: We have recently 13 requested our colleague from CE to give us their ECAR 14 II code and we have received it and given it to our 15 contractor at Idaho National Laboratory to try to run 16 it and benchmark it against our code. 17 In addition to that, we've received an 18 older version of the Russian code, RASPLAV, but we 19 have not received any update since about two years ago 20 and we're looking for other updates in that area. 21 The other codes that are available, most 22 of the other countries are using our codes so we rely i 23 on them, if they have for example an international 24 center problem and things like that, to participate 25 and we'll get feeaback about user input, the effect of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE N W. (202) 234-4433 " ASHINGTON. D C. 20005 (202) 2344 433

77 1 user on the analyses-and so on, but we have been 2 getting these codes. 3 Some of the Japanese codes, we have not 4 received them yet because they have not released them 5 out to the public because they are not complete. g 6 COMMISSIONER ROGERS: Thank you. 7 COMMISSIONER REMICK: I had not intended e 8 to raise it,

but, since you mentioned thermal

/ 9 hydraulic codes and he talked about international 10 sharing, anything new on CATHARE? 11 DOCTOR SHERON: I made a note here to find 12 out if we've sent the request over. 13 COMMISSIONER REMICK: Okay. 14 DOCTOR SHERON: Let me take a commitment 15 to get back to you on that. t 16 COMMISSIONER REMICK: Fine. I just want 17 to say that I'm very pleased with the presentation. 18 I congratulate the staff on what they've done in the 19 severe accident program area. -l l 20 I think the refocusing that went on in 21 1988, and I think, Brian, you played a major role in 22 that, was a significant development in bringing a 23 chance of closure on many of these issues. It's 24 obvious we've come a long way and I agree that 25 internationally people look to these because it's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE, N W. (202) 234-4433 WASHINGTON, D C. 20005 (202) 234 4 33

78 1 setting the pace throughout the world on severe 2 accident programs and it seems to be now that, 3 although even five years ago some countries didn't 4 want to talk about the possibility of severe 5 accidents, there seems to be a general consensus that 6 one must look at severe accidents. / 7 The work that has been done here I think 8 has been quite significant. In preparing for your 9 presentation, I went back and read an ACRS letter of 10 about a year ago and I found it quite interesting 11 because they were quite complimentary and ACRS doesn't 12 tend to always complimentary. They say, "In general, 13 the presentations were well organized and well 14 presented and our questions were dealt with patiently 15 and with good humor," and I thought, gee, that's a 16 good idea to provide it with ice cream and make them 17 happy. 18 CHAIRMAN SELIN: Commissioner de Planque? 19 COMMISSIONER de PLANQUE: I have no o 20 further questions, just to commend you on excellent 21 work and an excellent presentation. 22 DOCTOR SHERON: Thank you. 23 CHAIRMAN SELIN: Thank you very much. 24 (Whereupon, at 11:43 a.m., the above-1 25 entitled matter was adjourned.) NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVENUE. N W. (202) 234 4433 WASHINGTON, D C. 20005 (202) 2344433

L E CERTIFICATE OF-TRANSCRIBER ^ This is to certify that the attached events of a meeting of the United States Nuclear Regulatory Commission entitled: TITLE OF MEETING: BRIEFING ON SEVERE ACCIDENT RESEARCH PROGRAM r PLACE OF MEETING: ROCKVILLE; MARYLAND 1 ? DATE OF MEETING: OCTOBER 26, 1993 were transcribed by me. I further certify that said transcript' ion is accurate and complete. to the best of my ability, and-that the transcript is a true and accurate record of the foregoing events. /Ad MS (I Reporter's name: Peter Lynch I r NEAL R. GROSS cousnr esemns Ase mascessas 1333 RM004 ISLAle AVDeut. M.W. ggay) m WA5BGBISTOes. D4 TODOS (202) M M L w -,-e --r wr e,-

2 SEVERE ACCIDENT RESEARCH STATUS BRIEFING TO THE COMMISSION Presented by i. l-i Eric S. Beckjord. Brian W. Sheron Farouk Eltawila ' Office.of Nuclear Regulatory Research October 26,1993 Cont'acti .F. Eltawila l -- 492-3525 l.

y e 1 3 SEVERE ACCIDENT RESEARCH PROGRAM OUTLINE

Background

Revised Severe Accident Research Program Issues and Results Severe Accident Research Program Update since September 1992 Issues and Results Detailed Presentations on: Fuel coolant interactions Debris coolability e Utilization of Research Results Future Plan e International Programs e Additional Information Core melt progression Hydrogen Combustion 2

e o 'l 2) SEVERE ACCIDENTRESEARCH PROGRAM BACKGROUND e Severe Accident Considerations: WASH-1400, other PRAs, TMI-2 and Chernobyl accidents, all tell us that severe accidents represent major contribution to risk from' commercial nuclear power plants Margins (for severe accident challenges) in existing e plants Even before the Chernobyl accident, importance.of e containment was recognized. Source terms (accident. sequences / phenomena) Severe Accident Research Program (SARP) e Severe Accident Policy Statement e e NUREG-- 1150 - e .SECY 88-147:(IPE, CPI, 'AM,SARP) e NUREG-1365. 3-1 w ~s u--. .nvw -.,s. ew-- w+m...---#e e -.,..e,- v m-> m _ _ _.__- _ --_ _-_--,_i___.___ _._ ___ _._--__._ _._.-.- sa

a REVISED SEVERE ACCIDENT RESEARCH PROGRAM In 1988, the staff undertook a complete re-examination of the severe accident research program. The objective of this re-examination was to refocus the research on those areas where remaining uncertainties still had a potentially large influence on risk, and where important residual phenomenological uncertainties still remained high. This re-examination of the severe accident program e was documented as NUREG-1365, and was issued in August 1989. The program was extensively reviewed both internally as well as externally. Supportive letters were received from the ACRS and NSRRC. 4

y i i REVISED SEVERE ACCIDENT RESEARCH PROGRAM L (CONTINUED 1 i i The major elements of NUREG-1365 were: Emphasis on resolving uncertainties associated with scenarios.that produced early containment failure (Specific issues: Mark I containment liner failure, direct containment heating, source term update, core concrete interaction, Hicombustion) Eliminating duplication of analytical codes-Developing a scaling rationale for severe accident experiments Refocusing research in selected areas (i.e., core melt-progression, fuel coolant interactions and debris coolability, hydrogen combustion) s

.a i 1 CONTAINMENT LOADS P(t) Missile (H + CO) + O - P 4 2 2 f h) Melt + Air -+ P h Steam. H2 Fuel Path is _d not Unique h Energetic vs.. Quasi-Static FCI 0 "2Mi w//d~ W *'b': u'f?n Coolable Debris Bed i -T " cci $Er !$ $') N h 2..l.'."i 'd u Penetration I l i 4 ~r x .- -~u r e n-a v

s 1 REVISED SEVERE ACCIDENT RESEARCH PROGRAM i RESULTS 4 In the intervening period from 1988 to_1992, the staff i completed research on a number of these issues: e BWR Mark I liner failure a Demonstrated integrity of Mark I containments l under core melt conditions provided water can' be added to. cover the core debris (NUREG/CR-5423) -- follow up work is complete. NUREG/CR-6025'was peer reviewed and issued in October 1993. 1 7 .i 'l I

4 A REVISED SEVERE ACCIDENT 4 RESEARCH PROGRAM RESULTS (CONTINUED? 1 Severe Accident Scaling. Methodology Effort was completed (NUREG/CR-5809) -- . application to DCH by SNL and ANL Developed method.for applying scaled experimental results to full size plants i e. Source Term issues Research program completed (source term, . quantity, timing, chemistry) We are participating in the Phebus project Source-term update is imminent [SECY - 127], NUREG-1465 Final report.to be ' published end of 1993: L a 8 .. I

1s REVISED SEVERE ACCIDENT RESEARCH PROGRAM RESULTS (CONTINUED? Core Concrete interaction Completed all-dry core concrete interaction experimental research. Validated models are available for assessing plants' response Additional validation (against experiments conductedLworldwide) is ongoing in USA and Russia e Hydrogen Transport and Combustion Major effort completed (deflagration- . detonation criteria under variety of conditions). l Research on residual issues related to high- . temperature hydrogen-combustion -- Joint l NRC-NUPEC effort continues-l l l .9 l-i ... ~.

4 2 REVISED SEVERE ACCIDENT RESEARCH PROGRAM RESULTS (CONTINUED? TMI-2 Vessel investigation Project (OECD) 1 Program completed in October 1993 Provided information to benchmark some aspects of core melt progression and RV integrity Evaluated margin to failure of reactor vessel. Insights on accident management-10 i

a SARP UPDATE A SARP update (SECY-92-329; NUREG-1365 Rev.1) was issued in September 1992: Identifies issues that have been completed or are near completion Describes the progress in our understanding of important severe accident phenomena Defines the research that would lead to closure of core melt progression, direct containment heating (potential early containment failure mode) and fuel-coolant interactions Includes research on issues applicable to ALWRs, especially the containment performance criteria during severe accidents 11

s SARP MAJOR ACTIVITIES AND RECENT RESULTS Focus research on three phenomenological issues and code development and assessment An important aspect of the Code Development Program is the Peer Review. Each of our major codes has undergone a peer review to identify deficiencies and any necessary further work. Direct containment heating closure New insights and improved data base allowed quantification of DCH loads for Zion and Surry. Developed generic criteria for other types of containments and assess the need for any additional work Peer review of the closure reports ongoing 12 L. mm.. m .a .m -. mm m

P ~ SARP MAJOR ACTIVITIES AND RECENT RESULTS (CONTINUED? r I Core Melt Progression Early-phase core melt progression is well understood Remaining uncertainty related to Blockage vs. drainage Melt pool formation, growth and failure Fuel-Coolant interactions and Debris Coolability e Completed most of the research on the fundamentals of FCI' Work at FARO, (Italy) involving large: scale integral test Ex-vessel debris coolability research results remain inconclusive 4 13 1 a --a --e----m-e

• 4 w

~ w-- e-o w s~- .rw ,m e -s'm-, ~o.6 +-m-e -w me-,-,, r .a --e-w

SARP MAJOR ACTIVITIES AND RECENT RESULTS (CONTINUEDD 1' Severe Accident Codes MELCOR - risk assessment code -- peer review complete; program to address high priority-peer reviewers comments is complete 1 SCDAP/RELAP - detailed core melt progression code -- peer review complete; program to address peer reviewers comments is underway CONTAIN - detailed containment 1 analysis code, peer review started i -April 1993 and will be completed March 1994' Other special codes for specific functions include HMS (H transport) and COMMIX- ~ 2 (AP600) 14 s f

h SEVERE ACCIDENT RESEARCH a e Facilities The staff has funded significant experimental research in facilities both in the U.S. as well as. around the world. In-vessel crust behavior studies in Annular j Core Research Reactor (SNL) Ex-reactor experiments to study blockage vs, drainage issue. (SNL). i Full length core melt experiment in the Canadia'n NRU reactor Fuel Coolant interaction experiments in the FARO facility-(Ispra, Italy) L Lower vessel head coolability in the RASPLAV facility (Moscow) 15' l 4 a --.m-o.~... -e. er -w. w- ,s,.-~w, ,v.v .+r o e --c-b- + -e~* e- .c-:.

h SEVERE ACCIDENT RESEARCH (CONTINUED) DCH testing in SURTSEY' facility, CTT facility (SNL), and the COREXIT facility (ANL) Debris coolability experiments (MACE test-at ANL) Hydrogen detonation testing at BNL l (cooperative with Japanese); and Russia (Kurchatov) Fission Product transport in-the core region, RCS, and containment (Phebus reactor, j France) i c i 4 16 ____._ __ _. _.. _ _ _.. _ _. _... _. _.. _ _ _. _. _.. _ _ _,,. _ _ _... ~. _ _

am.wsm,, z,,_ a. w w w a-4aa,. aa.- u-m s a n a a h I i l. 1 i i 3 + l E.9'.1 * '. i y ;.1<49. :,., p '<ftsel

• '. 'y

.d 7 g^

• * $h*, ', {*

i

• f ',*Y

~ 4 5 ,.e N. "$,ft 's b = y,, E y ,. ), .r, .' k-, .~ t :y (.'~.1,

i..:L l

.c,4,,,a. '1 / / ,..p . ::.->;i. l i ' e, J I 1 m 3,7, 9.,%. *. .4.. ., t. *,; .t s.,.sk !.k s.. e /.. l, ! 'I -w ,a. t [.* % ' he: f'. E*g. o c-i o. ...i.... y,s e,e - l,'," '*,t. / g .W. + - - {

x t,,

i l gg .a i 1 a il I i r 5 l \\

a ._-.,3 c. a m s 1 h s -g, i ~-~w.=.- :L9=;iW'T ~ v. I. de g g,h,h w ~ a>=. - .me %:,s=. - w t "t, a f.e,. g R4 W Janerm [mjfy%xag4 E c E g.i ~ . s, w , E a 8 ^* C 0 at. '... c, T>y.. =. :..: :. -mu, w s., ww:. n... .,N(. i t. .;, c. :;u, s. :; ~ q 7, g., ,...pcsW.. .i N,,,.l SJ n. .. wi ' ^ ^ J '. y, '=e' ..,$,.%.fl.'iY ,,Yag,y}?.; ' ,c,. ,_ u %[.IfeN hlh,I;Is,4. - -E + [.,',% .:,.,......7

• [*

fM."-],g w,[ w c... .3..... / .{ N {?h, i ' i w

sh4 , s as._ .m .a.a--.m6 e-__ ..aa_ a.am Am. .a ,aa a.w de 4-4&hw-+ J +m- ..n -.#4 E +m,.. - _A ,,s .*,.4 9 1, 1, ie k f .\\ "9' h l u + n k j %k-g. l vs e- = +q } 5, 3 og., sf 3 4 4 W l

==

.i l

)'

1 a

• t;. -

eg.. l l m ..g' I { .a .'2: f, 4,... >. 1 ~ ~ 3 n .., 'f g :.w -

• 4" s
  • • =.*3 g

t ....* ~ ' l ,,',d, 6-h. 's 4 e a, g 'n ( .._.,..J.'." l ~ ?.M l d. %T ..r ??... h y 4 49 i t. n~;i & ; #.. wn'

4 l

....x.... - .= . x. x c. a -- C.c ' - - , [#, ,.,j s \\ ((,4,. 'f n i 4 i o / l g l l p. ~ \\- (- ", # Q%. \\ .n.... ')

m.... a c;G,..

J, e . - r y Yo g l /- 6.- e. 16 t ' v. { l T ,/ y~~. g. ^: g ,e --wr p wy,---,.-.r-.,-,r-. =m.m.. g.-

n m ^ w ^ D,EN5 IO ~, ' A'; i ly n a lid':pild h S[k 18'%h,' $*I.. m, = w.~m y.; y,;g. g ~ ' ~ h5

• N kf;d,

.l sl I g'f"d / 'N @g.4 [ isri efeit I A N u I pgM 4 t r k '4 Volom -"% s r';

ibas is ~*

t ' ~ ~ * ~ ' ' l l'. Ory.';k f ," Gh I Sif APA g p GENE R A1CR Ettr(F [ D [# I2E55UElZ{# s3 inrini.tNut sj f' :q """5 K> Dt E-Is3 itsi II AllJ F', 'G g 9 ~~~~ M 'Qg,% fj s N V gSQy x. G# con Am,.,rni N WITH E NSE E ' d-IAtJ sh. LifJE IP cot 4f AINMFf 4f v EXPERIMENTAL SYSTEM V WITH THREE DIFFERENT O removabte components PRIMARY CIRCUIT COMPONENTS m.i co,e -n., s-g

-.s; # i s FUEL COOLANT INTERACTIONS Fuel coolant interactions do not imply only steam explosions (a-mode containment failure); TMI-2 event exhibits another, in this' case, beneficial, example of effects - of FCI on melt quenching I Issues: e: What are the relocation-scenarios.likely to lead to coolable configuration? - e'. - What are the: effects of FCI'on lower head integrity andL - cavity / pedestal integrity? What is the likely energy release and resulting load on e-containment? -21 ? .a 4_, s,4-4 --.re .-n-w. e .->-*3-+-n .e e <m h h ---

• eE.e'.vs.:

.w..E < sew-

• nm.<

.m<, e mw. m. 3 -n ,, w -r-h.'c~ ~ - -. + - - $. h v, ..v..~

y o. FUEL COOLANT INTERACTIONS FCI (CONTINUED? Approach: Explore certain key, generic.phenomenological aspects of FCI to address. the complex variations found in core melt progression and accident management. l Small-scale experiments to explore the fundamentals of FCI'(premixing, - propagation / fragmentation, yield) Develop / assess analytical models 22 l + .,_...r..-,. ~ -..... -

s__,

.m 4 ~...... --,.,-~.<,-f4,

d MISSILE BOLT FAILURE /b#' W//A V//) W74 Vf4 V//#5 Y* s !i l i j UPPER INTERNAL s STRUCTURES (UIS) t7 DOWNCOMER I t.!OUID SLUG g j _ 3, * ~ Y [,' "- m-- - q 'A{a$f,s 5, A y '.. Y., /;h -LOWER CORE PLATE y r 1umx=zr11.1. rrr r1z. z r:n@ $6 4 M CIRCUMFERENTIAL FAILURE / Lk.7 l DOWNWARD MISSILE r i

m FUEL-COOLANT INTERACTIONS FCI (CONTINUED) Large-scale experiments with prototypic melt and prototypic conditions to investigate melt jet fragmentation, quenching, vapor generation, thermal load and possible erosion on the lower head -- FARO (Ispra) Use probabilistic framework to couple the different stages of FCI -- usec' to quantify the a-mode failure 24

' y FUEL-COOLANT INTERACTIONS FCI (CONTINUED) Results: i a-mode of containment failure is low probability e Initial assessment of loads due to FCI is possible Premixing and fragmentation phenomena are adequately understood More research may be needed on ex-vessel fuel coolant interactions and to establish accident management options i . Plan: -Complete the ongoing. programs Convene a group of experts to. identify..the need for and value of any: additional:research -25

t

...m. .m. _____.___..,_m mm. m _,_._..____.__m___- ..m. s. [.. me,.E._, .E.,', ...__,.,,E',,

~ DEBRIS COOLABILITY PROGRAM issue: What is the effect on containment performance from debris-structure-water interactions? What are the mechanisms, if any, by which molten e corium, released from the RPV, may be maintained as a coolable debris bed? Approach: Limited number of ex-vessel coolability tests conducted at small scales Results: Following water addition, significant amount of energy was extracted from the melt No firm demonstration of yet long-term coolability, e and results still inconclusive i l l 26

/ t DEBRIS COOLABILITY PROGRAM LCONTINUED) Demonstration of debris coolability may.be e hampered by physical scale of test apparatus; at larger scale any top crust more likely.to.be fractured by thermal / mechanical stresses Plan: e MACE project planning for next integral test (M3) l (February 1994) l Large scale 1.2m x -1.2m (2000 Kg reactor. melt, 20 cm deep debris bed) . Currently assessing usefulness of additional testing o versus accepting current uncertainty. 27 '*,m-_..___ _ _ _ _ _ _ _--1-_ -.i-. m. .m .m. .-w.. ,_.w,,._. ,m,4 s ,.,e m.,. ,w y....

4

c5 E

3 3 =e-2 c s a A yO C3 0 k n ~n n 0 -n r e-2 e e _: a w 0$ 5 I ! b Et S sT 5 Ei 9 BE E s j g E !) Ii / / s' \\c / / i ! /'/ J /\\/ 1 il \\// / (/ ? / j, ,r 1 !! i l/ // / / ll I I 1 m j , o,* I ti l3 ,l e,i, ,o h:: l l

c o

m 7:: ,.,o 2, g p 2 lpl e .c / ////,/ / //// ll1 1 u ll///////// F"! y u 4 o 6 C 30 (U 85g Ed. e g u x ~ 8 5 g .O 8-o Bf si 5 !H I em g3 R a NY g 1l s t 91 te s t ie c iil I ggg, \\r n O o .:s i \\l n 8 o \\ l y \\ W L i (t) -c

:...... n

.s.. a =- ... u$.'. n a::e:. : x., . =....... = s

. ::h'* e.

0 e 0 ..:l,:l:ll@., 1.', 1 )=a' / / ) 0

n y a a 4 UTILIZATION OF THE RESEARCH RESULTS -Source Term Research Modification to 10CFR-Part 100 Updated source term for ALWR (NUREG-1465) improved-understanding of containment performance e 4 Individual Plant Examinations, Containment Performance improvements, and Accident Management Closure of severe accident issues that could potentially e lead to early containment failure'(Mark I liner failure .and direct containment heating) l i 29 = .. =

g ~ UTILIZATION OF RESEARCH RESULTS (CONTINUED) Provide data for severe accident plant performance criteria for ALWRs L SECY 90-016 Advance Notice of Proposed Rulemaking (Part 50) Specifically SECY 90-016 and ANPR require technical s basis for assessing capability of the containment to i - withstand the: severe accident loads: Direct containment heating Energetic fuel coolant interactions Direct melt attack' Hydrogen combustion.and detonation Slow pressurization (steam, non-condensibles)~ l 30-l. l: 1 m . m ..m-w -.- ec, .**w*- 1 w v-< -m_,-

~ UTILIZATION OF RESEARCH RESULTS (CONTINUED) Understanding the severe accident phenomena and e their range of behavior is important for taking advantage of existing reactor and containment capabilities; exploring accident management strategies for different containment types; developing plant performance criteria for future LWRs against severe accident challenges 31

~ FUTURE PLANS We plan to complete research needed to close on major remaining issues by 1996 e However, staff intends to maintain reduced level of capability to evaluate and predict severe accident behavior beyond 1996 staff will identify necessary disciplines for which expertise should be maintained staff will identify minimum suite of analysis codes to be maintained future efforts will involve moderate level of continuing research to maintain expertise (involvement in international programs, maintain code development and assessment efforts, possible small scale experiments) 32

~ FUTURE PLANS (CONTINUED? Staff will continue to address issues as they arise Plan to maintain cooperation with foreign participants Long term planning needed to determine requisite funding level to maintain capability -in progress I i l 33 I l

n INTERNATIONAL PROGRAMS NRC has bilateral agreements with the following countries who have joined the Cooperative Severe Accident Research Program: Bulgaria Czech Republic Finland France Germany Hungary Italy Japan JRC Korea Lithuania Netherlands NUPEC Russia Slovak Republic Spain Sweden Switzerland Taiwan UK e PaiGcipants receive NRC Severe Accident Research Results, codes, etc. NRC receives some cash and in-kind contribution NRC codes are used extensively internationally 34

,e r <[ h t ~ 1 w D d h T r w -p = -e 'r v l 5/ N O e ITAM e R W O M w w F 'w N 5 I 3 L -4 C AN O IT I W D D p A ~ I N A N W w w W + 4 I 41

g CORE MELT PROGRESSION issue: What are the characteristics of the melt released from the core and the vessel? Need melt mass, temperature composition (Metallic or ceramic), and rate and time of release Used as input for assessment of vessel integrity, containment loads, hydrogen production, fission products. What are threshold and location of ceramic pool meltthrough from a blocked core? Does metallic melt drainage form a core blockage in BWR dry core accidents? Approach: Ex-reactor experiments (blockage vs. drainage) in-reactor experiments (ceramic pool meltthrough) e Analysis 36

CORE MELT PROGRESSION (CONTINUED) Results: Blocked core gives large ceramic melt release like e TMI-2; very low metal content (~ 1% at TMI-2) e Ceramic melt release occurs on failure of the growing ceramic crust that contains the melt pool. e if metallic melt drains fro the reactor core, the later release from the lowuc aead (following dry out) is lower temperature metallic melt 37

.e .y

• A-y CORE MELT PROGRESSION (CONTINUED?

Plan: Metallic melt drainage or core blockage Plan two simplified XR1 tests and four XR2 experiments within the next year Meltthrough from a blocked core Performed MP-1 and MP-2 experiments in ACRR Analysis of the data from experiments is ongoing e e in late 1993, review group will. consider current results and the adequacy of understanding, and the need for and likely benefits of further research 38

y ro ~ HYDROGEN COMBUSTION issue: At high temperature the limiting concentration for detonability may approach the deflagration limit (for downward propagation) High-temperature high-speed hydrogen combustion phenomena: Effect of elevated temperatures on inherent detonability and on deflagration to detonation transition Results: Higher temperatures increase the mixture sensitivity Steam addition to hydrogen-air mixture decreases the mixture sensitivity Detonations observed to propagate in hydrogen-air mixtures with 10% hydrogen at 650K (no steam) 39 l

lY .e w o HIGH-SPEED HIGH-SPEED HYDROGEN COMBUSTION PROGRAM Plans 1 I Complete construction of the high-temperature e combustion facility (HTCF) e Use the HTCF to-perform-Inherent detonability experiments, Deflagration to detonation transition experiments with and without venting, Experiments will start in January 1994 e e Program will be completed in April-1996 i 40 1 ~ w e. .-.,,,.,--i - m --. - - e---,-6.-.w,- -.+---m> r +-mw -,}}