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
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 2
3 DISCLAIMER 4
5 6
UNITED STATES NUCLEAR REGULATORY COMMISSIONS 7
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 8
9 10 The contents of this transcript of the 11 proceeding of the United States Nuclear Regulatory 12 Commission Advisory Committee on Reactor Safeguards, 13 as reported herein, is a record of the discussions 14 recorded at the meeting.
15 16 This transcript has not been reviewed, 17 corrected, and edited, and it may contain 18 inaccuracies.
19 20 21 22 23
1 UNITED STATES OF AMERICA 1
NUCLEAR REGULATORY COMMISSION 2
+ + + + +
3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
(ACRS) 5
+ + + + +
6 METALLURGY AND REACTOR FUELS SUBCOMMITTEE 7
+ + + + +
8 WEDNESDAY 9
NOVEMBER 17, 2021 10
+ + + + +
11 The Subcommittee met via Video 12 Teleconference, at 1:00 p.m. EST, Ronald Ballinger, 13 Chairman, presiding.
14 COMMITTEE MEMBERS:
15 RONALD G. BALLINGER, Chair 16 DENNIS BLEY, Member 17 CHARLES H. BROWN, JR. Member 18 GREG HALNON, Member 19 DAVID PETTI, Member 20 JOY L. REMPE, Member 21 22 ACRS CONSULTANT:
23 STEPHEN SCHULTZ 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
2 DESIGNATED FEDERAL OFFICIAL:
1 ZENA ABDULLAHI 2
3 ALSO PRESENT:
4 MICHELLE BALES, RES 5
JAMES CORSON, RES 6
ALADAR CSONTOS, Public Participant 7
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 P R O C E E D I N G S 1
1:00 p.m.
2 CHAIR BALLINGER: The meeting will now 3
come to order, this is a meeting of the Materials in 4
Metallurgy and Reactor Fuels Subcommittee of the 5
Advisory Committee on Reactor Safeguards. I'm Ron 6
Ballinger, Chairman of today's Subcommittee meeting.
7 ACRS Members in attendance are Charles 8
Brown, Greg Halnon, Joy Rempe, I have to go down here, 9
Dave Petti, and if I've missed somebody I'm sure I'll 10 hear about it.
11 We have our consultant also in attendance, 12 Steven Schultz. There may be others that will chime 13 in. Judging from the list of attendees, there's a 14 fair amount of interest here.
15 The purpose of this meeting is for the 16 Staff to brief the Subcommittee on the regulatory 17 information letter RIL 2021-13, interpretation of 18 research on fuel fragmentation, relocation and 19 dispersal at high burnout.
20 The Subcommittee will hear presentations 21 and buy and hold discussions with the research Staff 22 on this matter. A number of other people have just 23 chimed in, not yet, okay.
24 By way of background, the RIL is an update 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
4 to work in this area that depending on how old you 1
are, it's been ongoing for at least 40 years and Steve 2
Schultz says 50, related to cladding and brittle 3
performance during LOCA.
4 And then more recently, in the early 5
2000s, evolving to fuel fragmentation and dispersal at 6
high burnup.
7 There have been a number of documents that 8
have been released, NUREGs, another RIL previous to 9
this, 0801, and the latest to my knowledge fuel 10 fragmentation relocation and dispersal was NUREG 2121.
11 The ACRS has also issued a number of 12 letters related to this as well as embrittlement. The 13 latest one related to draft final rule of 5046C, which 14 is still under consideration by the Commission.
15 This RIL is a little bit unusual to my 16 mind because it not only summarizes research that's 17 been done in the past and does analysis on the 18 results, but also in Appendix A suggests a model that 19 might be used to accommodate that.
20 In effect, reset the burnup initiation 21 time from about 65 gigawatt days for metric tons --
22 sorry, I'm getting old -- to around 55. So, there's 23 a fair amount of meat in this that will be fruits for 24 discussion.
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
5 The ACRS was established by statute and is 1
governed by the Federal Advisory Committee Act, FACA.
2 That means the Committee can only speak through its 3
published letter reports. We hold Subcommittee and 4
full Committee meetings to gather information to 5
support our deliberations.
6 I might add that since it's a Subcommittee 7
meeting, the opinions of the members are just that, 8
personal opinions.
9 The ACRS Section of the U.S. NRC public 10 website provides our charter, bylaws, agendas and 11 letter reports and transcripts and all full and 12 Subcommittee meetings, including slides presented at 13 open meetings.
14 The interested parties who wish to provide 15 comments can contact the office requesting time. I 16 think we have had a request for a public comment from 17 at least one member of the public.
18 Closed meeting transcripts are not posted 19 to protect any information that might be proprietary.
20 The Subcommittee will gather information, analyze 21 relevant issues and facts, and formulate proposed 22 positions.
23 The full Committee for the research is 24 scheduled to be held during December full committee 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
6 meeting, which is scheduled for November 30th through 1
December 3, 2021. The transcript of the meeting is 2
being kept and will be made available as stated in the 3
Federal Register notice through the COVID-19 pandemic.
4 Today's meeting is being held over 5
Microsoft Teams. For NRC Staff and the stakeholder 6
and public attendees, there's also an MS Teams 7
telephone number that allows participation of the 8
public to make comments in the comment period 9
specified in the posted agenda.
10 When addressing the Subcommittee, 11 participants should first identify themselves and 12 speak with sufficient clarity and volume so that they 13 may be readily heard. When not speaking, we request 14 that participants mute their MS Teams microphone or 15 phone number.
16 We will now proceed with the meeting. Let 17 me call on I think Kim Webber hopefully? Yes, Kim 18 Webber, the Director of the Division of Systems 19 Analysis of the NRC's Research Office to deliver her 20 opening remarks.
21 MS. WEBBER: Yes, thank you, Chair 22 Ballinger, ACRS Members.
23 It's really a pleasure to be here. Just 24 for the record, my name is Kim Webber, I'm the 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
7 Director of the Division of Systems Analysis and we're 1
here to talk to you today about the research 2
information letter, or RIL, that's related to fuel 3
fragmentation, relocation, and dispersal, or FFRD.
4 The presentation is timely as it addresses 5
the performance of high burnup fuel at a time when 6
industry is interested in pursuing licensing of 7
current fuel designs to higher burnups.
8 And as you can see and as you noted in 9
your opening
- remarks, we have many industry 10 representatives and members of the public who have 11 interest in this topic.
12 So, it's good to see the broad interest 13 and participation in today's meeting.
14 The Office of Nuclear Regulatory Research, 15 or RES, has sponsored research of the behavior of high 16 burnup under loss of coolant conditions for over two 17 decades. I think you pointed out that the research 18 has been ongoing for even longer than that.
19 Over the last 10 years, the research has 20 increasingly focused on FFRD so today you'll hear from 21 the Staff about recent research which addresses 22 phenomena in which high burnup fuel has been observed 23 to fragment, relocate within the fuel rod, and 24 disperse into the coolant under loss of coolant 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
8 conditions.
1 The RIL is an NRC document and 2
specifically an Office of Research document that 3
communicates research findings between the Staff in 4
RES and other offices. In this particular case, the 5
RIL summarizes and communicates to NRR the available 6
research on FFRD.
7 And it provides insights regarding the 8
bases for empirical limits. In the past, RILS have 9
been used to support the development of regulatory 10 guidance and for this situation, the determination to 11 develop guidance associated with high burnup and/or 12 FFRD lies with NRR.
13 My Staff have worked very closely with 14 their counterparts in NRR to discuss the body of 15 research and the information that you will hear in 16 today's presentation.
17 We hope that in some way this will help 18 the licensing reviews of high burnup fuel in the 19 future. Before turning the presentation over to the 20 Staff, I want to say a few words about the third phase 21 of the cladding integrity program, or SCIP-III.
22 The SCIP-III program is one example of a 23 timely research investment that resulted in findings 24 that directly address the open regulatory safety 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
9 questions. It's a multi-party international research 1
program that allows NRC to accomplish complex research 2
in a highly leveraged way, both in terms of cost and 3
expertise.
4 For some topics addressed in the RIL, the 5
Staff incorporated conservatisms or stop short of 6
proposing a specific limit due to limited information 7
in the research. Many of these topics are being 8
investigated in the SCIP-III program, of which NRC is 9
a member.
10 The research conducted under the SCIP 11 program is proprietary to the members involved in that 12 program, however, the NRC sought permission from the 13 SCIP-III Management board, which granted our request 14 very recently to site-specific SCIP-III results in the 15 RIL.
16 And so prior to a week or two ago, this 17 information was not yet available to be published and 18 discussed publicly.
19 But we're grateful to the program and to 20 the Management Board because it enables us to make the 21 RIL publicly available, providing maximal transparency 22 about the research on this important topic.
23 And so with that, I'd like to turn the 24 presentation over to Michelle Bales.
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
10 MS. BALES: Thank you, Kim, I appreciate 1
those introductory remarks from both the Chairman and 2
from Kim. I'm going to now share my screen for the 3
presentation. Can you see the presentation?
4 CHAIR BALLINGER: Very well.
5 MS. BALES: Okay, perfect. My name is 6
Michelle Bales and I will be starting the presentation 7
today and then turning it over to colleague who also 8
works with me in the Office of Research in a few 9
slides.
10 So, I'll today be presenting the research 11 information letter 2021-13. It's an interpretation of 12 research on fuel fragmentation, relocation, and 13 dispersal at high burnup.
14 The controls are a little bit different 15 than I expected. Today's presentation, I'll start 16 with a brief regulatory history of FFRD at NRC, some 17 of which has been mentioned in the introductory 18 remarks.
19 I'll then spend some time talking about 20 the program cited in the RIL and the peer review 21 process that we used. I'm then going to speak to the 22 outcome of the RIL, in other words what the RIL makes 23 possible in terms of safety analysis.
24 And after doing that, I will go back and 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
11 speak about each empirical threshold and the basis for 1
it as documented in the research information letter.
2 There are a few other matters that we'll cover at the 3
end and then we will be wrapping up.
4 So, first, FFRD history at the NRC. The 5
first mention of fuel relocation and dispersal in a 6
regulatory context was in 2008 with the issuance of 7
RIL 0801.
8 This RIL was written just after fuel 9
dispersal was first observed in the Halden Reactor 10 Project on a road with over 90 gigawatts of data per 11 time.
12 The RIL 0801 discussed axial fuel 13 relocation and the loss of fuel particles through a 14 rupture opening, and recommended further research in 15 these areas. However, at that time it was documented 16 expected phenomena were occurring at burnups well 17 above current operating limits.
18 In 2012, the Staff conducted an extensive 19 literature review and published NUREG 2121. This 20 literature review captured the results of over 90 LOCA 21 tests performed in 8 different programs over 35 years.
22 Putting all of that information together, 23 the Staff concluded that the occurrence of FFRD 24 couldn't be precluded during the LOCA and really 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
12 required additional research.
1 In 2015 the Staff was in the middle of 2
rulemaking for loss of coolant accidents in the 5046C 3
rulemaking. And the Staff wrote a secy informing the 4
Commission that we did not plan to include or address 5
FFRD in that rulemaking.
6 But SECY also documented the Staff's 7
evaluation of FFRD and our basis for not including 8
FFRD in the proposed rulemaking.
9 At that time, the Staff also concluded 10 that immediate regulatory action was not needed to 11 address FFRD, however, as stated in that SECY, the 12 conclusion was closely linked to existing fuel design 13 limits and assumptions on how high burnup fuel would 14 be operated.
15 That brings us to today where the Staff 16 has written a new research information letter to 17 document the interpretation of fuel fragmentation, 18 relocation, and dispersal at high burnup.
19 We choose to do this now because industry 20 is pursuing extension of fuel design limits and also 21 because a large body of research has become available.
22 Documenting NRC's interpretation of available FFRD 23 research provides regulatory predictability and 24 technical review consistency.
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
13 So, now I'll go into a little bit of 1
background on the program cited in the RIL and the 2
peer review process used. With this illustration, I 3
want to communicate that we've learned a lot about 4
FFRD since the 2015 SECY.
5 The 2015 SECY is shown on this bottom 6
timeline and you can see that both the SCIP-III 7
program and a program at Oak Ridge started and 8
finished since the SECY was issued.
9 The circles that you see are scaled to the 10 size of the program and the shading indicates whether 11 results have been public prior to the RIL, or solid 12 shading indicates that results are in the public 13 domain.
14 As Kim mentioned in her introductory 15 remarks, prior to the RIL, the SCIP-III data was 16 completely proprietary towards members and therefore, 17 the publication of this RIL marks the first time this 18 information will be in the public domain.
19 MEMBER REMPE: Michelle, this is Joy. If 20 you'll go back a slide? Again, there's a lack of 21 reduction in prototypic-ness when you go out of pile, 22 isn't there? Can you kind of give us a feel for what 23 phenomena are considered?
24 Like radiation effects when you're in 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
14 Halden versus what you would get from the Studsvik and 1
the Oak Ridge test.
2 MS. BALES: Yes, I have a slide coming up 3
that provides some information on the hot cell setup, 4
which is actually very similar between the SCIP, Oak 5
Ridge and NRC's program.
6 So, when I get to that slide I'll be sure 7
to come back to that point because it's important that 8
some of the features of the test should be well 9
understood as we look at their results.
10 MEMBER REMPE: That's fine, whenever it's 11 appropriate is great. Thank you.
12 MS. BALES: I will come back to it. I 13 wanted to say a few words about our peer review group 14 because a peer review is not always part of --
15 (Telephonic interference.)
16 I don't know if I just got feedback but it 17 sounded like someone was asking a question.
18 CHAIR BALLINGER: I would remind people to 19 keep their computer muted. Thank you.
20 MEMBER REMPE: So, peer review is not a 21 necessary part of a research information letter but 22 when we look at the data resources and their recency 23 of the data, I want to explain that the nuclear fuel 24 community has not yet developed consensus around FFRD 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
15 phenomenon.
1 Significant research is still ongoing and 2
there's active work trying to develop a mechanistic 3
understanding of FFRD.
4 So, as you wrote the RIL, we recognized 5
that the empirical limits that we were proposing are 6
not consensus-based, mechanistic, or physics-based 7
models, but rather they represent the interpretation 8
of available data at this time from a safety authority 9
perspective.
10 Considering this, we wanted to pursue a 11 review and solicit perspective outside of NRC. So, we 12 specifically select reviewers with extensive 13 familiarity with FFRD research at national labs, 14 international labs, at national regulatory bodies, as 15 well as EPRI.
16 The peer reviewers were asked to identify 17 relevant research results that we might have missed to 18 identify alternative interpretation of the research 19 results presented in the RIL and identify gaps and 20 inadequacies of the basis that we proposed.
21 Ultimately, I want to make it clear that 22 the Staff was responsible for the positions taken in 23 the RIL and we were not subject to the peer review in 24 the sense that they could counter what we were saying.
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
16 At the end of the day, there were some 1
instances where the Staff took a conservative position 2
based on our perspective, but we took the comments 3
that we received from the peer review very seriously.
4 And as a whole, the peer review process 5
significantly strengthened the document.
6 The peer reviewers provided very 7
thoughtful and detailed comments, they made us go and 8
sharpen our pencils in a number of areas, and in one 9
highlight I want to put in is in the RIL you'll see a 10 definition in the terms section now.
11 And I think that really came out of the 12 peer review where many of the reviewers wanted us to 13 be more definitive about this new terminology that we 14 were introducing in the RIL.
15 In the RIL, there's an appendix which we 16 will keep when it's finally published that summarizes 17 the peer review comments as well as the resolution.
18 So, it's written at a high level but that is included 19 in the publication to provide some transparency for 20 the peer review process.
21 So, as we move into the meat of the 22 presentation, I'm actually going to speak about the 23 outcome of the RIL before I speak to the basis for 24 each of the thresholds in the RIL.
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
17 It's critical to communicate at a high 1
level how the RIL supports targeted safety evaluations 2
so I'm going to do that first. And Joy, this is the 3
slide that I mentioned. I'm going to say a few words 4
about the experimental methods used in all of the hot 5
cell testing programs.
6 I don't have a similar slide for Halden 7
but I'll speak to it a little bit as I walk through 8
the slides.
9 The RIL actually provides a
short 10 orientation to test procedures in each of the programs 11 that are discussed because some details of these tests 12 are important to understand before examining the 13 results.
14 The majority of the test programs that we 15 cite were conducted in a hot cell and subjected to a 16 time temperature profile similar to the one in the 17 lower right-hand corner of the slide.
18 Test segments were placed in accordance 19 and heated in a clamshell furnace with four heating 20 elements. Again, this is characteristic of the hot 21 cell programs. In the Halden reactor, individual rods 22 were placed in a closed loop.
23 There were heaters, electrical heaters in 24 that loop but the whole entire loop was placed in the 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
18 core and the fuel rods were operated at low power 1
prior to the transient, so that means that the fuel 2
was producing decay heat during the LOCA transient.
3 So, in the case of the hot cell tests, all 4
of the heat that was experienced by the rod is from 5
external heaters and the thermal couple on the 6
cladding really set the furnace profile to target a 7
specific cladding temperature transient.
8 CHAIR BALLINGER: Michelle, this is Ron.
9 I have discovered painfully that knowing 10 the temperature gradient around the cladding is 11 important for these tests and when you use a 12 quadalyptic furnace, which is what this looks like, 13 and you ramp it up very quickly, are people sure --
14 I'm assuming this is a SCIP thing.
15 Are they sure the temperature around the 16 cladding is uniform? Because in past times, when 17 people have done oxidation experiments they've 18 discovered that for this type of furnace, you get kind 19 of a scalloped oxide thickness around.
20 And you can take an average but it's not 21 really representative of what you're saying. The 22 helical texture that's in the zircaloid tubing 23 combined with if there's an uneven temperature around 24 the circumference can complicate things.
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
19 MS. BALES: So, I can speak to the SCIP 1
program. Before launching their irradiated testing 2
program, they did a lot of benchmarking of the 3
equipment, and that included some measurements of the 4
circumferential temperature profile.
5 They also looked at oxide measurements 6
after significant transients to look at whether there 7
was any scalloping or non-uniformity. So, the 8
equipment was qualified to produce sufficiently 9
consistent and uniform circumferential temperatures.
10 I guess there was a possibility that 11 individual tests that were done long after that 12 calibration was performed might have some unique 13 effects.
14 In some cases, some of the rods experience 15 bending after rupture, for example, in which case once 16 the rod goes off center, you can't confirm the 17 uniformity of the heating.
18 But I think that would be a unique 19 situation that we can look at to explain anomalies.
20 But generally, the equipment was looked at for that 21 and confirmed to not have any significant deviations 22 that would affect the results.
23 CHAIR BALLINGER: Because that usually 24 translates into an uncertainty bar on the temperature 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
20 versus time because the average sometimes does not 1
reflect the dispersion. Anyway, okay, thank you.
2 MS. BALES: I will say that in this 3
phenomenon, cladding temperature is not the most 4
critical part of the test.
5 That is certainly important for predicting 6
whether rupture occurs and the moment of rupture, but 7
as we go into the phenomenon that we're looking at of 8
relocation and dispersal, any non-uniformity of the 9
heating is probably going to be in the noise relative 10 to the things we're trying to measure.
11 Once you get above 760 which is the base 12 transformation, all the entropy goes away anyway I 13 suppose.
14 MS. BALES: The other things I wanted to 15 point out about the hot cell test are that the test 16 segments are about 30 to 50 centimeters. This 17 includes the Halden test and Powell was also similar 18 in length.
19 And the hot cell test but also the Halden 20 test were re-pressurized. They were segments cut from 21 commercially irradiated rods, repressurized through a 22 pressure line to various conditions dependent on the 23 test design.
24 So, as I said, I don't have a picture of 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
21 the setup in the Halden reactor in this slide deck but 1
the RIL very briefly mentions this and we can speak 2
about it maybe separately if there's a lot of 3
questions about it.
4 There's been some work done to look at the 5
temperature profile across the cladding and across the 6
pellet in comparing hot cell tests and Halden tests.
7 And while obviously the hot cell testing 8
started the rod cold and there's a big difference 9
between what the rod temperature profile looks like 10 before the test.
11 There is some analysis that is showing the 12 temperature profile after about 20 seconds in the 13 transient is very similar between hot cell tests and 14 irradiated tests.
15 And particularly when you look just at the 16 temperature profile and it's ability to introduce 17 thermal gradients.
18 So, even if the absolutely temperature is 19 a little bit different, the fact that it's flat means 20 that the thermal stresses that would be induced across 21 the pellet should be pretty similar even for the hot 22 cell test and in-cell test.
23 As we get into further, that is part of 24 are understanding of the mechanism, that the 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
22 temperature profile across the pellet would impact or 1
could impact fragmentation.
2 So, I don't know if that was as much as 3
you wanted know or if you have any other questions 4
about the test setups. There's a little bit of 5
discussion about this in the RIL but are there other 6
things that I should mention or that you'd like to 7
address before I go on?
8 MEMBER REMPE: Not at this time. Again, 9
it was something that puzzled me as I reviewed the RIL 10 and I'll ponder it a bit more if I have more 11 questions. But thank you.
12 MS. BALES: The test that I'm referring 13 to, they produce images of fragmented fuel, sometimes 14 fragments that are quite small in size. Dispersal was 15 observed in many of these tests.
16 Looking at the result separately, it can 17 seem that something alarming is happening at some 18 unknown point and it may at first seem intractable.
19 However, as documented in the RIL, a large body of 20 research is available to understand when this happens.
21 There is also repeatable and conforming 22 basis for clear and empirical limits. The RIL 23 provides the basis for when FFID does not occur and 24 allows for a well focused analysis of a relatively 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
23 small region of the core vulnerable to FFRD.
1 So, that brings me to the main outcomes of 2
the RIL. The Staff examined available research in 3
order to address five elements of FFRD. First, a 4
threshold for when fine fragmentation begins.
5 Next, a threshold for when we can expect 6
fuel axial relocation. We also sought to quantify a 7
model for predicting the amount of dispersal that you 8
might expect as a function of burnup.
9 The RIL also documents the phenomenon of 10 transient fission gas release distinct from the steady 11 state fission gas release that is already well 12 modeled. And finally, quantifying packing fractions 13 of axially relocated fuel in a balloon region.
14 So, putting these elements together, the 15 RIL forms the basis for targeted analysis of only 16 select rods that are of concern for FFRD. This is a 17 fundamental outcome of the RIL and therefore, I will 18 describe this outcome first and return to the basis 19 for each element after that.
20 To communicate how the RIL supports 21 targeted analysis, I will use an oversimplified 22 schematic to depict the rod population of an LWR core.
23 The schematic represents each assembly at 24 a given time in mind by its average burnup on the X 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
24 axis and operating power on the Y axis.
1 Burnup is a key parameter in FFRD 2
phenomena as I will explain. An operating power prior 3
to the start of a LOCA is a key determinant for 4
whether a rod will balloon and rupture, which is 5
another key parameter in FFRD phenomenon.
6 The RIL provides a basis to conclude that 7
only rods over 55 are susceptible to fine 8
fragmentation. The RIL also provides a basis to 9
conclude that only rods with strains greater than 10 three percent are susceptible to axial fuel 11 relocation.
12 Therefore, only high burnup rods with an 13 operating power prior to the transient that is high 14 enough to result in ballooning presented concerns for 15 relocation.
16 Since fine fragmentation and axial 17 relocation are precursors to fuel dispersal, only fuel 18 rods within this box are a concern for dispersal. In 19 this schematic that I'm showing here, there's only a 20 few rods that end up in this dispersal box.
21 However, the number of rods in this box is 22 highly dependent on core design, fuel design, and 23 plant characteristics. This box could be considered 24 just one of many factors that would influence core 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
25 design.
1 As you saw in the animation, there's other 2
core design maps that might have relatively lower heat 3
generation rates or have differences in how second 4
cycle and third cycle fuel rods are operated and may 5
have no overlap between the core loading map the 6
region of concern for FFRD.
7 MEMBER PETTI: This is Dave, I had a 8
question.
9 I understand this, it just seems the way 10 the rule is written, it mentions 55 as the burnup but 11 that's just sort of a necessary but not necessarily 12 sufficient metric to characterize if you've got an 13 issue.
14 And that doesn't come through as strongly 15 as what you just talked about here verbally with us.
16 When they read 55, people are going to go 62 is the 17 limit and now you're saying it's 55.
18 You're not really saying that, you're 19 saying that's just the first conditions is these other 20 things that could impact.
21 MS. BALES: It's an excellent point and I 22 think what we need to focus on is the RIL clearly 23 states that fine fragmentation, and I'm going to show 24 some results to demonstrate this, has been seen as 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
26 early as 55.
1 Fine fragmentation in and of itself is not 2
obviously presented as a safety concern. If that fuel 3
disperses, that's when you start to need analysis to 4
confirm safety. In order to disperse, there has to be 5
ballooning and rupture.
6 So, a high burnup rod above 55 that 7
doesn't balloon and doesn't rupture, you won't have 8
relocation or dispersal. And it's really the 9
dispersal that should be the focus of the safety 10 evaluations and relocation.
11 But I think your point is right that in 12 the RIL we tried to explain that fine fragmentation 13 and relocation are precursors to dispersal.
14 But perhaps it's not as clean, as I'm 15 walking through it now, that the 55 for fine 16 fragmentation only is not by itself the issue.
17 MR. SCHULTZ: This is Steve Schultz. Can 18 you hear me?
19 MS. BALES: Yes.
20 MR. SCHULTZ: You also mentioned the 21 strain number that was identified in the RIL 22 associated with the small-sized fragmentation. What 23 you seem to be showing here are fuel rods that are at 24 power levels in the core.
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
27 And I presume what you showed in terms of 1
the data set that's included at that strain level is 2
for the assumed LOCA strain, in other words during the 3
transient. Is that correct?
4 MS. BALES: Correct, you mean as opposed 5
to operating?
6 MR. SCHULTZ: Yes.
7 MS. BALES: Yes, the strain that we're 8
speaking about that is a precursor to relocation is a 9
strain that occurs during the transient.
10 MR. SCHULTZ: And that's of course in 11 terms of what you're showing here like the burnup 12 limit value? That's affected by the assumptions and 13 the evaluation of the LOCA transient and the dynamics 14 of the fuel performance in the LOCA.
15 MS. BALES: Yes, I think the schematic is 16 really an illustration to describe the zones at a very 17 high level. But I think when we get into the 18 technical basis we can speak a little bit more cleanly 19 about what is precisely changing between relocation 20 and dispersal.
21 But one of the things that I was trying to 22 communicate with this, you see this box, any fuel rods 23 above 55 are potentially rods that can finally 24 fragment.
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
28 But when we look at defining the box for 1
relocation, the horizontal line is my attempt to 2
communicate that only rods that are hot enough to 3
undergo ballooning would experience relocation.
4 So, on the left there's an image that 5
shows the strain limit which I haven't presented yet 6
but we'll get to in a second.
7 But the fact that the box has a horizontal 8
line to only include rods with high enough power to 9
strain under a LOCA is to communicate that this 10 relocation box is less than the population of rods 11 above 55 and subject to fine fragmentation.
12 And without seeing the boxes overlap, if 13 you just watch the figure, the dispersal box is 14 slightly higher, the idea that the difference between 15 relocation and dispersal.
16 There are some rods that are predicted to 17 balloon but not rupture. The dispersal box is 18 intended to only include rods which are predicted to 19 rupture.
20 MR. SCHULTZ: Thank you.
21 CHAIR BALLINGER: This is Ron again. I 22 keep wondering whether we need to be a little bit 23 careful in that the observed behavior is what it is, 24 it's observed.
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
29 But a calculation for the LOCA calculation 1
is at least Appendix K or is a stylized calculation 2
where there's a lot of conservatism built in.
3 And so do we need to be careful that the 4
stylized calculation, if that's what it is, doesn't 5
get us burst or greater than 3 percent strain when 6
that's due to the conservatism in the calculation 7
that's built in.
8 MS. BALES: I think that's an excellent 9
point and I want to explain that the meat of the RIL, 10 well, the entirety of the RIL is focused on when this 11 phenomenon has been observed in testing, and it's 12 trying to establish empirical limits to define when 13 FFRD will occur.
14 Everything that I'm presenting now is 15 really just in service of communicating that we're not 16 talking about the entire core, that we're not even 17 talking about all high burnup rods.
18 But the schematic is a
complete 19 oversimplification not only of a core loading map, but 20 also to what you're saying, where that horizontal line 21 cuts off is going to be extremely complicated to 22 calculate.
23 It's not easy to draw that lower line of 24 the box, that is dependent on the calculation method, 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
30 the conservatisms used. And so the RIL doesn't 1
attempt to define a ballooning model, for example, or 2
attempt to draw that bottom leg of the box.
3 I'm using it in its presentation because 4
I want to communicate that the RIL is linked to 5
analysis but as we go into the technical basis, and 6
we'll spend more time over there, that's really the 7
meat of the RIL.
8 What has research shown in terms of the 9
onset of these phenomena?
10 CHAIR BALLINGER: I'm sure we'll further 11 discuss this.
12 MS. BALES: So, there's just two more 13 points that are addressed in the RIL that I wanted to 14 walk through the schematic to illustrate.
15 Again, it's just a schematic and it only 16 serves to explain the implications of the RIL, it's 17 just an illustration.
18 Trained fission gas release is another 19 thing that's addressed during steady state operation, 20 diffusion-based fission gas release is expected and 21 it's very well modeled by fuel performance codes.
22 However, transient testing has shown that 23 significant additional gas release can occur during 24 the transient, changing the rod internal pressure.
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
31 The rod internal pressure is critical for actively 1
predicting whether rods burst when subject to LOCA 2
conditions.
3 And therefore, accounting for transient 4
fission gas release is critical for defining the 5
dispersal box. So, I think this further emphasizes 6
your point, whether the models are accounting for 7
transient fission gas release affects replacement of 8
this box.
9 Whether the models are accounting for 10 other factors in a conservative or best-estimate way 11 will affect the positioning of this box.
12 The final element the RIL addresses is 13 packing fraction and so this is particularly important 14 for rods where they're above the burnup limit so we 15 expect fine fragmentation.
16 They experience some ballooning and the 17 fine fragmentation can fall into the ballooned region.
18 Accurately accounting for that phenomenon 19 and having a density of relocation or packing fraction 20 is important for accurately predicting whether the rod 21 bursts because temperature is a critical element in 22 those predictions.
23 Also, even if the rod is not predicted to 24 burst, it's important that some accounting for 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
32 relocation and packing fraction values are used in a 1
calculation of peak cladding temperature because that 2
is a key metric in LOCA evaluation.
3 So, in summary, with that said on the 4
slides, I wanted to communicate that the RIL forms the 5
basis for targeted FFRD analysis. Only rods with the 6
following characteristics are a concern for FFRD.
7 Fine fragmentation, we have not seen it 8
below 55, for axial relocation we haven't seen 9
cladding with less than 3 percent strain. Fuel 10 dispersal requires axial relocation and therefore can 11 be limited to rods that are above 55 and strained more 12 than 3 percent.
13 But dispersal itself can't happen unless 14 there's rupture so also, the calculation of the 15 rupture event is critical to determining fuel 16 dispersal.
17 CHAIR BALLINGER: This is Ron, I keep 18 beating a dead horse but in this slide, the only thing 19 that's actually calculatable with any precision is the 20 burnup. Everything else is a calculation based on a 21 fuel performance model, which may or may not be 22 accurate.
23 MS. BALES: I think it's important to 24 emphasize that the RIL doesn't prescribe how to 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
33 analyze FFRD. The RIL simply defines when to analyze 1
it.
2 That means that vendors and their topical 3
reports are to define the models and the basis for 4
those models that need to be used in order to evaluate 5
these limits.
6 Having said that, I want to explain that 7
in some ways the means of analysis are not a dramatic 8
departure from the existing analysis. For example, as 9
you already mentioned, the burnup of each rod is 10 already tracked.
11 Evaluating whether a rod is above or below 12 that threshold is straightforward.
13 The empirical limits depend on ballooning 14 and burst and although there are variability in how 15 accurately that can be predicted, performance codes do 16 already have models for ballooning and the prediction 17 of rupture.
18 Some fuel performance codes even already 19 model axial fuel relocation. So, I said all of that 20 to say the RIL does not define how to analyze FFRD, 21 which models are acceptable, which models are even 22 recommended to match the experiments in the test.
23 It just simply leaves it at here's where 24 these phenomena had been observed and this is where 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
34 these phenomena should be analyzed, but the analysis 1
should largely map onto existing tools.
2 Whether they are assessment or ready for 3
use in this analysis is something that would be 4
addressed separately.
5 Now that I've given you a high-level 6
overview of what the outcome of the RIL is and what 7
the RIL makes possible, I'm going to step through each 8
of the empirical thresholds in turn and describe the 9
technical basis for each of the limits.
10 First the empirical threshold at which 11 fuel pellets become susceptible to fine fragmentation.
12 To define the threshold at which fuel pellets become 13 susceptible to fine fragmentation, we examine results 14 from 26 tests in which fuel fragment size was 15 carefully measured.
16 Fuel fragments were processed through a 17 series of sieves and the mass fraction of fragments 18 below 1 millimeter and 2 millimeters are shown in the 19 plot on the left. We see the mass fraction of fine 20 fragments increases with burnup, with some variability 21 between tests.
22 We have detailed fragment size results at 23 a burnup of about 43, showing no fragments smaller 24 than 2 millimeters, and then again at 6 days per ton, 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
35 showing fragments start to be observed at these fine 1
size levels.
2 Unfortunately, we don't have any result of 3
fragment size distribution between these levels using 4
that sieve fine resolution picture.
5 But we do have optical microscopy from 6
testing at Argonne National Lab for an average of 55-7 second burnup before and after LOCA testing that 8
indicates fragments on the order of 1 to 2 millimeters 9
were observed after testing.
10 The A&L results are the ones that anchor 11 our position with the onset of fine fragmentation 12 occurs at 55 gigawatts per ton.
13 CHAIR BALLINGER: This is Ron again. I 14 look at this and I look at, let's say, the lower 15 figure at 66. There are two data-points which look to 16 be almost exactly the same burnup but the variability 17 is plus or minus 30 percent.
18 So, is that a good number to think about 19 in terms of uncertainty?
20 MS. BALES: That's a complicated question 21 because you're previewing one of my concluding 22 remarks, which is we have used burnup as the Y axis in 23 the RIL. And most likely, the mechanism that is at 24 play that controls fine fragmentation has more 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
36 variables than just burnup.
1 And so in this we're just simply plotting 2
everything as a function of burnup so in some ways 3
that is true that 30 percent could be an uncertainty 4
factor on the burnup threshold, but I think it's 5
probably more accurate to say that burnup is somewhat 6
of an incomplete parameter in the onset of fine 7
fragmentation.
8 And as we look to the future, a true 9
mechanistic understanding will probably put burnup 10 amongst other factors that control the onset of fine 11 fragmentation.
12 Once we have those delineated, some of 13 this data might be more easy to explain.
14 CHAIR BALLINGER: I wonder whether or not 15 the flow conditions, vibration and such, might also be 16 a significant factor that is unpredictable?
17 MS. BALES: I think in a real LOCA that 18 could be the case.
19 Most of these tests, except for the red 20 Halden tests, and all of the other data-points would, 21 I expect, have been subject to the same forces and 22 conditions because they were all conducted in the same 23 test setup where you have flowing steam in accordance 24 with the furnace heating.
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
37 So, I wouldn't expect a lot of variability 1
between tests when it comes to the applied forces.
2 CHAIR BALLINGER: But the Halden results 3
say zero?
4 MS. BALES: For some of these burnups, 5
yes, that is true.
6 CHAIR BALLINGER: High burnup fuel, zero.
7 MS. BALES: I'm trying to bring your 8
attention to the one that is at 64.
9 CHAIR BALLINGER: I see that one but I 10 also see three non-bursts, but let's take the burst 11 ones. At least 2 at 73 or so, or thereabouts?
12 MS. BALES: There is certainly a lot of 13 variability in the fragmentation behavior and in 14 particular, some notable low fragmentation 15 observations in the Halden test.
16 So, I think this gets to the fact that we 17 don't have a fully mechanistic understanding.
18 This is really just empirical threshold 19 that is based only on what we have looked at right now 20 which is the effective burnup.
21 CHAIR BALLINGER: Thanks.
22 MR. SCHULTZ: The other question I had on 23 that slide was -- this is Steve Schultz -- each of 24 these data-points are representative of results 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
38 regarding fragmentation that are from a test where the 1
fuel rod has been exposed to a particular tailored 2
time, steam, temperature, pressure, condition presumed 3
to be appropriate for LOCA of a type.
4 In other words, rods and reactors, if 5
they're exposed to a LOCA, are going to behave 6
differently depending on their power history and what 7
their power level and burnup is at the time of the 8
event.
9 I presume there's a lot of variability in 10 the conditions to which these rods have been exposed 11 in the testing.
12 MS. BALES: Yes, there is variability in 13 the conditions of the tests. The time temperature 14 transient is a little bit different amongst these 15 tests. The rod internal pressure was different 16 between some of the tests.
17 In fact, in the SCIP-III program that was 18 one of the parameters that was being investigated, the 19 effect of rod internal pressure. So, that was 20 deliberately varied and so some of these tests have 21 different internal conditions.
22 In addition, I thought what you were 23 mentioning there is their operating history could also 24 be different and so getting to some of these burnups, 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
39 some of these rods are coming from lead test assembly 1
campaigns.
2 Some are even reinserted rods so they were 3
experiencing a very unique power history during 4
operation. So, there's a lot going on between the 5
tests that with layers of resolution may explain some 6
of the scatter.
7 But at this point, when we looked at what 8
we could say with this level of knowledge, we plotted 9
it all against burnup and generally see an increasing 10 trend although there's a lot of variability.
11 Again, you're previewing some of my 12 comments at the end to wrap up but I think that some 13 of the future research will look at how those 14 different parameters affect fragmentation and could 15 lead to a reduction in the conservatism that's 16 depicted by this burnup threshold.
17 MR. SCHULTZ: You're speaking to my point, 18 I appreciate that. I'll wait for your conclusions 19 later too. Thank you.
20 CHAIR BALLINGER: Do we know what the 21 temperature profile history of this pellet was prior 22 to the test?
23 MS. BALES: Are you referring to the 24 optical microscopy that is shown?
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
40 CHAIR BALLINGER: Yes.
1 MS. BALES: I don't have that handy with 2
me. The testing at Argonne was on commercially 3
irradiated rods and if it was at 55 I wouldn't have 4
expected it to be an LTA rod. I can look that up but 5
I don't have it handy with me.
6 I will say that amongst the tests that are 7
shown in the plot on the left, there was a lot of 8
variability in the operating history during commercial 9
irradiation. So, that represents quite a big span of 10 normal and possibly some extreme operating histories.
11 CHAIR BALLINGER: It's hard to tell from 12 looking at a computer screen of a micrograph of a 13 micrograph, but generally it's not just relocation 14 that occurs, it's restructuring that occurs sometimes.
15 And I'm looking at this circle at about 30 16 percent in or 40 percent in and wondering whether or 17 not that's generally an indication of something of 18 some restructuring that may have occurred.
19 And that also delineates the area on the 20 right where you get a lot of finds in radiuses greater 21 than that little ring. So, again, it's fodder for 10 22 or 20-page theses.
23 MS. BALES: It certainly is, and I'll just 24 say that I think what you're pointing to is the 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
41 slightly darker circle.
1 It's a little hard because this is a map 2
of many different images so the quality to discern a 3
dark zone, for example, is a little hard in this 4
particular image.
5 The SCIP program has done some really 6
great characterization of pretty transient conditions 7
as well as post-transient conditions.
8 And I'll just say that this phenomena 9
where we see some more fragmented region in a circle 10 and less fragmented in the center and even on the 11 periphery relative to that mostly fragmented in this 12 ring has been seen additional places.
13 This is not the only time that pattern was 14 observed.
15 So, the SCIP reports that are cited in the 16 RIL are still proprietary to the SCIP members but 17 there's quite a bit of information that has been 18 documented to look at before and after and what 19 patterns, porosity, fragment size, and such to see if 20 there's a link to what the final fragmentation results 21 were that were observed.
22 And NRC's program at Studsvic, we actually 23 looked at a collection of fine fragments that were 24 collected in the sieves after the LOCA test, and we 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
42 did isotopic analysis to try to determine where they 1
had come from with respect to the pellet radius.
2 And so we saw that the fine fragments that 3
were collected originated not just from the pellet 4
periphery but even some of the interior portions of 5
the rod were origins for those fine fragments.
6 So, that is all to say there has been work 7
to understand where relative to the pellet radius fine 8
fragmentation is occurring, and there's quite a bit of 9
variability.
10 Okay, as I suggested earlier when we were 11 talking about the 55 limit, fragmentation by itself 12 doesn't present a safety concern. However, if fine 13 fragments can relocate axially within a rod, there can 14 be safety implications.
15 On the issue of fuel relocation, tests 16 repeatedly showed that even when fuel was 17 significantly fragmented, not all fuel was able to 18 relocate axially.
19 In the image on the left, we see 20 measurements of one of the NRC's tests at Studsvic.
21 During this test, there was significant fuel dispersal 22 and we're actually going to show a video of this test 23 a little bit later, and I'll say a little bit more 24 about what was observed.
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
43 But for this point, I want to say that 1
after the LOCA tests a wire probe was inserted into 2
the test segment through a rupture opening and could 3
travel quite far, as indicated by the green span that 4
you see in the lower portion of the graph.
5 After shaking the rodlet, which again I'm 6
going to show in a video in a little bit, additional 7
fuel was collected and the wire probe could travel 8
even further as indicated by the pink span.
9 These measurements suggest a zone of empty 10 cladding but after all testing was completed, a gamma 11 scan was performed and revealed a slightly larger span 12 empty of fuel.
13 So, perhaps more importantly, the gamma 14 scan revealed a section of the rodlet where fuel 15 remained and appears in tact.
16 Pellet interfaces can be discerned in the 17 lower portion of the rod down here, so this span is 18 about a pellet length and so we see a pellet interface 19 and relatively in tact fuel judging by the gamma scan 20 alone.
21 Similar examinations were performed in 22 additional rodlets and these measurements formed the 23 basis for saying that fuel relocation is limited to 24 regions of the fuel rod experiencing greater than 3 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
44 percent strain.
1 So, one of the reasons that we care about 2
fuel relocation is that it's a precursor to dispersal.
3 Before I speak to the measurements made of 4
fuel dispersal, I want to provide details of the 5
testing so that we can understand how dispersal 6
measurements were made, especially in the SCIP-III 7
program.
8 The video that I'm going to show will at 9
first be a closeup just below the furnace in the test 10 train. At this location we can see the court's tube 11 and a rodlet is above the frame.
12 At the moment of burst we will see fuel 13 fragments fall into the field of view below the 14 rodlet. The LOCA test is happening and then the burst 15 occurred and then the black dust that you see is fuel 16 material that fell out of the rupture opening and to 17 the bottom of the court's tube.
18 The scan is just showing to illustrate 19 what the rest of the test train looked like and then 20 this is in slow motion. So, after LOCA testing, the 21 fuel material that you see in the bottom of the test 22 train was collected and weighed.
23 And we have 19 measurements showing the 24 mass of fuel collected at this point in the 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
45 experiment. After the LOCA tests, the rodlets were 1
shaken and this video reveals what that looked like.
2 I show this because in the RIL we ready to 3
mobile fuel and dispersed fuel. In the first video 4
the mass of fuel that was collected just after the 5
test that was found at the bottom of the test train, 6
that's what we talk about as the dispersed fuel.
7 And when we talk about all mobile fuel, 8
we're talking about any fuel material that was able to 9
be shaken out of the rod. And when comparing 10 dispersed fuel to mobile fuel, we see that it all 11 comes down to the burst opening size.
12 The figure on the left highlights two 13 particular tests but the size breakdown of all mobile 14 fuel is quite similar. So, in these bar graphs, the 15 height represents the mass of fuel collected.
16 Blue was fuel that was collected after the 17 LOCA tests and the orange was additional fuel that was 18 collected during the shaking.
19 And you can see that the top graph and the 20 bottom bar graphs have similar situations where we 21 have a lot of particles that are between 2 and 4 22 millimeters, and above 4 millimeters in this case, and 23 then relatively smaller amounts of fine fragments in 24 each of the size bins.
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
46 However, on the right of this figure I've 1
drawn a representation where the size of the two burst 2
openings is scaled, and so in the upper region the 3
burst opening was much, much larger than that of the 4
lower one.
5 And that is most likely an explanation for 6
why we saw such a difference in the dispersed mass 7
versus what was mobile in total. But if burst opening 8
is a key parameter we face a problem because burst 9
opening size varied between tests.
10 And there was not an obvious burnup trend 11 so when we are correlating in the RIL when we're 12 correlating this phenomenon with burnup, we can't 13 necessarily rule out large rupture openings in either 14 low or high burnup fuel.
15 And so for that reason, the dispersal 16 model that we propose in Element 3 of the RIL was not 17 based on the dispersal observed during the LOCA test 18 that I showed in the first video.
19 The model actually assumes all mobile fuel 20 could disperse and, therefore, it's tied to the limits 21 established for fine fragmentation and relocation.
22 The model proposed in the RIL is for a finely 23 fragmented fuel, all fuel above 55, that can relocate.
24 In other words, strain greater than 50 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
47 percent, all of that fuel could disperse.
1 Using this model to examine how well it 2
would predict all mobile fuel in seven LOCA 3
experiments in SCIP-III, we see that it is sometimes 4
conservative and sometimes not.
5 But we're interested in core wide 6
assessment
- and, therefore, considering these 7
predictions at a core-wide level we expect the results 8
to be conservative.
9 And other models were examined to some 10 extent and that's the documentation that you'll find 11 in Appendix A where we looked at different trend lines 12 that were taken from the size graph that was shown 13 earlier in the presentation, where fragmentation size 14 seems to sync with burnup.
15 But at the end of the evaluation, the 16 Staff concluded that conservatively at this point, 17 because burst opening is stacastic that we want to 18 create a model which was assuming all mobile fuel 19 could disperse.
20 So, there's two additional elements that 21 we're going to go into some details and for that I'm 22 actually going to turn it over to my colleague, James, 23 who is --
24 James, I'm assuming you want me to keep 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
48 running the slides or do you want me to turn the 1
presentation over to you?
2 MR. CORSON: You can keep driving the 3
slides. Hello, everyone, I'm James Corson and I work 4
in the same branch of the Office of Research as 5
Michelle.
6 So, I'm going to be talking about the 7
fourth element of the RIL and that's transient fission 8
gas release, which could impact ballooning and burst 9
behavior of rods under LOCA conditions.
10 As Michelle said, the modeling of 11 fission gas release during normal operations is pretty 12 well understood but during transients, there aren't 13 really a lot of good models to represent this 14 behavior.
15 So, you can see on the data there seems to 16 be an increasing trend where transient fission gas 17 release becomes more prominent at higher burnup but 18 there's more going on here, there's other factors than 19 just burnup.
20 The fuel temperature plays an important 21 role. Some tests have shown that you need a threshold 22 of about 600 degrees Centigrade before you get any 23 transient fission gas release.
24 Stresses in the fuel also play a role so 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
49 the cladding mechanical constraint or the internal 1
pressure, those can also influence transient fission 2
gas release, which is why you see quite a bit of 3
scatter on this graph here.
4 Since releasing more fission gas is going 5
to increase the rod internal pressure, that can impact 6
ballooning and bursts, it's something that needs to be 7
accounted for when trying to predict whether or not 8
your rod is building to burst.
9 And as we discussed that, whether you 10 rupture that determines whether or not you can get 11 dispersal.
12 So, right now there are some models that 13 have been proposed for a transient fission gas release 14 but they need a lot more validation at this point 15 before they're ready for use.
16 Next slide, the fifth and final element is 17 related to the packing fraction of the relocated fuel 18 in the balloon region. This is something that's been 19 studies quite extensively over the years, even for 20 lower burnup fuel.
21 But for this higher burnup fuel with fine 22 fragmentation you can get a higher packing fraction 23 and that's just related to the fact that if you have 24 a poly-dispersed size distribution, you can get a 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
50 higher theoretically packing fraction.
1 And so SCIP data shows packing fractions 2
ranging from about 70 to 85 percent for fuel that's 3
susceptible to fine fragmentation. And then for the 4
lower burnup fuel they looked at, the one test had a 5
lower packing fraction, which is more consistent with 6
previous measurements.
7 So, again, it's important to account for 8
this because it can affect the LOCA heat transfer 9
behavior and thus impact whether or not you balloon 10 and burst. Next slide.
11 That's it for the five elements the RIL 12 addressed. To wrap up we're going to talk about a few 13 other matters we addressed. Next slide.
14 So, all of this, the RIL really focuses 15 on, again, what rods are susceptible to ballooning and 16 burst, fine fragmentation, relocation, dispersal. It 17 doesn't really talk about how this would be applied on 18 a core-wide basis or what the implications of 19 dispersal would be.
20 Instead it just points back to previous 21 publications in this area to discuss the potential 22 safety concerns.
23 And the ones that were listed and 24 discussed in this previous publications are energetic 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
51 fuel coolant interactions, re-criticality of dispersed 1
fragments, core coolability and long-term decay heat 2
removal, and potential radiological and source term 3
impacts.
4 The latter topic is being addressed as 5
part of an update to Regulatory Guide 1.183. Next 6
slide. Lastly, the RIL talks about some of the 7
limitations of the empirical database.
8 Michelle has hinted at some of these 9
things already but I just want to make clear that all 10 these tests were performed on standard uranium dioxide 11 fuel pellets in zirconium alloy cladding.
12 So, they weren't performed on any doped 13 fuel or coated cladding and so the limits aren't 14 really applicable to those fuel and cladding types.
15 Additional research may demonstrate that 16 the limits in the RIL may apply to these new fuel 17 types so that they're bounding but more work needs to 18 be done in this area.
19 I should note that doped fuel has 20 different fission gas release behavior so that can 21 impact fine fragmentation relocation and dispersal.
22 And on the other hand, coated claddings at 23 least in unfueled samples have been shown to have less 24 strain and smaller burst openings during burst 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
52 testing. So, if you have a smaller burst opening that 1
could significantly limit the amount of fuel 2
dispersal.
3 So, coated cladding could potentially have 4
some beneficial impacts but, again, more research is 5
needed to verify those findings. Next slide.
6 Again, as Michelle said, these are 7
simplistic limits. They were only derived as a 8
function of burnup but in reality, there's more going 9
on than just the burnup.
10 There's the power of the rod right at the 11 start of the LOCA or right before the tests were 12 conducted. Those sorts of things impact things like 13 porosity or stresses within the fuel, it impacts grain 14 growth and subgrain formation.
15 And all these things are heavily 16 influenced by the operating history so as Michelle 17 said, that could potentially explain some of the 18 scatter that was in the plots shown earlier.
19 But we need more research to allow for a 20 more mechanistic treatment. Right now, all we really 21 have is burnup and we have some preliminary 22 indications that there's something else going on but 23 we need more research to verify that.
24 And then finally, these limits anticipate 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
53 that we can accuracy predict the cladding stream along 1
the rod. Historically, that hasn't been a real focus 2
of ballooning models.
3
- Rather, they focused on the burst 4
temperature and burst time or the maximum burst strain 5
at the burst location. They haven't necessarily 6
focused on the actual extent of the balloon region.
7 So, it may be necessary to make some 8
conservative assumptions to identify the mass of fuel 9
susceptible to relocation and dispersal. And as 10 Michelle said, we treated the burst opening size as 11 stacastic.
12 And so we based our limits assuming there 13 would be a large burst opening size.
14 However, there has been some research done 15 by the national labs in the U.S. to look at the 16 influence of different parameters and how they impact 17 burst opening size.
18 So, it's possible that in the future there 19 may be a more mechanistic model to predict burst 20 opening size. Next slide. Where does that leave us?
21 As Kim had mentioned in her introduction, 22 we're participating in the next phase of the SCIP 23 program and this actually includes some testing to 24 address the gaps that are in the existing database.
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
54 So, it has some tests on doped fuel, some 1
additional tests to characterize transient fission gas 2
release, and more testing in the immediate burnup 3
range between 45 and 60 gigawatt days for MTU, where 4
there was a bit of a gap in some of the figures that 5
we showed.
6 Separately, at NRC we're working on 7
refining some of our analysis tools to perform coal-8 wide FFRD analysis.
9 We perform some preliminary calculations 10 in the lead-up to the SECY in 2015 where we used a 11 combination of scale and parks neutronics tools that 12 traced thermal hydraulic systems analysis code and 13 FRAPCON, fuel performance code, to estimate the total 14 mass of fuel dispersed during the LOCA.
15 Since then we've used similar workflows 16 for MELLLA+ analysis on BWRs, so we're looking to 17 extend those to more realistic core designs for LOCAs 18 at higher burnup and to use some of the newer modeling 19 features in our FRAPCON code and in improvements in 20 trace and so on.
21 I'm going to turn it back over to 22 Michelle.
23 MS. BALES: Thanks, James. This is 24 effectively the conclusion slide and I wanted to end 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
55 the presentation by explaining to you that the RIL is 1
a foundation for next steps.
2 As you would have maybe gathered from some 3
of the questions but also what James just talked about 4
in terms of the limitations, you could consider this 5
an awkward time to write the conclusions of FFRD.
6 There's a lot that we still don't 7
understand and if when we sit down to write an update 8
or a revision in five years, I bet it will be much 9
more precise and much more satisfying.
10 But at this time we wanted to establish 11 what we know today, especially as the industry is 12 looking to increase burnup limits and explain what we 13 know from the available research up until this point.
14 We fully expect research to continue and 15 we would anticipate, for example, industry building 16 from the RIL based on their licensing needs, 17 justifying less conservative limits with more detailed 18 or problem-specific arguments where they're able to do 19 that.
20 We also expect that research at the 21 national labs or in international collaborative 22 programs can build from the basis that the RIL 23 provides and produce information that is needed to go 24 further, either to refine the limits or make them a 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
56 function of more properties than just burnup.
1 MEMBER PETTI: Michelle, question? What's 2
the timeframe for the additional testing that you're 3
talking about? Is it five years or less?
4 MS. BALES: The SCIP-IIII program is on 5
Year 2 I believe and that is where we'll get a little 6
bit more information about the midrange burnup as well 7
as some information of doped fuel.
8 I think there are additional research 9
programs that will continue for much longer than that.
10 We know about the SCIP-IV program because 11 we are members of that program but I believe there is 12 research well beyond that that could add to this body 13 of information that might come out in the next couple 14 years.
15 I can only speak to the SCIP-IV program 16 because that's the program we are members of.
17 CHAIR BALLINGER: Thank you for that 18 presentation.
19 I'm going to venture into some stream of 20 consciousness talking here I suppose, but we probably 21 should wait until we hear from Members first and then 22 we'll go and ask for comments from the public.
23 So, other Members, unfortunately we have 24 only four or five of our Members present and this is 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
57 a very detailed presentation of a multi-variant 1
problem where without being able to be present at the 2
presentation, it's going to make it pretty difficult 3
for other Members to understand what's been going on 4
with the information.
5 So, in any case, Members, questions and 6
comments, please?
7 MEMBER PETTI: Ron, my comment, having 8
read the CAPS paper in JNM and the RIL is there's 9
still a heck of a lot of uncertainty around this.
10 There's questions about prototypicality of the entire 11 experimental database in a lot of ways.
12 You'd expect the in pile to be the most 13 prototypic but it has some problems, no experiment is 14 perfect. And the only way you get a resolution on 15 this sort of stuff is you do lots of testing and lots 16 of different experimental approaches.
17 And like an Impressionist painting you 18 step back. If you look real close you get lose. You 19 step back and you may see some better trends.
20 But there's still a lot that we don't know 21 and it all relies on the mechanical properties of the 22 cladding to find out when it's going to burst, the 23 size it's going to burst.
24 And we don't know the mechanical 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
58 properties, the relations well enough. So, there's 1
models that exist, I understand that but they weren't 2
looking at this issue.
3 One may need a hell of a lot more 4
precision and better models than what's already there 5
to get a good calculation.
6 CHAIR BALLINGER: Now I'll start the 7
stream of consciousness discussion.
8 The problem is defined in the sense that 9
we have a right circular cylinder with dished pellets 10 and we have a more or less bounded power history for 11 Light Water Reactors.
12 But that's it, that's what constrains it 13 and like you say, it's a multi-variate problem with 14 several variables which are just undefined and maybe 15 unknowable in terms of what the values are.
16 And it's possible that you could develop 17 a program that would go on for the next 20 years and 18 still not get enough information, it's one of these 19 forever things.
20 And you wonder whether or not we should 21 come back and look at the consequences of the 22 phenomena to see if the consequences can somehow focus 23 the next or ongoing or future research efforts so that 24 it doesn't become a never-ending problem.
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
59 Because there's probably 15 or 20 1
variables there.
2 There was an infamous figure that was 3
produced by I think Adrian Roberts, who has since 4
passed away, where he was being facetious and he drew 5
a chart which had the relationships between all the 6
variables and fuel performance.
7 And I think it was Adrian Roberts.
8 MEMBER PETTI: I've seen the figure, yes, 9
it was Adrian Roberts I believe.
10 CHAIR BALLINGER: And you sit back and you 11 laugh at it but it was correct. And so I'll just stop 12 talking and ask for other comments from Members.
13 MEMBER BLEY: Ron, this is Dennis. Your 14 stream of consciousness is worthwhile. It seems to me 15 at points some real discussion inside the Committee 16 and I think your scheduled for a letter but I think 17 you should be if you're not.
18 CHAIR BALLINGER: We are scheduled for a 19 letter but I'm not sure this is the right time for a 20 letter.
21 MEMBER BLEY: Maybe not but it's probably 22 time, given the number of people who were here, to at 23 least have the full Committee meeting and the others 24 know more about this --
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
60 I'm not sure if it's time, it could be a 1
good time for a cautionary letter based on what you 2
were just saying. Let's look for places we can bound 3
the consequences and see how much it's worth, how much 4
and when it's worth really going a lot further in 5
this.
6 And most of the time in the past before 7
COVID-19, this was a typical turnout for the 8
Subcommittee and we had to have a pretty good 9
briefing, a couple hours over the same stuff. I think 10 that would be appropriate here.
11 Anyway, enough from me.
12 MEMBER REMPE: This is Joy, I would like 13 to follow up more on the question about the prototypic 14 aspects and non-prototypic aspects of the sources of 15 data and what Dave said I agree with, that there's no 16 perfect test.
17 And if we could ask the Staff before the 18 full Committee meeting to generate some sort of list 19 of pluses and minuses for each source of data to drive 20 home that point, I think it would be a useful task for 21 them to do.
22 What do you think?
23 CHAIR BALLINGER: I think you're correct 24 as far as it goes, but again, I keep coming back to 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
61 how prototypic does it need to be? In other words, if 1
we need a very precise description of the problem, 2
then test data that's the -- I've got figure out the 3
right word.
4 How prototypic the data needs to be is 5
dictated by the level of detail that you need to base 6
decisions upon. Is that not correct?
7 MEMBER REMPE: I'm with you about we need 8
to bound the problem and I guess I'm doing that as an 9
exercise to drive home the point there's no perfect 10 test and maybe instead of doing more tests that will 11 still not be the perfect test and answer the question, 12 look somewhere else.
13 That's why I think it would be good to 14 help see why there's some problems with each source of 15 data and I didn't get that enough from the RIL. Okay?
16 CHAIR BALLINGER: The other wild card in 17 this is I'm aware of at least one vendor, a fuel 18 vendor, that's looking at this and they have had 19 meetings that I've participated in at least as an 20 observer with the Staff.
21 And so I know that at least one vendor is 22 looking at this and communicating with the Staff. One 23 wonders whether or not the others are not doing the 24 same thing, maybe they are, maybe they aren't.
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
62 But that's a perspective which when you 1
deal with consequence is an important thing to listen 2
to.
3 And this meeting hasn't done that and I 4
wonder whether or not we shouldn't think about having 5
a follow-up meeting, regardless of whether we write a 6
letter or not, where if the vendors have enough 7
information that they can talk about, we get their 8
perspective as well.
9 What do Members think about that?
10 MEMBER PETTI: I think it's important to 11 do.
12 CHAIR BALLINGER: So, again, that's the 13 stream of consciousness thing. Who else do we have 14 here?
15 (Simultaneous Speaking.)
16 MEMBER PETTI: Steve has his hand up.
17 CHAIR BALLINGER: Who?
18 MEMBER PETTI: Steve Schultz.
19 CHAIR BALLINGER: Okay.
20 MR. SCHULTZ: I just wanted to comment on 21 your last note there.
22 And it also was brought up in James' 23 presentation that there is in fact a lot of work on 24 it, there has been a lot of work going on the, if you 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
63 will, forcing function side of things over the last 20 1
years as new information and new evaluation and new 2
analyses have been done associated with the LOCA event 3
itself.
4 As you said, Ron, the experimental 5
database associated with LOCA does go back 40, 50 6
years but there's been a lot of work to look at what 7
is anticipated to be actual conditions in LOCA 8
analysis for the units as well as for the fuel.
9 In addition to that, and again James 10 mentioned it, new fuel types are going to have 11 different performance than those that had been tested 12 here.
13 And as I look at the research papers that 14 are associated with the testing, the researchers have 15 tried extremely hard to differentiate between the 16 cladding materials and the fuel materials that they 17 have in fact tested.
18 But it's very difficult to do, and the 19 technology in fuel design and cladding design has 20 really changed in the last 10 or 15 years and there 21 have been important changes that have improved at 22 least steady state fuel performance.
23 And most of those do in fact affect the 24 transient fuel performance as well. So, just in 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
64 summary, there is a lot that could be done and if a 1
letter were to be written at this point, it certainly 2
could focus on what we've discussed today.
3 That's the connection between the 4
experimental evaluations and the conclusions that can 5
be drawn from it. But there's three or four other 6
areas that need to be explored including the overall 7
consequence evaluation.
8 CHAIR BALLINGER: Greg, what do you think?
9 You come from the operational side and so this has a 10 significant implication for the operational side.
11 MEMBER HALNON: Hold on, I've got 12 something going on. I'm sorry, Ron, I had a personal 13 issue. So, I'm coming up to speed on this, obviously 14 it's not in my expertise to look at the research and 15 whatnot, but there are some questions that enter my 16 mind.
17 And you all do a lot of research and 18 modeling and prototypical modeling and whatnot. It 19 seems that there's a lot of data out there in the 20 industry that could be ascertained from actual 21 configurations of fuel once it's been removed.
22 And I'm not sure if it's even physically 23 possible to slice and dice a fuel rod as a high burnup 24 and take a look at it.
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
65 Maybe there is and maybe there isn't but 1
it just seems like the empirical data could be brought 2
in to the equations to confirm where we're going, that 3
would be good.
4 But overall, what I took away from this is 5
that at the very minimum, we as a Committee should 6
encourage prompt continuation of this research and get 7
to a conclusion where we can all stand back and say, 8
yes, that's a good limit.
9 And so the fuel fabricators and the 10 designers going forward, especially for the fuels 11 coming out like ATF, they really have to be --
12 My learned opinion is probably not as 13 useful as some of the you guys who get into more often 14 but that's where my head came away from.
15 CHAIR BALLINGER: I keep coming back, as 16 David has, to the multi-variability of the problem.
17 I was once on a Committee where we were analyzing and 18 doing stress analysis for a complex shape of a device.
19 And I was mystified by the results and I 20 pulled the finite element guy aside and I said, hey, 21 what are the real stresses here? He looked at me with 22 a straight face and said what do you want them to be?
23 And that looks to me like this is so 24 multi-variate, if we're not careful we'll end up 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
66 getting what we want them to be. And that could turn 1
out to be not a good thing.
2 So, okay, if we haven't heard from anybody 3
else --
4 MEMBER BROWN: Ron? I've just waited 5
since I'm obviously not a fuel person and I enjoy the 6
graph and data-points. When I look at a research 7
effort such as this, I try to look and see what it's 8
going to be used for, the end result.
9 And I guess Slide 31 or 32, 33 talked 10 about what's next, but all to me all it said was 11 justifying less conservative limits with more detailed 12 product-specific arguments.
13 And I guess I would have looked at this 14 data to say does this allow me to extract more power 15 from a particular fuel design or can I get longer life 16 from that fuel design based on a better understanding 17 of these characteristics?
18 But I didn't see that. Maybe I missed it 19 in the discussion, but it seems to me that if this is 20 going to be done, it ought to have something other 21 than just information value, other than the few 22 statements here about less conservative limits.
23 I wasn't sure how that was going to be 24 used. So, that was what I took away from the 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
67 discussions.
1 CHAIR BALLINGER: That's a good comment 2
because that's another reason to want to understand 3
what the industry perspective is on this, and that we 4
haven't heard.
5 And so I keep thinking whatever we write, 6
that's definitely a recommendation at least from my 7
point of view. I'd be curious to know what other 8
members think about that?
9 MEMBER PETTI: I'm quickly skimming that 10 long paper from CAPS again and the thing that strikes 11 me -- I guess there's proposed testing in TREAT -- is 12 what you really have to do now --
13 Okay, we've done everything that's been 14 done to date, how do you do a better experiment?
15 Because it looks like the Halden stuff was 16 done at lower linear power rates than typical but they 17 were done at higher temperatures in terms of the 18 temperature which you get to than a lot of the cores.
19 And so how do you do something when you 20 kind of look at things a little more systematically 21 and start with stuff that you think is going to be 22 good but then things that are going to end up being 23 not good, and try to find that boundary.
24 Is there something that can be done taking 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
68 everything you've learned to date to design the next 1
generation of experiments, if you will? Because it's 2
hard to tell reading the paper how different was SCIP 3
from Oak Ridge.
4 If they're all very similar, then it's 5
just more data with slightly different run-ups and the 6
like. But there's a lot of stuff in the CAPS paper 7
that says that you could change the conditions a 8
little to be more prototypic and get answers that are 9
more prototypic.
10 CHAIR BALLINGER: Coming back to the 11 consequences issue, let's say everything that we've 12 learned today happens, let's say that in doing a LOCA 13 you do get ballooning, which we know would happen, you 14 do get dispersal, you do get relocation and all the 15 same.
16 How does that affect the answer? If it 17 has a great effect on the answer, and given all the 18 uncertainties, we have to factor that in, what are the 19 temperatures and all that stuff?
20 If it has a very strength effect, that 21 points you in one direction but if in effect, the 22 consequences are within the uncertainty of the 23 analysis, then it may be a nice thing to know that 24 happens but the consequences are such that you don't 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
69 need to have more detail.
1 Or maybe you need one specific set of 2
experiments to fill out something, to bracket 3
something, but those are two very different paths and 4
they're all dictated by what the consequences are 5
likely to be should it actually happen.
6 MEMBER PETTI: I don't disagree, Ron, I 7
just worry that as you propagate down the 8
calculational train here and you try to do more and 9
more, the uncertainty bounds tend to grow and in some 10 ways I worry that the phenomena that you try to model 11 get even more complex or uncertain.
12 I'd certainly be interested to know what 13 the impact is. If it seems to always be not a big 14 deal but that may be hard in the end. But I think 15 it's worth hearing from industry on that for sure.
16 MEMBER HALNON: This is Greg. What is the 17 burnup that we see in the three cycle fuel now? Isn't 18 it much less than 55?
19 CHAIR BALLINGER: We're over that.
20 MEMBER HALNON: I didn't have a feel for 21 it.
22 CHAIR BALLINGER: Peak rods are well above 23
- 65. Am I right there? Who should know? Michelle 24 should know.
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
70 MS. BALES: Currently burnup lead average 1
is limited to 62, so there are some rods that are 2
above 55.
3 MEMBER HALNON: And that has been 4
increased, I assume, over the years just based on 5
better core designs and longer cycles.
6 So, the impact is obviously just going to 7
be in that realm, the life that cycles or the amount 8
of capacity factors that you're able to get, or all of 9
the above.
10 So, certainly the impact is going to be 11 from a physical perspective there, and in the fuel 12 design, I'm not real good on that so I don't know in 13 the fuel fabrication if there's going to be additional 14 margins that have to be put in there or not.
15 MEMBER PETTI: I think some of these 16 things can be done in parallel. It sounds like these 17 two experiments are going to go forward, they're going 18 to try to be more prototypic since the TREAT guys are 19 co-authors of the CAPS paper.
20 And then hearing from industry, they'll be 21 a nexus somewhere here whenever all this stuff gets 22 completed. It will be worth another look.
23 CHAIR BALLINGER: That's a very good segue 24 and I'm assuming that public comments will come from 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
71 members of the industry.
1 So, if there aren't any other comments 2
right now from Members, I guess the public line is 3
actually open so can we ask for comments from members 4
of the public? I'm not sure how we do this in terms 5
of who controls what.
6 But I guess if you want to make a public 7
comment, just do star 6 and start talking.
8 MR. SONTOS: Ron, can you hear me?
9 CHAIR BALLINGER: I can hear somebody but 10 I don't know who it is.
11 MR. SONTOS: This is Al Santos. Can I go 12 ahead?
13 CHAIR BALLINGER: You certainly can.
14 MR. SONTOS: Al Santos, NEI. I want to 15 thank the Staff for turning around this report so 16 quickly. We've only had about a week to review it but 17 the industry has a lot of detailed technical comments 18 on the report.
19 So, the industry requests an opportunity 20 and specific time to provide the detailed technical 21 expert comments.
22 Even if we considered a revision to the 23 RIL or to present at a follow-up ACRS Subcommittee or 24 full Subcommittee, full Committee meeting regarding 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
72 the RIL.
1 So, those are the first points we wanted 2
to make. The second one is that similar to what I 3
hear, it seems like there's a lot of discussion on 4
this.
5 The RIL as written today is a very 6
considered treatment of the data, even choosing to 7
apply additional conservatisms when presented with 8
evidence that a
test may have already been 9
unrealistically conservative.
10 The industry continues to ask the NRC to 11 incorporate the research that has been summarized here 12 in a way that provides reasonable assurance of 13 adequate protection of public health and safety.
14 And lastly, we just want to reiterate that 15 the report, interpretive test data, and conservatisms 16 are in a manner that should be construed as 17 subjective.
18 And industry technical experts request an 19 opportunity to provide a comment and feedback to be 20 incorporated in the report or for the ACRS Committee.
21 We just wanted to have some high-level talking points 22 here to provide fuel.
23 CHAIR BALLINGER:
Because of our 24 protocols, our rules, we can only have comments, but 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
73 I have a question for Michelle, she might know the 1
answer to this hopefully.
2 If we were to treat this RIL as if it were 3
a Reg Guide or a NUREG, it sounds to me like the 4
industry might offer public comments which may in the 5
end result in a modification of the document.
6 Is that one way to look at this? Somebody 7
is doing something here. So, would the Staff consider 8
effectively putting it out for public comments -- I 9
don't know how they treat RILs -- and take those 10 comments into consideration.
11 It sounds to me like the RIL was issued in 12 a very expeditious way but there are some comments 13 that may be important.
14 MS. BALES: It's a good question and I 15 think that in order to answer it I'm going to explain 16
-- and Kim alluded to this in her introductory 17 remarks, but maybe we can go back to the line of 18 thought that she was raising.
19 From the nature of documents the NRC 20 issues, a research information letter is designed to 21 be a communication between the Office of Research 22 Staff to NRR Staff.
23 And we wanted to do that very quickly in 24 a timely manner because we know that NRR Staff is 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
74 receiving applications to extend burnup limits right 1
now and we wanted to put information in the hands of 2
those NRR reviewers so they could understand in a 3
concise way what information was available on FFRD.
4 So, that was our motivation for the 5
schedule that we kept and our target audience is NRR 6
reviewers. As Kim mentioned in her introductory 7
remarks, in the past RILs have been the start of other 8
regulatory documents.
9 So, they could be the basis for a later 10 guidance document or information that's contained in 11 a RIL could integrate into the standard review plan.
12 Any of those steps, which are so premature at this 13 stage, necessitate a public engagement.
14 So, if those steps were to be taken, it 15 would naturally happen that the industry and members 16 of the public would be able to engage because those 17 are clearly requirements of those types of documents.
18 So, a RIL typically would not have a 19 public comment period. I think I'm only aware of one 20 instance where there's been a public comment period 21 for RILs.
22 So, I just say that to explain what the 23 RIL is relative to other regulatory tools and to 24 explain that it would not surprise me if a year from 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
75 now we were talking about something that was built on 1
the RIL that does have public comment.
2 Now, that does not mean that we don't want 3
to continue to engage with stakeholders, I think 4
that's a critical part of NRC's principles of good 5
regulation is to continue to have transparency and 6
engage with our stakeholders.
7 So, I don't want to preclude that in any 8
way and I think that when it comes to the research 9
information letter, to the extent that we understood 10 the importance of going external, that's why we 11 proposed the peer review process, because we wanted to 12 get outside perspective.
13 We wanted to ensure that we weren't 14 operating in a vacuum with such a strict safety 15 authority perspective that we were missing something.
16 And so we sought outside peer review, recognizing the 17 potential for us to be too narrowly focused.
18 So, I'll just stop there and say I think 19 this is the start of additional documents and I see 20 Joe Donahue in NRR has his hand raised. I think he 21 would probably be more appropriate to finish my 22 thought.
23 MR. DONOGHUE: Chairman Ballinger, is that 24 okay with you?
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
76 CHAIR BALLINGER: Sure.
1 MR. DONOGHUE: I'm the Division Director 2
in the Division of Safety Systems in NRR. I think 3
Michelle put it very eloquently and very well. I'll 4
just add we have been interacting as you know with 5
individual vendors.
6 We've also been interacting with 7
stakeholders in general on high burnup. We've held 8
two public workshops on the topic and I think Michelle 9
already explained the background and the motivation 10 for the RIL being reduced.
11 As far as getting public comments, she let 12 it out very well. This is a step in a direction that 13 is going to include public comment in some form. What 14 I don't want to do today is go out in front of my 15 staff who has to evaluate --
16 As Michelle said, our Staff gets the RIL, 17 we have to decide what's next specifically. Is it 18 just an update to this SECY paper that she mentioned 19 from 2015 or is it more than that?
20 Could it be a guidance document as you've 21 all alluded to? Not exactly sure yet. So, I think 22 that there's fast-paced activities in high burnup and 23 accident-tolerant fuel.
24 I don't want to ask Research to go through 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
77 a public comment period on this because that's going 1
to delay our ability at NRR to take the information 2
and start taking next steps. That's just my own 3
feeling.
4 It's up to you and the Committee on how 5
much you want to hear. The last thing I'll say is the 6
vendors are all preparing -- you've heard about some 7
of them, but in some ways they're considering -- in 8
one case we have a similar interim increased burnup 9
from one vendor.
10 But topical report reviews are anticipated 11 here and I think you will have the opportunity, the 12 ACRS will have the opportunity, of course, to interact 13 on all of those reviews.
14 CHAIR BALLINGER: Thank you. If I've 15 spoken out of turn in terms of procedure, I apologize.
16 But I wonder whether or not it would not be timely to 17 have a Subcommittee meeting on this particular topic?
18 Not the RIL but high burnup fuel and the 19 issues related to high burnup fuel, where we could 20 have both the NRC Staff as well as the vendors and 21 other stakeholders be able to speak.
22 And it sounds to me like there have been 23 a few public meetings that are sort of on this topic 24 in the recent path. What do Members think about that?
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
78 MR. DONOGHUE: That would be helpful for 1
me because clearly an industry perspective and a 2
vendor perspective would round out the holes in my 3
head about this.
4 CHAIR BALLINGER: Joy, Dennis, Steve?
5 MEMBER BLEY: I liked the idea when you 6
first brought it up a while ago, I think that would be 7
a useful thing.
8 CHAIR BALLINGER: I'm trying to get a 9
little bit of my thoughts together and what to say at 10 the full Committee meeting, what a document would look 11 like should we produce one or whether or not we should 12 just recommend that we --
13 Thank you very much, we understand the 14 issues that have been presented to us, the 15 uncertainties are such that we think we should have 16 further discussions at a subsequent meeting, whenever 17 that should occur.
18 MEMBER BLEY: The only thing I'd toss in, 19 and you hinted at it, is if we decide to do what you 20 just suggested, our staff and the NRC Staff should get 21 together and we ought to make sure we're not 22 procedurally getting in the way of anything associated 23 with the RIL.
24 I think bringing the general information 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
79 back to the Committee is probably very helpful.
1 CHAIR BALLINGER: The RIL is what it is.
2 It's issued so there's no question of the Committee, 3
at least I don't think so, saying don't do this.
4 That's just not the way it works.
5 So, our normal letter comments would be 6
different in this case.
7 MEMBER HALNON: Ron, it just seems 8
premature at this point in that we would be getting 9
into some internal processes that we might influence 10 in the direction that is not necessarily fully 11 informed.
12 CHAIR BALLINGER: We could simply treat 13 this meeting as an information briefing, which is 14 exactly what it is and make comments assuming it was 15 an information briefing, with suggestions that we've 16 all bantered about.
17 That may be one way to look at it and that 18 would seem like -- oh, Michelle has got her hand up.
19 MS. BALES: I didn't want to interrupt but 20 I wanted to clarify the RIL isn't issued yet. It was 21 in concurrence and it was made publicly available in 22 draft form in order to facilitate this meeting.
23 But technically, it is not issued so I 24 wanted to just clarify that.
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
80 CHAIR BALLINGER: Thanks, I don't know how 1
often the Committee has looked at or had meetings on 2
a RIL but I don't think it's been very frequent. Al 3
has his hand up or maybe it's been up for a while and 4
he didn't put it down.
5 MR. SONTOS: I just wanted to say that to 6
Michelle's point, there has been a past present of RIL 7
being commented on by technical experts from the 8
industry. So, putting that out there, that's all.
9 I just wanted to let that be known.
10 CHAIR BALLINGER: Thank you.
11 MEMBER BLEY: Here's a question, can we as 12 a Committee look at RILs? I was trying to search my 13 memory.
14 There have been a number of cases where 15 we've looked at your section of the SRP or a reg guide 16 or a NUREG that also had associated with it a RIL and 17 came under our review as part of the other documents.
18 So, we've seen them before coming before 19 the Committee, I don't think just by themselves but 20 I'm not positive of that.
21 CHAIR BALLINGER: You've been here a lot 22 longer than I have so you would know. Okay, I'm going 23 to have to try to put thoughts together and I'll 24 solicit --
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
81 Now there's a flashing blue circle around 1
Dave. What does that mean?
2 MEMBER PETTI: I don't know, I just got 3
unmuted to ask you are we going to recommend that we 4
write a letter or not?
5 CHAIR BALLINGER: I'm casting about for 6
suggestions, I'll send an email out to everybody was 7
well because my memory is not as good as me writing 8
something down. And people are taking notes.
9 But
- yes, I'm casting about for 10 recommendations of whether we should write a letter of 11 some kind and what it should say.
12 MEMBER PETTI: I'm struggling with a 13 letter that would help the draft RIL be, quote, 14 better, that's one thing.
15 The letter could be much broader and talk 16 about the issue in broader terms but it wouldn't 17 necessarily affect the subsequent steps of the RIL.
18 Same thing, we need to get industry's 19 perspective. Some of the things we've heard, I don't 20 know that they affect the next step. So, I'm with 21 you, I'm struggling as well. I'm trying to figure out 22 exactly what we'd say.
23 MEMBER REMPE: Could we ask Michelle or 24 maybe Kim what would a letter from ACRS do? We didn't 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
82 ask you at the beginning of this meeting, which I 1
sometimes do. What does the Staff anticipate from 2
ACRS and how does it affect your schedule?
3 What would you do with a letter if there 4
were some recommendations or comments in it? Is it 5
too late, this thing is going out the door and you'd 6
consider it as you go forward?
7 What didn't does it make in your opinion?
8 MS. WEBBER: Go ahead, Michelle, I'll add 9
in if I need to.
10 MS. BALES: I was just going to explain 11 that in our office instructions for RILs, in the 12 section on peer review, it acknowledges the ACRS 13 Committee is a body that can serve as peer review.
14 So, in one element we can just think about 15 the benefit of the ACRS writing a letter that examines 16 the RIL from a peer review standpoint and makes 17 comments about your technical review.
18 And I think with respect to the timing of 19 it, the RIL is done from the Staff's perspective, it 20 is complete with concurrence but we're very conscious 21 of not issuing it until the ACRS meetings were 22 complete.
23 Because we anticipated the possibility 24 that ACRS members would have important comments that 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
83 we needed to consider. So, I think in one respect a 1
letter that serves to expand the sense of peer review 2
is certainly one element.
3 I'll just say for myself, and I'm just 4
reacting to the things in real time so I haven't 5
thought about this too much.
6 Nut some of the ACRS Members were 7
commenting on pulling back to see where does this 8
information fall relative to a broader either research 9
need, or where these phenomena fall relative to 10 looking at upstream LOCA possibility events or 11 downstream consequences.
12 I think that's another area that I thought 13 the ACRS comments were very important to put the RIL 14 in perspective.
15 So, even just in this meeting I think that 16 we've gathered that but it could be that putting that 17 into a letter helps to just place the RIL in the 18 context of a bigger picture.
19 MEMBER REMPE: Michelle, I'm a little slow 20 and so I heard you say at one point from the Staff's 21 perspective the RIL is done.
22 But then I heard you say, which I am more 23 sympathetic and encouraging of, is that you considered 24 us a technical peer review, you've heard things but it 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
84 was from individual members that you had thought you 1
didn't want to issue the RIL until the full Committee 2
had written a letter because there might be some 3
changes.
4 Is the latter statements that I've 5
summarized where you're at, that, yes, there is the 6
opportunity that our letter might make you hold off in 7
the final publication and you might consider some of 8
our comments?
9 MS. BALES: Yes, sorry, when I said the 10 letter was done from a Staff's perspective, what I 11 mean is there are no action items that we have on our 12 list that we are still waiting to incorporate, or no 13 additional data that we're waiting for.
14 The only input that we have not received 15 that we know we want to receive are these ACRS 16 meetings. So, what I mean from the saying that the 17 Staff is done, there's nothing else that we know to do 18 that we haven't yet done except for these ACRS 19 meetings.
20 So, the latter is true.
21 MEMBER REMPE: Thank you, that helps.
22 Kim, did you want to add anything else?
23 MS. WEBBER: No, actually, Michelle did a 24 fantastic job explaining it much better than I could 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
85 because she's very familiar with the RIL process.
1 MEMBER REMPE: Ron, I would like to have 2
a presentation. I'm just one Member but I'd like to 3
see a presentation at full Committee and I think it 4
would be a worthwhile letter for us to put together at 5
this time.
6 Because we do have this opportunity and 7
the Staff has said they're waiting to hear our 8
comments.
9 CHAIR BALLINGER: We've assumed that would 10 be the case and that's why we scheduled the full 11 Committee meeting. So, we're slotted into have this 12 take place so I think for sure, the only questions 13 that I have are the tone and content and scope of the 14 letter.
15 I think I've got a few ideas and I'll be 16 soliciting within the next five minutes written 17 comments from Members that are in attendance so that 18 I can strike while the iron is hot, or Zana and I can 19 strike while the iron is hot.
20 So, that would be the path forward, unless 21 people have other suggestions. Dennis, the blue 22 circle is flashing around you now. I don't know what 23 it means.
24 MEMBER BLEY: I had my mic on for a while 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
86 then I turned it off.
1 CHAIR BALLINGER: Kim, your hand is up.
2 MS. WEBBER: I was just thinking about 3
Michelle's response and the value of your review and 4
ideas around the RIL.
5 I just wanted to add that I fully agree 6
with Michelle that having your perspectives on the RIL 7
in a letter would be very valuable to us.
8 I know it takes more time, there's a 9
presentation writing the letter, but it would be of 10 value to us and so I think we're looking forward to 11 that.
12 CHAIR BALLINGER: Thank you, again. So, 13 now it comes to the full Committee. I don't have it 14 exactly in front of me so I don't know how much time, 15 Zana might know, we have slotted. Is it an hour or 16 two?
17 MS. ABDULLAHI: Let me look, I'll get it 18 to you in a few minutes.
19 MEMBER BLEY: Ron, whatever it is, I would 20 suggest that we try to make sure it's at least two 21 hours2.430556e-4 days <br />0.00583 hours <br />3.472222e-5 weeks <br />7.9905e-6 months <br /> between now and then, because I don't think an 22 hour2.546296e-4 days <br />0.00611 hours <br />3.637566e-5 weeks <br />8.371e-6 months <br /> would be long enough to --
23 (Simultaneous Speaking.)
24
-- the other Members. I think you need a 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
87 couple hours.
1 CHAIR BALLINGER: We started at 1:00 p.m.
2 and it's 3:00 p.m. and we've had --
3 MEMBER REMPE: This is Joy and I'm just 4
going to interrupt you. Right now you're scheduled 5
for Wednesday morning at 8:30 a.m. to 10:00 a.m. for 6
the presentation.
7 But Matt often puts in an hour right after 8
so he's got, really, from 8:30 a.m. to 11:00 a.m. for 9
the combined time to have the presentation and read 10 him the letter. That's something you might be able to 11 negotiate with him.
12 That's where it's at right now.
13 CHAIR BALLINGER: My guess is that if we 14 allowed that time for presentation and questions as 15 opposed to reading in a letter and then read the 16 letter later on, that might free up some time.
17 Because my guess is that by that time 18 we'll also get requests for members of the public to 19 make comments. So, I think if we go two and a half 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> -- did I do the math right?
21 MEMBER REMPE: Yes, you did and by the 22 way, I'd also note that from 1:00 p.m. to 2:30 p.m.
23 was the research review and we got done with that so 24 we do have some flexibility with the schedule in 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
88 December.
1 I'm just pointing out some things and what 2
Matt prepared, I'm not the chair.
3 CHAIR BALLINGER: That's good. I don't 4
see anybody's hands or anything up, I guess we've got 5
to notify Matt that we shortened up the meeting once 6
again. So, if there aren't any other comments, I 7
think we are finished.
8 On that basis, the meeting is adjourned.
9 MEMBER REMPE: Thanks everyone for the 10 presentation and discussion.
11 (Whereupon, the above-entitled matter 12 went off the record at 3:08 p.m.)
13 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 2000 2020 2010 SCIP-III, Hot-cell NRC
@Studsvik, Hot-cell
- ORNL, Hot-cell S
E C
Y
- Halden, In-Pile
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 30-50 cm refabricated fueled segment from commercially irradiated rod Pressure line establishes segment pressure 4 heating elements in furnace
Outcome of the RIL:
Identify when FFRD occurs Segment from NRCs ANL LOCA program at 55 GWd/MTU before and after testing
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 when FFRD occurs
Outcome of the RIL:
Identify which rods are a concern for FFRD Burnup Power (LHGR) 1st cycle 2nd cycle 3rd cycle Fine fragmentation requires ~55 GWd/MTU
Outcome of the RIL:
Identify which rods are a concern for FFRD Burnup Power (LHGR) 1st cycle 2nd cycle 3rd cycle Relocation requires ballooning
Outcome of the RIL:
Identify which rods are a concern for FFRD Burnup Power (LHGR) 1st cycle 2nd cycle 3rd cycle Dispersal requires fine fragmentation, relocation and burst
Burnup Power (LHGR) 1st cycle 2nd cycle 3rd cycle Dispersal requires fine fragmentation, relocation and burst For the dispersal zone, some core loadings may result in no region of overlap Overlap influenced by:
- ECCS response
- Plant design
- Loading pattern
- Fuel and cladding 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 Power (LHGR) tFGR influences burst probability 1st cycle 2nd cycle 3rd cycle Transient Fission Gas Release - Increasingly important at higher burnup; Important to accurately predict burst Dispersal requires fine fragmentation, relocation and burst
Relocation packing fraction - Important to accurately predict burst and peak temperature for non-burst high burnup rods Assume the balloon region is 70-85% fuel (15-30% void) by volume for calculating rod behavior Burnup Power (LHGR) 1st cycle 2nd cycle 3rd cycle
- 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 The RIL supports targeted FFRD analysis
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 Research supports a pellet-average burnup limit of 55 GWd/MTU as the onset of fine fuel fragmentation Segment from NRCs ANL LOCA program at 55 GWd/MTU before and after testing FFRD
Element 2: A local cladding strain threshold below which relocation is limited Research suggests fuel relocation is limited in regions of the fuel rod experiencing less than 3% cladding strain.
NRC test #
Strain threshold, top (%)
Strain threshold, 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 FFRD
What do dispersal measurements look like?
Element 3: A conservative value for the mass of dispersible fuel as a function of burnup Dispersal during the test
Element 3: A conservative value for the mass of dispersible fuel as a function of burnup FFRD
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.
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%
ALL collected fuel Element 3: A conservative value for the mass of dispersible fuel as a function of burnup 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
RIL 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
- Utilizing new modeling features
RIL 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 needed to go further
Questions?