ML21356A854
| ML21356A854 | |
| Person / Time | |
|---|---|
| Issue date: | 11/17/2021 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| Abdullahi, Z, ACRS | |
| References | |
| NRC-1752 | |
| Download: ML21356A854 (124) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Metallurgy and Reactor Fuels Subcommittee Docket Number: (n/a)
Location: teleconference Date: Wednesday, November 17, 2021 Work Order No.: NRC-1752 Pages 1-88 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1716 14th Street, N.W., Suite 200 Washington, D.C. 20009 (202) 234-4433
1 1
2 3
4 DISCLAIMER 5
6 7 UNITED STATES NUCLEAR REGULATORY COMMISSIONS 8 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 9
10 11 The contents of this transcript of the 12 proceeding of the United States Nuclear Regulatory 13 Commission Advisory Committee on Reactor Safeguards, 14 as reported herein, is a record of the discussions 15 recorded at the meeting.
16 17 This transcript has not been reviewed, 18 corrected, and edited, and it may contain 19 inaccuracies.
20 21 22 23 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3 + + + + +
4 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5 (ACRS) 6 + + + + +
7 METALLURGY AND REACTOR FUELS SUBCOMMITTEE 8 + + + + +
9 WEDNESDAY 10 NOVEMBER 17, 2021 11 + + + + +
12 The Subcommittee met via Video 13 Teleconference, at 1:00 p.m. EST, Ronald Ballinger, 14 Chairman, presiding.
15 COMMITTEE MEMBERS:
16 RONALD G. BALLINGER, Chair 17 DENNIS BLEY, Member 18 CHARLES H. BROWN, JR. Member 19 GREG HALNON, Member 20 DAVID PETTI, Member 21 JOY L. REMPE, Member 22 23 ACRS CONSULTANT:
24 STEPHEN SCHULTZ 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
2 1 DESIGNATED FEDERAL OFFICIAL:
2 ZENA ABDULLAHI 3
4 ALSO PRESENT:
5 MICHELLE BALES, RES 6 JAMES CORSON, RES 7 ALADAR CSONTOS, Public Participant 8 JOSEPH DONOGHUE, NRR 9 KIMBERLY WEBBER, RES 10 11 12 13 14 15 16 17 18 19 20 21 22 23 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
3 1 P R O C E E D I N G S 2 1:00 p.m.
3 CHAIR BALLINGER: The meeting will now 4 come to order, this is a meeting of the Materials in 5 Metallurgy and Reactor Fuels Subcommittee of the 6 Advisory Committee on Reactor Safeguards. I'm Ron 7 Ballinger, Chairman of today's Subcommittee meeting.
8 ACRS Members in attendance are Charles 9 Brown, Greg Halnon, Joy Rempe, I have to go down here, 10 Dave Petti, and if I've missed somebody I'm sure I'll 11 hear about it.
12 We have our consultant also in attendance, 13 Steven Schultz. There may be others that will chime 14 in. Judging from the list of attendees, there's a 15 fair amount of interest here.
16 The purpose of this meeting is for the 17 Staff to brief the Subcommittee on the regulatory 18 information letter RIL 2021-13, interpretation of 19 research on fuel fragmentation, relocation and 20 dispersal at high burnout.
21 The Subcommittee will hear presentations 22 and buy and hold discussions with the research Staff 23 on this matter. A number of other people have just 24 chimed in, not yet, okay.
25 By way of background, the RIL is an update NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
4 1 to work in this area that depending on how old you 2 are, it's been ongoing for at least 40 years and Steve 3 Schultz says 50, related to cladding and brittle 4 performance during LOCA.
5 And then more recently, in the early 6 2000s, evolving to fuel fragmentation and dispersal at 7 high burnup.
8 There have been a number of documents that 9 have been released, NUREGs, another RIL previous to 10 this, 0801, and the latest to my knowledge fuel 11 fragmentation relocation and dispersal was NUREG 2121.
12 The ACRS has also issued a number of 13 letters related to this as well as embrittlement. The 14 latest one related to draft final rule of 5046C, which 15 is still under consideration by the Commission.
16 This RIL is a little bit unusual to my 17 mind because it not only summarizes research that's 18 been done in the past and does analysis on the 19 results, but also in Appendix A suggests a model that 20 might be used to accommodate that.
21 In effect, reset the burnup initiation 22 time from about 65 gigawatt days for metric tons --
23 sorry, I'm getting old -- to around 55. So, there's 24 a fair amount of meat in this that will be fruits for 25 discussion.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
5 1 The ACRS was established by statute and is 2 governed by the Federal Advisory Committee Act, FACA.
3 That means the Committee can only speak through its 4 published letter reports. We hold Subcommittee and 5 full Committee meetings to gather information to 6 support our deliberations.
7 I might add that since it's a Subcommittee 8 meeting, the opinions of the members are just that, 9 personal opinions.
10 The ACRS Section of the U.S. NRC public 11 website provides our charter, bylaws, agendas and 12 letter reports and transcripts and all full and 13 Subcommittee meetings, including slides presented at 14 open meetings.
15 The interested parties who wish to provide 16 comments can contact the office requesting time. I 17 think we have had a request for a public comment from 18 at least one member of the public.
19 Closed meeting transcripts are not posted 20 to protect any information that might be proprietary.
21 The Subcommittee will gather information, analyze 22 relevant issues and facts, and formulate proposed 23 positions.
24 The full Committee for the research is 25 scheduled to be held during December full committee NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
6 1 meeting, which is scheduled for November 30th through 2 December 3, 2021. The transcript of the meeting is 3 being kept and will be made available as stated in the 4 Federal Register notice through the COVID-19 pandemic.
5 Today's meeting is being held over 6 Microsoft Teams. For NRC Staff and the stakeholder 7 and public attendees, there's also an MS Teams 8 telephone number that allows participation of the 9 public to make comments in the comment period 10 specified in the posted agenda.
11 When addressing the Subcommittee, 12 participants should first identify themselves and 13 speak with sufficient clarity and volume so that they 14 may be readily heard. When not speaking, we request 15 that participants mute their MS Teams microphone or 16 phone number.
17 We will now proceed with the meeting. Let 18 me call on I think Kim Webber hopefully? Yes, Kim 19 Webber, the Director of the Division of Systems 20 Analysis of the NRC's Research Office to deliver her 21 opening remarks.
22 MS. WEBBER: Yes, thank you, Chair 23 Ballinger, ACRS Members.
24 It's really a pleasure to be here. Just 25 for the record, my name is Kim Webber, I'm the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
7 1 Director of the Division of Systems Analysis and we're 2 here to talk to you today about the research 3 information letter, or RIL, that's related to fuel 4 fragmentation, relocation, and dispersal, or FFRD.
5 The presentation is timely as it addresses 6 the performance of high burnup fuel at a time when 7 industry is interested in pursuing licensing of 8 current fuel designs to higher burnups.
9 And as you can see and as you noted in 10 your opening remarks, we have many industry 11 representatives and members of the public who have 12 interest in this topic.
13 So, it's good to see the broad interest 14 and participation in today's meeting.
15 The Office of Nuclear Regulatory Research, 16 or RES, has sponsored research of the behavior of high 17 burnup under loss of coolant conditions for over two 18 decades. I think you pointed out that the research 19 has been ongoing for even longer than that.
20 Over the last 10 years, the research has 21 increasingly focused on FFRD so today you'll hear from 22 the Staff about recent research which addresses 23 phenomena in which high burnup fuel has been observed 24 to fragment, relocate within the fuel rod, and 25 disperse into the coolant under loss of coolant NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
8 1 conditions.
2 The RIL is an NRC document and 3 specifically an Office of Research document that 4 communicates research findings between the Staff in 5 RES and other offices. In this particular case, the 6 RIL summarizes and communicates to NRR the available 7 research on FFRD.
8 And it provides insights regarding the 9 bases for empirical limits. In the past, RILS have 10 been used to support the development of regulatory 11 guidance and for this situation, the determination to 12 develop guidance associated with high burnup and/or 13 FFRD lies with NRR.
14 My Staff have worked very closely with 15 their counterparts in NRR to discuss the body of 16 research and the information that you will hear in 17 today's presentation.
18 We hope that in some way this will help 19 the licensing reviews of high burnup fuel in the 20 future. Before turning the presentation over to the 21 Staff, I want to say a few words about the third phase 22 of the cladding integrity program, or SCIP-III.
23 The SCIP-III program is one example of a 24 timely research investment that resulted in findings 25 that directly address the open regulatory safety NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
9 1 questions. It's a multi-party international research 2 program that allows NRC to accomplish complex research 3 in a highly leveraged way, both in terms of cost and 4 expertise.
5 For some topics addressed in the RIL, the 6 Staff incorporated conservatisms or stop short of 7 proposing a specific limit due to limited information 8 in the research. Many of these topics are being 9 investigated in the SCIP-III program, of which NRC is 10 a member.
11 The research conducted under the SCIP 12 program is proprietary to the members involved in that 13 program, however, the NRC sought permission from the 14 SCIP-III Management board, which granted our request 15 very recently to site-specific SCIP-III results in the 16 RIL.
17 And so prior to a week or two ago, this 18 information was not yet available to be published and 19 discussed publicly.
20 But we're grateful to the program and to 21 the Management Board because it enables us to make the 22 RIL publicly available, providing maximal transparency 23 about the research on this important topic.
24 And so with that, I'd like to turn the 25 presentation over to Michelle Bales.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
10 1 MS. BALES: Thank you, Kim, I appreciate 2 those introductory remarks from both the Chairman and 3 from Kim. I'm going to now share my screen for the 4 presentation. Can you see the presentation?
5 CHAIR BALLINGER: Very well.
6 MS. BALES: Okay, perfect. My name is 7 Michelle Bales and I will be starting the presentation 8 today and then turning it over to colleague who also 9 works with me in the Office of Research in a few 10 slides.
11 So, I'll today be presenting the research 12 information letter 2021-13. It's an interpretation of 13 research on fuel fragmentation, relocation, and 14 dispersal at high burnup.
15 The controls are a little bit different 16 than I expected. Today's presentation, I'll start 17 with a brief regulatory history of FFRD at NRC, some 18 of which has been mentioned in the introductory 19 remarks.
20 I'll then spend some time talking about 21 the program cited in the RIL and the peer review 22 process that we used. I'm then going to speak to the 23 outcome of the RIL, in other words what the RIL makes 24 possible in terms of safety analysis.
25 And after doing that, I will go back and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
11 1 speak about each empirical threshold and the basis for 2 it as documented in the research information letter.
3 There are a few other matters that we'll cover at the 4 end and then we will be wrapping up.
5 So, first, FFRD history at the NRC. The 6 first mention of fuel relocation and dispersal in a 7 regulatory context was in 2008 with the issuance of 8 RIL 0801.
9 This RIL was written just after fuel 10 dispersal was first observed in the Halden Reactor 11 Project on a road with over 90 gigawatts of data per 12 time.
13 The RIL 0801 discussed axial fuel 14 relocation and the loss of fuel particles through a 15 rupture opening, and recommended further research in 16 these areas. However, at that time it was documented 17 expected phenomena were occurring at burnups well 18 above current operating limits.
19 In 2012, the Staff conducted an extensive 20 literature review and published NUREG 2121. This 21 literature review captured the results of over 90 LOCA 22 tests performed in 8 different programs over 35 years.
23 Putting all of that information together, 24 the Staff concluded that the occurrence of FFRD 25 couldn't be precluded during the LOCA and really NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
12 1 required additional research.
2 In 2015 the Staff was in the middle of 3 rulemaking for loss of coolant accidents in the 5046C 4 rulemaking. And the Staff wrote a secy informing the 5 Commission that we did not plan to include or address 6 FFRD in that rulemaking.
7 But SECY also documented the Staff's 8 evaluation of FFRD and our basis for not including 9 FFRD in the proposed rulemaking.
10 At that time, the Staff also concluded 11 that immediate regulatory action was not needed to 12 address FFRD, however, as stated in that SECY, the 13 conclusion was closely linked to existing fuel design 14 limits and assumptions on how high burnup fuel would 15 be operated.
16 That brings us to today where the Staff 17 has written a new research information letter to 18 document the interpretation of fuel fragmentation, 19 relocation, and dispersal at high burnup.
20 We choose to do this now because industry 21 is pursuing extension of fuel design limits and also 22 because a large body of research has become available.
23 Documenting NRC's interpretation of available FFRD 24 research provides regulatory predictability and 25 technical review consistency.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
13 1 So, now I'll go into a little bit of 2 background on the program cited in the RIL and the 3 peer review process used. With this illustration, I 4 want to communicate that we've learned a lot about 5 FFRD since the 2015 SECY.
6 The 2015 SECY is shown on this bottom 7 timeline and you can see that both the SCIP-III 8 program and a program at Oak Ridge started and 9 finished since the SECY was issued.
10 The circles that you see are scaled to the 11 size of the program and the shading indicates whether 12 results have been public prior to the RIL, or solid 13 shading indicates that results are in the public 14 domain.
15 As Kim mentioned in her introductory 16 remarks, prior to the RIL, the SCIP-III data was 17 completely proprietary towards members and therefore, 18 the publication of this RIL marks the first time this 19 information will be in the public domain.
20 MEMBER REMPE: Michelle, this is Joy. If 21 you'll go back a slide? Again, there's a lack of 22 reduction in prototypic-ness when you go out of pile, 23 isn't there? Can you kind of give us a feel for what 24 phenomena are considered?
25 Like radiation effects when you're in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
14 1 Halden versus what you would get from the Studsvik and 2 the Oak Ridge test.
3 MS. BALES: Yes, I have a slide coming up 4 that provides some information on the hot cell setup, 5 which is actually very similar between the SCIP, Oak 6 Ridge and NRC's program.
7 So, when I get to that slide I'll be sure 8 to come back to that point because it's important that 9 some of the features of the test should be well 10 understood as we look at their results.
11 MEMBER REMPE: That's fine, whenever it's 12 appropriate is great. Thank you.
13 MS. BALES: I will come back to it. I 14 wanted to say a few words about our peer review group 15 because a peer review is not always part of --
16 (Telephonic interference.)
17 I don't know if I just got feedback but it 18 sounded like someone was asking a question.
19 CHAIR BALLINGER: I would remind people to 20 keep their computer muted. Thank you.
21 MEMBER REMPE: So, peer review is not a 22 necessary part of a research information letter but 23 when we look at the data resources and their recency 24 of the data, I want to explain that the nuclear fuel 25 community has not yet developed consensus around FFRD NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
15 1 phenomenon.
2 Significant research is still ongoing and 3 there's active work trying to develop a mechanistic 4 understanding of FFRD.
5 So, as you wrote the RIL, we recognized 6 that the empirical limits that we were proposing are 7 not consensus-based, mechanistic, or physics-based 8 models, but rather they represent the interpretation 9 of available data at this time from a safety authority 10 perspective.
11 Considering this, we wanted to pursue a 12 review and solicit perspective outside of NRC. So, we 13 specifically select reviewers with extensive 14 familiarity with FFRD research at national labs, 15 international labs, at national regulatory bodies, as 16 well as EPRI.
17 The peer reviewers were asked to identify 18 relevant research results that we might have missed to 19 identify alternative interpretation of the research 20 results presented in the RIL and identify gaps and 21 inadequacies of the basis that we proposed.
22 Ultimately, I want to make it clear that 23 the Staff was responsible for the positions taken in 24 the RIL and we were not subject to the peer review in 25 the sense that they could counter what we were saying.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
16 1 At the end of the day, there were some 2 instances where the Staff took a conservative position 3 based on our perspective, but we took the comments 4 that we received from the peer review very seriously.
5 And as a whole, the peer review process 6 significantly strengthened the document.
7 The peer reviewers provided very 8 thoughtful and detailed comments, they made us go and 9 sharpen our pencils in a number of areas, and in one 10 highlight I want to put in is in the RIL you'll see a 11 definition in the terms section now.
12 And I think that really came out of the 13 peer review where many of the reviewers wanted us to 14 be more definitive about this new terminology that we 15 were introducing in the RIL.
16 In the RIL, there's an appendix which we 17 will keep when it's finally published that summarizes 18 the peer review comments as well as the resolution.
19 So, it's written at a high level but that is included 20 in the publication to provide some transparency for 21 the peer review process.
22 So, as we move into the meat of the 23 presentation, I'm actually going to speak about the 24 outcome of the RIL before I speak to the basis for 25 each of the thresholds in the RIL.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
17 1 It's critical to communicate at a high 2 level how the RIL supports targeted safety evaluations 3 so I'm going to do that first. And Joy, this is the 4 slide that I mentioned. I'm going to say a few words 5 about the experimental methods used in all of the hot 6 cell testing programs.
7 I don't have a similar slide for Halden 8 but I'll speak to it a little bit as I walk through 9 the slides.
10 The RIL actually provides a short 11 orientation to test procedures in each of the programs 12 that are discussed because some details of these tests 13 are important to understand before examining the 14 results.
15 The majority of the test programs that we 16 cite were conducted in a hot cell and subjected to a 17 time temperature profile similar to the one in the 18 lower right-hand corner of the slide.
19 Test segments were placed in accordance 20 and heated in a clamshell furnace with four heating 21 elements. Again, this is characteristic of the hot 22 cell programs. In the Halden reactor, individual rods 23 were placed in a closed loop.
24 There were heaters, electrical heaters in 25 that loop but the whole entire loop was placed in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
18 1 core and the fuel rods were operated at low power 2 prior to the transient, so that means that the fuel 3 was producing decay heat during the LOCA transient.
4 So, in the case of the hot cell tests, all 5 of the heat that was experienced by the rod is from 6 external heaters and the thermal couple on the 7 cladding really set the furnace profile to target a 8 specific cladding temperature transient.
9 CHAIR BALLINGER: Michelle, this is Ron.
10 I have discovered painfully that knowing 11 the temperature gradient around the cladding is 12 important for these tests and when you use a 13 quadalyptic furnace, which is what this looks like, 14 and you ramp it up very quickly, are people sure --
15 I'm assuming this is a SCIP thing.
16 Are they sure the temperature around the 17 cladding is uniform? Because in past times, when 18 people have done oxidation experiments they've 19 discovered that for this type of furnace, you get kind 20 of a scalloped oxide thickness around.
21 And you can take an average but it's not 22 really representative of what you're saying. The 23 helical texture that's in the zircaloid tubing 24 combined with if there's an uneven temperature around 25 the circumference can complicate things.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
19 1 MS. BALES: So, I can speak to the SCIP 2 program. Before launching their irradiated testing 3 program, they did a lot of benchmarking of the 4 equipment, and that included some measurements of the 5 circumferential temperature profile.
6 They also looked at oxide measurements 7 after significant transients to look at whether there 8 was any scalloping or non-uniformity. So, the 9 equipment was qualified to produce sufficiently 10 consistent and uniform circumferential temperatures.
11 I guess there was a possibility that 12 individual tests that were done long after that 13 calibration was performed might have some unique 14 effects.
15 In some cases, some of the rods experience 16 bending after rupture, for example, in which case once 17 the rod goes off center, you can't confirm the 18 uniformity of the heating.
19 But I think that would be a unique 20 situation that we can look at to explain anomalies.
21 But generally, the equipment was looked at for that 22 and confirmed to not have any significant deviations 23 that would affect the results.
24 CHAIR BALLINGER: Because that usually 25 translates into an uncertainty bar on the temperature NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
20 1 versus time because the average sometimes does not 2 reflect the dispersion. Anyway, okay, thank you.
3 MS. BALES: I will say that in this 4 phenomenon, cladding temperature is not the most 5 critical part of the test.
6 That is certainly important for predicting 7 whether rupture occurs and the moment of rupture, but 8 as we go into the phenomenon that we're looking at of 9 relocation and dispersal, any non-uniformity of the 10 heating is probably going to be in the noise relative 11 to the things we're trying to measure.
12 Once you get above 760 which is the base 13 transformation, all the entropy goes away anyway I 14 suppose.
15 MS. BALES: The other things I wanted to 16 point out about the hot cell test are that the test 17 segments are about 30 to 50 centimeters. This 18 includes the Halden test and Powell was also similar 19 in length.
20 And the hot cell test but also the Halden 21 test were re-pressurized. They were segments cut from 22 commercially irradiated rods, repressurized through a 23 pressure line to various conditions dependent on the 24 test design.
25 So, as I said, I don't have a picture of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
21 1 the setup in the Halden reactor in this slide deck but 2 the RIL very briefly mentions this and we can speak 3 about it maybe separately if there's a lot of 4 questions about it.
5 There's been some work done to look at the 6 temperature profile across the cladding and across the 7 pellet in comparing hot cell tests and Halden tests.
8 And while obviously the hot cell testing 9 started the rod cold and there's a big difference 10 between what the rod temperature profile looks like 11 before the test.
12 There is some analysis that is showing the 13 temperature profile after about 20 seconds in the 14 transient is very similar between hot cell tests and 15 irradiated tests.
16 And particularly when you look just at the 17 temperature profile and it's ability to introduce 18 thermal gradients.
19 So, even if the absolutely temperature is 20 a little bit different, the fact that it's flat means 21 that the thermal stresses that would be induced across 22 the pellet should be pretty similar even for the hot 23 cell test and in-cell test.
24 As we get into further, that is part of 25 are understanding of the mechanism, that the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
22 1 temperature profile across the pellet would impact or 2 could impact fragmentation.
3 So, I don't know if that was as much as 4 you wanted know or if you have any other questions 5 about the test setups. There's a little bit of 6 discussion about this in the RIL but are there other 7 things that I should mention or that you'd like to 8 address before I go on?
9 MEMBER REMPE: Not at this time. Again, 10 it was something that puzzled me as I reviewed the RIL 11 and I'll ponder it a bit more if I have more 12 questions. But thank you.
13 MS. BALES: The test that I'm referring 14 to, they produce images of fragmented fuel, sometimes 15 fragments that are quite small in size. Dispersal was 16 observed in many of these tests.
17 Looking at the result separately, it can 18 seem that something alarming is happening at some 19 unknown point and it may at first seem intractable.
20 However, as documented in the RIL, a large body of 21 research is available to understand when this happens.
22 There is also repeatable and conforming 23 basis for clear and empirical limits. The RIL 24 provides the basis for when FFID does not occur and 25 allows for a well focused analysis of a relatively NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
23 1 small region of the core vulnerable to FFRD.
2 So, that brings me to the main outcomes of 3 the RIL. The Staff examined available research in 4 order to address five elements of FFRD. First, a 5 threshold for when fine fragmentation begins.
6 Next, a threshold for when we can expect 7 fuel axial relocation. We also sought to quantify a 8 model for predicting the amount of dispersal that you 9 might expect as a function of burnup.
10 The RIL also documents the phenomenon of 11 transient fission gas release distinct from the steady 12 state fission gas release that is already well 13 modeled. And finally, quantifying packing fractions 14 of axially relocated fuel in a balloon region.
15 So, putting these elements together, the 16 RIL forms the basis for targeted analysis of only 17 select rods that are of concern for FFRD. This is a 18 fundamental outcome of the RIL and therefore, I will 19 describe this outcome first and return to the basis 20 for each element after that.
21 To communicate how the RIL supports 22 targeted analysis, I will use an oversimplified 23 schematic to depict the rod population of an LWR core.
24 The schematic represents each assembly at 25 a given time in mind by its average burnup on the X NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
24 1 axis and operating power on the Y axis.
2 Burnup is a key parameter in FFRD 3 phenomena as I will explain. An operating power prior 4 to the start of a LOCA is a key determinant for 5 whether a rod will balloon and rupture, which is 6 another key parameter in FFRD phenomenon.
7 The RIL provides a basis to conclude that 8 only rods over 55 are susceptible to fine 9 fragmentation. The RIL also provides a basis to 10 conclude that only rods with strains greater than 11 three percent are susceptible to axial fuel 12 relocation.
13 Therefore, only high burnup rods with an 14 operating power prior to the transient that is high 15 enough to result in ballooning presented concerns for 16 relocation.
17 Since fine fragmentation and axial 18 relocation are precursors to fuel dispersal, only fuel 19 rods within this box are a concern for dispersal. In 20 this schematic that I'm showing here, there's only a 21 few rods that end up in this dispersal box.
22 However, the number of rods in this box is 23 highly dependent on core design, fuel design, and 24 plant characteristics. This box could be considered 25 just one of many factors that would influence core NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
25 1 design.
2 As you saw in the animation, there's other 3 core design maps that might have relatively lower heat 4 generation rates or have differences in how second 5 cycle and third cycle fuel rods are operated and may 6 have no overlap between the core loading map the 7 region of concern for FFRD.
8 MEMBER PETTI: This is Dave, I had a 9 question.
10 I understand this, it just seems the way 11 the rule is written, it mentions 55 as the burnup but 12 that's just sort of a necessary but not necessarily 13 sufficient metric to characterize if you've got an 14 issue.
15 And that doesn't come through as strongly 16 as what you just talked about here verbally with us.
17 When they read 55, people are going to go 62 is the 18 limit and now you're saying it's 55.
19 You're not really saying that, you're 20 saying that's just the first conditions is these other 21 things that could impact.
22 MS. BALES: It's an excellent point and I 23 think what we need to focus on is the RIL clearly 24 states that fine fragmentation, and I'm going to show 25 some results to demonstrate this, has been seen as NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
26 1 early as 55.
2 Fine fragmentation in and of itself is not 3 obviously presented as a safety concern. If that fuel 4 disperses, that's when you start to need analysis to 5 confirm safety. In order to disperse, there has to be 6 ballooning and rupture.
7 So, a high burnup rod above 55 that 8 doesn't balloon and doesn't rupture, you won't have 9 relocation or dispersal. And it's really the 10 dispersal that should be the focus of the safety 11 evaluations and relocation.
12 But I think your point is right that in 13 the RIL we tried to explain that fine fragmentation 14 and relocation are precursors to dispersal.
15 But perhaps it's not as clean, as I'm 16 walking through it now, that the 55 for fine 17 fragmentation only is not by itself the issue.
18 MR. SCHULTZ: This is Steve Schultz. Can 19 you hear me?
20 MS. BALES: Yes.
21 MR. SCHULTZ: You also mentioned the 22 strain number that was identified in the RIL 23 associated with the small-sized fragmentation. What 24 you seem to be showing here are fuel rods that are at 25 power levels in the core.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
27 1 And I presume what you showed in terms of 2 the data set that's included at that strain level is 3 for the assumed LOCA strain, in other words during the 4 transient. Is that correct?
5 MS. BALES: Correct, you mean as opposed 6 to operating?
7 MR. SCHULTZ: Yes.
8 MS. BALES: Yes, the strain that we're 9 speaking about that is a precursor to relocation is a 10 strain that occurs during the transient.
11 MR. SCHULTZ: And that's of course in 12 terms of what you're showing here like the burnup 13 limit value? That's affected by the assumptions and 14 the evaluation of the LOCA transient and the dynamics 15 of the fuel performance in the LOCA.
16 MS. BALES: Yes, I think the schematic is 17 really an illustration to describe the zones at a very 18 high level. But I think when we get into the 19 technical basis we can speak a little bit more cleanly 20 about what is precisely changing between relocation 21 and dispersal.
22 But one of the things that I was trying to 23 communicate with this, you see this box, any fuel rods 24 above 55 are potentially rods that can finally 25 fragment.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
28 1 But when we look at defining the box for 2 relocation, the horizontal line is my attempt to 3 communicate that only rods that are hot enough to 4 undergo ballooning would experience relocation.
5 So, on the left there's an image that 6 shows the strain limit which I haven't presented yet 7 but we'll get to in a second.
8 But the fact that the box has a horizontal 9 line to only include rods with high enough power to 10 strain under a LOCA is to communicate that this 11 relocation box is less than the population of rods 12 above 55 and subject to fine fragmentation.
13 And without seeing the boxes overlap, if 14 you just watch the figure, the dispersal box is 15 slightly higher, the idea that the difference between 16 relocation and dispersal.
17 There are some rods that are predicted to 18 balloon but not rupture. The dispersal box is 19 intended to only include rods which are predicted to 20 rupture.
21 MR. SCHULTZ: Thank you.
22 CHAIR BALLINGER: This is Ron again. I 23 keep wondering whether we need to be a little bit 24 careful in that the observed behavior is what it is, 25 it's observed.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
29 1 But a calculation for the LOCA calculation 2 is at least Appendix K or is a stylized calculation 3 where there's a lot of conservatism built in.
4 And so do we need to be careful that the 5 stylized calculation, if that's what it is, doesn't 6 get us burst or greater than 3 percent strain when 7 that's due to the conservatism in the calculation 8 that's built in.
9 MS. BALES: I think that's an excellent 10 point and I want to explain that the meat of the RIL, 11 well, the entirety of the RIL is focused on when this 12 phenomenon has been observed in testing, and it's 13 trying to establish empirical limits to define when 14 FFRD will occur.
15 Everything that I'm presenting now is 16 really just in service of communicating that we're not 17 talking about the entire core, that we're not even 18 talking about all high burnup rods.
19 But the schematic is a complete 20 oversimplification not only of a core loading map, but 21 also to what you're saying, where that horizontal line 22 cuts off is going to be extremely complicated to 23 calculate.
24 It's not easy to draw that lower line of 25 the box, that is dependent on the calculation method, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
30 1 the conservatisms used. And so the RIL doesn't 2 attempt to define a ballooning model, for example, or 3 attempt to draw that bottom leg of the box.
4 I'm using it in its presentation because 5 I want to communicate that the RIL is linked to 6 analysis but as we go into the technical basis, and 7 we'll spend more time over there, that's really the 8 meat of the RIL.
9 What has research shown in terms of the 10 onset of these phenomena?
11 CHAIR BALLINGER: I'm sure we'll further 12 discuss this.
13 MS. BALES: So, there's just two more 14 points that are addressed in the RIL that I wanted to 15 walk through the schematic to illustrate.
16 Again, it's just a schematic and it only 17 serves to explain the implications of the RIL, it's 18 just an illustration.
19 Trained fission gas release is another 20 thing that's addressed during steady state operation, 21 diffusion-based fission gas release is expected and 22 it's very well modeled by fuel performance codes.
23 However, transient testing has shown that 24 significant additional gas release can occur during 25 the transient, changing the rod internal pressure.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
31 1 The rod internal pressure is critical for actively 2 predicting whether rods burst when subject to LOCA 3 conditions.
4 And therefore, accounting for transient 5 fission gas release is critical for defining the 6 dispersal box. So, I think this further emphasizes 7 your point, whether the models are accounting for 8 transient fission gas release affects replacement of 9 this box.
10 Whether the models are accounting for 11 other factors in a conservative or best-estimate way 12 will affect the positioning of this box.
13 The final element the RIL addresses is 14 packing fraction and so this is particularly important 15 for rods where they're above the burnup limit so we 16 expect fine fragmentation.
17 They experience some ballooning and the 18 fine fragmentation can fall into the ballooned region.
19 Accurately accounting for that phenomenon 20 and having a density of relocation or packing fraction 21 is important for accurately predicting whether the rod 22 bursts because temperature is a critical element in 23 those predictions.
24 Also, even if the rod is not predicted to 25 burst, it's important that some accounting for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
32 1 relocation and packing fraction values are used in a 2 calculation of peak cladding temperature because that 3 is a key metric in LOCA evaluation.
4 So, in summary, with that said on the 5 slides, I wanted to communicate that the RIL forms the 6 basis for targeted FFRD analysis. Only rods with the 7 following characteristics are a concern for FFRD.
8 Fine fragmentation, we have not seen it 9 below 55, for axial relocation we haven't seen 10 cladding with less than 3 percent strain. Fuel 11 dispersal requires axial relocation and therefore can 12 be limited to rods that are above 55 and strained more 13 than 3 percent.
14 But dispersal itself can't happen unless 15 there's rupture so also, the calculation of the 16 rupture event is critical to determining fuel 17 dispersal.
18 CHAIR BALLINGER: This is Ron, I keep 19 beating a dead horse but in this slide, the only thing 20 that's actually calculatable with any precision is the 21 burnup. Everything else is a calculation based on a 22 fuel performance model, which may or may not be 23 accurate.
24 MS. BALES: I think it's important to 25 emphasize that the RIL doesn't prescribe how to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
33 1 analyze FFRD. The RIL simply defines when to analyze 2 it.
3 That means that vendors and their topical 4 reports are to define the models and the basis for 5 those models that need to be used in order to evaluate 6 these limits.
7 Having said that, I want to explain that 8 in some ways the means of analysis are not a dramatic 9 departure from the existing analysis. For example, as 10 you already mentioned, the burnup of each rod is 11 already tracked.
12 Evaluating whether a rod is above or below 13 that threshold is straightforward.
14 The empirical limits depend on ballooning 15 and burst and although there are variability in how 16 accurately that can be predicted, performance codes do 17 already have models for ballooning and the prediction 18 of rupture.
19 Some fuel performance codes even already 20 model axial fuel relocation. So, I said all of that 21 to say the RIL does not define how to analyze FFRD, 22 which models are acceptable, which models are even 23 recommended to match the experiments in the test.
24 It just simply leaves it at here's where 25 these phenomena had been observed and this is where NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
34 1 these phenomena should be analyzed, but the analysis 2 should largely map onto existing tools.
3 Whether they are assessment or ready for 4 use in this analysis is something that would be 5 addressed separately.
6 Now that I've given you a high-level 7 overview of what the outcome of the RIL is and what 8 the RIL makes possible, I'm going to step through each 9 of the empirical thresholds in turn and describe the 10 technical basis for each of the limits.
11 First the empirical threshold at which 12 fuel pellets become susceptible to fine fragmentation.
13 To define the threshold at which fuel pellets become 14 susceptible to fine fragmentation, we examine results 15 from 26 tests in which fuel fragment size was 16 carefully measured.
17 Fuel fragments were processed through a 18 series of sieves and the mass fraction of fragments 19 below 1 millimeter and 2 millimeters are shown in the 20 plot on the left. We see the mass fraction of fine 21 fragments increases with burnup, with some variability 22 between tests.
23 We have detailed fragment size results at 24 a burnup of about 43, showing no fragments smaller 25 than 2 millimeters, and then again at 6 days per ton, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
35 1 showing fragments start to be observed at these fine 2 size levels.
3 Unfortunately, we don't have any result of 4 fragment size distribution between these levels using 5 that sieve fine resolution picture.
6 But we do have optical microscopy from 7 testing at Argonne National Lab for an average of 55-8 second burnup before and after LOCA testing that 9 indicates fragments on the order of 1 to 2 millimeters 10 were observed after testing.
11 The A&L results are the ones that anchor 12 our position with the onset of fine fragmentation 13 occurs at 55 gigawatts per ton.
14 CHAIR BALLINGER: This is Ron again. I 15 look at this and I look at, let's say, the lower 16 figure at 66. There are two data-points which look to 17 be almost exactly the same burnup but the variability 18 is plus or minus 30 percent.
19 So, is that a good number to think about 20 in terms of uncertainty?
21 MS. BALES: That's a complicated question 22 because you're previewing one of my concluding 23 remarks, which is we have used burnup as the Y axis in 24 the RIL. And most likely, the mechanism that is at 25 play that controls fine fragmentation has more NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
36 1 variables than just burnup.
2 And so in this we're just simply plotting 3 everything as a function of burnup so in some ways 4 that is true that 30 percent could be an uncertainty 5 factor on the burnup threshold, but I think it's 6 probably more accurate to say that burnup is somewhat 7 of an incomplete parameter in the onset of fine 8 fragmentation.
9 And as we look to the future, a true 10 mechanistic understanding will probably put burnup 11 amongst other factors that control the onset of fine 12 fragmentation.
13 Once we have those delineated, some of 14 this data might be more easy to explain.
15 CHAIR BALLINGER: I wonder whether or not 16 the flow conditions, vibration and such, might also be 17 a significant factor that is unpredictable?
18 MS. BALES: I think in a real LOCA that 19 could be the case.
20 Most of these tests, except for the red 21 Halden tests, and all of the other data-points would, 22 I expect, have been subject to the same forces and 23 conditions because they were all conducted in the same 24 test setup where you have flowing steam in accordance 25 with the furnace heating.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
37 1 So, I wouldn't expect a lot of variability 2 between tests when it comes to the applied forces.
3 CHAIR BALLINGER: But the Halden results 4 say zero?
5 MS. BALES: For some of these burnups, 6 yes, that is true.
7 CHAIR BALLINGER: High burnup fuel, zero.
8 MS. BALES: I'm trying to bring your 9 attention to the one that is at 64.
10 CHAIR BALLINGER: I see that one but I 11 also see three non-bursts, but let's take the burst 12 ones. At least 2 at 73 or so, or thereabouts?
13 MS. BALES: There is certainly a lot of 14 variability in the fragmentation behavior and in 15 particular, some notable low fragmentation 16 observations in the Halden test.
17 So, I think this gets to the fact that we 18 don't have a fully mechanistic understanding.
19 This is really just empirical threshold 20 that is based only on what we have looked at right now 21 which is the effective burnup.
22 CHAIR BALLINGER: Thanks.
23 MR. SCHULTZ: The other question I had on 24 that slide was -- this is Steve Schultz -- each of 25 these data-points are representative of results NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
38 1 regarding fragmentation that are from a test where the 2 fuel rod has been exposed to a particular tailored 3 time, steam, temperature, pressure, condition presumed 4 to be appropriate for LOCA of a type.
5 In other words, rods and reactors, if 6 they're exposed to a LOCA, are going to behave 7 differently depending on their power history and what 8 their power level and burnup is at the time of the 9 event.
10 I presume there's a lot of variability in 11 the conditions to which these rods have been exposed 12 in the testing.
13 MS. BALES: Yes, there is variability in 14 the conditions of the tests. The time temperature 15 transient is a little bit different amongst these 16 tests. The rod internal pressure was different 17 between some of the tests.
18 In fact, in the SCIP-III program that was 19 one of the parameters that was being investigated, the 20 effect of rod internal pressure. So, that was 21 deliberately varied and so some of these tests have 22 different internal conditions.
23 In addition, I thought what you were 24 mentioning there is their operating history could also 25 be different and so getting to some of these burnups, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
39 1 some of these rods are coming from lead test assembly 2 campaigns.
3 Some are even reinserted rods so they were 4 experiencing a very unique power history during 5 operation. So, there's a lot going on between the 6 tests that with layers of resolution may explain some 7 of the scatter.
8 But at this point, when we looked at what 9 we could say with this level of knowledge, we plotted 10 it all against burnup and generally see an increasing 11 trend although there's a lot of variability.
12 Again, you're previewing some of my 13 comments at the end to wrap up but I think that some 14 of the future research will look at how those 15 different parameters affect fragmentation and could 16 lead to a reduction in the conservatism that's 17 depicted by this burnup threshold.
18 MR. SCHULTZ: You're speaking to my point, 19 I appreciate that. I'll wait for your conclusions 20 later too. Thank you.
21 CHAIR BALLINGER: Do we know what the 22 temperature profile history of this pellet was prior 23 to the test?
24 MS. BALES: Are you referring to the 25 optical microscopy that is shown?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
40 1 CHAIR BALLINGER: Yes.
2 MS. BALES: I don't have that handy with 3 me. The testing at Argonne was on commercially 4 irradiated rods and if it was at 55 I wouldn't have 5 expected it to be an LTA rod. I can look that up but 6 I don't have it handy with me.
7 I will say that amongst the tests that are 8 shown in the plot on the left, there was a lot of 9 variability in the operating history during commercial 10 irradiation. So, that represents quite a big span of 11 normal and possibly some extreme operating histories.
12 CHAIR BALLINGER: It's hard to tell from 13 looking at a computer screen of a micrograph of a 14 micrograph, but generally it's not just relocation 15 that occurs, it's restructuring that occurs sometimes.
16 And I'm looking at this circle at about 30 17 percent in or 40 percent in and wondering whether or 18 not that's generally an indication of something of 19 some restructuring that may have occurred.
20 And that also delineates the area on the 21 right where you get a lot of finds in radiuses greater 22 than that little ring. So, again, it's fodder for 10 23 or 20-page theses.
24 MS. BALES: It certainly is, and I'll just 25 say that I think what you're pointing to is the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
41 1 slightly darker circle.
2 It's a little hard because this is a map 3 of many different images so the quality to discern a 4 dark zone, for example, is a little hard in this 5 particular image.
6 The SCIP program has done some really 7 great characterization of pretty transient conditions 8 as well as post-transient conditions.
9 And I'll just say that this phenomena 10 where we see some more fragmented region in a circle 11 and less fragmented in the center and even on the 12 periphery relative to that mostly fragmented in this 13 ring has been seen additional places.
14 This is not the only time that pattern was 15 observed.
16 So, the SCIP reports that are cited in the 17 RIL are still proprietary to the SCIP members but 18 there's quite a bit of information that has been 19 documented to look at before and after and what 20 patterns, porosity, fragment size, and such to see if 21 there's a link to what the final fragmentation results 22 were that were observed.
23 And NRC's program at Studsvic, we actually 24 looked at a collection of fine fragments that were 25 collected in the sieves after the LOCA test, and we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
42 1 did isotopic analysis to try to determine where they 2 had come from with respect to the pellet radius.
3 And so we saw that the fine fragments that 4 were collected originated not just from the pellet 5 periphery but even some of the interior portions of 6 the rod were origins for those fine fragments.
7 So, that is all to say there has been work 8 to understand where relative to the pellet radius fine 9 fragmentation is occurring, and there's quite a bit of 10 variability.
11 Okay, as I suggested earlier when we were 12 talking about the 55 limit, fragmentation by itself 13 doesn't present a safety concern. However, if fine 14 fragments can relocate axially within a rod, there can 15 be safety implications.
16 On the issue of fuel relocation, tests 17 repeatedly showed that even when fuel was 18 significantly fragmented, not all fuel was able to 19 relocate axially.
20 In the image on the left, we see 21 measurements of one of the NRC's tests at Studsvic.
22 During this test, there was significant fuel dispersal 23 and we're actually going to show a video of this test 24 a little bit later, and I'll say a little bit more 25 about what was observed.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
43 1 But for this point, I want to say that 2 after the LOCA tests a wire probe was inserted into 3 the test segment through a rupture opening and could 4 travel quite far, as indicated by the green span that 5 you see in the lower portion of the graph.
6 After shaking the rodlet, which again I'm 7 going to show in a video in a little bit, additional 8 fuel was collected and the wire probe could travel 9 even further as indicated by the pink span.
10 These measurements suggest a zone of empty 11 cladding but after all testing was completed, a gamma 12 scan was performed and revealed a slightly larger span 13 empty of fuel.
14 So, perhaps more importantly, the gamma 15 scan revealed a section of the rodlet where fuel 16 remained and appears in tact.
17 Pellet interfaces can be discerned in the 18 lower portion of the rod down here, so this span is 19 about a pellet length and so we see a pellet interface 20 and relatively in tact fuel judging by the gamma scan 21 alone.
22 Similar examinations were performed in 23 additional rodlets and these measurements formed the 24 basis for saying that fuel relocation is limited to 25 regions of the fuel rod experiencing greater than 3 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
44 1 percent strain.
2 So, one of the reasons that we care about 3 fuel relocation is that it's a precursor to dispersal.
4 Before I speak to the measurements made of 5 fuel dispersal, I want to provide details of the 6 testing so that we can understand how dispersal 7 measurements were made, especially in the SCIP-III 8 program.
9 The video that I'm going to show will at 10 first be a closeup just below the furnace in the test 11 train. At this location we can see the court's tube 12 and a rodlet is above the frame.
13 At the moment of burst we will see fuel 14 fragments fall into the field of view below the 15 rodlet. The LOCA test is happening and then the burst 16 occurred and then the black dust that you see is fuel 17 material that fell out of the rupture opening and to 18 the bottom of the court's tube.
19 The scan is just showing to illustrate 20 what the rest of the test train looked like and then 21 this is in slow motion. So, after LOCA testing, the 22 fuel material that you see in the bottom of the test 23 train was collected and weighed.
24 And we have 19 measurements showing the 25 mass of fuel collected at this point in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
45 1 experiment. After the LOCA tests, the rodlets were 2 shaken and this video reveals what that looked like.
3 I show this because in the RIL we ready to 4 mobile fuel and dispersed fuel. In the first video 5 the mass of fuel that was collected just after the 6 test that was found at the bottom of the test train, 7 that's what we talk about as the dispersed fuel.
8 And when we talk about all mobile fuel, 9 we're talking about any fuel material that was able to 10 be shaken out of the rod. And when comparing 11 dispersed fuel to mobile fuel, we see that it all 12 comes down to the burst opening size.
13 The figure on the left highlights two 14 particular tests but the size breakdown of all mobile 15 fuel is quite similar. So, in these bar graphs, the 16 height represents the mass of fuel collected.
17 Blue was fuel that was collected after the 18 LOCA tests and the orange was additional fuel that was 19 collected during the shaking.
20 And you can see that the top graph and the 21 bottom bar graphs have similar situations where we 22 have a lot of particles that are between 2 and 4 23 millimeters, and above 4 millimeters in this case, and 24 then relatively smaller amounts of fine fragments in 25 each of the size bins.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
46 1 However, on the right of this figure I've 2 drawn a representation where the size of the two burst 3 openings is scaled, and so in the upper region the 4 burst opening was much, much larger than that of the 5 lower one.
6 And that is most likely an explanation for 7 why we saw such a difference in the dispersed mass 8 versus what was mobile in total. But if burst opening 9 is a key parameter we face a problem because burst 10 opening size varied between tests.
11 And there was not an obvious burnup trend 12 so when we are correlating in the RIL when we're 13 correlating this phenomenon with burnup, we can't 14 necessarily rule out large rupture openings in either 15 low or high burnup fuel.
16 And so for that reason, the dispersal 17 model that we propose in Element 3 of the RIL was not 18 based on the dispersal observed during the LOCA test 19 that I showed in the first video.
20 The model actually assumes all mobile fuel 21 could disperse and, therefore, it's tied to the limits 22 established for fine fragmentation and relocation.
23 The model proposed in the RIL is for a finely 24 fragmented fuel, all fuel above 55, that can relocate.
25 In other words, strain greater than 50 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
47 1 percent, all of that fuel could disperse.
2 Using this model to examine how well it 3 would predict all mobile fuel in seven LOCA 4 experiments in SCIP-III, we see that it is sometimes 5 conservative and sometimes not.
6 But we're interested in core wide 7 assessment and, therefore, considering these 8 predictions at a core-wide level we expect the results 9 to be conservative.
10 And other models were examined to some 11 extent and that's the documentation that you'll find 12 in Appendix A where we looked at different trend lines 13 that were taken from the size graph that was shown 14 earlier in the presentation, where fragmentation size 15 seems to sync with burnup.
16 But at the end of the evaluation, the 17 Staff concluded that conservatively at this point, 18 because burst opening is stacastic that we want to 19 create a model which was assuming all mobile fuel 20 could disperse.
21 So, there's two additional elements that 22 we're going to go into some details and for that I'm 23 actually going to turn it over to my colleague, James, 24 who is --
25 James, I'm assuming you want me to keep NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
48 1 running the slides or do you want me to turn the 2 presentation over to you?
3 MR. CORSON: You can keep driving the 4 slides. Hello, everyone, I'm James Corson and I work 5 in the same branch of the Office of Research as 6 Michelle.
7 So, I'm going to be talking about the 8 fourth element of the RIL and that's transient fission 9 gas release, which could impact ballooning and burst 10 behavior of rods under LOCA conditions.
11 As Michelle said, the modeling of 12 fission gas release during normal operations is pretty 13 well understood but during transients, there aren't 14 really a lot of good models to represent this 15 behavior.
16 So, you can see on the data there seems to 17 be an increasing trend where transient fission gas 18 release becomes more prominent at higher burnup but 19 there's more going on here, there's other factors than 20 just burnup.
21 The fuel temperature plays an important 22 role. Some tests have shown that you need a threshold 23 of about 600 degrees Centigrade before you get any 24 transient fission gas release.
25 Stresses in the fuel also play a role so NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
49 1 the cladding mechanical constraint or the internal 2 pressure, those can also influence transient fission 3 gas release, which is why you see quite a bit of 4 scatter on this graph here.
5 Since releasing more fission gas is going 6 to increase the rod internal pressure, that can impact 7 ballooning and bursts, it's something that needs to be 8 accounted for when trying to predict whether or not 9 your rod is building to burst.
10 And as we discussed that, whether you 11 rupture that determines whether or not you can get 12 dispersal.
13 So, right now there are some models that 14 have been proposed for a transient fission gas release 15 but they need a lot more validation at this point 16 before they're ready for use.
17 Next slide, the fifth and final element is 18 related to the packing fraction of the relocated fuel 19 in the balloon region. This is something that's been 20 studies quite extensively over the years, even for 21 lower burnup fuel.
22 But for this higher burnup fuel with fine 23 fragmentation you can get a higher packing fraction 24 and that's just related to the fact that if you have 25 a poly-dispersed size distribution, you can get a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
50 1 higher theoretically packing fraction.
2 And so SCIP data shows packing fractions 3 ranging from about 70 to 85 percent for fuel that's 4 susceptible to fine fragmentation. And then for the 5 lower burnup fuel they looked at, the one test had a 6 lower packing fraction, which is more consistent with 7 previous measurements.
8 So, again, it's important to account for 9 this because it can affect the LOCA heat transfer 10 behavior and thus impact whether or not you balloon 11 and burst. Next slide.
12 That's it for the five elements the RIL 13 addressed. To wrap up we're going to talk about a few 14 other matters we addressed. Next slide.
15 So, all of this, the RIL really focuses 16 on, again, what rods are susceptible to ballooning and 17 burst, fine fragmentation, relocation, dispersal. It 18 doesn't really talk about how this would be applied on 19 a core-wide basis or what the implications of 20 dispersal would be.
21 Instead it just points back to previous 22 publications in this area to discuss the potential 23 safety concerns.
24 And the ones that were listed and 25 discussed in this previous publications are energetic NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
51 1 fuel coolant interactions, re-criticality of dispersed 2 fragments, core coolability and long-term decay heat 3 removal, and potential radiological and source term 4 impacts.
5 The latter topic is being addressed as 6 part of an update to Regulatory Guide 1.183. Next 7 slide. Lastly, the RIL talks about some of the 8 limitations of the empirical database.
9 Michelle has hinted at some of these 10 things already but I just want to make clear that all 11 these tests were performed on standard uranium dioxide 12 fuel pellets in zirconium alloy cladding.
13 So, they weren't performed on any doped 14 fuel or coated cladding and so the limits aren't 15 really applicable to those fuel and cladding types.
16 Additional research may demonstrate that 17 the limits in the RIL may apply to these new fuel 18 types so that they're bounding but more work needs to 19 be done in this area.
20 I should note that doped fuel has 21 different fission gas release behavior so that can 22 impact fine fragmentation relocation and dispersal.
23 And on the other hand, coated claddings at 24 least in unfueled samples have been shown to have less 25 strain and smaller burst openings during burst NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
52 1 testing. So, if you have a smaller burst opening that 2 could significantly limit the amount of fuel 3 dispersal.
4 So, coated cladding could potentially have 5 some beneficial impacts but, again, more research is 6 needed to verify those findings. Next slide.
7 Again, as Michelle said, these are 8 simplistic limits. They were only derived as a 9 function of burnup but in reality, there's more going 10 on than just the burnup.
11 There's the power of the rod right at the 12 start of the LOCA or right before the tests were 13 conducted. Those sorts of things impact things like 14 porosity or stresses within the fuel, it impacts grain 15 growth and subgrain formation.
16 And all these things are heavily 17 influenced by the operating history so as Michelle 18 said, that could potentially explain some of the 19 scatter that was in the plots shown earlier.
20 But we need more research to allow for a 21 more mechanistic treatment. Right now, all we really 22 have is burnup and we have some preliminary 23 indications that there's something else going on but 24 we need more research to verify that.
25 And then finally, these limits anticipate NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
53 1 that we can accuracy predict the cladding stream along 2 the rod. Historically, that hasn't been a real focus 3 of ballooning models.
4 Rather, they focused on the burst 5 temperature and burst time or the maximum burst strain 6 at the burst location. They haven't necessarily 7 focused on the actual extent of the balloon region.
8 So, it may be necessary to make some 9 conservative assumptions to identify the mass of fuel 10 susceptible to relocation and dispersal. And as 11 Michelle said, we treated the burst opening size as 12 stacastic.
13 And so we based our limits assuming there 14 would be a large burst opening size.
15 However, there has been some research done 16 by the national labs in the U.S. to look at the 17 influence of different parameters and how they impact 18 burst opening size.
19 So, it's possible that in the future there 20 may be a more mechanistic model to predict burst 21 opening size. Next slide. Where does that leave us?
22 As Kim had mentioned in her introduction, 23 we're participating in the next phase of the SCIP 24 program and this actually includes some testing to 25 address the gaps that are in the existing database.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
54 1 So, it has some tests on doped fuel, some 2 additional tests to characterize transient fission gas 3 release, and more testing in the immediate burnup 4 range between 45 and 60 gigawatt days for MTU, where 5 there was a bit of a gap in some of the figures that 6 we showed.
7 Separately, at NRC we're working on 8 refining some of our analysis tools to perform coal-9 wide FFRD analysis.
10 We perform some preliminary calculations 11 in the lead-up to the SECY in 2015 where we used a 12 combination of scale and parks neutronics tools that 13 traced thermal hydraulic systems analysis code and 14 FRAPCON, fuel performance code, to estimate the total 15 mass of fuel dispersed during the LOCA.
16 Since then we've used similar workflows 17 for MELLLA+ analysis on BWRs, so we're looking to 18 extend those to more realistic core designs for LOCAs 19 at higher burnup and to use some of the newer modeling 20 features in our FRAPCON code and in improvements in 21 trace and so on.
22 I'm going to turn it back over to 23 Michelle.
24 MS. BALES: Thanks, James. This is 25 effectively the conclusion slide and I wanted to end NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
55 1 the presentation by explaining to you that the RIL is 2 a foundation for next steps.
3 As you would have maybe gathered from some 4 of the questions but also what James just talked about 5 in terms of the limitations, you could consider this 6 an awkward time to write the conclusions of FFRD.
7 There's a lot that we still don't 8 understand and if when we sit down to write an update 9 or a revision in five years, I bet it will be much 10 more precise and much more satisfying.
11 But at this time we wanted to establish 12 what we know today, especially as the industry is 13 looking to increase burnup limits and explain what we 14 know from the available research up until this point.
15 We fully expect research to continue and 16 we would anticipate, for example, industry building 17 from the RIL based on their licensing needs, 18 justifying less conservative limits with more detailed 19 or problem-specific arguments where they're able to do 20 that.
21 We also expect that research at the 22 national labs or in international collaborative 23 programs can build from the basis that the RIL 24 provides and produce information that is needed to go 25 further, either to refine the limits or make them a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
56 1 function of more properties than just burnup.
2 MEMBER PETTI: Michelle, question? What's 3 the timeframe for the additional testing that you're 4 talking about? Is it five years or less?
5 MS. BALES: The SCIP-IIII program is on 6 Year 2 I believe and that is where we'll get a little 7 bit more information about the midrange burnup as well 8 as some information of doped fuel.
9 I think there are additional research 10 programs that will continue for much longer than that.
11 We know about the SCIP-IV program because 12 we are members of that program but I believe there is 13 research well beyond that that could add to this body 14 of information that might come out in the next couple 15 years.
16 I can only speak to the SCIP-IV program 17 because that's the program we are members of.
18 CHAIR BALLINGER: Thank you for that 19 presentation.
20 I'm going to venture into some stream of 21 consciousness talking here I suppose, but we probably 22 should wait until we hear from Members first and then 23 we'll go and ask for comments from the public.
24 So, other Members, unfortunately we have 25 only four or five of our Members present and this is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
57 1 a very detailed presentation of a multi-variant 2 problem where without being able to be present at the 3 presentation, it's going to make it pretty difficult 4 for other Members to understand what's been going on 5 with the information.
6 So, in any case, Members, questions and 7 comments, please?
8 MEMBER PETTI: Ron, my comment, having 9 read the CAPS paper in JNM and the RIL is there's 10 still a heck of a lot of uncertainty around this.
11 There's questions about prototypicality of the entire 12 experimental database in a lot of ways.
13 You'd expect the in pile to be the most 14 prototypic but it has some problems, no experiment is 15 perfect. And the only way you get a resolution on 16 this sort of stuff is you do lots of testing and lots 17 of different experimental approaches.
18 And like an Impressionist painting you 19 step back. If you look real close you get lose. You 20 step back and you may see some better trends.
21 But there's still a lot that we don't know 22 and it all relies on the mechanical properties of the 23 cladding to find out when it's going to burst, the 24 size it's going to burst.
25 And we don't know the mechanical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
58 1 properties, the relations well enough. So, there's 2 models that exist, I understand that but they weren't 3 looking at this issue.
4 One may need a hell of a lot more 5 precision and better models than what's already there 6 to get a good calculation.
7 CHAIR BALLINGER: Now I'll start the 8 stream of consciousness discussion.
9 The problem is defined in the sense that 10 we have a right circular cylinder with dished pellets 11 and we have a more or less bounded power history for 12 Light Water Reactors.
13 But that's it, that's what constrains it 14 and like you say, it's a multi-variate problem with 15 several variables which are just undefined and maybe 16 unknowable in terms of what the values are.
17 And it's possible that you could develop 18 a program that would go on for the next 20 years and 19 still not get enough information, it's one of these 20 forever things.
21 And you wonder whether or not we should 22 come back and look at the consequences of the 23 phenomena to see if the consequences can somehow focus 24 the next or ongoing or future research efforts so that 25 it doesn't become a never-ending problem.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
59 1 Because there's probably 15 or 20 2 variables there.
3 There was an infamous figure that was 4 produced by I think Adrian Roberts, who has since 5 passed away, where he was being facetious and he drew 6 a chart which had the relationships between all the 7 variables and fuel performance.
8 And I think it was Adrian Roberts.
9 MEMBER PETTI: I've seen the figure, yes, 10 it was Adrian Roberts I believe.
11 CHAIR BALLINGER: And you sit back and you 12 laugh at it but it was correct. And so I'll just stop 13 talking and ask for other comments from Members.
14 MEMBER BLEY: Ron, this is Dennis. Your 15 stream of consciousness is worthwhile. It seems to me 16 at points some real discussion inside the Committee 17 and I think your scheduled for a letter but I think 18 you should be if you're not.
19 CHAIR BALLINGER: We are scheduled for a 20 letter but I'm not sure this is the right time for a 21 letter.
22 MEMBER BLEY: Maybe not but it's probably 23 time, given the number of people who were here, to at 24 least have the full Committee meeting and the others 25 know more about this --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
60 1 I'm not sure if it's time, it could be a 2 good time for a cautionary letter based on what you 3 were just saying. Let's look for places we can bound 4 the consequences and see how much it's worth, how much 5 and when it's worth really going a lot further in 6 this.
7 And most of the time in the past before 8 COVID-19, this was a typical turnout for the 9 Subcommittee and we had to have a pretty good 10 briefing, a couple hours over the same stuff. I think 11 that would be appropriate here.
12 Anyway, enough from me.
13 MEMBER REMPE: This is Joy, I would like 14 to follow up more on the question about the prototypic 15 aspects and non-prototypic aspects of the sources of 16 data and what Dave said I agree with, that there's no 17 perfect test.
18 And if we could ask the Staff before the 19 full Committee meeting to generate some sort of list 20 of pluses and minuses for each source of data to drive 21 home that point, I think it would be a useful task for 22 them to do.
23 What do you think?
24 CHAIR BALLINGER: I think you're correct 25 as far as it goes, but again, I keep coming back to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
61 1 how prototypic does it need to be? In other words, if 2 we need a very precise description of the problem, 3 then test data that's the -- I've got figure out the 4 right word.
5 How prototypic the data needs to be is 6 dictated by the level of detail that you need to base 7 decisions upon. Is that not correct?
8 MEMBER REMPE: I'm with you about we need 9 to bound the problem and I guess I'm doing that as an 10 exercise to drive home the point there's no perfect 11 test and maybe instead of doing more tests that will 12 still not be the perfect test and answer the question, 13 look somewhere else.
14 That's why I think it would be good to 15 help see why there's some problems with each source of 16 data and I didn't get that enough from the RIL. Okay?
17 CHAIR BALLINGER: The other wild card in 18 this is I'm aware of at least one vendor, a fuel 19 vendor, that's looking at this and they have had 20 meetings that I've participated in at least as an 21 observer with the Staff.
22 And so I know that at least one vendor is 23 looking at this and communicating with the Staff. One 24 wonders whether or not the others are not doing the 25 same thing, maybe they are, maybe they aren't.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
62 1 But that's a perspective which when you 2 deal with consequence is an important thing to listen 3 to.
4 And this meeting hasn't done that and I 5 wonder whether or not we shouldn't think about having 6 a follow-up meeting, regardless of whether we write a 7 letter or not, where if the vendors have enough 8 information that they can talk about, we get their 9 perspective as well.
10 What do Members think about that?
11 MEMBER PETTI: I think it's important to 12 do.
13 CHAIR BALLINGER: So, again, that's the 14 stream of consciousness thing. Who else do we have 15 here?
16 (Simultaneous Speaking.)
17 MEMBER PETTI: Steve has his hand up.
18 CHAIR BALLINGER: Who?
19 MEMBER PETTI: Steve Schultz.
20 CHAIR BALLINGER: Okay.
21 MR. SCHULTZ: I just wanted to comment on 22 your last note there.
23 And it also was brought up in James' 24 presentation that there is in fact a lot of work on 25 it, there has been a lot of work going on the, if you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
63 1 will, forcing function side of things over the last 20 2 years as new information and new evaluation and new 3 analyses have been done associated with the LOCA event 4 itself.
5 As you said, Ron, the experimental 6 database associated with LOCA does go back 40, 50 7 years but there's been a lot of work to look at what 8 is anticipated to be actual conditions in LOCA 9 analysis for the units as well as for the fuel.
10 In addition to that, and again James 11 mentioned it, new fuel types are going to have 12 different performance than those that had been tested 13 here.
14 And as I look at the research papers that 15 are associated with the testing, the researchers have 16 tried extremely hard to differentiate between the 17 cladding materials and the fuel materials that they 18 have in fact tested.
19 But it's very difficult to do, and the 20 technology in fuel design and cladding design has 21 really changed in the last 10 or 15 years and there 22 have been important changes that have improved at 23 least steady state fuel performance.
24 And most of those do in fact affect the 25 transient fuel performance as well. So, just in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
64 1 summary, there is a lot that could be done and if a 2 letter were to be written at this point, it certainly 3 could focus on what we've discussed today.
4 That's the connection between the 5 experimental evaluations and the conclusions that can 6 be drawn from it. But there's three or four other 7 areas that need to be explored including the overall 8 consequence evaluation.
9 CHAIR BALLINGER: Greg, what do you think?
10 You come from the operational side and so this has a 11 significant implication for the operational side.
12 MEMBER HALNON: Hold on, I've got 13 something going on. I'm sorry, Ron, I had a personal 14 issue. So, I'm coming up to speed on this, obviously 15 it's not in my expertise to look at the research and 16 whatnot, but there are some questions that enter my 17 mind.
18 And you all do a lot of research and 19 modeling and prototypical modeling and whatnot. It 20 seems that there's a lot of data out there in the 21 industry that could be ascertained from actual 22 configurations of fuel once it's been removed.
23 And I'm not sure if it's even physically 24 possible to slice and dice a fuel rod as a high burnup 25 and take a look at it.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
65 1 Maybe there is and maybe there isn't but 2 it just seems like the empirical data could be brought 3 in to the equations to confirm where we're going, that 4 would be good.
5 But overall, what I took away from this is 6 that at the very minimum, we as a Committee should 7 encourage prompt continuation of this research and get 8 to a conclusion where we can all stand back and say, 9 yes, that's a good limit.
10 And so the fuel fabricators and the 11 designers going forward, especially for the fuels 12 coming out like ATF, they really have to be --
13 My learned opinion is probably not as 14 useful as some of the you guys who get into more often 15 but that's where my head came away from.
16 CHAIR BALLINGER: I keep coming back, as 17 David has, to the multi-variability of the problem.
18 I was once on a Committee where we were analyzing and 19 doing stress analysis for a complex shape of a device.
20 And I was mystified by the results and I 21 pulled the finite element guy aside and I said, hey, 22 what are the real stresses here? He looked at me with 23 a straight face and said what do you want them to be?
24 And that looks to me like this is so 25 multi-variate, if we're not careful we'll end up NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
66 1 getting what we want them to be. And that could turn 2 out to be not a good thing.
3 So, okay, if we haven't heard from anybody 4 else --
5 MEMBER BROWN: Ron? I've just waited 6 since I'm obviously not a fuel person and I enjoy the 7 graph and data-points. When I look at a research 8 effort such as this, I try to look and see what it's 9 going to be used for, the end result.
10 And I guess Slide 31 or 32, 33 talked 11 about what's next, but all to me all it said was 12 justifying less conservative limits with more detailed 13 product-specific arguments.
14 And I guess I would have looked at this 15 data to say does this allow me to extract more power 16 from a particular fuel design or can I get longer life 17 from that fuel design based on a better understanding 18 of these characteristics?
19 But I didn't see that. Maybe I missed it 20 in the discussion, but it seems to me that if this is 21 going to be done, it ought to have something other 22 than just information value, other than the few 23 statements here about less conservative limits.
24 I wasn't sure how that was going to be 25 used. So, that was what I took away from the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
67 1 discussions.
2 CHAIR BALLINGER: That's a good comment 3 because that's another reason to want to understand 4 what the industry perspective is on this, and that we 5 haven't heard.
6 And so I keep thinking whatever we write, 7 that's definitely a recommendation at least from my 8 point of view. I'd be curious to know what other 9 members think about that?
10 MEMBER PETTI: I'm quickly skimming that 11 long paper from CAPS again and the thing that strikes 12 me -- I guess there's proposed testing in TREAT -- is 13 what you really have to do now --
14 Okay, we've done everything that's been 15 done to date, how do you do a better experiment?
16 Because it looks like the Halden stuff was 17 done at lower linear power rates than typical but they 18 were done at higher temperatures in terms of the 19 temperature which you get to than a lot of the cores.
20 And so how do you do something when you 21 kind of look at things a little more systematically 22 and start with stuff that you think is going to be 23 good but then things that are going to end up being 24 not good, and try to find that boundary.
25 Is there something that can be done taking NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
68 1 everything you've learned to date to design the next 2 generation of experiments, if you will? Because it's 3 hard to tell reading the paper how different was SCIP 4 from Oak Ridge.
5 If they're all very similar, then it's 6 just more data with slightly different run-ups and the 7 like. But there's a lot of stuff in the CAPS paper 8 that says that you could change the conditions a 9 little to be more prototypic and get answers that are 10 more prototypic.
11 CHAIR BALLINGER: Coming back to the 12 consequences issue, let's say everything that we've 13 learned today happens, let's say that in doing a LOCA 14 you do get ballooning, which we know would happen, you 15 do get dispersal, you do get relocation and all the 16 same.
17 How does that affect the answer? If it 18 has a great effect on the answer, and given all the 19 uncertainties, we have to factor that in, what are the 20 temperatures and all that stuff?
21 If it has a very strength effect, that 22 points you in one direction but if in effect, the 23 consequences are within the uncertainty of the 24 analysis, then it may be a nice thing to know that 25 happens but the consequences are such that you don't NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
69 1 need to have more detail.
2 Or maybe you need one specific set of 3 experiments to fill out something, to bracket 4 something, but those are two very different paths and 5 they're all dictated by what the consequences are 6 likely to be should it actually happen.
7 MEMBER PETTI: I don't disagree, Ron, I 8 just worry that as you propagate down the 9 calculational train here and you try to do more and 10 more, the uncertainty bounds tend to grow and in some 11 ways I worry that the phenomena that you try to model 12 get even more complex or uncertain.
13 I'd certainly be interested to know what 14 the impact is. If it seems to always be not a big 15 deal but that may be hard in the end. But I think 16 it's worth hearing from industry on that for sure.
17 MEMBER HALNON: This is Greg. What is the 18 burnup that we see in the three cycle fuel now? Isn't 19 it much less than 55?
20 CHAIR BALLINGER: We're over that.
21 MEMBER HALNON: I didn't have a feel for 22 it.
23 CHAIR BALLINGER: Peak rods are well above 24 65. Am I right there? Who should know? Michelle 25 should know.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
70 1 MS. BALES: Currently burnup lead average 2 is limited to 62, so there are some rods that are 3 above 55.
4 MEMBER HALNON: And that has been 5 increased, I assume, over the years just based on 6 better core designs and longer cycles.
7 So, the impact is obviously just going to 8 be in that realm, the life that cycles or the amount 9 of capacity factors that you're able to get, or all of 10 the above.
11 So, certainly the impact is going to be 12 from a physical perspective there, and in the fuel 13 design, I'm not real good on that so I don't know in 14 the fuel fabrication if there's going to be additional 15 margins that have to be put in there or not.
16 MEMBER PETTI: I think some of these 17 things can be done in parallel. It sounds like these 18 two experiments are going to go forward, they're going 19 to try to be more prototypic since the TREAT guys are 20 co-authors of the CAPS paper.
21 And then hearing from industry, they'll be 22 a nexus somewhere here whenever all this stuff gets 23 completed. It will be worth another look.
24 CHAIR BALLINGER: That's a very good segue 25 and I'm assuming that public comments will come from NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
71 1 members of the industry.
2 So, if there aren't any other comments 3 right now from Members, I guess the public line is 4 actually open so can we ask for comments from members 5 of the public? I'm not sure how we do this in terms 6 of who controls what.
7 But I guess if you want to make a public 8 comment, just do star 6 and start talking.
9 MR. SONTOS: Ron, can you hear me?
10 CHAIR BALLINGER: I can hear somebody but 11 I don't know who it is.
12 MR. SONTOS: This is Al Santos. Can I go 13 ahead?
14 CHAIR BALLINGER: You certainly can.
15 MR. SONTOS: Al Santos, NEI. I want to 16 thank the Staff for turning around this report so 17 quickly. We've only had about a week to review it but 18 the industry has a lot of detailed technical comments 19 on the report.
20 So, the industry requests an opportunity 21 and specific time to provide the detailed technical 22 expert comments.
23 Even if we considered a revision to the 24 RIL or to present at a follow-up ACRS Subcommittee or 25 full Subcommittee, full Committee meeting regarding NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
72 1 the RIL.
2 So, those are the first points we wanted 3 to make. The second one is that similar to what I 4 hear, it seems like there's a lot of discussion on 5 this.
6 The RIL as written today is a very 7 considered treatment of the data, even choosing to 8 apply additional conservatisms when presented with 9 evidence that a test may have already been 10 unrealistically conservative.
11 The industry continues to ask the NRC to 12 incorporate the research that has been summarized here 13 in a way that provides reasonable assurance of 14 adequate protection of public health and safety.
15 And lastly, we just want to reiterate that 16 the report, interpretive test data, and conservatisms 17 are in a manner that should be construed as 18 subjective.
19 And industry technical experts request an 20 opportunity to provide a comment and feedback to be 21 incorporated in the report or for the ACRS Committee.
22 We just wanted to have some high-level talking points 23 here to provide fuel.
24 CHAIR BALLINGER: Because of our 25 protocols, our rules, we can only have comments, but NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
73 1 I have a question for Michelle, she might know the 2 answer to this hopefully.
3 If we were to treat this RIL as if it were 4 a Reg Guide or a NUREG, it sounds to me like the 5 industry might offer public comments which may in the 6 end result in a modification of the document.
7 Is that one way to look at this? Somebody 8 is doing something here. So, would the Staff consider 9 effectively putting it out for public comments -- I 10 don't know how they treat RILs -- and take those 11 comments into consideration.
12 It sounds to me like the RIL was issued in 13 a very expeditious way but there are some comments 14 that may be important.
15 MS. BALES: It's a good question and I 16 think that in order to answer it I'm going to explain 17 -- and Kim alluded to this in her introductory 18 remarks, but maybe we can go back to the line of 19 thought that she was raising.
20 From the nature of documents the NRC 21 issues, a research information letter is designed to 22 be a communication between the Office of Research 23 Staff to NRR Staff.
24 And we wanted to do that very quickly in 25 a timely manner because we know that NRR Staff is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
74 1 receiving applications to extend burnup limits right 2 now and we wanted to put information in the hands of 3 those NRR reviewers so they could understand in a 4 concise way what information was available on FFRD.
5 So, that was our motivation for the 6 schedule that we kept and our target audience is NRR 7 reviewers. As Kim mentioned in her introductory 8 remarks, in the past RILs have been the start of other 9 regulatory documents.
10 So, they could be the basis for a later 11 guidance document or information that's contained in 12 a RIL could integrate into the standard review plan.
13 Any of those steps, which are so premature at this 14 stage, necessitate a public engagement.
15 So, if those steps were to be taken, it 16 would naturally happen that the industry and members 17 of the public would be able to engage because those 18 are clearly requirements of those types of documents.
19 So, a RIL typically would not have a 20 public comment period. I think I'm only aware of one 21 instance where there's been a public comment period 22 for RILs.
23 So, I just say that to explain what the 24 RIL is relative to other regulatory tools and to 25 explain that it would not surprise me if a year from NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
75 1 now we were talking about something that was built on 2 the RIL that does have public comment.
3 Now, that does not mean that we don't want 4 to continue to engage with stakeholders, I think 5 that's a critical part of NRC's principles of good 6 regulation is to continue to have transparency and 7 engage with our stakeholders.
8 So, I don't want to preclude that in any 9 way and I think that when it comes to the research 10 information letter, to the extent that we understood 11 the importance of going external, that's why we 12 proposed the peer review process, because we wanted to 13 get outside perspective.
14 We wanted to ensure that we weren't 15 operating in a vacuum with such a strict safety 16 authority perspective that we were missing something.
17 And so we sought outside peer review, recognizing the 18 potential for us to be too narrowly focused.
19 So, I'll just stop there and say I think 20 this is the start of additional documents and I see 21 Joe Donahue in NRR has his hand raised. I think he 22 would probably be more appropriate to finish my 23 thought.
24 MR. DONOGHUE: Chairman Ballinger, is that 25 okay with you?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
76 1 CHAIR BALLINGER: Sure.
2 MR. DONOGHUE: I'm the Division Director 3 in the Division of Safety Systems in NRR. I think 4 Michelle put it very eloquently and very well. I'll 5 just add we have been interacting as you know with 6 individual vendors.
7 We've also been interacting with 8 stakeholders in general on high burnup. We've held 9 two public workshops on the topic and I think Michelle 10 already explained the background and the motivation 11 for the RIL being reduced.
12 As far as getting public comments, she let 13 it out very well. This is a step in a direction that 14 is going to include public comment in some form. What 15 I don't want to do today is go out in front of my 16 staff who has to evaluate --
17 As Michelle said, our Staff gets the RIL, 18 we have to decide what's next specifically. Is it 19 just an update to this SECY paper that she mentioned 20 from 2015 or is it more than that?
21 Could it be a guidance document as you've 22 all alluded to? Not exactly sure yet. So, I think 23 that there's fast-paced activities in high burnup and 24 accident-tolerant fuel.
25 I don't want to ask Research to go through NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
77 1 a public comment period on this because that's going 2 to delay our ability at NRR to take the information 3 and start taking next steps. That's just my own 4 feeling.
5 It's up to you and the Committee on how 6 much you want to hear. The last thing I'll say is the 7 vendors are all preparing -- you've heard about some 8 of them, but in some ways they're considering -- in 9 one case we have a similar interim increased burnup 10 from one vendor.
11 But topical report reviews are anticipated 12 here and I think you will have the opportunity, the 13 ACRS will have the opportunity, of course, to interact 14 on all of those reviews.
15 CHAIR BALLINGER: Thank you. If I've 16 spoken out of turn in terms of procedure, I apologize.
17 But I wonder whether or not it would not be timely to 18 have a Subcommittee meeting on this particular topic?
19 Not the RIL but high burnup fuel and the 20 issues related to high burnup fuel, where we could 21 have both the NRC Staff as well as the vendors and 22 other stakeholders be able to speak.
23 And it sounds to me like there have been 24 a few public meetings that are sort of on this topic 25 in the recent path. What do Members think about that?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
78 1 MR. DONOGHUE: That would be helpful for 2 me because clearly an industry perspective and a 3 vendor perspective would round out the holes in my 4 head about this.
5 CHAIR BALLINGER: Joy, Dennis, Steve?
6 MEMBER BLEY: I liked the idea when you 7 first brought it up a while ago, I think that would be 8 a useful thing.
9 CHAIR BALLINGER: I'm trying to get a 10 little bit of my thoughts together and what to say at 11 the full Committee meeting, what a document would look 12 like should we produce one or whether or not we should 13 just recommend that we --
14 Thank you very much, we understand the 15 issues that have been presented to us, the 16 uncertainties are such that we think we should have 17 further discussions at a subsequent meeting, whenever 18 that should occur.
19 MEMBER BLEY: The only thing I'd toss in, 20 and you hinted at it, is if we decide to do what you 21 just suggested, our staff and the NRC Staff should get 22 together and we ought to make sure we're not 23 procedurally getting in the way of anything associated 24 with the RIL.
25 I think bringing the general information NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
79 1 back to the Committee is probably very helpful.
2 CHAIR BALLINGER: The RIL is what it is.
3 It's issued so there's no question of the Committee, 4 at least I don't think so, saying don't do this.
5 That's just not the way it works.
6 So, our normal letter comments would be 7 different in this case.
8 MEMBER HALNON: Ron, it just seems 9 premature at this point in that we would be getting 10 into some internal processes that we might influence 11 in the direction that is not necessarily fully 12 informed.
13 CHAIR BALLINGER: We could simply treat 14 this meeting as an information briefing, which is 15 exactly what it is and make comments assuming it was 16 an information briefing, with suggestions that we've 17 all bantered about.
18 That may be one way to look at it and that 19 would seem like -- oh, Michelle has got her hand up.
20 MS. BALES: I didn't want to interrupt but 21 I wanted to clarify the RIL isn't issued yet. It was 22 in concurrence and it was made publicly available in 23 draft form in order to facilitate this meeting.
24 But technically, it is not issued so I 25 wanted to just clarify that.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
80 1 CHAIR BALLINGER: Thanks, I don't know how 2 often the Committee has looked at or had meetings on 3 a RIL but I don't think it's been very frequent. Al 4 has his hand up or maybe it's been up for a while and 5 he didn't put it down.
6 MR. SONTOS: I just wanted to say that to 7 Michelle's point, there has been a past present of RIL 8 being commented on by technical experts from the 9 industry. So, putting that out there, that's all.
10 I just wanted to let that be known.
11 CHAIR BALLINGER: Thank you.
12 MEMBER BLEY: Here's a question, can we as 13 a Committee look at RILs? I was trying to search my 14 memory.
15 There have been a number of cases where 16 we've looked at your section of the SRP or a reg guide 17 or a NUREG that also had associated with it a RIL and 18 came under our review as part of the other documents.
19 So, we've seen them before coming before 20 the Committee, I don't think just by themselves but 21 I'm not positive of that.
22 CHAIR BALLINGER: You've been here a lot 23 longer than I have so you would know. Okay, I'm going 24 to have to try to put thoughts together and I'll 25 solicit --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
81 1 Now there's a flashing blue circle around 2 Dave. What does that mean?
3 MEMBER PETTI: I don't know, I just got 4 unmuted to ask you are we going to recommend that we 5 write a letter or not?
6 CHAIR BALLINGER: I'm casting about for 7 suggestions, I'll send an email out to everybody was 8 well because my memory is not as good as me writing 9 something down. And people are taking notes.
10 But yes, I'm casting about for 11 recommendations of whether we should write a letter of 12 some kind and what it should say.
13 MEMBER PETTI: I'm struggling with a 14 letter that would help the draft RIL be, quote, 15 better, that's one thing.
16 The letter could be much broader and talk 17 about the issue in broader terms but it wouldn't 18 necessarily affect the subsequent steps of the RIL.
19 Same thing, we need to get industry's 20 perspective. Some of the things we've heard, I don't 21 know that they affect the next step. So, I'm with 22 you, I'm struggling as well. I'm trying to figure out 23 exactly what we'd say.
24 MEMBER REMPE: Could we ask Michelle or 25 maybe Kim what would a letter from ACRS do? We didn't NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
82 1 ask you at the beginning of this meeting, which I 2 sometimes do. What does the Staff anticipate from 3 ACRS and how does it affect your schedule?
4 What would you do with a letter if there 5 were some recommendations or comments in it? Is it 6 too late, this thing is going out the door and you'd 7 consider it as you go forward?
8 What didn't does it make in your opinion?
9 MS. WEBBER: Go ahead, Michelle, I'll add 10 in if I need to.
11 MS. BALES: I was just going to explain 12 that in our office instructions for RILs, in the 13 section on peer review, it acknowledges the ACRS 14 Committee is a body that can serve as peer review.
15 So, in one element we can just think about 16 the benefit of the ACRS writing a letter that examines 17 the RIL from a peer review standpoint and makes 18 comments about your technical review.
19 And I think with respect to the timing of 20 it, the RIL is done from the Staff's perspective, it 21 is complete with concurrence but we're very conscious 22 of not issuing it until the ACRS meetings were 23 complete.
24 Because we anticipated the possibility 25 that ACRS members would have important comments that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
83 1 we needed to consider. So, I think in one respect a 2 letter that serves to expand the sense of peer review 3 is certainly one element.
4 I'll just say for myself, and I'm just 5 reacting to the things in real time so I haven't 6 thought about this too much.
7 Nut some of the ACRS Members were 8 commenting on pulling back to see where does this 9 information fall relative to a broader either research 10 need, or where these phenomena fall relative to 11 looking at upstream LOCA possibility events or 12 downstream consequences.
13 I think that's another area that I thought 14 the ACRS comments were very important to put the RIL 15 in perspective.
16 So, even just in this meeting I think that 17 we've gathered that but it could be that putting that 18 into a letter helps to just place the RIL in the 19 context of a bigger picture.
20 MEMBER REMPE: Michelle, I'm a little slow 21 and so I heard you say at one point from the Staff's 22 perspective the RIL is done.
23 But then I heard you say, which I am more 24 sympathetic and encouraging of, is that you considered 25 us a technical peer review, you've heard things but it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
84 1 was from individual members that you had thought you 2 didn't want to issue the RIL until the full Committee 3 had written a letter because there might be some 4 changes.
5 Is the latter statements that I've 6 summarized where you're at, that, yes, there is the 7 opportunity that our letter might make you hold off in 8 the final publication and you might consider some of 9 our comments?
10 MS. BALES: Yes, sorry, when I said the 11 letter was done from a Staff's perspective, what I 12 mean is there are no action items that we have on our 13 list that we are still waiting to incorporate, or no 14 additional data that we're waiting for.
15 The only input that we have not received 16 that we know we want to receive are these ACRS 17 meetings. So, what I mean from the saying that the 18 Staff is done, there's nothing else that we know to do 19 that we haven't yet done except for these ACRS 20 meetings.
21 So, the latter is true.
22 MEMBER REMPE: Thank you, that helps.
23 Kim, did you want to add anything else?
24 MS. WEBBER: No, actually, Michelle did a 25 fantastic job explaining it much better than I could NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
85 1 because she's very familiar with the RIL process.
2 MEMBER REMPE: Ron, I would like to have 3 a presentation. I'm just one Member but I'd like to 4 see a presentation at full Committee and I think it 5 would be a worthwhile letter for us to put together at 6 this time.
7 Because we do have this opportunity and 8 the Staff has said they're waiting to hear our 9 comments.
10 CHAIR BALLINGER: We've assumed that would 11 be the case and that's why we scheduled the full 12 Committee meeting. So, we're slotted into have this 13 take place so I think for sure, the only questions 14 that I have are the tone and content and scope of the 15 letter.
16 I think I've got a few ideas and I'll be 17 soliciting within the next five minutes written 18 comments from Members that are in attendance so that 19 I can strike while the iron is hot, or Zana and I can 20 strike while the iron is hot.
21 So, that would be the path forward, unless 22 people have other suggestions. Dennis, the blue 23 circle is flashing around you now. I don't know what 24 it means.
25 MEMBER BLEY: I had my mic on for a while NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
86 1 then I turned it off.
2 CHAIR BALLINGER: Kim, your hand is up.
3 MS. WEBBER: I was just thinking about 4 Michelle's response and the value of your review and 5 ideas around the RIL.
6 I just wanted to add that I fully agree 7 with Michelle that having your perspectives on the RIL 8 in a letter would be very valuable to us.
9 I know it takes more time, there's a 10 presentation writing the letter, but it would be of 11 value to us and so I think we're looking forward to 12 that.
13 CHAIR BALLINGER: Thank you, again. So, 14 now it comes to the full Committee. I don't have it 15 exactly in front of me so I don't know how much time, 16 Zana might know, we have slotted. Is it an hour or 17 two?
18 MS. ABDULLAHI: Let me look, I'll get it 19 to you in a few minutes.
20 MEMBER BLEY: Ron, whatever it is, I would 21 suggest that we try to make sure it's at least two 22 hours2.546296e-4 days <br />0.00611 hours <br />3.637566e-5 weeks <br />8.371e-6 months <br /> between now and then, because I don't think an 23 hour2.662037e-4 days <br />0.00639 hours <br />3.80291e-5 weeks <br />8.7515e-6 months <br /> would be long enough to --
24 (Simultaneous Speaking.)
25 -- the other Members. I think you need a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
87 1 couple hours.
2 CHAIR BALLINGER: We started at 1:00 p.m.
3 and it's 3:00 p.m. and we've had --
4 MEMBER REMPE: This is Joy and I'm just 5 going to interrupt you. Right now you're scheduled 6 for Wednesday morning at 8:30 a.m. to 10:00 a.m. for 7 the presentation.
8 But Matt often puts in an hour right after 9 so he's got, really, from 8:30 a.m. to 11:00 a.m. for 10 the combined time to have the presentation and read 11 him the letter. That's something you might be able to 12 negotiate with him.
13 That's where it's at right now.
14 CHAIR BALLINGER: My guess is that if we 15 allowed that time for presentation and questions as 16 opposed to reading in a letter and then read the 17 letter later on, that might free up some time.
18 Because my guess is that by that time 19 we'll also get requests for members of the public to 20 make comments. So, I think if we go two and a half 21 hours2.430556e-4 days <br />0.00583 hours <br />3.472222e-5 weeks <br />7.9905e-6 months <br /> -- did I do the math right?
22 MEMBER REMPE: Yes, you did and by the 23 way, I'd also note that from 1:00 p.m. to 2:30 p.m.
24 was the research review and we got done with that so 25 we do have some flexibility with the schedule in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
88 1 December.
2 I'm just pointing out some things and what 3 Matt prepared, I'm not the chair.
4 CHAIR BALLINGER: That's good. I don't 5 see anybody's hands or anything up, I guess we've got 6 to notify Matt that we shortened up the meeting once 7 again. So, if there aren't any other comments, I 8 think we are finished.
9 On that basis, the meeting is adjourned.
10 MEMBER REMPE: Thanks everyone for the 11 presentation and discussion.
12 (Whereupon, the above-entitled matter 13 went off the record at 3:08 p.m.)
14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
Research Information Letter 2021-13:
Interpretation of Research on Fuel Fragmentation, Relocation, and Dispersal at High Burnup Briefing of the Metallurgy & Reactor Fuels Subcommittee of the Advisory Committee for Reactor Safeguards November 17, 2021
This presentation will address FFRD history at NRC The programs cited in the RIL and the peer reviewers used The outcome of the RIL The basis for empirical thresholds included in the RIL Other Matters
This presentation will address FFRD history at NRC The programs cited in the RIL and the peer reviewers used The outcome of the RIL The basis for empirical thresholds included in the RIL Other Matters
FFRD history at NRC
- RIL 2008-01, Technical Basis for Revision of Embrittlement Criteria in 10 CFR 50.46
- NUREG-2121, Fuel Fragmentation, Relocation, and Dispersal During the Loss-of-Coolant Accident
- SECY-15-0148, Evaluation of Fuel Fragmentation, Relocation and Dispersal under Loss-of-Coolant Accident (LOCA) Conditions Relative to the Draft Final Rule on Emergency Core Cooling System Performance during a LOCA (50.46c)
- RIL 2021-13, Interpretation of Research on Fuel Fragmentation, Relocation, and Dispersal at High Burnup
This presentation will address FFRD history at NRC The programs cited in the RIL and the peer reviewers used The outcome of the RIL The basis for empirical thresholds included in the RIL Other Matters
Data Sources for RIL ORNL, Hot-cell SCIP-III, Halden, Hot-cell In-Pile NRC
@Studsvik, Hot-cell S
E C
Y 2000 2010 2020
Peer Review Group
- Ad hoc, issue-focused group. Reviewers selected with extensive familiarity of FFRD research
- Nathan Capps, ORNL - Author of two significant publications on FFRD, extensive experience modeling aspects of FFRD and collaborator in ORNL FFRD experimental program
- Tatiana Taurines, IRSN - Extensive experience modeling aspects of FFRD and leader in SCIP program review group discussions
- Fabiola Cappia, INL - Collaborator in FFRD publications, lead for INL PIE campaigns of HBU fuel, extensive work examining evolution of fuel microstructure with burnup
- Ken Yueh, EPRI - Designed and led FFRD research campaigns at Studsvik for EPRI and NSUF at ORNL/INL, leader in SCIP program review group discussions
- Daniel Jderns, Studsvik - Collaborator in SCIP experimental design and expert in SCIP results
Peer Review Outcome
- Peer Reviewers provided thoughtful and detailed comments
- Reviewers pointed to a number of places where our positions could be stronger. The edits result in a RIL that is more solid.
- Reviewers suggested a definitions and terms section.
- We have a 5 page Summary of Peer Review Comments and Resolution that may accompany the RIL as documentation of the peer review process.
This presentation will address FFRD history at NRC The programs cited in the RIL and the peer reviewers used The outcome of the RIL The basis for empirical thresholds included in the RIL Other Matters
Hot-cell testing to simulate Loss-of Coolant Accident condition Pressure line establishes segment pressure 30-50 cm refabricated fueled segment from 4 heating commercially elements in irradiated rod furnace
Outcome of the RIL:
Identify when FFRD occurs Segment from NRCs ANL LOCA program at 55 GWd/MTU before and after testing
Outcome of the RIL:
Identify when FFRD occurs Address five elements of the RES staffs interpretation of FFRD research and describe the technical basis for these elements:
- 1. Fine fragmentation threshold
- 2. Fuel relocation threshold
- 3. Model to quantify dispersal
- 4. Document transient fission gas release
- 5. Quantify packing fractions in the balloon region
Outcome of the RIL:
Identify which rods are a concern for FFRD Fine fragmentation requires ~55 GWd/MTU Power (LHGR) 1st cycle 2nd cycle 3rd cycle Burnup
Outcome of the RIL:
Identify which rods are a concern for FFRD Relocation requires ballooning Power (LHGR) 1st cycle 2nd cycle 3rd cycle Burnup
Outcome of the RIL:
Identify which rods are a concern for FFRD Dispersal requires fine fragmentation, relocation and burst Power (LHGR) 1st cycle 2nd cycle 3rd cycle Burnup
For the dispersal zone, some core loadings may result in no region of overlap Overlap influenced by: Dispersal requires fine fragmentation,
- Plant design
- Loading pattern
- Fuel and 1st cycle 2nd cycle cladding 3rd cycle design This information is prototypical of PWR. BWRs will have few if any rods susceptible to dispersal due to different operating practices, system pressure, etc. Burnup
Transient Fission Gas Release - Increasingly important at higher burnup; Important to accurately predict burst Dispersal requires fine fragmentation, relocation and burst Power (LHGR) tFGR influences burst probability 1st cycle 2nd cycle 3rd cycle Burnup
Relocation packing fraction - Important to accurately predict burst and peak temperature for non-burst high burnup rods Power (LHGR) 1st cycle 2nd cycle 3rd cycle Assume the balloon region is 70-85% fuel (15-30% void) by volume for calculating Burnup rod behavior
The RIL supports targeted FFRD analysis
- Fine fragmentation
- Not seen below 55
- Fuel axial relocation
- Not seen in cladding with less than 3% strain
- Fuel dispersal
- Fine fragmentation and relocation are prerequisites; Therefore, dispersal requires BU>55, strain>3%
- Doesnt happen unless theres rupture
- Transient Fission Gas Release
- Increasingly important at higher burnup
- Relocation packing fraction
- Important for non-burst high burnup rods
This presentation will address FFRD history at NRC The programs cited in the RIL and the peer reviewers used The outcome of the RIL The basis for empirical thresholds included in the RIL Other Matters
Element 1: Empirical threshold at which fuel pellets become susceptible to fine fragmentation Segment from NRCs ANL LOCA program at 55 GWd/MTU before and after testing Research supports a pellet-average burnup limit of 55 GWd/MTU as the onset of fine fuel fragmentation FFRD
Element 2: A local cladding strain threshold below which relocation is limited Strain Strain NRC test # threshold, threshold, top (%) bottom (%)
189 6.0 3.0 191 6.0 4.0 192 5.0 4.0 193 1.0 4.0 196 3.0 5.0 198 4.5 9.0 Research suggests fuel relocation is limited in regions of the fuel rod experiencing less than 3% cladding strain.
Element 3: A conservative value for the mass of dispersible fuel as a function of burnup What do dispersal measurements look like?
Dispersal during the test
Element 3: A conservative value for the mass of dispersible fuel as a function of burnup FFRD
Element 3: A conservative value for the mass of dispersible fuel as a function of burnup Difference between dispersal predicted by the model and all mobile fuel observed in the experiment SCIP test Mass (g) Prediction/Measured OL1L04-LOCA-2 125 250%
N05-LOCA -19 76%
VUR1-LOCA-1 15 109%
WZR0067-LOCA -16 83%
VUL2-LOCA1 -7 94%
VUL2-LOCA3 8 105%
VUL2-LOCA4 5 102%
Recommend a model to predict the mass of fuel dispersal to be all fuel above the burnup threshold of 55 GWd/MTU in the length of the rod with greater than 3% cladding strain to disperse.
ALL collected fuel FFRD
Element 4: Provide evidence of significant tFGR that may impact ballooning and burst behavior of high burnup fuel under LOCA conditions Data shows increasing transient fission gas release with burnup. However, many other factors besides burnup impact tFGR (e.g., fuel temperature, stresses in fuel).
Licensees will need to address tFGR in their LOCA evaluation models. Some models exist for tFGR, but more validation of those models is needed.
Element 5: Establish a value for the packing fraction of relocated but non-dispersed fuel in the balloon region It is reasonable to use packing fraction values between 70 to 85 percent for fuel susceptible to fine fragmentation. (Fuel at lower burnup would likely have a lower packing fraction).
To determine the impact on ballooning and burst, it is important to examine a range of packing fractions to account for these effects.
This presentation will address FFRD history at NRC The programs cited in the RIL and the peer reviewers used The outcome of the RIL The basis for empirical thresholds included in the RIL Other Matters
Discussion of Consequences and Consequence Modeling
- Refers to SECY-15-0148 and NUREG-2121
- Reiterates potential safety concerns associated with FFRD:
- energetic fuel-coolant interactions
- recriticality of dispersed fragments
- core coolability and long-term decay heat removal
- radiological impacts, including control room dose and equipment qualification*
- Being addressed outside of the RIL, as part of an update to Regulatory Guide 1.183, Alternative Radiological Source Terms for Evaluating Design Basis Accidents at Nuclear Power Reactors.
Limitations of the Empirical Database
- Limits are not applicable to doped fuel or coated cladding.
- Additional research could demonstrate that the limits in the RIL apply or are bounding
- Note doped fuels have different FGR behavior, so it would be important to understand implications on FFRD
- Note coated claddings have been shown to have less strain and smaller burst openings, which could mean better performance with respect to FFRD
Limitations of the Empirical Database
- Limits are simplistic, derived as a function of burnup only.
- Burnup is likely a surrogate for more direct variables such as porosity, stresses within the fuel pellet, grain growth and subgrain formation. These features are likely influenced by operating history
- Additional research to allow for more mechanistic treatment of these variables could allow for refinement of the limits
- Limits anticipate accurate prediction of cladding strain along the axial length of a fuel rod. Burst opening size is presumed to be stochastic and therefore limits assume large opening size.
- Additional research to validate balloon height, axial strain profile and burst opening could allow for refinement of the limits
Whats next?
- Participating in SCIP IV
- Includes testing of non-standard fuel
- Additional testing to characterize tFGR
- Testing in the mid-level burnup range
- Refining analysis tools to improve core-wide FFRD analysis
- Building from 2015 and MELLLA+ experience
- Enhancing resolution and realism RIL
- Utilizing new modeling features
Whats next?
- RIL will establish foundation for next steps
- Industry can build from the RIL based on their licensing needs, justifying less conservative limits with more detailed/product specific arguments
- Researchers can build programs from the RIL that produce the information RIL needed to go further
Questions?