ML23352A393
| ML23352A393 | |
| Person / Time | |
|---|---|
| Issue date: | 11/16/2023 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| NRC-2628 | |
| Download: ML23352A393 (1) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Radiation Protection and Nuclear Materials Subcommittee Meeting Docket Number: (n/a)
Location: teleconference Date: Thursday, November 16, 2023 Work Order No.: NRC-2628 Pages 1-170 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1716 14th Street, N.W.
Washington, D.C. 20009 (202) 234-4433
1 1
2 3
4 DISCLAIMER 5
6 7 UNITED STATES NUCLEAR REGULATORY COMMISSIONS 8 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 9
10 11 The contents of this transcript of the 12 proceeding of the United States Nuclear Regulatory 13 Commission Advisory Committee on Reactor Safeguards, 14 as reported herein, is a record of the discussions 15 recorded at the meeting.
16 17 This transcript has not been reviewed, 18 corrected, and edited, and it may contain 19 inaccuracies.
20 21 22 23 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3 + + + + +
4 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5 (ACRS) 6 + + + + +
7 RADIATION PROTECTION AND NUCLEAR MATERIALS 8 SUBCOMMITTEE 9 + + + + +
10 THURSDAY, NOVEMBER 16, 2023 11 + + + + +
12 The Subcommittee met via Teleconference, 13 at 1:00 p.m. EST, David A. Petti, Chair, presiding.
14 COMMITTEE MEMBERS:
15 DAVID A. PETTI, Chair 16 RONALD G. BALLINGER, Member 17 VICKI M. BIER, Member 18 CHARLES H. BROWN, JR., Member 19 GREGORY H. HALNON, Member 20 WALTER L. KIRCHNER, Member 21 JOSE A. MARCH-LEUBA, Member 22 ROBERT MARTIN, Member 23 JOY L. REMPE, Member 24 THOMAS ROBERTS, Member 25 MATTHEW W. SUNSERI, Member NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
2 1 ACRS CONSULTANT:
2 STEPHEN SCHULTZ 3
4 5 DESIGNATED FEDERAL OFFICIAL:
6 WEIDONG WANG 7
8 9
10 11 12 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
3 1 TABLE OF CONTENTS 2
3 Opening Remarks . . . . . . . . . . . . . . . . . 4 4 Introductions . . . . . . . . . . . . . . . . . . 6 5 Discussion on SAND 2023-01313 . . . . . . . . . . 11 6 Discussion on Additional Source Term Analysis . . 96 7 Opportunity for Public Comment . . . . . . . . 168 8 Meeting Adjourns 9
10 11 12 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
4 1 P-R-O-C-E-E-D-I-N-G-S 2 1:01 p.m.
3 CHAIR PETTI: Good afternoon, the meeting 4 will now come to order. This is a meeting of the 5 Radiation Protection and Nuclear Materials 6 Subcommittee of the Advisory Committee on Reactor 7 Safeguards.
8 I'm Dave Petti, chairman of the 9 subcommittee. Members in attendance are Charles 10 Brown, Joy Rempe, Matt Sunseri, Ron Ballinger, Walt 11 Kirchner, Vesna Dimitrijevic, Vicki Bier, Greg Halnon, 12 Tom Roberts, Bob Martin, and I believe Jose March-13 Leuba may be on.
14 MEMBER MARCH-LEUBA: I am on.
15 CHAIR PETTI: Great. And Steve Schultz, 16 our consultant, is also with us today. Weidong Wang 17 is the Designated Federal Official for this meeting.
18 As posted in the agenda and on the ACRS 19 website, the topic for today is to hear information --
20 an information briefing on Sandia National 21 Laboratory's report, high burnup fuel source term 22 accident sequence analysis.
23 The subcommittee will hear presentations 24 by and hold discussions with the NRC staff, Sandia, 25 and other interested persons regarding this matter.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
5 1 The meeting is open to the public. Rules 2 for participation in all ACRS meetings, including 3 today's, were announced in the Federal Register on 4 June 13, 2019.
5 The ACRS section of the U.S. NRC public 6 website provides our charter, bylaws, agendas, letter 7 reports, and full transcripts of all full and 8 subcommittee meetings, including slides presented 9 there. The meeting notice and agenda for this meeting 10 were posted to there.
11 We've received no written statements or 12 requests to make an oral statement from the public.
13 The subcommittee will gather information, 14 analyze all of the issues and facts, and formulate a 15 post positions and actions as appropriate today.
16 Transcript of the meeting is being kept 17 and will be made available. Today's meeting is being 18 held in person and over Microsoft Teams through ACRS 19 staff and members, NRC staff, and other attendees.
20 There's also a telephone bridge line and 21 a Microsoft Teams link allowing participation for the 22 public.
23 When addressing the subcommittee, 24 participants should first identify themselves and 25 speak with sufficient clarity and volume so they may NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
6 1 be readily heard. When not speaking, we request that 2 participants mute their computers or microphone by 3 pressing star six.
4 We will now proceed with the meeting. And 5 before we call up our staff management, I'd like to 6 just put some context in. As you know, we reviewed 7 Reg Guide 1.183 here about a month ago. And 8 questions, as part of that discussion, talked about 9 the calculations that were done by Sandia.
10 And we thought it would be useful to have 11 a briefing on this so members get a more complete 12 picture of the depth of the technical basis upon which 13 1.183 relied.
14 With that, whoever's going to -- you are?
15 Okay, great.
16 MS. WEBBER: All right, good afternoon, 17 everybody. My name is Kim Webber. I'm the Director 18 of the Division of Systems Analysis in the NRC's 19 Office of Nuclear Regulatory Research.
20 It's a pleasure to be here today to talk 21 about a topic that is of really broad interest, not 22 only for the NRC, but also for our external 23 stakeholders.
24 There's a lot of interest as it relates to 25 Reg Guide 1.183 and, you know, a future update to that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
7 1 regulatory guide. So, I'm glad to see that there are 2 a lot of online participants as well, as those who are 3 in the room.
4 And this is going to be a focus on 5 research that was recently done over the last few 6 years. And it's really not focused on the regulatory 7 application of it, although it's probably a key 8 component of the technical basis that may be used to 9 develop a Reg Guide, a future revision of the Reg 10 Guide.
11 So, we do also have our regulatory 12 partners here with us, Elijah Dickson and Michelle 13 Hart. And if there are questions related to the 14 regulatory aspects, hopefully they can answer those 15 types of questions.
16 But also here with me today, I have my 17 staff, Hossein Esmaili, who's the chief of the fuel 18 source terms code branch, along with Shawn Campbell 19 and Mike Salay who are experts in severe accidents.
20 And then, we also have our colleagues from 21 Sandia National Lab at the table, too, who are also 22 going to do some presentations.
23 But before we get into that, you know, I 24 just wanted to say that, you know, this analysis that 25 the staff and contractors have been working on for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
8 1 quite some time is really important when it comes to 2 providing technical basis for the use of high burnup 3 fuel and also accident tolerant fuel in the future.
4 And as many of you know, the MELCOR codes, 5 severe accident code, was used to perform these 6 analyses. And it's a system level code that simulates 7 the entire spectrum of accidents and phenomena from 8 accident initiation to core and fuel degradation and 9 fission product gas release from the fuel and 10 transportation to containment and the environment.
11 It has a large user base, both 12 domestically and internationally, with about 30 13 participants or 30 countries participating in our 14 CSARP, Cooperative Severe Accident Research Program, 15 co-chairing program.
16 And you know, it's critical to have their 17 participation because they identify code bugs. They 18 highlight important aspects of the scenarios that are 19 included in those codes. They contribute, you know, 20 their own studies with using those codes. So, that 21 cooperation is critical.
22 And then, also, MELCOR uses inputs from 23 our SCALE neutronics code for decay heat and fission 24 product inventories.
25 And so, because of the flexibilities of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
9 1 the combination of MELCOR and SCALE, we also use them 2 quite extensively for analysis that will support the 3 non-light water reactor technical bases and 4 confirmatory analysis going forward.
5 So, over the years, we've used MELCOR for 6 a number of regulatory applications, as many of you 7 know.
8 Some of the high visibility projects 9 include the state of the art reactor consequence 10 analysis, or SOARCA study, and post-Fukushima analysis 11 such as the containment protection and release 12 reduction documented in a NUREG -- in one of the 13 NUREGs.
14 And we've also completed earlier MELCOR 15 analysis which formed the technical basis for the 16 Revision 1 to Reg Guide 1.183 which is called the 17 alternative radiologic source term for evaluating 18 design basis accidents at nuclear power reactors.
19 So, this research benefits not only away 20 from code physical model improvements, but also 21 improvements in best practices in generating code 22 input decks which are publically available 23 representations of plants and accident scenarios.
24 So, in 2020, we convened a panel of 25 international experts to create the phenomena NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
10 1 identification ranking table to address the 2 significant phenomenologic issues impacting core 3 degradation and radiological releases for various 4 accident tolerant and high burnup fuels.
5 And we compared those phenomena against 6 the traditional large light water reactor fuels.
7 The aim of the PIRT was to help NRC 8 understand how the ATF concepts and the high burnup 9 fuel may change core degradation and radiological 10 release behavior which provide information that is 11 useful in developing source terms for these designs.
12 And also, along with that PIRT, we did 13 publish a literature view that provided input to the 14 PIRT panelists and also the results of the panel 15 findings.
16 And both of those are documented in 17 NUREGs.
18 So, now, I'd like to turn the presentation 19 over to the staff and our colleagues from Sandia 20 National Lab who, Dave Luxat and Lucas Albright, and 21 they'll present the analysis that was associated with 22 the high burnup source term report and the peer 23 review.
24 And I think we mentioned that this will be 25 used as the, you know, probably part of the technical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
11 1 basis for the Reg Guide 1.183 Revision 2.
2 So, let me turn it over to Shawn, I think 3 you're next.
4 So, thank you for your attention.
5 MR. CAMPBELL: Yes, thank you very much.
6 So, hello, everyone. My name is Shawn 7 Campbell. I'm in the Office of Research in the Sandia 8 Branch and with Kim Webber.
9 And so, I'm actually going to turn it over 10 to Lucas Albright here in just a moment and our 11 colleagues at Sandia National Labs.
12 They are the ones that we commissioned to 13 do this work and put together this Sandia report.
14 So, we've asked them to come in and help 15 to explain the work that they've done. And we relish 16 your questions and feedback as we go.
17 So, Lucas?
18 MR. ALBRIGHT: Thanks, Shawn.
19 So, I think we can pull up the slides if 20 those are ready. Is that something we have access to 21 here? Okay.
22 MEMBER REMPE: Sorry, I've been tied up 23 with other things, so Dave's in charge.
24 But yes, we thought, unless we're told, we 25 rely on the presenters to pull up their slides.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
12 1 Weidong, can you do it for them? Is that 2 what you'd prefer?
3 MR. ALBRIGHT: Yes, I think -- yes.
4 (Off-microphone comments.)
5 MEMBER REMPE: Weidong, can you -- yes, 6 can you share your screen? We can do that.
7 MR. ALBRIGHT: No, sorry about that, that 8 was our misunderstanding. I apologize.
9 (Off-microphone comments.)
10 MR. ALBRIGHT: Yes, no problem. If 11 they're not readily available, we can pull them up and 12 --
13 MEMBER REMPE: He just needs to share his 14 screen and they're coming up now.
15 MR. ALBRIGHT: All right, thank you all.
16 Okay, so, my name is Lucas Albright. I 17 work at Sandia National Laboratories in the severe 18 accident analysis and modeling group performing 19 analyses with the MELCOR code and also developing the 20 MELCOR code.
21 This presentation that I'll be giving 22 today is sort of an overview of the technical details 23 of the high burnup fuel accident source terms that we 24 developed.
25 This was a multi-year effort to basically NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
13 1 extend the NUREG-1465 source terms to higher burnups.
2 What you'll notice in this presentation is 3 that I'm sticking to an explanation of what we did, 4 how we did it, and why we did it that way, not 5 necessarily, you know, implementation or how these 6 numbers would be used on the regulatory side.
7 Next slide, please? Thank you.
8 So, a brief overview of some contents that 9 we have in this presentation.
10 I'll go into the motivation and background 11 that fed into this work.
12 Then, I'll give a high level overview of 13 the key takeaways from the work before diving into the 14 technical details during the deep dive.
15 Then, we'll have a little bit of a summary 16 of what we went over for the high burnup source terms 17 before going into the independent peer review and the 18 upcoming work.
19 All right, next slide, please? So, the 20 high burnup fuel source term analysis, like I said, 21 this was a multi-year effort published in 2023.
22 The objective here was to develop 23 alternative source terms that were applicable to 24 higher burnup light water reactor cores with extended 25 enrichment, or HALEU, fuel.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
14 1 And this, like I said, is extending that 2 NUREG-1465 source term and is in the SAND2011 source 3 terms to this higher level of burnup and extended 4 enrichment.
5 Next slide, please? Some historically 6 relevant studies that we wanted to just sort of just 7 give a high level overview of were TID-14844, the 8 calculation of distance factors for power and test 9 reactors.
10 This was published back in 1962 and 11 focused really on some experimental data. This was 12 sort of prior to the introduction of the use of 13 computer codes to inform our source terms.
14 The next major study was NUREG-1465, the 15 accident source terms for light water nuclear power 16 plants.
17 This was the first source term to use 18 computer codes. This one used the STCP code, which 19 was the forefather of the MELCOR code.
20 The next major relevant study that we come 21 across is the SAND2011 study.
22 This was the first source term study to 23 use the MELCOR code.
24 And this study actually looked at accident 25 source terms for, again, light water reactors, but NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
15 1 using high burnup or MOX fuels.
2 This was an older version of MELCOR 3 published in 2011. So, many advancements have 4 occurred since that study was published.
5 Next slide, please? So, this is a little 6 time line that I put together for today's talk just 7 sort of going over the major developments as we march 8 through time from NUREG-1465 to today's SAND2023 9 report.
10 What you'll see are that between the 11 NUREG-1465 and SAND 2011 reports, we had a number of 12 major developments, including the MELCOR code becoming 13 sort of -- or coming online.
14 NUREG-1560, this was the plant 15 examinations that sort of gave us the description of 16 scenarios that we look at in SAND2011 and SAND2023.
17 Reg Guide 1.183, which we're all familiar 18 with.
19 Phebus FP occurred in this time frame as 20 well which gave us some insights into severe accident 21 progression that we didn't necessarily have as clear 22 of an understanding of it at the time.
23 In the time since 2011, we actually see 24 that a lot more work has been done.
25 We note first, Fukushima Daiichi happened NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
16 1 a couple months after SAND2011 was published.
2 The SOARCA studies, for those of you who 3 are unfamiliar, these are the state of the art reactor 4 consequence analyses.
5 This is where we took a given plant, we 6 have two volumes for the original body of work which 7 was Surry and Peach Bottom, where we basically ran a 8 set of sequences and sort of advanced the state of 9 practice using the MELCOR code to model these severe 10 accidents.
11 During this time, the BSAF project, of 12 course, began.
13 This is where we demonstrated, excuse me, 14 where we demonstrated the MELCOR code in modeling the 15 Fukushima accidents.
16 A number of improvements were also made to 17 the code during this period of time to incorporate 18 those findings.
19 The last two that I want to make clear 20 here, but before the SAND2023 report were produced, 21 were the SOARCA UAs which were, essentially, 22 extensions of the original two SOARCA documents to 23 look at uncertainty analysis where they investigated 24 the parametric uncertainties associated with severe 25 accidents to sort of explore that uncertainty space NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
17 1 that we see.
2 The other major point here is the HBU, 3 HALEU, and ATF severe accident PIRT that was mentioned 4 earlier.
5 This body of work was very important in 6 terms of influencing both this work and the ATF source 7 terms which are ongoing right now.
8 Next slide, please? Okay, severe accident 9 modeling advancements. So, I mentioned that a number 10 of advancements have occurred in the time since 11 SAND2011 as well as the prior studies.
12 Two major modeling advancements that we 13 want to highlight today are the heterogeneous 14 integrated reactor core modeling.
15 What mean is that we have discretized the 16 core.
17 So, what this does in discretizing the 18 core to multiple nodes instead of a single node, we 19 actually promote a progressive and extended core 20 degradation period.
21 And this actually sort of has shifted the 22 way we look at accidents. We no longer a distinct gap 23 release phase because we've got different stages of 24 degradation in the various nodes.
25 We have prolonged core damage progression NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
18 1 because we have more efficient transfer of heat out of 2 the core as its degrading.
3 And this all translates to longer times 4 for lower head failure.
5 The next key point in terms of modeling 6 advancements that I want to touch on today is the 7 prevalence of low pressure scenarios.
8 So, basically, this was a finding of the 9 SOARCA analysis that, during the early in-vessel 10 phase, we will reach a larger proportion of low 11 pressure scenarios through either thermally induced 12 safety release valve seizure of hot leg creep rupture 13 for pressurized water reactors.
14 So, this is, basically, both the BWRs and 15 PWRs are basically more likely to go through a low 16 pressure scenario then reach lower head failure before 17 the depressurize.
18 MEMBER REMPE: I have a couple of 19 questions.
20 MR. ALBRIGHT: Yes?
21 MEMBER REMPE: Don't you think that there 22 may be some additional insights that are needed -- or 23 expected to come as we go -- we learn more from 24 Fukushima and other tests?
25 For example, maybe we should learn a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
19 1 little more about vessel failure as we look at the 2 fuel assemblies and control rods that are ex-vessel 3 and the photos coming from Fukushima?
4 And then, the SRVs, the SRV failure, I 5 don't think that's been accommodated yet.
6 And so, maybe you might want to make some 7 comments about that, even though you improved MELCOR, 8 there might be more that's coming.
9 MR. ALBRIGHT: Yes, yes, I think those are 10 very good comments, very good points.
11 I definitely think that, you know, we're 12 on the march towards progress here. And there's 13 always improvements to be made and sort of further 14 refinements.
15 And I think the way I would contextualize 16 this body of work was that it was performed according 17 to the current state of practice.
18 And as that data becomes more available, 19 we absolutely would be interested in taking a look at 20 those and incorporating them into these types of 21 analyses. Thank you.
22 Next slide? So, just a quick overview for 23 the impact of early depressurization on these two 24 different reactor types.
25 We have two sort of concept nodalizations NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
20 1 of a BWR Mark I containment structure on the left, 2 with the reactor vessel in the center there.
3 And then, on the right, we have a 4 pressurized water reactor and a large dry containment.
5 What I want to draw your eyes to are the 6 red Xs in each of these cases.
7 This is where the early loss of the 8 pressure boundary occurs in each of these reactor 9 types.
10 And basically, the point here is that, 11 once we open up this flow path, once this break 12 happens, or in the case of the valve seizure occurs, 13 this is a direct release pathway for radionuclides to 14 transfer directly into containment during early in-15 vessel degradation.
16 So, this is a key point for why we're 17 seeing the numbers we are today.
18 Next slide, please? Thank you. So, the 19 next point we want to talk about here are the severe 20 accident data sets that have sort of developed.
21 In recent years, we have sort of 22 highlighted a few of the data sets here. These --
23 this is not a comprehensive list.
24 But the idea is that, as time has gone on, 25 particularly since NUREG-1465, these severe accident NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
21 1 data sets from different experiments or different 2 events have given us a clearer idea and a clearer 3 picture of core degradation or core damage progression 4 and the radionuclide releases that occur as a result.
5 The ones we call out here for the 6 experiments are both Phebus and VERCORS.
7 These two experiments, both sort of 8 demonstrated early fuel failure.
9 What I mean by early fuel failure is 10 failure at lower temperatures than the sort of 11 constituent materials would suggest.
12 And then, the Phebus experiment is where 13 we saw the hypothesized cesium molybdate being the 14 dominant chemical form of cesium.
15 And then, for cores, we saw -- we actually 16 used that MELCOR as a validation basis for a high 17 burnup fission product release rates model.
18 The next data set here is the Fukushima 19 Daiichi accident.
20 And right now, as Joy said, this is 21 ongoing work. We're still learning things about this 22 accident as they sort of continue with the 23 decommissioning process.
24 But one of the key findings that actually 25 came from one of our peer reviewers with their innate NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
22 1 knowledge or intimate knowledge of what's going on 2 over there, was that they're actually seeing 3 confirmation that cesium molybdate is the dominant 4 chemical form of cesium.
5 So, next slide, please?
6 DR. SCHULTZ: Lucas, the other elements 7 that Joy mentioned with regard to Fukushima, they were 8 not brought forward in the PIRT evaluation that led to 9 the work that you've done?
10 MR. ALBRIGHT: No, the PIRT work that sort 11 of fed into this work were phenomenological 12 differences between high burnup fuels and conventional 13 fuels.
14 So, it wasn't necessarily looking at plant 15 wide behavior so much as the fuel behaviors.
16 DR. SCHULTZ: Thank you.
17 MR. ALBRIGHT: Thank you.
18 So, the next background slide here is 19 severe accident knowledge advancements.
20 These are some of the high level key 21 insights over the past decade of development, two 22 decades, really.
23 And the first one we see here is the 24 chemical form of iodine.
25 The treatment of iodine in our severe NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
23 1 accident codes is different than back in 1995 when we 2 were using STCP to develop the NUREG-1465 study.
3 They had assumed 95 percent of iodine in 4 the form of cesium iodide.
5 We, today, according to the current best 6 practice -- published best practices in MELCOR, assume 7 all iodine is bound in cesium iodide.
8 We still assume that 5 percent of that 9 iodine inventory is present in the gap which is 10 consistent between the two studies.
11 The next point is the chemical form of 12 cesium.
13 This is also a new practice relative to 14 NUREG-1465. In NUREG-1465, the assumption was 15 predominant presence of volatile cesium hydroxide.
16 Today's practice is to assume, 17 essentially, 5 percent of the total cesium inventory 18 is in the gap. And that's made up of both cesium 19 iodide and cesium hydroxide based on the available 20 mass inventory for cesium iodide.
21 The next change since NUREG-1465 is the 22 predominance of cesium molybdate, which I talked about 23 in the last slide being the dominant chemical species 24 and being confirmed in the current Fukushima efforts.
25 The next slide -- the next point here, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
24 1 excuse me, are the molybdenum releases.
2 We found that, since NUREG-1465, 3 especially in the presence of the cesium molybdate, 4 that molybdenum releases are much larger than the 5 other metallic products such as ruthenium and 6 palladium.
7 And this is sort of reflected in the new 8 way that we break out our chemical classes which we'll 9 see later in this presentation.
10 Yes?
11 MEMBER REMPE: Since you pointed out the 12 peer review, which, by the way, I know you're going to 13 talk about it later, and I thought it was admirable 14 that you not only got their comments and addressed it, 15 you went back to them to see how they did it.
16 But DDR and Louis, the comments they made, 17 it seems like that they mentioned some of the more 18 recent data from OECD projects than you have said.
19 It was state of the practice. We did what 20 we could.
21 Is there a path forward that's clear that 22 NRC will be updating the code MELCOR to take into 23 consideration some of the changes that they've 24 recommended?
25 MR. ALBRIGHT: I think that's something NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
25 1 that Shawn can answer.
2 MR. CAMPBELL: Yes, the answer is yes, 3 always.
4 I mean, we're actively involved in all of 5 these international programs. Right? And that's 6 always our purpose in being involved in these programs 7 is try to wait until they are mature. They've been --
8 and the testing is complete.
9 And then, yes, to try and go on and 10 incorporate that.
11 So, we're always seeking to make MELCOR 12 align with this best practice, right, and to align 13 with what's coming out of the research programs.
14 So, yes, for sure.
15 MEMBER REMPE: Then, is going to lead to 16 another question. I was going to wait until later.
17 But after Fukushima happened, there was 18 the benchmark, right, between MAAP and MELCOR.
19 And if you had to do that, even with the 20 changes you've made now, plus the future changes, 21 would you come up to some point where things were 22 still progressing along the same path or are you going 23 to have some divergence?
24 And, again, hopefully, we don't have 25 another accident and we're trying to figure out what's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
26 1 going on, but if it were, you would understand, well, 2 okay, MAAP hasn't incorporated this data or they did 3 incorporate something as we understand the 4 differences?
5 MR. CAMPBELL: I'd say that's a research 6 project in and of itself.
7 But yes, I mean, we are -- especially, 8 meaning Dave, has a wonderful experience in MAAP as 9 well and understands the ins and outs of that code.
10 And so, yes, we do actively try to 11 understand what's involved in MELCOR on that.
12 So, are you saying that, would we be able 13 to understand the differences between the two codes?
14 Is --
15 MEMBER REMPE: If you had to do something 16 again, like the benchmark where you had to do a 17 comparison between MAAP and MELCOR, and understand why 18 there's some differences, which differences would you 19 expect and to be able to say, okay, yes, that's 20 because such and such a model was or wasn't 21 incorporated in our two different codes?
22 Because I know MAAP's going at a different 23 rate on how they incorporate some Fukushima insights.
24 And I'm just wondering how that that 25 comparison would go if you had to do it?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
27 1 Which is a bit off the topic here, I know, 2 but I'm interested and I'm curious.
3 MR. ESMAILI: So, Dave, do you want to --
4 sorry, Hossein Esmaili.
5 So, Dave, you've been involved in that 6 benchmarking of MELCOR versus MAAP.
7 And so, what do you think was the upshot 8 of that and where do we go from there?
9 MR. LUXAT: So, a fair amount of that.
10 There are some differences that emerge in 11 terms of how the codes treat in-vessel degradation, 12 particularly in core degradation.
13 And utilize, particularly for the BWR, 14 paths for relocation downward.
15 So, MELCOR tends to have a propensity to 16 use the bypass more effectively to allow to relocate 17 down.
18 MAAP, at times, has tended to form crusts 19 that can hold up debris above it, so to speak, and 20 promotes more of a crucible like geometry.
21 I can't speak to more recent updates 22 necessarily, but some of the indications from 23 Fukushima did tend to highlight the potential for a, 24 shall we call it, a more incoherent degradation that 25 is the downward relocation of debris that we typically NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
28 1 see in MELCOR and a more segmented relocation of 2 debris in to the lower plenum as a result.
3 And I think we have generally been moving 4 down the direction of making better use across all the 5 codes of a bypass to promote downward relocation, and 6 if you will, a more incoherent release of debris into 7 lower plenum for particularly BWRs.
8 The challenge is still one where, how do 9 I put this, we still -- when it comes to a PWR, when 10 it comes to events with say less water addition where 11 that the strength of those crusts are not as clear, 12 there are still uncertainties we just don't have the 13 reactor scale data to understand if these sort of 14 lower crusts will form, be strong enough to hold up a 15 crucible like geometry like we saw at TMI-2.
16 If you recall, the TMI-2 was a very, very 17 different type of event in the perspective of water 18 injection over to Fukushima.
19 But generally, we -- with both codes, 20 we've been moving in the direction, to wrap this up, 21 of, you know, understanding and promoting the idea of 22 downward relocation towards the core plate as a 23 potentially more dominant relocation mechanism.
24 MEMBER REMPE: So, where I'm going with 25 what I'm asking is, because of, again, there's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
29 1 international comparisons which MELCOR is involved 2 with as well as other countries codes, is that, as you 3 improve the code which has been recommended, and 4 you're saying your going to keep doing it, keep in 5 mind the other codes in industry.
6 Because it helped that we kind of keep, as 7 well as we can understand about in-vessel early 8 relocation.
9 And later on, there's a lot of 10 uncertainties.
11 And it just seems like different models 12 are being put in and it just doesn't mean that you can 13 influence what industry's doing, but to just kind of 14 keep track is what I was kind of going with the 15 question.
16 MR. LUXAT: There's overall a general 17 sense that when it comes to source terms, there are 18 going to be difference, obviously, between the codes.
19 But a lot of the dominant releases occur 20 for both codes during the same period of time where 21 you've got largely, if you will, debris that has 22 surface area through which fission products can be 23 released.
24 It's generally later in the in-vessel 25 accident progression where you start to either build NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
30 1 up relative to realized debris or relatively, shall we 2 say, multi-cool like debris.
3 But I would say from at least to the 4 perspective of the early in-vessel phase, when it 5 comes to source terms, the codes tend to come close to 6 each other, generally.
7 And when we look at the Fukushima data via 8 MAAP or MELCOR, the general releases that both codes 9 are predicting or estimating for Fukushima are 10 generally consistent overall.
11 MEMBER REMPE: The chemistry doesn't 12 change, but the iodine is in all of those things?
13 MR. LUXAT: Absolutely.
14 MEMBER REMPE: Okay.
15 MR. LUXAT: Absolutely.
16 MEMBER REMPE: Thank you.
17 MEMBER ROBERTS: Just a question of the 18 detail.
19 Just looking at your slides with knowledge 20 investment and then the statements that have the word 21 assumes, they all reflect, say, a chemical reaction.
22 It is assumed, in this case, right, there 23 is not, in other models, that are incorporated in 24 these reactions and reaction rates and energy ins and 25 outs?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
31 1 MR. LUXAT: The speciation in particular, 2 yes, that is an input, shall we say, that is assumed.
3 It's not based on, shall we say, the execution during 4 the simulation, but we'll talk about the next, say, a 5 thermochemical equilibrium calculation.
6 The speciation is, if you will, to use 7 some of our lingo, assumed as frozen catastrivia 8 (phonetic), if you will.
9 MEMBER MARTIN: And you don't feel like, 10 you know, neglecting that has this negative impact on 11 progression in one of the other?
12 MR. LUXAT: For the conditions that we 13 see, generally, we're in a steam environment, so to 14 speak. Obviously, we would make different 15 considerations if this was an accident with a spent 16 fuel pool with a different atmosphere reducing versus 17 oxidizing conditions, so to speak.
18 And that will influence it. But for this, 19 and this was sort of a point of discussion that we had 20 with the peer review, the frozen chemistry or the 21 phase of the accident in a steam rich, if you will, 22 reactor vessel, during the phase of release is a 23 reasonable approximation, particularly when it comes 24 to the cesium and iodine releases.
25 MEMBER ROBERTS: I'm wondering if you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
32 1 could speak to the assumption of a 100 percent of the 2 iodine and cesium iodide?
3 It seems like ignoring the gaseous iodine 4 could be significantly nonconservative, depending on 5 the analysis.
6 And I was just wondering, you know, why 7 that assumption is justified, given that at least the 8 report says it's highly uncertain as to what the 9 behavior of gaseous iodine is?
10 MR. ALBRIGHT: Yes, that's a good point.
11 So, the SOARCA uncertainty analyses have 12 actually looked at the impact of some of these 13 speciation uncertainties.
14 And we sort of point to those reports as 15 being like the place to find that information and 16 being outside of the scope because it's really 17 investigating what kinds of practices one could 18 perform rather than the current state of practice, if 19 that makes sense.
20 In those analyses, they actually had a, 21 essentially, the iodine -- elemental iodine mass was 22 a function of burnup.
23 And this was informed by experiments.
24 And if I remember correctly, the iodine --
25 the percent of the iodine mass that ended up in this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
33 1 elemental form was single digit percents, so, one, 2 two, three.
3 And these percentages are very low and 4 well within the uncertainties that are actually being 5 released to containment based on the study -- this 6 current study.
7 So, while the uncertainty is there, I 8 think we actually have, in this analysis, covered it 9 with our uncertainty bands on these results, if that 10 makes sense.
11 MEMBER ROBERTS: I don't suppose it would 12 depend on how much credit you're getting in 13 containment.
14 MR. ALBRIGHT: So, the analysis that we're 15 presenting today actually doesn't account for 16 scrubbing.
17 We report the total radionuclide inventory 18 reaching containment in the tables that we'll be 19 talking about today.
20 So, the scrubbing is something that's 21 considered traditionally in downstream codes, if at 22 all. And that's something that, actually, in our 23 follow up presentation we'll be talking about it in 24 much more detail.
25 MEMBER ROBERTS: Okay, thank you.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
34 1 So, it might be more of a question for 2 Elijah, then.
3 It seems like once you start to look at 4 containment, decontamination phenomena, it makes a big 5 difference to, in my past experience, of what the form 6 is as how much credit you take for decontamination.
7 MR. DICKSON: Yes, that's right.
8 And we do the transport portion of these 9 analyses utilizing that source, we're taking those 10 credits based off of chemical speciation in the 11 regulatory guidance space.
12 So, you would see those models in effect 13 in Appendix A of Reg Guide 1.183.
14 MEMBER ROBERTS: All right.
15 So, it seems like a couple percent, you 16 know, gaseous iodine can make a big difference once 17 you come into containment response phase.
18 So, if you get into that later, that's 19 great, but I was just wondering how that played out 20 because I'd have to have .15 percent that's currently 21 in the Reg Guide, I've seen that dominate, depending 22 on what the analysis is because it's not scrubbed.
23 MR. ALBRIGHT: I think the last point, I 24 don't think we've covered this one yet, are the 25 tellurium releases.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
35 1 Based on current practices, we see much 2 more extensive tellurium releases during the early in-3 vessel phase.
4 This was a finding from the Phebus 5 experiments that basically we've got a fission 6 transportation of the tellurium because it's not being 7 bound up with the zirconium as was previously assumed.
8 Next slide, please?
9 CHAIR PETTI: Do they know why?
10 MR. ALBRIGHT: I think this is an area of 11 investigation still, is my understanding.
12 CHAIR PETTI: Because, in the day, when we 13 measured it, we were pretty convinced there was 14 tellurium in the cloud. And that's something that 15 doesn't move around that easily. And yet --
16 MR. SALAY: This is Mike Salay. Yes, the 17 -- in both of our cores in Phebus, I think they both 18 observed that once the cloud was oxidized, the --
19 CHAIR PETTI: Okay, so these were -- yes.
20 We predicted that if you got to full 21 oxidation, these are highly oxidic melts as opposed to 22 more metallic melts which were, probably some of the 23 earlier testing done in the U.S. was much more.
24 Yes, you relocated a lot of metal because 25 you've got a liquefaction which is, you know, taking NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
36 1 the highly metallic.
2 MR. SALAY: Yes, they tried to stop the 3 experiment at the Phebus FP, but that's practically 4 like on the level of a PWR when they start getting 5 significant melting.
6 So, they were getting the tellurium 7 releases before and then, the core ones are just the 8 fuel pellets in the furnace.
9 And also, they stop before it melted.
10 MEMBER REMPE: So, Mike, move your mic 11 closer to you.
12 People like me, I have to shout across the 13 room, but it'll help.
14 MR. ALBRIGHT: Next slide, please? Thank 15 you.
16 This is a quick overview of the findings 17 of that HBU, HALEU, ATF, PIRT.
18 So, this was an investigation into the 19 severe accident behavior for these different fuel 20 types.
21 And the findings -- major findings from 22 this report were that there were no significant 23 differences between HUB and HBU HALEU fuels. So, 24 those are going to perform more or less similarly 25 under severe accident conditions.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
37 1 There are some different thermophysical 2 properties that could be expects, thermal conductivity 3 of the fuel being one of them.
4 Fuel fragmentation and centering is 5 expected to impact core degradation and the sort of 6 rate at which that occurs.
7 The next point here being that the fission 8 part of chemistry may change followed by a possible 9 cladding embrittlement.
10 And then, -- and the cladding 11 embrittlement, just for some context here, was more 12 related to the impacts of a reflooding scenario than 13 necessarily leading to different fuel failure 14 behaviors.
15 And the last point here being the 16 potential for recriticality if there were to reflood 17 without unborated water.
18 So, that sort of covers the high level 19 findings of that HBU, HALEU, PIRT.
20 Some of these findings made them -- made 21 their way into our report through sensitivity analyses 22 or sensitivity calculations that we'll be covering 23 later in this presentation.
24 Next slide, please? All right, key 25 findings. So, this is going to be a quick overview of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
38 1 the main findings of our analysis. In our report, we 2 sort of have three key findings that we claim at the 3 beginning of this report.
4 The first being that the increased burnup 5 and enrichment is not strongly impacting our in 6 containment source terms and that the most significant 7 variation that we see in source terms is due to 8 sequence variations.
9 The next finding is that the larger early 10 releases to containment are the result of early 11 pressure boundary failures, primary pressure boundary 12 failures.
13 In our analysis, we had a higher 14 predominance of those low pressure accident sequences 15 that were sort of being mechanistically predicated by 16 the MELCOR code. And this is in contrast to the 17 NUREG-1465 document that had high pressure sequences.
18 The third finding here is that releases to 19 containment are significantly reduced if you can keep 20 that primary pressure boundary intact.
21 So, if we prevent that low pressure 22 scenario from evolving, we're going to have smaller in 23 containment source terms.
24 And this is sort of related to some of 25 those findings from the SOARCA analysis, and we'll NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
39 1 touch on that again in a couple of slides.
2 MEMBER MARCH-LEUBA: This is Jose March-3 Leuba. So, the key findings -- sorry, I've got 4 something in my throat -- key findings two or three 5 are not underlying. I mean, I don't need MELCOR to 6 tell me that.
7 But key finding one is counterintuitive, 8 right? If you have high burnup, you have a higher 9 amount of inventory inside the fuel.
10 Could you -- I assume you're going to 11 expand on this, can you give me a high level 12 explanation of why increased burnup doesn't affect the 13 source term?
14 MR. ALBRIGHT: Yes. So, when we talk 15 about the in containment source term, we're not 16 looking at the magnitude of mass that's being released 17 to the containment. We're looking at the release 18 fraction.
19 So, we'll go into that in more details in 20 the next few slides. But I think that's the clearest 21 cut way at this stage of the presentation to clarify 22 that.
23 MEMBER REMPE: So, that's actually a 24 comment I was going to have. And I hate to nitpick on 25 words, but I think that you -- what you said is what NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
40 1 you mean there is release fractions, not magnitude.
2 And you might, I guess, this is a final 3 report, but you might think about not having that 4 wording next time.
5 MEMBER MARCH-LEUBA: Yes, being a little 6 facetious here, if this is the release fraction, it's 7 another -- I mean, it's a no, never mind.
8 If those are your three key findings, you 9 should be proud of yourselves. You didn't find any 10 surprises, I don't know. Okay, keep going.
11 MR. ALBRIGHT: Thank you.
12 MEMBER ROBERTS: Because I don't want to 13 nitpick, either, but I will.
14 The key finding three, releases to 15 containment is going to be reduced. Isn't that true 16 only for the early in-vessel phase? Because when I 17 read the report, it seems like the total release to 18 containment is probably the same in any of these 19 scenarios.
20 MR. ALBRIGHT: Yes, yes, yes, that's a 21 good qualifier for this statement is that we are 22 focusing on the gap release and early in-vessel phases 23 and we'll actually touch on the clarification of the 24 phases later in this presentation.
25 MEMBER ROBERTS: Okay, thank you.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
41 1 MR. LUXAT: I'll just very quickly say, 2 when it comes to severe accident progression and 3 burnup, what's primarily driving us the containment 4 and what the heat level is.
5 For the fractional releases, you know, 6 they're driven by decay heat. And what we saw from 7 the Oak Ridge work on the decay heat is that, for the 8 early phase, or essentially, the early times post 9 accident, early cooling times, there isn't a 10 significant difference in decay heat.
11 MR. CAMPBELL: Well, we can go to the next 12 slide and you can see exactly that.
13 MR. LUXAT: Yes, thanks, Shawn.
14 MR. CAMPBELL: Spoiler alert. Thank you, 15 Dave.
16 Yes, so, this next slide sort of 17 highlights the SCALE analyses that were used as the 18 initial conditions for our reactor core inventories.
19 And what we see on the left are the decay 20 heats in terms of relative percent to the reference 21 core here.
22 And I've highlighted in the black box 23 there the time region of interest. And we see that 24 we're always within 5 percent. In fact, less than 5 25 percent of the same decay heat during the reference NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
42 1 period of interest for our analyses.
2 And so, the main point here is that our 3 burnup and enrichment aren't changing the decay heat, 4 which is one of the major drivers of the accident 5 progression.
6 On the second half of this slide, on the 7 right side, we actually are looking at the 8 bootstrapped release fractions for the different 9 nuclide classes of each of the core types, being 60 10 and 80 gigawatt low enriched, and 60 and 80 gigawatt 11 high enriched, or HALEU.
12 And what this slide is showing, or what 13 this figure is showing us, is that the differences in 14 the source term across these different cores are 15 actually very small so that the increased burnup and 16 enrichment does not strongly impact the in containment 17 source term.
18 Now, this is taken -- or this statement is 19 really much more clearly sort of re-emphasized or 20 reinforced by the next slide, if we can go there, 21 where we look at the actual source terms based on the 22 sequences.
23 Next slide, please? So, on the left side 24 here, what we see are the same source terms for BWRs 25 and PWRs looking at the different accident sequences.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
43 1 And what we see are that we're actually 2 seeing differences in those bar charts now.
3 So, the main point here is that accident 4 progression and the in containment source terms are 5 actually being -- the numbers we're getting are driven 6 by the accident sequences themselves, not necessarily 7 the reactor cores or the different burnups that we can 8 have.
9 On the right half of this slide, we then 10 look at the impact of that early depressurization of 11 the primary pressure boundary.
12 So, the purple lines in both of these 13 plots are going to be our reference cases where the 14 reference -- or where the hot leg creep rupture is 15 enabled.
16 And the orange dashed lines are going to 17 be the cases where we disabled hot leg creep rupture.
18 And what we see is that that prevention of 19 early depressurization of the primary pressure 20 boundary or early loss of the primary pressure 21 boundary, during these critical early phases of the 22 accident are actually decreasing the source term, the 23 in containment source term significantly.
24 Next slide, please? All right, so now, we 25 get into sort of the high level tables or the main NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
44 1 sort of cuts through our source term tables that we've 2 developed here.
3 The first point that I want to make is we 4 talked a little bit about the late in-vessel and the 5 ex-vessel phases, you know, after lower head failure.
6 And those are reported in this analysis in 7 keeping with the current state of practice and giving 8 us the ability to compare to previous source terms 9 like SAND2011.
10 But the NRC has actually determined that 11 the design basis source terms won't include these two 12 phases.
13 So, we're going to focus in our 14 presentation today on the gap and early in-vessel 15 phase values.
16 Now, the first point that I want to make 17 here with these yellow highlighted boxes are that we 18 have significantly longer in-vessel phase durations 19 due to that progressive core degradation that I 20 mentioned earlier.
21 This is a MELCOR advancement in the way 22 we've modeled these accidents.
23 Next slide, please? Now, the next point 24 here is looking --
25 MEMBER ROBERTS: Quick question.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
45 1 Acknowledging what Kim said at the outset, I'm just 2 curious, maybe for logic, are you reconsidering that 3 1994 cutoff?
4 MR. DICKSON: That's a Commission policy.
5 MEMBER ROBERTS: Oh yes, I recognize that, 6 but it seems to me that going from 1.5 to 6.7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> 7 seems like a pretty major change in terms of what the 8 overall outlook is of the progression of the plant 9 transient and that.
10 I was wondering if that's something you're 11 looking at?
12 Because the principle from that '94 SECY 13 seemed to be, A, the release is about the same as the 14 TID; and B, that's about the time it take before the 15 operators could do anything where the casualty would 16 become so bad that they can't do anything about it.
17 I think those are the two main preventions 18 that were in there.
19 And again, it just seems like we go from 20 times like 1.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> to 6.7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br />, that -- and this 21 range now is roughly double in TID.
22 And it seems like the basis for that 1994 23 judgment might be something worth revisiting.
24 And I was just wondered if you're thinking 25 about that?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
46 1 MR. DICKSON: Yes, we can think about.
2 MEMBER MARTIN: I don't know what the 3 right answer is, because clearly, it's a 30-year-old 4 judgment that's been out there for a while.
5 But it just seems like it's something 6 worth thinking about.
7 Thank you.
8 MR. ALBRIGHT: Okay, this current slide 9 here, what we're looking at are the highlighted gap 10 releases or the gap release phase values.
11 And what we're seeing is that the enhanced 12 reactor coolant system modeling that we have in MELCOR 13 today allows for the progressive releases to 14 containment.
15 And this is where start to see that the 16 gap release phase, essentially, now we're seeing those 17 fission parts actually transporting through the 18 reactor coolant system out to containment.
19 And this is, again, related to that gap 20 release phase sort of no longer being distinct from 21 the early in-vessel phase.
22 Next slide, please? Finally, I want to 23 highlight the larger release magnitudes, release 24 fraction magnitudes that we're seeing in the current 25 study. We've highlighted them here.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
47 1 Now, primary reasons that we've already 2 discussed for these are the progressive core 3 degradation that we're seeing in MELCOR and the longer 4 durations to lower head failure.
5 So, essentially, we've got our fuel and 6 our debris are being held at longer time periods at 7 high temperatures during these early in-vessel phases 8 in the current MELCOR calculations.
9 And that's driving basically larger 10 releases that are then captured in containment based 11 on this early loss of that primary pressure valve 12 we're seeing in our simulations today.
13 So, that sort of gives a quick overview of 14 some of the major differences between the 2023 source 15 terms and the NUREG-1465 in containment source terms.
16 Next slide? Oh?
17 MEMBER BIER: Sorry. I have a very high 18 level question but on this kind of modeling at all.
19 When the new analysis was done, was it 20 done kind of from, I don't want to say post principles 21 like basic physics, but, you know, modeling the whole 22 scenario from scratch?
23 Or was it done by kind of looking at the 24 previous analysis and figuring out where you can take 25 advantage of improved fuel characteristics?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
48 1 MR. ALBRIGHT: Yes, that's a great 2 question. So, the current analysis, what we used as 3 our starting point were the actual reactor models and 4 practices from the SAND2011 report.
5 And the first thing that we did was bring 6 those inputs up to modern MELCOR best practices.
7 And then, we actually used the same 8 scenario models, except for where best practices have 9 evolved, so these early pressure failures -- or early 10 primary pressure boundary failures.
11 And we maintained as much consistency 12 there as we could with the 2011 values so that we 13 could actually compare apples to apples.
14 MEMBER BIER: Okay. I guess the reason 15 I'm asking is, in my area of PRA, I know there's 16 always a tendency to look for like where you can 17 sharpen your pencil to get better numbers and not, you 18 know, is there some unexpected phenomenon that gives 19 you worse numbers.
20 But it sounds like you tried to do it 21 pretty even handed.
22 CHAIR PETTI: So, but like you talked 23 about how you modeled -- how you incorporated line 24 from Phebus and like.
25 So, the release rates are fundamentally NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
49 1 different, right, in these calculations and in the 2 1465?
3 MR. ALBRIGHT: Yes, so, in terms of 4 release rates, we actually use a different release 5 rate correlation for high burnup fuels.
6 And that is incorporated and based on the 7 validation of the RT-6 VERCORS experiment using the 8 MELCOR code.
9 CHAIR PETTI: Yes, so, it's a combination 10 of better accident progression and release of what's 11 --
12 MR. ALBRIGHT: Absolutely, yes, yes, yes.
13 What we've tried to do is maintain 14 consistency where we could and advance the previous 15 practices to modern practices where anything has 16 evolved.
17 CHAIR PETTI: Kind of just scary that 18 these numbers are heading back towards the TID source 19 terms. We spent billions of dollars on this.
20 MEMBER REMPE: That's why I was asking 21 about how MAAP would compare if --
22 CHAIR PETTI: Right.
23 MEMBER REMPE: -- they have a different 24 failure time. It's earlier, they could get back down 25 again.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
50 1 CHAIR PETTI: Yes, that's another question 2 about how you decided what sequences to look at in 3 terms of establishing this amalgamated source terms, 4 if you will, given there are thousands of severe 5 accident scenarios, right, that you could have?
6 MR. ALBRIGHT: Sure.
7 And specifically in the selection of 8 scenarios, you know, we fall back on that NUREG-1560 9 and I think we talk about it later in a later slide 10 where we talk about that.
11 What we're trying -- that's where we try 12 to stay with consistency. Right? We tried to stay 13 consistent with the 2011 practice of scenario 14 selection. And we have a NUREG basis for those 15 choices.
16 MEMBER MARTIN: The numbers that get 17 reported here, they reflect, of course, the 18 calculation will talk about -- mention, of course, in 19 certain analysis.
20 And it's been a cold month since I've 21 looked at the 2023 Sandia report.
22 But I believe uncertainties were, as I 23 recall in that document, uncertainties were discussed, 24 addressed.
25 When we look at numbers like this, I mean, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
51 1 these are kind of mean. Whether they -- that is the 2 -- it's implied or stated, correct?
3 MR. ALBRIGHT: Yes.
4 So, we can -- we'll go into the more 5 detail on that statistical process that was 6 implemented in this study.
7 But these numbers are the median of the 8 distributions that we selected. And that's based on 9 the accepted practice from the -- I believe it was the 10 peer review for the 2011, actually, that sort of was 11 the first assertion I'm aware of that said that that 12 was the most representative so that we weren't 13 unequally weighting certain scenarios.
14 MEMBER MARTIN: Will you be discussing 15 what all went into the uncertainty analysis part?
16 MR. ALBRIGHT: Yes, we'll have a number of 17 slides, I'm forgetting off the of my head right now, 18 but I think it's a handful of slides on the process 19 that we used to arrive at the distributions that 20 informed the numbers that we present.
21 MEMBER MARTIN: And maybe you can say it 22 ahead of time, I mean, is the biggest uncertainty 23 really the event itself, to Dave's point, that --
24 MR. ALBRIGHT: Yes, yes, the largest 25 uncertainty that we have in this study is the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
52 1 sequence.
2 The variation in sequences is larger than 3 any other variation that we observed.
4 MR. LUXAT: Yes.
5 So, it's sort of common, if you look at a 6 Level 2 PRA. Your end states and your releases are 7 typically a distribution, but it's really going from 8 one branch to the next. One end state to the next 9 that really causes the big changes in release.
10 And any phenomenological variation about 11 that particular sort of branch or sequence is 12 typically gives you a Gaussian, gives you 13 distribution, but it isn't enough to necessarily push 14 you from one end state release category into a 15 completely different release category for another 16 state.
17 And so, it's -- what we found is typical 18 of what we always find in Level 2 PRAs.
19 MEMBER MARTIN: If you played around that 20 space, I mean, we talk about cusp events in different 21 context a bit earlier today, I mean, if you had a 22 survey to look at, you know, event outcomes and with 23 an eye towards, you know, those events that you would 24 otherwise classify as like cusp events, I mean, do you 25 think we've crossed a line, a cusp of such?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
53 1 Of course, a severe accident by nature, 2 we've crossed the line.
3 But when it comes to releases, are there 4 tiers where you see kind of clustering of, you know, 5 these events, you know, kind of land in this cluster 6 and the more and more severe as you look at worse and 7 worse conditions?
8 MR. LUXAT: I think for this particular 9 set of scenarios, we're dealing with unmitigated 10 scenarios.
11 And so, really, the big -- the main 12 changes are, A, the initiating event, is it a LOCA or 13 is it an SBO?
14 And also, the other main issue is the 15 integrity of the RCS or the nuclear steam's ply system 16 pressure valve.
17 And it's those two features that typically 18 give you, if you will, the underlying bifurcations 19 that push you in a direction of one cluster versus 20 another.
21 And obviously, if we were to go down 22 further and further into event trees with mitigating 23 actions, you would see other types of bifurcations.
24 But within the scope of essentially 25 unmitigated events early in-vessel source terms, it's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
54 1 really those two characteristics that are principally 2 influencing the nature of the source term.
3 MEMBER MARTIN: Thanks.
4 MR. ALBRIGHT: Okay, I think we've covered 5 this slide, so next slide, please?
6 This next set of -- couple of slides is 7 going to be related to the release rates.
8 And we just wanted to sort of highlight 9 here that, with the longer phase durations, the 10 release rates, when we assume uniform release across 11 the phase duration are decreasing quite significantly 12 relative to NUREG-1465 for many key radionuclides.
13 Next slide, please? So, these are release 14 fractions per hour. So, this is a very simple 15 calculation. Take the total release fraction, divide 16 it by the phase duration.
17 And again, what we're seeing is that, in 18 general, these are much smaller.
19 We see -- next slide, please? Thank you.
20 We see that, for the tellurium group, 21 these numbers are increasing. And the main reason 22 here is because we didn't used to assume what 23 tellurium was going to transport efficiently to 24 containment.
25 So, this is that advancement in our NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
55 1 understanding of severe accidents that's resulting in 2 this change or larger release rate tellurium.
3 Next slide, please? Thank you.
4 Okay, so -- yes?
5 CHAIR PETTI: Just a question.
6 So, there's no reaction with any of the 7 still surfaces in the primary system highly reactive 8 for metal?
9 MR. ALBRIGHT: So, our tellurium release 10 rates are being informed based on the validation 11 matrix for radionuclide transport.
12 So, this includes, in particular, that 13 VERCORS RT-6 experiment.
14 And what we do is we're actually, how do 15 I put this, the tellurium release rate is not going to 16 look at any downstream chemistry. Right?
17 We have frozen chemistry in MELCOR. So, 18 the only way for the tellurium to be release is direct 19 release. Right?
20 We're not actually looking at any 21 chemistry on its way out of the fuel, if that makes 22 sense. We've assumed tellurium is going to transport 23 out of the fuel at this rate and it will have these 24 transport properties.
25 And that goes back to our frozen chemistry NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
56 1 assumption in MELCOR.
2 CHAIR PETTI: But physically, one of the 3 most reactive metals, you would think that it would 4 probably react with steel as it's, you know, coming 5 between the core and the break.
6 MR. SALAY: Mike Salay.
7 It was before my time, in Phebus, they 8 actually -- and if they'd observed a chemical reaction 9 in Phebus, they would have done it.
10 And Phebus had a model steam generator and 11 RCS and containment. So, they would have seen 12 something, seen retention, specific retention there.
13 And if they'd seen it, they would have accounted for 14 it.
15 CHAIR PETTI: So, all these old 16 discussions I can remember with Dana on fission 17 product revaporization, because it's on the surface of 18 the primary system and it's going to self-heat.
19 That's all gone, we don't do that anymore?
20 MR. LUXAT: We do.
21 CHAIR PETTI: Okay, we do it for some 22 fission products but not for others? I'm confused.
23 But Phebus probably didn't show that, did 24 it?
25 MR. LUXAT: We have a set -- we do model NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
57 1 revaporization and we also do the chem absorption on 2 surfaces and tellurium --
3 CHAIR PETTI: So, you do model a tellurium 4 chem absorption?
5 MR. LUXAT: There is tellurium chem 6 absorption and the coefficiency in the validation 7 data.
8 And we do actually have them as a --
9 CHAIR PETTI: Okay.
10 MR. LUXAT: -- show on surfaces.
11 Now, tellurium is probably not -- it's not 12 a very dominant one from what I recall in terms of 13 chem absorption. I think some of the cesium that, 14 obviously, cesium hydroxide is a more dominant one 15 that we typically consider.
16 But we do have chem absorption models and 17 --
18 CHAIR PETTI: But since your position now 19 assuming it's all cesium molybdate instead of cesium 20 hydroxide, there's not a lot of cesium chem 21 absorption.
22 MR. LUXAT: And cesium -- there is some 23 cesium iodide.
24 CHAIR PETTI: Yes, cesium iodide, sure.
25 MR. ALBRIGHT: Okay, I think that covers NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
58 1 this slide.
2 Next slide, please? Going into the deep 3 dive, so just getting everyone, including our members 4 in the virtual audience on the same page, what do we 5 mean by in containment source terms?
6 This is a little late in the presentation, 7 but what we're talking about is the total radioactive 8 inventory in containment.
9 So, what that means is, we combine all of 10 the different sort of phases of radionuclides, 11 airborne, liquid, or anything that's escaped the 12 containment in the case of the later vessel phases.
13 We combine that all into one single value, 14 the in containment source term.
15 What you see here on the figure in the 16 right is some of these different values, deposited 17 airborne escaped. And then, total, which is that top 18 black line at the very top here.
19 For one example case, the halogens. And 20 it just goes to show, you know, the many processes 21 that MELCOR is tracking and modeling that we then have 22 collapsed into the numbers that we're presenting in 23 the tables for this analysis.
24 The reason we do this is because we have 25 downstream codes in the process that are meant to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
59 1 handle these mechanisms.
2 So, we have to actually, you know, post 3 process them out of our MELCOR simulations for those 4 codes to do what they need to do.
5 MR. SALAY: Just to clarify, you don't 6 mean Sandia has downstream codes?
7 MR. ALBRIGHT: Yes, I apologize.
8 Yes, yes, particularly the RADTRAD code 9 that's mentioned on this slide which is done later in 10 the regulatory process for these regulatory source 11 terms.
12 Next slide, please? So, what is an 13 alternative source term? Basically, the concept and 14 requirements of alternative source terms were defined 15 in Reg Guide 1.183.
16 We've sort of boiled down the five 17 criteria here that it has to be based on major 18 accidents involving substantial meltdown of the 19 reactor core.
20 It needs to be represented in terms of 21 quantities, times, rates, and speciation of a fission 22 product release.
23 It needs to be based on a representative 24 set of accident scenarios.
25 It needs to have a defensible technical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
60 1 basis.
2 And it must be peer reviewed.
3 These criteria were all met according to 4 the peer review process that was completed in 2023 and 5 published in like at the same time as the SAND2023 6 source term report.
7 It's listed on the left there for anyone 8 who wants that reference.
9 Next slide, please? So, how do we develop 10 these source terms? We're looking at light water 11 reactors, so we go and we find the accident sequences 12 that are relevant for a BWR or PWR.
13 We develop a radionuclide inventory and 14 decay heat for the particular reactors of interest 15 using the SCALE code package.
16 Then, we perform our accident progression 17 and source term analysis using MELCOR.
18 That's the part that we do, the SCALE code 19 package was completed by another team.
20 And then, finally, we develop the 21 statistically representative source terms based on 22 that data. And that's another part of our job.
23 Next slide, please? So, as I mentioned, 24 we were in this process starting from SAND2011 inputs 25 and trying to maintain consistency so that the two NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
61 1 reports could be sort of compared reasonably.
2 In that SAND2011 report, we -- sorry, 3 excuse me -- in our 2023 report, we focused on 4 extending the SAND2011 source terms to look at higher 5 burnups and HALEU fuel.
6 The main sort of overlap that we 7 maintained between these two studies were the power 8 plants that we modeled, the scenarios -- the accident 9 scenarios that we simulated, the chemical classes 10 represented, and the sort of phase criteria.
11 So, when does a given phase start and when 12 does that given phase end?
13 And finally, the sort of statistical 14 process that was used to develop the final values was 15 maintained across these two studies.
16 Next slide, please? Okay --
17 DR. SCHULTZ: Lucas, while you're --
18 MR. ALBRIGHT: Yes.
19 DR. SCHULTZ: Was there any reason to make 20 changes in those elements of analysis?
21 MR. ALBRIGHT: Any reason to make changes 22 to the methodology?
23 DR. SCHULTZ: As you described them, you 24 maintained those from first analysis to the second.
25 Was there anything pulling on you to make NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
62 1 changes to those?
2 You indicated you had maintained them, so 3 you had comparative.
4 MR. ALBRIGHT: Yes, yes.
5 DR. SCHULTZ: Is there any reason why you 6 would, but I just wondered because you did the 7 analyses, or the team did, and it didn't work.
8 Did anything pull you in a direction to 9 change those fundamentals?
10 MR. ALBRIGHT: So, I think at the outset 11 of this project, the goal was to extend those source 12 terms with the current state of practice.
13 So, we weren't necessarily drawn to evolve 14 any of the practices because we were trying to sort of 15 do things according to the current state of practice, 16 not necessarily develop new practices as part of this 17 process, if that makes sense.
18 I think -- oh, go ahead.
19 DR. SCHULTZ: The PIRT basically validated 20 that? The evaluations that were performed?
21 MR. ALBRIGHT: Yes, the PIRT findings 22 really informed the way we interrogated the 23 uncertainties with high burnup fuels here.
24 And in terms of the methodology that was 25 used, that SAND2011 report was also peer reviewed.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
63 1 And in that peer review itself, that was 2 where sort of that state of practice was accepted and 3 that we tried to sort of maintain for that comparison 4 of apples to apples for this set of data.
5 MR. CAMPBELL: And just to piggyback off 6 of that.
7 And then, confirmed by our own peer 8 review. Right?
9 So then, for the 2023 one of the things 10 that they stated was methodology was acceptable.
11 So, we were jumping off of the basis of 12 the 2011 peer review and then, confirmed by the 2023 13 peer review.
14 MR. ALBRIGHT: Thank you, Shawn.
15 Let's see, okay. So, in this slide, we're 16 sort of giving a summary of some of those details that 17 we just talked about.
18 So, we looked at BWRs and PWRs, looked at 19 four different containment types for these two 20 reactors, a Mark I containment with the Peach Bottom 21 reactor and a Mark III containment with Grand Gulf.
22 For the PWRs, the pressurized water 23 reactors, we looked at an ice condenser containment 24 looking at Sequoyah, and a large dry containment 25 looking at Surry.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
64 1 In terms of accident scenarios that we 2 used in this analysis, we focused on the same that 3 were used in 2011, the small break LOCAs, the large 4 break LOCAs, the short-term station blackouts, and 5 then, two additional scenarios, the long-term station 6 blackout and anticipated transient without scrams for 7 the BWRs.
8 The table that we're looking at here is a 9 summary of the release phase criteria or the accident 10 phase criteria that were used in the analysis. These 11 were refined back in 2011 and used here so that our 12 numbers were comparable.
13 Again, I think the sort of main points 14 here are the gap release phase starts when the RPV 15 water level reaches the top of active fuel and it ends 16 after 5 percent of the total xenon inventory release 17 from the fuel.
18 The early in-vessel phase releases 5 19 percent -- starts when we've released that 5 percent 20 of xenon and it ends at lower head failure.
21 The two other phases that are not 22 considered in these design basis source terms, the ex-23 vessel and late in-vessel both start when that lower 24 head fails.
25 And then, they have their own criteria NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
65 1 based on the cesium releases pertaining to those 2 different bases.
3 Now, the peer review findings in regards 4 to these details were that the ex-vessel and late in-5 vessel phase criteria have limited technical 6 justification.
7 Again, these two phases are not really 8 used for the design basis source terms. And that's 9 sort, I guess, something for our discussion that you 10 brought up earlier.
11 So, I think that covers the details for 12 this slide, if we want to move to --
13 Oh, yes? Sorry, no, go ahead.
14 MEMBER REMPE: I have a question on that, 15 is it unfair to ask you, but I think maybe the staff 16 could comment on it.
17 A couple slides back, you talk about what 18 the AST should be based on Reg Guide 1.183.
19 And you talk about not a single accident 20 scenario.
21 Risk guidance insights can be used, but 22 there's a phrase you didn't -- or a sentence you 23 didn't mention, however, risk insights alone are not 24 an acceptable basis for excluding a particular event.
25 And do you think, I mean, just saying, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
66 1 well, we did what the folks in 2011 did and they did 2 it based on the IPE.
3 Have we met this last point that's still 4 appeared in Reg Guide 1.183 Rev 1? Have you gone back 5 and thought about, did we exclude some things because 6 of risk assessment?
7 I don't quite understand why that sentence 8 is still in Reg Guide 1.183, but have we done that?
9 MR. CAMPBELL: There wasn't even a 10 consideration of trying to look at scenarios. People 11 have mentioned that perhaps we should look at 12 scenarios more, but yes, so --
13 MEMBER REMPE: Containment bypass, for 14 example, well, you've kind of said, well, we don't 15 think we're going to have it because the hot leg fails 16 earlier.
17 But --
18 MR. CAMPBELL: Well, the containment 19 bypass, I think we're -- I think explicitly excluded 20 or from the set of scenarios because the whole purpose 21 is to develop a source term to test your equipment in 22 containment.
23 And so, it literally excluded those.
24 MEMBER REMPE: Yes.
25 MR. CAMPBELL: And so, they tried to get NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
67 1 the -- based on the IDs, as big a fraction of the core 2 damage frequency as possible for both Bs and Ps.
3 MEMBER REMPE: Good.
4 I'm just wondering if maybe either that 5 sentence ought to go away in Rev 2 or maybe we'd 6 better have a reason why we didn't think about things 7 that might have been excluded Because of risk 8 insights.
9 MR. ALBRIGHT: Are you suggesting a 10 scenario that needs to be added?
11 MEMBER REMPE: I just didn't want it -- we 12 need to think about a scenario that should be added.
13 It's just an interesting thing that has caught my 14 attention doing the one letter on Reg Guide 1.183 and 15 here it is again.
16 MR. ALBRIGHT: Sure.
17 And like we said, that was kind of beyond 18 the scope of this work. We were trying to be 19 consistent. We were trying to have an apples to 20 apples comparison and extend the basis. Right?
21 It was an extension exercise at this 22 point. Right?
23 If there's a need or a desire or we see 24 the potential need to go and re-evaluate scenarios, 25 again, that's another research project that we can --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
68 1 MEMBER REMPE: Yes, maybe change the Reg 2 Guide.
3 MR. ALBRIGHT: Right, there you go.
4 MEMBER REMPE: It's just a comment.
5 MR. ALBRIGHT: But at this stage, we 6 haven't seen a need, I guess that's what I went back 7 to.
8 MR. ESMAILI: This is Hossein, can I say 9 something, Joy?
10 Okay, so, we think that these scenarios, 11 right, what's driving this things? What happens 12 during the scenario? Not what is the initiating 13 event, right?
14 And so, there are cases that, you know, 15 start at high pressure and they have shown that, you 16 know, the hot leg -- either the hot leg fails or they 17 have started.
18 But those have, by far, the biggest 19 influence as you have seen in distortion, how things 20 get released from the, you know, RCS into the 21 containment than whether you are considering a, you 22 know, whether a two-inch LOCA or whether it's a four-23 inch LOCA, et cetera. Right?
24 So, starting from the core damage, what 25 happens during the core damage? And you know this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
69 1 even better than I do, is what's driving this source?
2 And I'm just looking at Dave and Lucas 3 just, you know, to confirm it.
4 So, we can throw in other accident 5 sequences. You know, another LOCA, maybe another 6 thing. And I'm not suggesting we shouldn't, I'm just 7 saying that we know that what the answer would be.
8 They're going to fall somewhere in the --
9 between here. Right?
10 And so, we have captured the most 11 important phenomena that occurs during the accident 12 progression.
13 I want to make one thing clear, is that 14 the input deck, the other thing I wanted to make 15 clear, is that the input decks, as we use for in 2011, 16 we did SOARCA analysis, right, we did the uncertainty 17 analysis.
18 And so, the input decks actually evolved.
19 It was not just the cold experiment and everything 20 that we learned. We learned how these plants are 21 built, you know, what is the insulated, what's not 22 insulated.
23 And in some cases, we understood, you 24 know, that they make a lot of difference.
25 After all is said and done, so we have NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
70 1 improved over these past ten years or so, we have also 2 improved our input deck.
3 But that finding three says that what's 4 driving these things is the failure of the pressure 5 valve.
6 If you keep that intact, you are going to 7 be where you were, you know, in NUREG-1465.
8 Does that -- okay.
9 MR. LUXAT: I just wanted to say, like 10 Hossein said, the primary driver here is the release 11 of fission products from the fuel. And it's really 12 what pathways exist to get those fission products out 13 of the primary system into the containment that's sort 14 of driving the magnitude and timing of the buildup of 15 fission products in the containment.
16 And as we'll see a little bit later, one 17 of the discussion that we got in with respect to the 18 peer review was related to, you know, okay, there's 19 the core damage, but then, there's also looking more 20 deeply into the transport of fission products from the 21 core through the primary system and ultimately into 22 the containment.
23 And Shawn is going to be talking a little 24 bit later about some of the follow on work that came 25 out of the peer review looking at the transport NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
71 1 pathways in more detail to better understand where 2 fission products have migrated into different regions 3 of the containment or the primary system.
4 And then specifically, do different 5 regions that have or can influence different types of 6 release pathways be it through an MSID or through a 7 containment back area.
8 And we'll be discussing that in more 9 detail.
10 But it's really from the perspective of 11 the accident scenarios, we got core damage.
12 And then, it's really some of the 13 additional failures that are influencing the transport 14 pathways or transport of fission products to key 15 release pathways that can get them out of the 16 containment that are dominant.
17 And Shawn is going to be going into that 18 in a little more detail soon.
19 MR. ALBRIGHT: Thank you.
20 Next slide? Okay, so the accident 21 selection. The accidents that we used in this 22 analysis were informed by that NUREG-1560 that I 23 mentioned earlier, the IPEs, or the individual plant 24 examination program.
25 And basically, in 2011, they took that IPE NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
72 1 to develop the accident sequences that they looked at 2 so that they were consistent.
3 And then, we have followed those through.
4 And these are actually similar to what was 5 selected for NUREG-1465. So, doing this, it gives us 6 apples to apples comparison with NUREG-1465 as well.
7 And this provides coverage of all of our 8 major severe accident -- unmitigated severe accidents.
9 It incorporates, again, the station 10 blackouts, the LOCAs, and the ATWS scenarios.
11 And the peer review during that process 12 acknowledged that there are more recent PRAs that 13 would potentially show different core damage 14 contributors.
15 But overall, for the intended application, 16 this set of scenarios is appropriate with regard to 17 the progression of these unmitigated severe accidents 18 for evaluation of the in containment source term.
19 Next slide, please? So --
20 MEMBER ROBERTS: Can you provide any 21 insight as why they said that?
22 So, they concluded that the set of 23 reactions you had were sufficient?
24 MR. ALBRIGHT: Say that one more time?
25 MEMBER ROBERTS: So, why couldn't -- why NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
73 1 did the say the reactions you have were sufficient 2 given they may not be the right ones?
3 MR. ALBRIGHT: Well, I think -- yes, yes.
4 So, I think their characterization wasn't 5 that they were the wrong accident scenarios, but that 6 there may be more recent plant proprietary PRA studies 7 that have different contributors to that core damage.
8 So, again, I think the big picture is that 9 we're covering the spectrum of severe accidents here 10 and that, for the development of this representative 11 unmitigated severe accident source term, that the 12 current set of scenarios provides that coverage.
13 MEMBER ROBERTS: So, if something that you 14 didn't look at was worse, that's okay because of the 15 nature of why you do the studies now? Is that the way 16 to translate that?
17 MR. ALBRIGHT: Say that one more time?
18 MEMBER ROBERTS: Yes, what I think you 19 said is that they -- for the intended use, this set of 20 accidents is sufficient.
21 But I -- is having your why? And one 22 reason would be that, you know, they're bounding.
23 I don't think you said.
24 Another reason would be, they're not 25 bounding, but because of the way they're used, it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
74 1 doesn't matter that they're not bounding.
2 MR. ALBRIGHT: Okay.
3 MEMBER ROBERTS: I guess I'm trying to 4 figure out which it is. And you get the latter, you 5 know, what is the thought of what the intended use is?
6 Because there's a lot of intended uses.
7 MR. ALBRIGHT: I think I understand your 8 question now.
9 And you know, responding on behalf of the 10 peer review committee, I'm speculating here.
11 The peer review committee did not identify 12 scenarios that we missed. They identified that there 13 may be different contribution percentages to the core 14 damage frequencies.
15 So, it wasn't -- they didn't say that you 16 missed a scenario, they said, the relative percentage 17 of the different scenarios might be different than 18 what was done in NUREG-1560.
19 MEMBER ROBERTS: Okay, I see.
20 So, if you had a more complete set, they 21 don't think you would have a larger answer?
22 MR. ALBRIGHT: I don't think -- I don't 23 know that the peer review committee made any 24 statements regarding how the numbers would change.
25 I think the only statements that come to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
75 1 mind are that the current spectrum of accident 2 sequences provided the coverage of the potential 3 accident sequences that were expected to be covered.
4 CHAIR PETTI: But there's other PRAs that 5 show that the frequency of those events may be quite 6 different.
7 So, when you stack them all up into a CDF, 8 their percentage contributions, depending on how you 9 weight the sequences, are different.
10 MR. ALBRIGHT: Yes, I believe that was the 11 intention of that first statement and I guess the 12 final statement as well.
13 DR. ESMAILI: May I say something? This 14 is Hossein Esmaili, again.
15 So, yes, you're absolutely right, except 16 that they are not looking at the frequencies, you 17 know, when they're coming that they have 18 representative source terms.
19 And the other thing I'm speculating is 20 that, you know, the peer reviewers, you know, they 21 have been doing -- a lot of them have been doing this 22 for a long time, as I mentioned earlier.
23 The accident signature, the core damage 24 signature, right, it does not change as long as they 25 have representative station blackouts or NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
76 1 representative LOCAs, et cetera.
2 And even in PRA, you know, we combine, we 3 don't look at every accident sequence, we just combine 4 a lot of them into a single plant damage state because 5 we know the accident progression itself does not 6 change.
7 You know, we are still -- it's an 8 unmitigated accident scenario. We're still melting 9 the core. You're still looking at, you know, 10 oxidation energy.
11 So, a lot of these accident signatures 12 that we have, they are captured.
13 And again, as I said, you know, what 14 changes here is, you know, during this progression, 15 what happens to the pressure boundary and how do you 16 get these things from the reactor into the containment 17 itself.
18 So, those are the important main drivers 19 rather than, you know, what is the initiating event?
20 And how do you combine them? Et cetera.
21 So, I think they believe, and this is my 22 person opinion, that we are capturing, in terms of, 23 you know, accident signatures, et cetera within this.
24 And you are absolutely right, you know, we 25 have done SOARCA uncertainty analysis. Even when you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
77 1 do SOARCA uncertainty for a single accident, you're 2 seeing a rate and, you know, what parametric do you 3 use.
4 But our systems are not unbounded. Right?
5 I mean, we have a system that can produce this much 6 hydrogen, you know, you have this much metal, you've 7 got this much water, et cetera.
8 So, we understand, you know, what that 9 range of uncertainty is and whether you are doing 10 uncertainties with accident scenarios or parametric, 11 et cetera, so we know where we end up.
12 And I'm just, again, I'm just speculating 13 that that's what they are.
14 Thank you.
15 MEMBER REMPE: So, my question that was 16 the unfair question about the Reg Guide 1.183 actually 17 was motivated by some of the peer reviewer comments 18 where they were suggesting, did you look to see if 19 these scenarios that you've selected are really 20 capturing the dominant ones at this time from a risk 21 perspective?
22 And my thought at the time as well, even 23 if you had something that wasn't important, you should 24 leave it in because of that one sentence or if you've 25 missed something, well, you probably should.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
78 1 And I think it might not -- it wouldn't be 2 a bad idea to kind of look at that at some point, not 3 this version, but I think you're even planning to do 4 another updated AST in the future, somewhere I read or 5 someplace to consider more things. And the next time, 6 maybe think about that.
7 MR. CAMPBELL: We're going to have a new 8 AST coming, I'll just say that. We're not redoing 9 2023 at this point, but we're doing really additional 10 follow on work to explore additional sensitivity to 11 that type of thing.
12 MEMBER REMPE: Yes, it might be something 13 to look at.
14 MR. ALBRIGHT: Thank you.
15 So, the next few slides will sort of give 16 that overview of the accident scenarios that were 17 covered here.
18 We break them out into sort of four 19 categories.
20 We've got initiating events, coolant 21 injection, RPV status, and containment status.
22 In terms of Peach Bottom, we looked at 23 seven SBOs, four of them were short-term and three of 24 them were long-term SBOs with that prolonged DC power.
25 And then, we had two LOCA scenarios.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
79 1 For coolant injection, we had RCIC 2 operation in three of the scenarios, the long-term 3 station blackouts.
4 We had no coolant injection in the short-5 term station blackouts.
6 And we had actually no coolant of any kind 7 -- no coolant injection of any kind in six of those 8 scenarios.
9 For the RPV status, we had the potential 10 for, I will say, high pressure scenarios in that we 11 didn't prescribe failure of that primary pressure 12 boundary.
13 The accidents were allowed to evolve 14 according to the MELCOR calculations. And if that SRV 15 reached its failure criteria, then it would seize.
16 In terms of containment status, we had 17 early failures and late failures. We'll see this in 18 the following slides as well.
19 We looked at drywell liner melt through, 20 torus overpressure, drywell head flames leakage for 21 those early failures.
22 And then, for the late failure, we saw a 23 high temperature failure -- or I'm sorry, that should 24 say overpressure failure. I'm not sure why that one 25 got mixed up.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
80 1 Next slide, please? For Grand Gulf, we 2 see much the same in sort of the overview of the 3 scenarios here.
4 The main difference is that we considered 5 at ATWS scenario for the Grand Gulf plant.
6 And I'll just jump to the end here where 7 the main difference is in terms of the containment 8 failures.
9 Again, we had those early failures 10 including at a high containment pressure failure prior 11 to core degradation in the ATWS case.
12 And then, the same high containment 13 pressure failure for the late failure.
14 So, Grand Gulf and Peach Bottom scenarios, 15 SBOs, LOCAs and ATWS.
16 Next slide, please? In terms of the Surry 17 accident scenarios, we considered two station 18 blackouts and three LOCAs.
19 So, you'll notice that, in this case, 20 we've actually shifted the PWRs are showing more LOCAs 21 than station blackouts in this analysis.
22 We credited cooling in one of the 23 scenarios and didn't have any coolant injection in 24 four of them.
25 We, again, have the possibility of high NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
81 1 pressure scenarios for the Surry analysis where, if 2 the model would predict hot leg creep rupture, then we 3 would allow it to occur and it actually ended up 4 happening in all of those scenarios.
5 So, all scenarios were low pressure 6 scenarios.
7 In terms of the containment failures, we 8 see the same containment failure possibilities here 9 with hydrogen deflagration at head failure and high 10 containment pressure as a late failure mechanism.
11 Next slide, please? So, Sequoyah accident 12 scenarios, this is our last plant here.
13 Again, I'll point out, we've got more 14 LOCAs in this model as well than station blackouts.
15 Five of our scenarios credit coolant 16 injection, while two of them are not considering 17 coolant injection.
18 Sequoyah also had the possibility of the 19 high pressure scenarios but ended up predicting hot 20 leg creep rupture in all cases.
21 And then, we had the same containment 22 failures.
23 I will make a general note about the 24 different models. The containment failures occurred 25 at or after lower head failure for folks who are maybe NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
82 1 wondering about that.
2 The only case where that wasn't true was 3 that ATWS that I mentioned earlier where we -- an 4 overpressure during the ATWS scenario before core 5 degradation.
6 Next slide, please? Okay, so, the next 7 couple slides, we'll be going into the radionuclide 8 inventories. I know this is a wall of numbers.
9 MEMBER ROBERTS: Yes, Lucas, is that a new 10 insight that you could get containment failure before 11 a lower head failure?
12 MR. ALBRIGHT: No, that was actually 13 prescribed in SAND2011 as well.
14 MEMBER ROBERTS: How about the NUREG-1465?
15 MR. ALBRIGHT: Off the top of my head, I 16 do not know, but we could definitely go look at that.
17 MEMBER ROBERTS: Yes, it's very, again, 18 half my question for Elijah, is it -- if you've got 19 scenarios where lower head failure happens after 20 containment failure, that's yet another reason to 21 question that SECY-1994, you know, philosophical or 22 principle, or whatever assumption.
23 Because if it's based on that's the point 24 at which you could reverse the transient and make 25 things better and if you had containment failure NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
83 1 before lower head failure, that doesn't seem to make 2 a whole lot of sense.
3 MR. ALBRIGHT: Thank you. So, as I was 4 saying, this is a bit of a wall of numbers here, and 5 I recognize that. I think it's actually more 6 important that we consider the relative findings from 7 the peer review committee in that process.
8 So, what these tables on this slide and 9 the next slide are showing us are large changes in 10 mass for the different radionuclide groups.
11 And you know, up to 51 percent in this 12 particular slide.
13 What I want to make clear is that the 14 radionuclide mass differences are not equal to the 15 differences in activity that result from those 16 changes.
17 So, a 50 percent increase in mass does not 18 correlate to a 50 percent increase in activity.
19 And this is important because what the 20 downstream codes are going to look at is the dose 21 based on the activity that's released.
22 MELCOR is looking at the release fraction 23 according to the current state of practice. And that 24 release fraction is reported in terms of the mass 25 release.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
84 1 So, one of the peer review committee 2 members had a note that it's actually unlikely --
3 based on the actual changes in activity that are being 4 observed here, that it's unlikely that the siting 5 calculations will be impacted by these changes in 6 burnup because key radionuclides like iodine are not 7 actually changing in activity across these burnups, if 8 that makes sense.
9 Next slide, please? So, this is the same 10 table for the pressurized water reactors. I think 11 I've probably covered this in enough detail at this 12 point that it's important these changes in mass are 13 not equal to changes in activity.
14 And the impact of the larger masses is 15 essentially going to reveal itself in these downstream 16 calculations that are looking at the actual change in 17 activity as the result of the change in burnup an 18 enrichment.
19 Next slide, please? This slide covers our 20 iodine and cesium chemical forms. So, this can get a 21 little bit confusing with all of the numbers we've got 22 here, so I'll try and be clear.
23 The NUREG-1465 analysis assumed 5 percent 24 of the iodine inventory was gaseous, either elemental 25 iodine or organic iodines.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
85 1 Ninety-five percent of that inventory was 2 assumed to be in the form of cesium iodide and the 3 remaining cesium inventory was assumed to be 4 involatile cesium hydroxide.
5 This is a very large difference between 6 current practices which are reported in the I think 7 it's the NUREG-CR-7008 or something like that that's 8 the MELCOR best practices for iodine and cesium 9 chemical forms.
10 We've reported them here. They're 11 consistent with those base case SOARCAs.
12 And in this analysis, 100 percent of the 13 iodine inventory is assumed to react with cesium to 14 form cesium iodide.
15 Five percent of the iodine inventory, so 16 5 percent of cesium iodide is placed into the gap.
17 And then, the remaining 5 percent of cesium hydroxide 18 is made up of -- sorry, of cesium in the gap is made 19 up of cesium hydroxide.
20 So, we've got cesium iodide and cesium 21 hydroxide in the gap in our current MELCOR 22 calculations.
23 And then, 95 percent of the cesium is 24 represented as cesium molybdenum.
25 So, big picture, uncertainty in iodine NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
86 1 speciation is persists to this day despite several 2 experimental studies.
3 We had several discussions with our peer 4 review committee regarding the FPT3, DF-4, and BECARRE 5 experiments that highlight some of the remaining 6 uncertainties that's here.
7 Fukushima Daiichi, as I indicated in the 8 post accident analyses is confirming our assumption 9 that cesium molybdate is the dominant chemical form of 10 cesium.
11 And finally, the peer review committee, 12 Joy, for your information, I think you mentioned this 13 earlier, has recommended that we do go and look at 14 some of the other experiments like the VERDON 15 experiments in terms of how we look at these aging 16 analyses.
17 MEMBER REMPE: So, I think I heard yes, 18 we're going to today. But it wasn't clear then in the 19 actual report. So, it's still not clear when, but I 20 just wanted to bring it up.
21 MR. CAMPBELL: We'll get back to you.
22 Like I said, we're always trying to adopt the best 23 practices. But at the time of these calculations, 24 that was the best practice. But we're always 25 evolving.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
87 1 MR. ALBRIGHT: Next slide, please? So, 2 the next few slides are going to cover other analysis 3 assumptions. I'll try to get through these quick.
4 Basically, we aren't looking at any 5 variations in gap inventory at the start of the 6 accident. We're assuming in every one of these 7 scenarios the same speciation and gap inventories that 8 I mentioned a minute ago.
9 We're also not looking at the fraction of 10 aerosolized iodine in containment or any radionuclide 11 retention or removal mechanism.
12 So, if you'll remember, MELCOR is 13 calculating these numbers, but we are clapping them 14 all back together to the total inventory in 15 containment.
16 The source terms that we're presenting are 17 consistent with the state of the art or current state 18 of practice.
19 Many of these practices were established 20 under the SOARCA project and have been published under 21 that NUREG that I mentioned a little while ago, the 22 MELCOR best practices.
23 We did use the latest code version at the 24 time labeled MELCOR 2.2. And that version was 25 released back in 2021.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
88 1 Some modeling practices have evolved since 2 SOARCA that we've incorporated into the analysis that 3 we're presenting today, namely three items.
4 The time at temperature fuel rod failure 5 model. So, basically, it's a lifetime model that 6 defines how long a fuel rod can stand at a given 7 temperature has been modified to use our default model 8 that was informed by the VERCORS experiments.
9 We have changed the liquefaction and 10 oxidized fuel failure temperatures for UO2 and ZrO2 to 11 2479 Kelvin. This is the mean value from the SOARCA 12 analyses -- the SOARCA uncertainty analyses.
13 So, we took this as a more representative 14 value than the default MELCOR value.
15 Next slide, please? We have assumed in 16 this analysis that the relative contribution of 17 different accident sequences for both PWRs and BWRs is 18 not being changed by the different core types that 19 we've got that we're looking at. So, the high burnup 20 and the HALEU cores.
21 We've looked at the -- sorry, we have 22 analyzed the aleatory uncertainty or the range of 23 accidents here through our analysis.
24 But we're not really looking at any 25 parametric uncertainties except for the sensitivity 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
89 1 calculations where we investigate key phenomenological 2 uncertainties that were identified as part of the HBU, 3 HALEU, ATF, PIRT as being different or as being 4 distinct for high burnup or high burnup HALEU fuels 5 from conventional fuels.
6 So, we're not looking at uncertainties 7 that exist in conventional fuels. We're looking at 8 uncertainties due to the differences of the fuels that 9 are being analyzed here.
10 We are not looking at the containment 11 removal mechanisms, as I mentioned. So, containment 12 sprays, any sort of deposition, suppression pool 13 scrubbing, we're not crediting that in the numbers 14 we're reporting today.
15 But again, we will be talking about that 16 in the follow up presentation with the follow on 17 calculations that we've done since this release.
18 Finally, the release fractions, anything 19 below 1E to the negative 6 was considered negligible 20 and it was truncated to and reported at 1E to the 21 negative 6 in this analysis.
22 Next slide, please? So, here we are at 23 the non-parametric statistical analysis. I'm sure 24 some of you have been looking forward to this slide 25 based on previous comments.
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
90 1 So, this is a non-parametric bootstrap 2 methodology. And what we're trying to do is we're 3 trying to explore the uncertainty across these 4 scenarios.
5 And give an idea of the uncertainty 6 distribution that we might expect based on the 7 spectrum of scenarios that we have considered.
8 So, the strengths of this analysis or of 9 this method is that it can be applied to data that 10 follow any distribution.
11 We're not assuming normality or anything 12 like that in this analysis.
13 It's a bootstrap methodology. So, we're 14 repeatedly resampling the existing distribution of 15 results.
16 And then, we're using that sampling to 17 actually develop a mean empirical cumulative 18 distribution reduction.
19 And then, we have this ECDF for each 20 quantity of interest that we're interested in. So, 21 phase duration, cesium release fraction during a given 22 phase, et cetera.
23 Then, we look at that distribution and we 24 actually report the 50th percentile of the ECDF. This 25 has the bonus of equally weighting our simulations.
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
91 1 We're not sort of, you know, biasing our results 2 towards one end of the spectrum here.
3 On the plot -- in the plot on the right 4 here, you see the early in-vessel phase duration 5 results for SAND2023 relative to the NUREG results.
6 So, what we're looking at with the points 7 along the solid line are the actual percentiles that 8 we've calculated for each of these distributions 9 through this bootstrap process.
10 And then, we select the 50th percentile as 11 our reported representative source term, in 12 containment source term.
13 And the dashed lines that you see around 14 those solid lines are the uncertainty that's described 15 here as spanning plus or minus standard deviation at 16 each percentile. And the lines have been smoothed out 17 here so that it looks continuous.
18 Yes?
19 MEMBER BIER: So, I am far from a 20 bootstrapping expert, so if I spent more time on this, 21 I might have a different opinion.
22 But personally, I would really rather see 23 things based on mean values rather than medians.
24 Basically, I mean, I understand what 25 you're saying that you're weighting all cases equally.
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
92 1 But it matters whether the times you're 2 above the median, are you above by a little bit or by 3 a huge amount? And mean captures that.
4 So, you know, I don't want to, you know, 5 send you guys back to redo everything, but in future, 6 I would sort of suggest doing things both ways and 7 presenting both because it does give different 8 information.
9 My two cents.
10 MR. ALBRIGHT: Thank you.
11 Yes, this was definitely a point of 12 discussion with the peer review committee as well.
13 And ultimately, they determined that not 14 biasing the results using this 50th percentile was 15 what they found acceptable.
16 But thank you for, yes, I definitely agree 17 with the points that you're making about the 18 differences between choosing a median or a mean.
19 Oh, there's the point right there, 20 actually, at the bottom of the slide.
21 The peer review did find that those median 22 values are appropriate because they're not introducing 23 that bias that we were talking about a minute ago from 24 potential outliers.
25 Next slide, please?
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
93 1 MEMBER MARTIN: Just real quick. On the 2 previous slide, please, looking at the plot and, it's 3 probably just confusion or maybe not, if they're a 4 cumulative distribution ruptures and the PWR1 has a 5 shape that you kind of expect, nice S curve there.
6 And the BWR isn't. That does suggest, you 7 know, I asked the question about a cusp effect, how 8 should I interpret that shape there?
9 MR. ALBRIGHT: Yes, I think the -- if I 10 remember your comment earlier correctly, you were 11 asking, do we see clustering for certain accident 12 sequences of source terms, is that a correct summary?
13 MEMBER MARTIN: That's correct, yes.
14 MR. ALBRIGHT: Yes. So, and the answer is 15 yes, we see discrete clusters of source terms based on 16 the accident scenarios.
17 MEMBER MARTIN: Maybe more so for BWRs?
18 MR. ALBRIGHT: I would have to go look 19 again at what those clusters looked like, but in this 20 case, with the different shapes of the cumulative 21 distribution functions, yes, it does look like there 22 is a different weighting across the spread here.
23 MEMBER MARTIN: Okay.
24 MR. ALBRIGHT: Next slide, please? So, 25 this is the high level overview of this bootstrapping 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
94 1 procedure. Again, this was maintained from prior 2 studies and deployed in this study as well.
3 So, if we think about this, the first step 4 here is we run case simulations. Let's say we run 200 5 simulations, so that's about the order that we can get 6 here.
7 So, we have 200 values for the early in-8 vessel phase duration. And then, we actually generate 9 N samples of that many simulations from the original 10 distribution.
11 So, let's say we take a 1,000 samples from 12 that distribution of phase durations, right, and then, 13 we actually calculate the percentile for each of these 14 N samples.
15 So, we calculate the 5th up to the 95th 16 percentile. And now, we have a distribution of 1,000 17 50th percentiles. The next step is to compute the 18 mean value of that distribution so that it's 19 representative and the standard deviation of each of 20 those percentiles.
21 So, this goes back to the dots and the 22 dashed lines on the last plot. And then, we actually 23 will interpolate to obtain what the ECDF for that 24 particular quantity of interest looks like.
25 So, I hope that was a clear example for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
95 1 folks on what this process would like for an example 2 quantity.
3 The advantage here is that we can actually 4 look at the variability from the different plants and 5 accident scenarios inside of the representative source 6 term as a single sort of statistical method.
7 We're not looking at things differently 8 for the different plants or the different scenarios 9 unless we want to break out the blocks of data in that 10 way.
11 And a general note here is that, with this 12 process, basically, you're maximum and your minimum 13 values are going to be determined by the maximum and 14 minimum source terms that you observe in your data 15 set.
16 So, we're not trying to extrapolate out 17 beyond the observed data here.
18 Next slide, please? Okay, so results and 19 discussion. So, I don't know if we need to go over 20 this again. I wasn't sure when our break would be, so 21 I wanted to sort of restate some key aspects of the 22 analysis here.
23 CHAIR PETTI: So, our break is normally in 24 seven minutes.
25 MEMBER REMPE: It's up to the subcommittee 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
96 1 chairman's discretion.
2 CHAIR PETTI: Well, I have an agenda, so 3 I thought I had to live to the agenda because I've 4 been --
5 MEMBER REMPE: It's at your discretion.
6 CHAIR PETTI: So why don't we take a break 7 now, then? And we'll be back at ten after 3:00.
8 (Whereupon, the above-entitled matter went 9 off the record at 2:53 p.m. and resumed at 3:09 p.m.)
10 CHAIR PETTI: Okay, it's time to begin 11 again.
12 MR. ALBRIGHT: All right, so, now that 13 we're back from our break, I'll restate some of the 14 key aspects of our analysis.
15 So, again, our objective here was to 16 extend the NUREG-1465 for these higher burnup and 17 HALEU fuels.
18 And we looked at four accident -- or four, 19 sorry, nuclear power plants. We looked at two BWRs 20 and two PWRs with Mark I and Mark III containment for 21 the Bs and a ice condenser in large dry containment 22 for the Ps.
23 We had four reactor cores analyzed with 24 varying degrees of burnup enrichment.
25 And we considered accident scenarios, 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
97 1 including small break LOCAs, large break LOCAs, short-2 term station blackout, long-term station blackout, and 3 ATWS, or anticipated transient without scram.
4 The criteria for the different accident 5 phases are listed below. Again, I think we've 6 discussed these in a good deal of detail.
7 We're going to be focusing on gap release 8 and early vessel release for the remainder of this 9 presentation.
10 Next slide, please? So, we talked earlier 11 about the impact of the reactor core or the burnup and 12 enrichment on our in containment source term. This is 13 sort of summarizing those results again.
14 And basically, what we're seeing from our 15 analysis is that the different reactor cores are 16 really not significantly different in terms of the in 17 containment source term.
18 So, the conclusion here is that the 19 increase in burnup and enrichment does not strongly 20 impact the in containment source term because the 21 accident sequences are driving variability here.
22 Next slide, please? So, in terms of the 23 BWR in containment source term evolution, this is an 24 extended version of the table that we showed earlier 25 today.
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
98 1 What I want to highlight here is that both 2 SAND2023 and SAND2011 are using MELCOR to predict 3 these results. Whereas NUREG-1465 is using that STCP 4 code which was the forefather of MELCOR so to speak.
5 We've advanced our understanding of the 6 accident progression through the knowledge 7 advancements, modeling advancements, and input deck 8 advancements over the course of the SOARCA project.
9 And our results are actually consistent 10 with the findings of the different SOARCA projects.
11 So, what I'll highlight in this table here 12 are, again, that our in-vessel phase durations for 13 BWRs are quite a bit longer than NUREG-1465.
14 This is, again, because of the progressive 15 core damage that we -- or the prolonged core damage 16 progression, excuse me, that we see in the MELCOR code 17 due to greater discretization of the core region.
18 And we're seeing larger releases for 19 several key radionuclide groups, including the 20 halogens, the alkaloids metals tellurium as well as 21 molybdenum.
22 We've talked about several reasons why 23 these releases are longer -- or larger, excuse me.
24 Some of the primary reasons are the longer duration 25 until lower head failure as well as the early failure 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
99 1 of that primary pressure boundary.
2 One thing I want to highlight before we 3 move on to the next slide is that, again, these 4 results are total inventory in containment.
5 So, MELCOR is calculating deposition.
6 It's calculating the scrubbing in the suppression 7 pool. It's calculating all of the different 8 mechanisms that will remove or retain different 9 fission products at different locations in the power 10 plant.
11 And then, we are, in our analysis, taking 12 all of those values and summing them up into our total 13 inventory in containment. So --
14 MEMBER BIER: Quick question.
15 MR. ALBRIGHT: Yes?
16 MEMBER BIER: In terms of things we care 17 about like offsite consequences or whatever, is it 18 important to know the total in-vessel inventory? What 19 does it tell us in the end?
20 MR. ALBRIGHT: Yes, thank you for that 21 question.
22 So, the -- in the process of how these 23 numbers are used in regulatory practice, basically, we 24 calculate the total inventories and then, the 25 downstream codes will calculate the transport of those 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
100 1 inventories to the environment.
2 And the short answer is, yes, the 3 processes of retention and removal are very important.
4 These numbers here are much larger than 5 what you would expect because we have taken those 6 aspects of what MELCOR is calculating out of the 7 numbers that we're reporting, essentially, so that 8 those downstream codes can do what they're designed to 9 do.
10 MEMBER BIER: Yes, I guess I'm wondering 11 whether having different numbers for in containment 12 inventory is actually meaningful if they're parceled 13 out differently between the different streams like in 14 the old study versus the newer study.
15 You know, like if certain things are going 16 to be, you know, released faster in one study than 17 another, wouldn't we want to separate those out rather 18 than have them lumped?
19 MR. CAMPBELL: So, it's kind of -- not 20 kind of, it's a legal requirement like the downstream 21 codes, and so they need -- they have models for some 22 of these processes, but they're simplified and 23 aggregated.
24 And usually they have conservative models 25 that are chosen.
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
101 1 So, even though this best estimate, 2 sometimes the -- for the purpose of licensing, maybe 3 and our guys can elaborate on that, they need more 4 conservative models for these processes.
5 MEMBER MARTIN: Maybe related to what 6 Vicki was saying, and probably related to what I might 7 have said to Dave during our break.
8 I'd say the BWRs seem to get a penalty 9 here with that view. And I know the underlying 10 assumption and the whole mitigating features credit, 11 I can't help but feel that, you know, passive 12 mitigating features.
13 I mean, we spend a lot of time talking, 14 you know, about the non-LWRs and crediting passive 15 features for everything and we're not looking at our 16 old plants in the same light.
17 Do you have a feel for if you eliminated 18 the inventory that MELCOR predicts in the pool?
19 You're already ready, Shawn.
20 MR. CAMPBELL: I'm ready to answer, I'm so 21 sorry.
22 First of all, I'll say, wait until my 23 presentation.
24 MEMBER MARTIN: Okay.
25 MR. CAMPBELL: I've got a whole 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
102 1 presentation on that and it's in light of some of the 2 peer reviewer comments where they were talking about, 3 you know, there's a lot of retention there in the 4 suppression pool, it seems, and so, can you guys try 5 to explain that and help us to understand what's going 6 on. Right?
7 So, we have some work that we've done post 8 the 2023 report, and that's what I'm going to be 9 presenting on later.
10 To try -- and right now, we're not trying 11 to speak of anything in the regulatory sense or what's 12 being done downstream.
13 We're just trying to better understand, 14 what are the concentrations? What impact does the 15 suppression pool have? Is there places where you may 16 be bypassing the suppression pool and not fully 17 capturing the -- what's available for release if you 18 just let it all go into the suppression pool?
19 So, that's what I'm going to hope to 20 explain a little bit later.
21 And I'm hoping that it will also answer a 22 little bit of Vicki's questions as well on 23 understanding, you know, how are these things then 24 being used kind of downstream in these codes as well.
25 So, we try to explain that as well.
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
103 1 MEMBER MARTIN: I'm anxious to see that.
2 Maybe I'll be quiet up until then.
3 MR. ALBRIGHT: Thank you.
4 Next slide, please? So, I mentioned 5 before that the results that we have found in this 6 study are consistent with SOARCA.
7 And the major result that we want to 8 highlight here is that SOARCA identified limited in-9 vessel halogen retention during that early in-vessel 10 phase.
11 And we see it here in this plot. It's a 12 little bit busy with all of the different places that 13 we're seeing the radionuclides.
14 But essentially, everything that's being 15 captured in that suppression pool, everything that's 16 being released airborne into the drywell, or released 17 into the environment during the early in-vessel phase 18 which is towards the front of this plot, around that, 19 let's see, probably around that six-hour mark, if I 20 remember correctly.
21 What we're seeing is that you've got an 22 enormous amount of the halogen fraction inventory 23 release to containment. And that's what we've 24 observed in this study and it's because of that that 25 failure of the primary pressure boundary.
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
104 1 Next slide, please? Again, the same type 2 of data we're presenting for the PWRs here, so the 3 main points here are very similar to what we talked 4 about with the BWRs.
5 We've updated these practices. We started 6 with the 2011 decks. We modernized all of those 7 practices. We implemented the new best practices from 8 SOARCA. And then, we revised any other practices that 9 I mentioned earlier like the time and temperature 10 model, et cetera.
11 These practices represent an improvement 12 over SOARCA in many cases. And actually, are still 13 consistent with the SOARCA results in terms of the way 14 these releases are occurring and the timing and 15 accident phase.
16 I do want to highlight that the accident 17 phases are, again, longer than NUREG-1465 and that 18 we're seeing larger releases for the halogens, the 19 alkaloids metals, the tellurium group, and the 20 molybdenum group.
21 So, we're seeing much the same pattern of 22 changes here and that's because the changes are rooted 23 in those advancements and our understanding and our 24 modeling practices here.
25 So, we should be seeing the same changes 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
105 1 across these two reactor types.
2 Next slide, please? So, in this slide, 3 again, highlighting the overlap or agreement with 4 SOARCA. For PWRs, SOARCA also found limited halogen 5 release or limited halogen retention, excuse me, 6 during the early in-vessel phase. And this is due to 7 the hot leg creep rupture.
8 So, different failure mechanism of the 9 primary pressure boundary, but the same result. We 10 open up that pressure boundary and things start to 11 move into containment earlier and in larger quantities 12 than observed previously.
13 Next slide, please? In terms of the 14 release rate evolution, we talked a little bit about 15 this earlier.
16 The highlight that I want to make here is 17 that when we assume a uniform release rate, the 18 release rates from the 2023 report are actually lower 19 than NUREG-1465. And this is because of those longer 20 phase durations that we have for our early in-vessel 21 phase.
22 So, yes, larger in magnitude, but over the 23 duration assuming that uniform distribution, smaller 24 values.
25 Next slide, please? So, this next group 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
106 1 of slides will go over some of our sensitivity cases 2 or sensitivity calculations that were performed as 3 part of this report, very high level. For anyone 4 interested in the further details about some of the 5 different key radionuclide groups, we provide that in 6 the report.
7 But we've summarized here the main 8 findings. The first sensitivity calculation we'll 9 look at is the fuel thermal conductivity sensitivity.
10 The idea here being that increased burnup will tend to 11 decrease the fuel thermal conductivity.
12 So, what we did was we ran actually a fuel 13 conductivity sensitivity with a lower fuel 14 conductivity that was informed by the FAST code.
15 What we found was that there was no impact 16 from variation of the fuel thermal conductivity.
17 Next slide, please? In the next 18 sensitivity calculation, we looked at the in-vessel 19 particulate debris porosity.
20 The idea here is that the higher burnups 21 will promote disintegration of the fuel material.
22 We ran three sensitivity cases here with 23 a reference porosity, an increased porosity, and a 24 decreased porosity to sort of show the spread that 25 might result from different porosities.
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
107 1 What we found here is that there is no 2 major impact from variation of in-vessel particulate 3 debris porosity.
4 Next slide, please? This third 5 sensitivity calculation was looking at the diameter of 6 the in-vessel particulate debris.
7 The idea being that higher burnups will 8 promote breakup of the fuel resulting in smaller fuel 9 particulate debris.
10 So, what we did in this analysis, because 11 of actually, you know, the range of sizes that you 12 could see inside of a core was we ran a high and low 13 case scenario to get an idea of what the spread was.
14 And what we found was that the variation 15 in particulate debris diameter is impacting that in 16 containment source term, but that those impacts are 17 actually smaller than the variation we're seeing 18 across the scenarios.
19 And we'll highlight this particular point 20 in just a few slides.
21 Next slide, please? I think we're on 22 four. Particulate debris following velocity.
23 In this case, due to the change in size of 24 the particulate debris, you might expect a different 25 velocity of that debris as it falls in the lower 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
108 1 plenum to the lower head.
2 What we found in our analysis, even 3 decreasing that velocity by a significant amount, is 4 that there no impact on source term due to variation 5 in the particulate debris falling velocity.
6 Next slide, please? In terms of the fuel 7 relocation temperature sensitivity, this is a 8 particularly interesting sensitivity for those of you 9 familiar with the SOARCA uncertainty analyses, this 10 was a primary parameter that was varied in those 11 analyses.
12 The idea here is that material 13 interactions can cause early failure of fuel 14 assemblies as well as other components.
15 So, in MELCOR, we have two modeling 16 options for these types of material interactions.
17 We have the interactive materials model 18 and we have the eutectics model.
19 The interactive materials model is the 20 practice that was used in previous analyses and was 21 explored through that SOARCA uncertainty analysis.
22 And the eutectics model is sort of a newer 23 model that's been explored through other uncertainty 24 analyses.
25 In this calculation, we basically looked 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
109 1 at the SOARCA uncertainty analysis distribution of 2 fuel relocation temperatures. And we chose an 3 uncertainty range that spanned the SOARCA evaluation.
4 So, we've got a low value, a high value, 5 a reference value that I mentioned earlier.
6 And then, we also ran a case with the 7 eutectics model.
8 So, the importance of this sensitivity, 9 and the reason I've spent a little bit of time here, 10 is that material interactions are causing different 11 fuel failure timings.
12 And that's actually impacting the accident 13 progression. That's changing the timing of relocation 14 of fuel and it's actually moving the progression of 15 the accident around to an extent.
16 And in the SOARCA uncertainty studies, 17 this was found to have an impact on those in 18 containment source terms.
19 However, this uncertainty is also present 20 for conventional fuels.
21 For those of you who are familiar with 22 those SOARCA analyses, they were looking at 23 conventional fuels.
24 And we're actually looking at the same 25 distribution of uncertainty here for these high burnup 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
110 1 fuels in this case.
2 Next slide, please? Oh --
3 MEMBER REMPE: There's a lot of other 4 events like doesn't relocation also affect hydrogen 5 generation in core? And again, I think this peer 6 review committee emphasized the correlated variables 7 that I just wanted to bring that up.
8 MR. ALBRIGHT: Yes, no, that's a good 9 point. I'm a little tunnel-visioned in on in 10 containment source terms today. But that's a very 11 good point, that the change to the accident 12 progression that's occurring as part of this fuel 13 relocation, it's more widespread than just the in 14 containment source term for sure. Next slide, please.
15 So this next model if the fuel rod 16 lifetime model. And I mentioned this briefly earlier.
17 So this is a model that basically we allow fuel 18 components to stand for a certain amount of time at 19 different temperatures, and it accumulates damage over 20 time.
21 So through the temperature history, we 22 accumulate damage until the fuel fails. This is one 23 of the many fuel failure models that we have in 24 MELCOR. What we did in this analysis was we looked at 25 the reference which was the default time at 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
111 1 temperature model, an increased lifetime, reduced 2 lifetime, and then the SOARCA lifetime.
3 And basically what this is, is it's timing 4 where -- a lifetime model where at lower temperatures 5 your fuel is going to stay in for a really long time, 6 22 hours2.546296e-4 days <br />0.00611 hours <br />3.637566e-5 weeks <br />8.371e-6 months <br /> in the case of the increased lifetime. And 7 at really high temperatures, around 2,600, we're going 8 to decrease the amount of time that it can stand. And 9 in our most -- in our smallest lifetime case, we only 10 allow the fuel to stand for three minutes if it 11 reached that high temperature.
12 What you'll notice in the plot is that 13 we're not actually seeing major significant 14 differences in the releases to containment. And this 15 is because the fuel rod lifetime model just doesn't 16 drive fuel rod failure. What we're seeing in our 17 simulations is that the fuel relocation temperature or 18 the failure of oxidized fuel assemblies is generally 19 dominating our fuel failure.
20 So this is sort of competing models in 21 MELCOR interacting, and in this case, this one didn't 22 have a strong impact on the containment source term.
23 Next slide, please. This analysis shows the hot leg 24 creep rupture. We talked about this a little bit 25 earlier.
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
112 1 This was a key insight from SOARCA that 2 early failure of the hot leg due to this creep rupture 3 will result in larger releases to containment. What 4 we did in our analysis was we actually just disabled 5 the hot leg creep rupture model for our sensitivity 6 calculation. And this allowed the model to actually 7 persist at high pressures for the simulation to 8 persist at high pressures until that vessel failure 9 occurred.
10 And what we see is that there is a 11 significant -- I think it's on the order of, like, 40 12 percent or something -- difference in the mass 13 fraction released to the containment when we allow 14 this pressure boundary to remain intact. So this is 15 particularly important in terms of that in containment 16 source term. Next slide, please. I really like this 17 slide. For other folks, I hope you do too.
18 So this is showing our source term 19 variability from an orange, the sequences, and in 20 purple -- sorry, in orange, the sensitivities and in 21 purple, the sequences. This really highlights sort of 22 the idea that we've been talking about all day today, 23 that the sequences are driving our source term 24 variability, not parametric uncertainties in our 25 models. So what you'll see and the easiest way to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
113 1 explain this is the uncertainty around the purple 2 lines is a larger span than then the uncertainty 3 around the orange lines.
4 And basically what that means is we've got 5 more variability in outcomes from sequences than we do 6 in terms of individual model sensitivities. Now in 7 relation to this finding or this observation, the peer 8 reviewers did not the potential for the combined 9 effects of the sensitivities. We ran separate effects 10 sensitivity calculations.
11 The peer reviewers were concerned that if 12 you combined all of these sensitivities, you might see 13 larger differences than what we observed. From our 14 experience with these models and our understanding of 15 these codes, the non-linear processes in our models 16 tend to limit the amplification of any combined 17 sensitivities so that when we combine several 18 sensitivities, we're not going to see additive 19 movement in a single direction. We've got non-linear 20 movement here such that to explore the impact of any 21 of these sensitivities, a single scenario with a 22 single parameter sensitivity is representative of what 23 kind of variation we might observe from that 24 uncertainty.
25 DR. SCHULTZ: Did you happen to do that?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
114 1 I mean, did you happen to take a few sensitivities and 2 combine them, see what would happen?
3 MR. ALBRIGHT: We did not include any 4 combined sensitivities in the final report, no.
5 DR. SCHULTZ: And based upon what we've 6 seen, I think you're correct. You could test it out 7 pretty quickly. Well, quickly is perhaps an 8 exaggeration. I'm sorry.
9 MR. LUXAT: From past uncertainty analysis 10 students, certainly for an international uncertainty 11 analysis project, we typically see is that you can 12 collapse to a smaller set of sort of uncertain 13 parameters and still realize, if you will, the same 14 output uncertainty, if you will. There are sometimes 15 a few kind of parameters or output parameters that are 16 very sensitive and important. And those tend to have, 17 if you will, some amplifying effects, and it's usually 18 around hydrogen generation.
19 But generally what we do is that one way 20 to think about it is you're inducing variability 21 across a very complex network of calculations, 22 calculational steps in a code. So think of it, just 23 a myriad of floating point operations that you're 24 walking down this computational tree. And you put an 25 uncertainty up here or a variability up here and it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
115 1 kind of propagates through this not quite a butterfly 2 effect or a tree falling in Brazil or the Amazon, but 3 in some ways, very similar.
4 Because of the complexity of these 5 calculations, you wind up with a very small curvation 6 up here, propagating and realizing this vast sort of 7 space of realizations. And to a certain extent, you 8 lose a certain amount of correlation ultimately to 9 that starting variability as you go through this 10 complicated set of calculations. But it's just enough 11 to kind of, if you will, push you down and realize the 12 same sort of space of outcomes and variability. I 13 know that's probably a very abstract statement we 14 always see with these codes, but --
15 DR. SCHULTZ: It's a good description. A 16 nice thought experiment. Appreciate it.
17 MR. ALBRIGHT: So before I move on from 18 these sensitivity calculations, I want to take a 19 minute to sort of reemphasize that the point of these 20 sensitivity calculations was to investigate the 21 identified uncertainties that were distinct between 22 high burn-up fuels and conventional fuels. We're not 23 looking at all uncertainties. We're looking at what's 24 special for high burn-up or HALEU fuels.
25 With that, that's move to the next slide.
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
116 1 So summary of this report, I think these are fairly 2 familiar concepts at this point to everyone. But 3 basically, what we're seeing that increased burn-up 4 and extended enrichment are not significantly 5 impacting the source term based on our analysis and 6 that sequences are the most significant contributor to 7 variability in our data set.
8 The status of the RPV or of when that 9 lower head failure occurs is basically going to be an 10 important factor in terms of our early in-vessel 11 releases. That low pressures are exhibiting more 12 significant releases to the containment than high 13 pressure that were considered in previous analyses.
14 And finally, that those early in-vessel source terms 15 are greatly reduced if that pressure boundary remains 16 intact.
17 MEMBER REMPE: Well, I guess at this time 18 I want to ask my question about the status of reactor 19 pressure vessel. Not worried about the pressure 20 within the vessel but vessel failure. And I know we 21 did those tests -- or you did those test out at Sandia 22 years ago.
23 And we didn't think about things that 24 we're seeing nowadays, Fukushima. But maybe vessel 25 failure isn't at a distinct time. And it's really --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
117 1 is it really what you're interested in is when you 2 have a large mass of material ex-vessel on a 3 containment floor? And if that's the case, maybe it 4 doesn't happen automatically. And I was wondering if 5 maybe some additional thought is needed in that area.
6 MR. ALBRIGHT: So I think in terms of the 7 way we model severe accidents and in terms of the 8 current state of practice, lower hedge failure is sort 9 of a discrete bifurcation in terms of what we see in 10 our analysis results. And I think you were alluding 11 to that earlier, right? We see the release of 12 significant quantities of material, debris to the 13 containment.
14 In terms of what we're seeing from 15 Fukushima, there may be reason to be that failure 16 could occur more progressively than the lower head 17 failure tests at Sandia would suggest where you have 18 sort of a progressive relocation of debris into the 19 containment. In terms of how we split out our 20 accident phases, I think that's probably at the end of 21 the day we can do different things with our 22 calculations. But in terms of how it feeds into a 23 regulatory source term, I would have to defer to my 24 NRC colleagues to give me direction on what they need 25 for my calculations.
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
118 1 MR. CAMPBELL: Yeah, just going back to 2 what we said earlier that we're already going out to 3 seven hours here. And you'd have to have some sort of 4 start and stopping point. And so there's a 5 bifurcation of -- at that point of vessel failure.
6 And so we're trying to find a stopping point. When 7 else would we end, I guess.
8 MEMBER REMPE: So the question is what 9 you're going to do with it. And then should it be a 10 discrete point is what I'm asking. And maybe, again, 11 if you're looking at the effectiveness of the ECCS 12 systems, I'm not sure that a discrete point is -- this 13 is a larger question.
14 MR. CAMPBELL: It's a larger question that 15 is --
16 (Simultaneous speaking.)
17 MEMBER REMPE: -- I just think of 18 something that ought to be thought about because as 19 you get more information -- I mean, the Sandia test 20 didn't have fuel assemblies drop out of the -- and 21 they were -- again, that's what everybody wanted.
22 We used to worry about a penetration 23 versus global vessel failure and what does it mean.
24 And just again, now it's so important. At that time 25 is when we went from, like, one and half hours to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
119 1 eight hours and then six point whatever hours. It 2 seems like it's going all over the board. And maybe 3 it's not all at once.
4 MR. SALAY: This is Mike Salay. We didn't 5 even consider trying to redefine that because that's 6 another thing that we'd have to defend. And so we 7 stuck with the definition which is a point that you 8 can get, that they can -- that NRR and the regulators 9 can use.
10 MEMBER REMPE: But maybe the regulators 11 ought to -- is it a good point? Yes. Anyway, I've 12 made the question a long one.
13 MR. ALBRIGHT: Next slide, please. So now 14 we come to the end of a peer review. What I'll give 15 here is an overview of this peer review process and 16 some of the main findings, definitely direct people to 17 the peer review report which is highlighted here on 18 the left for any further details they may be 19 interested in. The idea behind this external peer 20 review was to review the technical basis of the 21 SAND2023 document that we just finished covering 22 today.
23 Recommend improvements to the draft form 24 of the document prior to its publication. And then 25 any additional recommendations after its publication 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
120 1 as well as assess the suitability of the source terms 2 for the regulatory applications that were intended.
3 Next slide, please. The overview of the organization 4 of this panel, there were, let's see, six panel 5 members here from different organizations.
6 We had Dr. Mohsen Khatib-Rahbar, excuse 7 me, from ERI. We had Dr. Richard Denning from -- he 8 was a consultant. We had Mr. Jeff Gabor from Jensen 9 Hughes, Dr. Didier Jacquemain from the OECD NEA, Dr.
10 Luis Herranze from CIEMAT, and Yu Maruyama from JAEA.
11 So an international group of severe 12 accident experts who are reviewing this document.
13 Their objectives were to assess those qualities of an 14 alternative source term that we mentioned earlier 15 today. So what was the technical adequacy of the 16 approach and the specific applications of the MELCOR 17 code to developing these source terms?
18 How appropriate were the sequences 19 selected? How the assumptions and applied models line 20 up with our current understanding of severe accidents 21 and source terms, and how adequate are these 22 approaches given current experimental and other 23 existing data sets. Finally, to assess whether our 24 source terms were representative rather than 25 conservative and bounding and then to basically make 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
121 1 a conclusion on the overall technical basis of the 2 approach. So next slide, please.
3 In terms of this review process, high 4 level, we prepared a draft high burnup fuel source 5 term that was completed in 2021. And then we had a 6 group of virtual meetings that began in 2022. The 7 first meeting was a briefing of the draft report that 8 was followed up by discussion and essentially comments 9 from the peer review committee.
10 During the second meeting, we provided 11 initial responses to the peer review committee's 12 comments, resolving some comments but not all of them.
13 So we revised the report. And during the third 14 meeting, we provided them with a final report that had 15 responses to all of their different comments which are 16 detailed in that peer review report for anyone who's 17 interesting in sort of following this track through of 18 comment and discussion. Next slide, please.
19 In terms of the acceptability of our 20 source term, I've taken some direct quotes here that 21 the peer review panel endorsed the approach. They've 22 stated that we've provided a defendable technical 23 basis for our source terms. We reasonably represent 24 the U.S. nuclear fleet, and we have a spectrum of 25 accidents that is sufficient to satisfy the RG 1.183 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
122 1 requirements. Next slide, please.
2 In terms of some qualities of our source 3 term, the peer review committee had a number of 4 comments that I thought relevant to bring to our 5 attention today. The first was the study's 6 significant technical improvement using state of the 7 art methods implemented in MELCOR. In containment 8 source terms for high burnup in HALEU fuels are 9 representative MELCOR estimates rather than 10 conservative and bounding estimates.
11 The peer review committee did not identify 12 any biases that would overestimate in containment 13 source terms in our analysis. And the sensitivity 14 studies were valuable in supporting this application, 15 particularly looking at the impact of the 16 depressurization of that primary pressure boundary or 17 the failure of the primary pressure boundary, excuse 18 me, so the HALEU pre-pressure sensitivity analysis.
19 Next slide, please.
20 In terms of recommendations, the peer 21 review committee had two major recommendations that I 22 have on this slide. The first is that the gap release 23 phase be incorporated into the early in-vessel phase.
24 I mentioned earlier that the advancements in how we 25 model severe accidents has led to the loss of distinct 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
123 1 gap and early in-vessel phases.
2 We now have sort of overlap between those 3 two phases depending on where we are in the core. The 4 second point that they made is that it would be more 5 appropriate to represent the impact of burn-up on core 6 inventories through expression of radiological 7 activities. And we have some details on this 8 particular note in the next presentation where we look 9 at this one in more detail. Next slide, please.
10 This list here is basically the 11 compilation of some of the peer review comments that 12 we've made throughout previous slides here. The first 13 one here is basically that we didn't consider bypass 14 or (audio interference) regression scenarios in the 15 development of these tabular source terms and that we 16 did not include any fission product removal and 17 retention mechanisms in the final reported results.
18 Again, MELCOR is capturing these, but we are providing 19 source terms in terms of the total inventory.
20 We talked about how the peer reviewers 21 acknowledged more recent PRAs that might give us a 22 different distribution of core damage contributors but 23 that the current analysis practice was suitable for 24 its intended purpose. There was the important note 25 that for most radionuclides there is not increase in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
124 1 activity with burn-up such that it would impact siting 2 calculations. Then we had comments on the uncertainty 3 in iodine speciation that we touched on earlier.
4 And then the last two bullets here are the 5 confirmation of the assumption that cesium molybdate 6 is the primary chemical form or dominate chemical form 7 of cesium coming out of Fukushima again. And finally, 8 that the use of the median estimate is appropriate for 9 avoiding bias in our final results here. Next slide, 10 please. This last comment from the peer review 11 committee sort of is driving some of our follow-up 12 calculations.
13 So I just wanted to put it up here and 14 quickly summarize it. The tabular source terms 15 provide a simplified tool for regulatory applications 16 but that there are limits to how these tabular source 17 terms can be used and the information that can be 18 provided in them. And the peer review committee 19 encouraged direct application of state of the art 20 severe accident codes like MELCOR to specific issues 21 when appropriate.
22 The issue that we will be talking about in 23 the next slide and the next presentation with Shawn is 24 the suppression pool scrubbing or the impact of 25 suppression pool and the radionuclide concentrations 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
125 1 in the steam line. Next slide, please. So this 2 table, I don't want to spend too much time on the 3 details because Shawn is going to go into this in much 4 greater detail that I could in a single slide here.
5 But what we're looking at are for the gap release 6 phase and the early in-vessel phase, the total 7 inventory including the suppression pool, and the 8 total inventory excluding the suppression pool.
9 So what you'll notice is that when we pull 10 the suppression -- the radionuclide inventory that's 11 in the suppression pool out so the right column for 12 each of these accident phases, the source term 13 decreases significantly. So the main point here is 14 that the suppression pool like some of you have 15 already mentioned today has a significant effect in 16 terms of radionuclide retention for key radionuclide 17 groups. It is basically immobilizing some of these 18 fission products while they are retained in the 19 suppression pool.
20 Two peer review findings are particularly 21 important for us to note here. The first -- and these 22 are quoted -- I think they're direct quotes, not 23 summaries. But the in-containment source term should 24 consider the impact of retention in the suppression 25 pools, especially for SBO scenarios that discharge 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
126 1 directly into the suppression pool.
2 And estimates of retention and suppression 3 pools provided in SAND2023 could be used in regulatory 4 guidance to establish suppression pool decontamination 5 factors. So this is something that we're still 6 looking at. And we've done some follow-on 7 calculations again that Shawn will be looking at in 8 the next presentation which I think we'll get to in 9 just a few minutes. Next slide, please.
10 The next two slides are going to be very 11 quick on upcoming work. The first is the chromium 12 coated ATF concept. We have been working on an 13 chromium coated accident tolerant fuel concept source 14 term that follows the same practices that we've 15 outlined for you all today looking a chromium coated 16 fuels.
17 And this analysis is also being informed 18 by that ATF part that we mentioned earlier today.
19 Next slide, please. We are also currently working on 20 a source term report for FeCrAl fuels. Again, this 21 analysis is being informed by that ATF severe accident 22 part. And these are following the same practices that 23 we've outlined today.
24 CHAIR PETTI: What's your timeline for 25 those?
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
127 1 MR. CAMPBELL: That's for me. We're very 2 close to these. So I'd say in the next couple of 3 months we're planning on having a draft complete of 4 these. And then we'll have another period of time of 5 just internal review and discussion. But in the next 6 coming months, we plan to have these finished.
7 CHAIR PETTI: We might be interested.
8 MR. ALBRIGHT: Next slide, please. Thank 9 you.
10 CHAIR PETTI: Okay. Watch the time.
11 Let's keep going.
12 MR. CAMPBELL: Okay. So that's the end of 13 that presentation. And if we don't mind switching to 14 the other presentation, I can get started on mine.
15 All right. So hello, my name is Shawn Campbell, and 16 I'm in the Office of Research.
17 So I want to provide you all with some 18 follow-on work that we've been doing in light of some 19 of those latter peer review comments that we were 20 talking about. And I think Sandia has done a great 21 job teeing this up a little bit. So this is follow-on 22 work that we've been doing, just to try to explore the 23 impact of suppression pool and its retention. So next 24 slide, please.
25 So overall background here, like I said, 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
128 1 the peer review panel has commented on potential 2 impact that a suppression pool could have on what's 3 getting into containment obviously. And that table, 4 it's Table 516 if anyone is interested. It's in the 5 SAND2023 report.
6 So it's there, and it provides the 7 containment release fractions both including and then 8 excluding the suppression pool. So we did some 9 supplemental investigations following those peer 10 review comments to try to investigate the impact of 11 the suppression pool. And in particular, try to look 12 at scenarios and pathways that might bypass the 13 suppression pool because that's really what's 14 important here, right?
15 So to that end, we modified the two BWR 16 input decks, Peach Bottom and Grand Gulf, try to 17 better capture the behavior that could be going on, 18 particularly in the steam line. And then we performed 19 a set of BWR source term calculations. So it's the 20 same scenarios that we did with 2023 that were 21 performed for this analysis. So next slide, please.
22 So before I go forward, there's been a lot 23 of confusion about what a source term versus 24 inventories versus all that. So I thought that this 25 could be kind of useful to make sure that we're all on 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
129 1 the same page. We all understand what is a 2 containment source term and it's being used.
3 This becomes really important later on in 4 my presentation. So up here at the top is kind of a 5 representative MELCOR calculation if you will where, 6 for example, if you're trying to do a SOARCA analysis 7 or a Fukushima analysis, we would use MELCOR to do the 8 full accident scenario. This means cladding 9 oxidation, relocation, transport of the fission 10 products out of the core, into the reactor vessel, out 11 into containment, containment failure and vessel 12 breach.
13 And then after vessel breach, you've got 14 MCCI and over-pressure and so on. So MELCOR is 15 calculating all of these things, right? And so not to 16 say that when we talk about the source term, MELCOR is 17 calculating all aspects of retention and deposition 18 and everything.
19 It's just what is then being reported as 20 the containment source term. So the containment 21 source term then is the cumulative amount of fission 22 products that enter the containment during these 23 phases. So in the gap phase, early and vessel and so 24 on.
25 It's the cumulative amount of fission 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
130 1 products that have entered into that volume. And 2 that's what we report down here in the green box.
3 That's our containment source term.
4 And it's a fraction of the overall 5 inventory in the core that has made its way into the 6 containment. This is a fraction, not -- because 7 MELCOR deals in overall mass, not inactivities. So in 8 order to get to activities as you were saying, what we 9 really care about here is the dose, right?
10 Or at least from a regulatory perspective, 11 what we care about is the dose in the end, right? And 12 so how that's done downstream in the regulatory space 13 then is that containment source term is then tried to 14 make by an applicant or whoever, make it more reactor 15 specific, right, because this is a representative 16 source term that we have here in the green that's 17 supposed to be representative of the fleet. And so to 18 make it more reactor specific, we want the 19 concentration of that source term.
20 So that's where you would divide by the 21 volume of your containment, your individual 22 containment, to get your concentration of the source 23 term into that containment structure. So it's a 24 simple ST divided by volume here. That concentration 25 is then used in this simplified approach using removal 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
131 1 mechanisms, whether that may be sprays or a natural 2 deposition or whatever that might be.
3 In a simplified approach, those removal 4 mechanisms are accounted for here. And then you have 5 a leak rate. And then this is when your inventory or 6 your activity comes into play, right? And so now 7 we're seeing how much activity goes from containment 8 and now into your people space. I'll pause there and 9 see if there's any questions or needed additional 10 clarifications.
11 MEMBER REMPE: Jose always says that he 12 has an evil mind to try and think of something again.
13 (Laughter.)
14 MEMBER REMPE: -- and go against the 15 system. Had a vessel that purposely would fail early.
16 And that's what I was saying really if the 17 vessel would fail earlier. And then you got a smaller 18 source term. And so to try and avoid that would be a 19 good motivator to rethink the question I asked 20 earlier. Again, because there are a lot of design 21 developers and it's a way to make sure things --
22 MR. CAMPBELL: I'll state the obvious.
23 Our aim is not to game the system.
24 MEMBER REMPE: It is --
25 (Simultaneous speaking.)
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
132 1 MEMBER REMPE: But somebody else's would 2 be.
3 MR. CAMPBELL: Right. But that's why we 4 provide the source term or that's written into the Reg 5 Guide is that we provide a source term that we think 6 is representative, that's peer reviewed, that we think 7 is defensible. And then that's what's used then by in 8 the regulatory space.
9 MEMBER REMPE: It's just a reason to think 10 about.
11 MR. CAMPBELL: Right. Okay. Any other 12 questions?
13 MEMBER ROBERTS: Yeah, just for 14 clarification. When you talked about suppression pool 15 scrubbing, I don't see that on this page. It's not on 16 the removal mechanisms, or --
17 MR. CAMPBELL: It is not part of the 18 removal mechanisms.
19 MEMBER ROBERTS: Okay. So if you don't 20 account for it, then --
21 MR. CAMPBELL: It is not accounted for in 22 this bottom part. And then up on the top, obviously 23 MELCOR is calculating it. But it's part of that whole 24 bookkeeping. It still gets lumped into that green 25 space and brought down.
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
133 1 MEMBER ROBERTS: Right. So it could be 2 part of that mechanism modeling. Mechanism modeling, 3 if you chose to do it that --
4 (Simultaneous speaking.)
5 MR. CAMPBELL: And that's -- yes and no.
6 And that's one of the things I want to kind of tease 7 out here because one of the concerns is, is there 8 anything that's bypassing. And that's one of the 9 aspects that you have to kind of tease out in this 10 whole aspect.
11 So you don't want to just simply say, oh, 12 let's just scrub and take 80 percent, 95 percent of it 13 is just gone. Are there pathways -- release pathways 14 to the people space. Let me bypass this whole thing.
15 And so that's what we're trying to investigate here.
16 MEMBER ROBERTS: Okay, thanks.
17 MR. CAMPBELL: Yeah. All right. Next 18 slide, please. So this is just trying to show where 19 we've tried to refine our models. So over here on the 20 left is a representative -- this is Peach Bottom. So 21 this is our original from 2023.
22 This is the nodalization that we have 23 within MELCOR. So we wanted more refined nodalization 24 of the steam line in particular because this is what 25 we're talking about when I'm saying particularly 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
134 1 through the MSIVs we're trying to understand what's 2 going on up there, what's happening within the steam 3 line, what's the concentration within the steam line.
4 And so we refined our modeling.
5 And I'm afraid I can't zoom in here. If 6 I had my mouse, I would. But in this purple space 7 here, you can see that top bar. It's two volumes.
8 You can see two boxes.
9 That was our steam line A. Thank you very 10 much. That's our steam line A. And so that's the 11 steam line that has the lowest pressure SRV. And so 12 that had a slightly more -- thank you so much.
13 That has a slightly more refined modeling, 14 I guess, with two CVs. But then everything else was 15 lumped together into a single CV, all three of the 16 other lines. So we wanted to break this out a little 17 bit so we could better capture the physics of what's 18 going on in the steam line.
19 So this is where we have our more refined 20 modeling. And the reason we need more refined 21 modeling if you look on the right-hand side over here, 22 you can see it's a rather long -- the steam line is 23 rather long. It's got bends and turns and everything, 24 not to mention the SRV cycling on it, HPCI, RCIC 25 pulling off and so on.
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
135 1 So we wanted to be able to better capture 2 what's going on if we're going to ask questions about 3 what is the fraction of inventory that's making its 4 way into the steam line. So next slide, please.
5 Thank you. So here's our more refined steam line 6 modeling. So for each of the BWRs, each steam line 7 has now been broken up into this nodalization.
8 So the first volume that I have here in 9 blue, this is going from the steam dome down to and 10 including the first SRV. And in our model, that first 11 SRV is the lowest pressure SRV. So that's the one 12 that's going to be cycling.
13 And then the next volume is this green 14 volume. This is everything downstream of that first 15 SRV up to the first MSIV. RCIC and HPCI pull off of 16 this line depending on the steam line obviously.
17 And so we have a much finer nodalization.
18 We also have as a separate volume then in between the 19 MSIVs downstream to the MSIV down to the stop valve 20 and condenser. So all of that has much more refined 21 modeling. And then so now going forward --
22 MEMBER HALNON: Is the only vent to the 23 environment the condenser because you have so many 24 others?
25 MR. CAMPBELL: We do. And we're not going 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
136 1 out to that level of detail to try to model everything 2 that can get to the environment. That's really don't 3 by those downstream codes, right? We're not trying to 4 capture -- you might have leakage through the valve 5 stems and so on. We're not trying to capture overall 6 --
7 (Simultaneous speaking.)
8 MEMBER HALNON: That's usually a 9 significant loss, megawatts.
10 MR. CAMPBELL: Sure.
11 MEMBER HALNON: So that we -- we're 12 always chasing. It could be one -- three percent.
13 Getting close though.
14 MR. CAMPBELL: Sure.
15 MEMBER HALNON: Quite a bit.
16 MR. CAMPBELL: Yeah, and as I'm about to 17 say, we kind of stopped our investigation at the first 18 MSIV. And so I'll explain why here in a moment. But 19 most of our investigation is here in this green 20 portion.
21 So in the rest of my presentation, I'm 22 going to be talking about source term fraction kind of 23 in the same way that we talk about our containment 24 source term. I'm going to be talking about source 25 term fraction in the steam line because one release 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
137 1 pathway for BWRs obviously is the MSIVs, right? And 2 this is a release path that has the potential for 3 bypassing the suppression pool.
4 And that's why we thought that it was 5 important to better characterize and understand what's 6 going on in the steam line. So as we go forward, a 7 few things that I need to talk about here about this 8 green portion. First of all, when I report source 9 term in the future, I'm reporting what's in the green 10 portion.
11 We thought this was representative kind of 12 in the same way that the containment source term is 13 what's available for release through a leakage pathway 14 in containment. This is kind of what's available for 15 release through an MSIV. Also distinct from how the 16 source term is reported in containment as a cumulative 17 release into containment, that's harder to do here in 18 this green portion because you don't have fission 19 products entering into this volume and then staying 20 there indefinitely due to cycling of the SRV, through 21 RCIC and HPCI operation, through leakage through the 22 MSIV.
23 This is a dynamic volume. It's not quite 24 as dynamic obviously as that blue portion. But this 25 is a dynamic volume. And so you're going to have a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
138 1 lot of in and out.
2 So we needed a better way to come up with 3 a value for source term. And what we came up with is 4 a time averaged source term. And I'll try to flush 5 that out a little bit more as we go.
6 But we talked about what we're going to 7 capture here is a time averaged airborne fission 8 product source term. And so emphasis on airborne 9 because we're already taking into account all of those 10 deposition mechanisms. This is another thing that's 11 distinct from how we report in containment where those 12 downstream codes then look at whatever might deposit 13 through sprays or through whatever.
14 We've already taken that into account here 15 in this green portion in order to try to capture all 16 of that physics, the ins and the outs. We're 17 reporting it all as an overall fraction of airborne 18 time averaged within that phase. So I'll pause there.
19 That was a lot.
20 Okay. Well, then I'll move on from there.
21 And if we need to cycle back to this, I can. So this 22 is what we're reporting here. We've seen most of 23 these values already many times.
24 You've got your Reg Guide 1.183, Rev. 0, 25 Rev. 1, and then 2023 values that my colleagues have 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
139 1 been talking a lot about. And then these next two, I 2 need to point out. These are just the Table 5-16 3 values.
4 These are not doing a bunch of additional 5 calculations. This is a bookkeeping exercise here of 6 what is or is not in the suppression pool. The only 7 thing that's truly unique on this slide versus what we 8 did in the 2023 report is in the last column, and this 9 is this fraction that is sitting in that green portion 10 of the steam line.
11 Now I want to point out here this is 12 fraction. So you may look at that and say it's 13 incredibly small, e to the -5. But you have to 14 remember this is a different volume.
15 Concentration is what's really important 16 here, not fraction. And so what we need to really do 17 is look at what's the overall concentration comparison 18 between what's in containment versus what may be 19 sitting in the steam line. And that's why the next 20 slide is going to be really important where now we 21 have to look at an example and divide by the volume, 22 then do a concentration comparison between what's in 23 containment versus what's in the steam line.
24 And that's why the next slide is going to 25 be really important where now we have to look at an 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
140 1 example and divide by the volume, then do a 2 concentration comparison between what's in containment 3 versus what's in the steam line. So that's what I 4 want to do in the next slide if you don't mind going 5 to the next slide. So here we are.
6 We got Grand Gulf on the left, Peach 7 Bottom on the right. All I've done here is taken 8 those values from the previous slide and divided by 9 the containment volume. So this is fraction of core 10 inventory per meter cubed. That's the units here.
11 And so you can see 1465 and 2011 and 2023 12 all listed here. And then off here in my brackets, 13 this is what's made it into containment minus what's 14 in the suppression pool. That's what's in that first 15 column of the brackets and then what's in the steam 16 line.
17 So all this is, is the SAND2023 values, 18 the first column. All it is, is the SAND2023 taking 19 away the suppression. That's all I've done here. But 20 then on the steam line side, I just need to point out 21 a few things.
22 Once again, you may look at this and think 23 that steam line containment are comparable. But be 24 careful. Remember these aren't apples and apples 25 source terms, right? The one on the right has already 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
141 1 taken into account any deposition mechanisms within 2 that phase. So this is the concentration throughout 3 that phase, if that makes sense.
4 MEMBER BIER: So I'm trying to understand 5 the whole picture. So the third set of bars is the 6 2023 results --
7 MR. CAMPBELL: Yes.
8 MEMBER BIER: -- that these guys just 9 talked about.
10 MR. CAMPBELL: Correct.
11 MEMBER BIER: And you parcel it out 12 further and get those bars to the right?
13 MR. CAMPBELL: That's right.
14 MEMBER BIER: Do you have a figure that 15 does that to the 2011 numbers, or --
16 MR. CAMPBELL: This has 2011 numbers on 17 there as well.
18 MEMBER BIER: No, but that parcel out 19 process on the right.
20 MR. CAMPBELL: Oh, no, no. I'm actually 21 separating what's in the suppression pool versus --
22 no, we don't have those values. I would assume it's 23 going to be kind of analogous here.
24 MEMBER BIER: Got it. That's what I 25 wanted.
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
142 1 MR. CAMPBELL: It's a fair assumption. I 2 don't have that in front of me.
3 MEMBER BIER: Okay.
4 MR. CAMPBELL: A couple things I'll point 5 out. Size of containment matters obviously. Grand 6 Gulf is a much larger containment. So your 7 concentration is diluted here.
8 Also, I'll point out that in the steam 9 line, you notice it really doesn't change too much 10 between the two plants. And that's really a 11 consequence of your steam lines aren't all that 12 different between the plants. It's a relative -- the 13 steam line is relatively similar between these two 14 plants. All right. Yes.
15 So then here the only purpose of this 16 slide was trying to do a comparison of BWR to PWR 17 then. So I'm not talking about the steam line in this 18 slide. So the purpose here is just to say how does 19 this compare to what's going on with Ps.
20 And so this is representative volumes for 21 a couple Mark I, Mark II, Mark III and then for PWRs 22 and ice condensers, sub-atmospheric and a large dry.
23 So I've just divided by some volumes here. And you 24 can see that time progression essentially from 1465 to 25 2011 to 2023, and then if you don't account for what's 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
143 1 in the suppression pool.
2 And you'll see that at least for the 3 halogens, then it's more similar to what you're seeing 4 in a PWR. All right. So the purpose of this one also 5 was -- and this was a recent addition. The purpose of 6 this one really was to find some kind of 7 representative example fission product activities for 8 a few case studies that we've done for high burnup and 9 high enrichment PWRs and BWRs.
10 So we're trying to calculate end of cycle 11 activities for that inventory, a piece of this, right?
12 Remember we're going back to that first slide.
13 Inventory matters in activities.
14 So that's what we're trying to get at here 15 versus the kilograms that we've been talking about in 16 the 2023 report. We're trying to better understand 17 what would it be in terms of activity for some of 18 these reactors going up to higher burn-ups. So here 19 we have for BWRs and PWRs, we have a reference case on 20 the first column.
21 So we've got a core average end of cycle 22 burnup for the reference case for a B of 36.2. We 23 have an average assembly discharge burnup of 52.6 with 24 an enrichment of 4.45. And then we're going to a more 25 higher burnup here.
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
144 1 And we're estimating about 58 average 2 assembly discharge burnup. And this is going from an 3 enrichment of 4.45 to 5.3. So we're trying to come to 4 something that's, like, a representative loading 5 pattern for a BWR that might want to go to a higher 6 burnup and tease out what kind of values they could 7 get.
8 What you see here, I find it kind of 9 interesting is that the iodine concentration trends 10 with the power. And since the power hasn't increased 11 here, your iodine concentration hasn't increased. Now 12 the alkalis, however, that increases with your burnup.
13 So you do see a proportional increase of 14 your cesium going to these higher burnups. And this 15 is the same for Bs and for Ps. And then same thing 16 for the tellurium.
17 There's really a weak dependency on burnup 18 enrichment for tellurium as well. Same thing with the 19 Ps. We went to a higher cycling for the Ps and had a 20 higher enrichment and higher burnup there as well.
21 So you're seeing that we're not getting --
22 maybe it's a little deceptive when you're looking at 23 some of the 2023 values and you're seeing a 40 percent 24 increase in cesium and iodine mass. That's not what 25 we're talking about as far as activity. And we're not 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
145 1 trying to say that this is the activity that you're 2 going to see from a licensee.
3 We're just saying this is possibly a 4 little bit more representative of something going to 5 a higher burnup. Okay. Next slide, please. So 6 conclusions, we did some refined modeling, and it 7 provided some better estimates of fission product 8 distribution in the steam line and then in 9 containment. And we found that concentration in the 10 steam line is distinct from that of containment when 11 you're not looking at the suppression pool.
12 MEMBER HALNON: On the steam line, I guess 13 I'm still befuddled a little bit.
14 MR. CAMPBELL: Please.
15 MEMBER HALNON: And you're probably just 16 going to say that's somebody else's job. But we've 17 been looking at license amendments that are trying to 18 take credit for the downstream sections of piping or 19 scrubbing --
20 MR. CAMPBELL: Yeah.
21 MEMBER HALNON: -- and other items. How 22 is that factoring in? Or is that somebody else's 23 issue?
24 MR. CAMPBELL: The answer is it is a 25 regulatory issue of how it's going to be applied.
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
146 1 That being said, one of the things we were trying 2 tease out here is can you apply -- could you apply the 3 same source term that's in containment sans 4 suppression pool? Would you apply that same source 5 term then to MSIV leakage? And I think that's what 6 I'm trying to tease out here is that wouldn't be quite 7 right. You really need to have a -- there is a 8 distinct and distinctly higher concentration in the 9 steam line than there is in containment minus 10 suppression.
11 MEMBER HALNON: Even if you add in all 12 that extra piping just because --
13 MR. CAMPBELL: And that's the thing, if 14 you look at all the downstream effects. But that's 15 typically teased out by the individual applicant or 16 licensees, right, because they're using their 17 downstream codes to look at all of that.
18 MEMBER HALNON: This is the verifying, 19 validating, whatever you want to call it, confirmed 20 starting point. If they can justify some other type 21 of scrubbing suppression pool or downstream, then they 22 might be able to bring it back to a point where their 23 design works.
24 MR. CAMPBELL: Right. And that's what 25 we're trying to tease out here.
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
147 1 MEMBER HALNON: Okay.
2 CHAIR PETTI: So longer term, bigger 3 picture is this stuff will inform and update to 1.183 4 on MSIV?
5 MR. CAMPBELL: I would have to defer to my 6 NRR colleagues on that. Our purpose here first of all 7 was just to be responsive to the peer review, right?
8 That was the purpose here was to say the peer review 9 said to go look at this. Let's be responsive to that 10 because they're saying that this suppression pool is 11 going to have an impact.
12 But we wanted to make sure that there has 13 been talk of suppression pool, suppression pool. We 14 wanted to make sure that we teased out can you just 15 apply the suppression pool or just do a DF completely 16 off of that same containment source term without 17 looking up the steam line separately. And hopefully 18 we've teased out that, no, you do need to look at that 19 steam line as a separate entity.
20 MR. DICKSON: Hey Shawn?
21 MR. CAMPBELL: Yeah.
22 MR. DICKSON: Elijah Dickson with the 23 staff. I'd like to connect some of the other, like, 24 regulatory initiatives that are going on right now.
25 So just a few weeks ago, I spoke in regards to the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
148 1 increased enrichment. And we're looking at the 2 control and design criteria in that case.
3 And part of those alternatives is looking 4 at these mechanistic transport models in support of 5 development of Regulatory Guidance 1.183, Rev. 2. So 6 we are considering all of this based off of the 7 current work, based off comments received from ACRS, 8 individuals in industry. So it's all part of that 9 work right now. And we're still in the process of 10 collecting public comments. Again, I just wanted to 11 kind of connect the dots in regards to other 12 regulatory initiatives that are being done right now.
13 MR. CAMPBELL: Great.
14 MR. DICKSON: Yeah, no problem.
15 MR. KORTGE: Can I ask a question?
16 MR. CAMPBELL: Yeah, please.
17 CHAIR PETTI: Who is --
18 MR. KORTGE: What is the mode of force for 19 the suppression pool?
20 CHAIR PETTI: Sorry, who's speaking?
21 MR. KORTGE: This is David Kortge.
22 CHAIR PETTI: From? Organization?
23 MR. KORTGE: Constellation, Safety 24 Analysis.
25 CHAIR PETTI: Yeah, no, you can't ask.
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
149 1 Only during public comments.
2 MR. KORTGE: Apologies.
3 MEMBER BROWN: As an uninitiated, may I 4 ask a question?
5 CHAIR PETTI: Maybe.
6 MEMBER BROWN: In excruciating detail.
7 I'm trying to figure out what the bottom line is.
8 It's wonderful analysis. It's not answering my 9 question.
10 MR. CAMPBELL: Please.
11 MEMBER BROWN: If I read this and did you 12 analysis, all the green pipes are longer than the 13 other pipes which means to me you've got higher source 14 terms you have to deal with under a severe accident.
15 Is that going to affect now the EPZs or the zones that 16 we have to deal with? Is that -- I mean, they're 17 bigger, a lot bigger.
18 MR. CAMPBELL: Which one is? Can we go up 19 one slide, please?
20 MEMBER BROWN: I'm looking at the one with 21 the bars, the little green, red, and gold, previous 22 slide.
23 MR. CAMPBELL: Previous, let's go up one.
24 Up another one. This one right here?
25 MEMBER BROWN: Yeah, that's the first 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
150 1 question. Does that affect --
2 MR. CAMPBELL: So I'll say that the orange 3 right there is Reg Guide 1.183, Rev 1 basis. Is that 4 correct? So the 2011 values there --
5 MEMBER BROWN: Yeah, with the little --
6 MR. CAMPBELL: -- in the orange?
7 MEMBER BROWN: Yeah.
8 MR. CAMPBELL: So then that is used by the 9 downstream codes as the regulatory basis.
10 MEMBER BROWN: So now it's going to get 11 bigger.
12 MR. CAMPBELL: If 2023 was the basis for 13 a reg guide, then --
14 MEMBER BROWN: But you would argue -- I'm 15 trying to be contrary a little bit here. But you 16 would argue that you've proved that the basis --
17 you've had peer review of the basis, an impressive 18 list of people with qualifications to do that review 19 which would seem to indicate that there's a 20 significant increase in what the source term you have 21 to deal with for computing EPZs which would apply to 22 present day plants. Now that's the takeaway that 23 somebody that's not -- I mean, don't ask me to do the 24 calculations. Okay? My mind was exploding while you 25 were doing that. The second question I would have --
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
151 1 the answer is yes, if you adopt it.
2 MR. CAMPBELL: That's where I would have 3 to defer to my friend at NRR.
4 MEMBER BROWN: I understand you're 5 deferring. That's fine. The second is this is high 6 burnup. People want to go to high burnup. Changes 7 your refueling, all that good stuff.
8 But then once you've taken the fuel out, 9 you now have a spent fuel. This would imply that you 10 have a bigger -- something you have to deal with in 11 the spent fuel pools as well, as well as in storage 12 casks once they've cooled down and you've got them in 13 storage casks. Does that compute?
14 You've got more source term you start 15 with. You've got more residual. It's not the 16 instantaneous stuff. But I'm trying to come to a 17 conclusion. How does this affect other things? Like, 18 now do I have to have different casks? I haven't 19 heard any of that in any of the previous high burnup 20 fuel discussions.
21 MR. CAMPBELL: Go back to the point of 22 Lucas's source term. But it's not the high burnup but 23 actually the change in how you're doing sequences and 24 going from high pressure to low pressure that causes 25 that increase primarily, not the increase burnup.
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
152 1 MR. CAMPBELL: In other words, we don't 2 anticipate there being a significant change going to 3 high burnup in the overall activity.
4 (Simultaneous speaking.)
5 MEMBER BIER: But there is a significant 6 change due to the analysis.
7 MR. CAMPBELL: In our source term. In our 8 source term.
9 (Simultaneous speaking.)
10 MEMBER BROWN: Existing low burnup then 11 would have the same results. And so that you're still 12 impacted in terms of the EPZs and/or (audio 13 interference). So if it's not a result of the burnup, 14 you just figured out that you didn't have the right 15 answers before for EPZ and for --
16 MR. CAMPBELL: Elijah is jumping at the 17 bit.
18 CHAIR PETTI: This is a research 19 presentation, then there is a licensing part.
20 MEMBER BROWN: I understand, okay.
21 CHAIR PETTI: This doesn't affect any of 22 the existing plans.
23 MR. CAMPBELL: That's right.
24 MEMBER BROWN: Well, you've got results 25 that has to be evaluated at some point that says, the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
153 1 analysis that was used is more comprehensive, has a 2 higher result, and it's independent. That's what he 3 just said, whether you use high burnup fuel or the 4 regular fuel we've got now which implies to me that 5 somebody is going to have to sit down with you all, 6 thrash out another -- you have to go address this in 7 the existing plans in terms of an analysis of their 8 occupations at speakeasies and our spent fuel pools 9 and/or past storage and/or transportation of those 10 casks.
11 More shielding is -- whatever. I'm just 12 a poor electrical guy. I do understand having numbers 13 come out significantly different regardless of how you 14 got there.
15 MR. LUXAT: So let me just quickly make a 16 comment that from the peer review, again, the method, 17 they appreciated the advance in the methods. But one 18 of the important comments from the peer review panel 19 was an RD contamination measures related to the 20 suppression pool that were not being --
21 (Simultaneous speaking.)
22 MR. LUXAT: -- for BWRs that were not 23 directly being addressed. And what Shawn has been 24 talking about is if we were to look at the effect of 25 those decontamination or those removable mechanisms in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
154 1 the suppression pool, A, it would have an impact on 2 what the containment source term is. But it would 3 also importantly have an effect on what the 4 concentration of fission products are nearby other 5 release pathways like the MSIV.
6 And that's what this is teasing out is 7 that the key comment from the peer review was there's 8 an important removal mechanism, a passive removal or 9 a inherent removal mechanism for BWRs. And what we've 10 been doing is we've been looking to expand the 11 technical basis and understand better the transport of 12 fission products out of the reactor system and how it 13 potentially -- how they could potentially be removed 14 by the suppression pool, what that effect would be on 15 the containment source term. But also importantly, 16 what the effect of that removal could be on other 17 release pathways that are not, if you will, 18 interfacing directly with the suppression pool.
19 MEMBER BROWN: There's other mechanisms 20 that remove some of this. That's fine under --
21 (Simultaneous speaking.)
22 MR. LUXAT: That was -- yeah, but that was 23 --
24 MEMBER BROWN: -- standpoint. But what 25 about the non-severe accident standpoint? I have to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
155 1 deal with removal. And you're still telling me that 2 I've got more stuff that I have to deal with that's 3 got source term in it when you pull it out regardless 4 if you have a severe accident or not.
5 That's the other takeaway I -- that's the 6 uninitiated, bottom line, takeaway for somebody like 7 me. And yet somewhere -- forget the severe accident.
8 I understand a PWR is a PWR.
9 They have other ways of reducing the 10 activity so it doesn't get spread all over the place.
11 But I still have to deal with regular fuel, regular 12 high burn-up, put it into the spent fuel, then put it 13 in a cask when it gets heated. And now I've got 14 larger source terms that I have to deal with in those 15 circumstances.
16 MR. DICKSON: This is Elijah Dickson.
17 With an increased enrichment, the rule making efforts, 18 they are looking at transportation and several other 19 rules as well. In regards to EPZ sizing, there are 20 several regulatory source terms that we use for 21 different purposes.
22 So this is an in containment source term 23 used to size mitigative systems to reduce radioactive 24 release to the environment and inside a facility, 25 certain distance away from a population center. For NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
156 1 EPZ sizing, they do use this in containment source 2 term as one of the analysis that went into the Ten 3 Mile EPZ rule, right? They also used source terms 4 that were derived from the PRAs, WASH-1400 as well.
5 And then also looked at different figures 6 of merit at that time to justify that Ten Mile EPZ.
7 So between the work that's been done in SOARCA with 8 the severe accident work and the more realistic 9 consequence analyses and this in containment source 10 term used as a design tool, right, to size safety-11 related mitigative equipment. There's no a whole lot 12 of difference that we're seeing between now and what 13 was done back in the late '60s when we did the EPZ 14 sizing.
15 MEMBER BROWN: If you take into these 16 other considerations and other things. I'm trying to 17 transition away from the severe accident. We haven't 18 had, quote, severe accidents.
19 MR. DICKSON: Three Mile Island was.
20 MEMBER BROWN: That was not as severe as 21 a real severe accident.
22 MR. DICKSON: Right.
23 MEMBER BROWN: That was like spitting in 24 the ocean.
25 MR. DICKSON: Right.
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
157 1 MEMBER BROWN: And a lot of dumb stuff.
2 I mean, my point being is what this tells me in your 3 results regardless you've got analyses, methodologies 4 in terms of routine operations. You still built up a 5 different spectrum of stuff that you had --
6 MR. DICKSON: Understood.
7 MEMBER BROWN: -- to deal with on a 8 routine refueling storage, then cask storage, and then 9 transportation issues aside from this severe accident.
10 That's all. I'm just trying to make sure that thought 11 process is in place as well or should be in place.
12 MR. DICKSON: It is. It is. We're 13 thinking a lot about --
14 MEMBER BROWN: There's --
15 MR. DICKSON: -- all of this.
16 MEMBER BROWN: -- routine operations, no 17 accidents. We haven't been getting the right answers.
18 If everybody accepts this as proper analyses --
19 MR. DICKSON: Right.
20 MEMBER BROWN: -- to get the right 21 results. Thank you. I think I've exhausted my brain 22 power.
23 MEMBER BIER: I have one more question 24 that's kind of a follow-up. Charlie, you were saying 25 that, okay, there are other mitigations that are not 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
158 1 reflected in the results here. Would those be 2 reflected in the results of a Level 3 PRA? Or are 3 Level 3s overestimated because they're not modeling 4 all of the (audio interference)?
5 MR. DICKSON: So I'm a practitioner of 6 this source term. I utilize this source term. These 7 are the experts that develop the source term, right?
8 Again, this is Elijah Dickson with the 9 staff. The way Reg Guide 1.183 is set up, Rev. 1 has 10 these tables of fractions of the reactor core source 11 term. In the appendices, Appendix A specifically is 12 the MHA LOCA appendix that tells you how to transport 13 this source term through all the different systems, 14 how to credit different type of removal mechanisms.
15 And that's where Shawn has been discussing how we can 16 make improvements in these particular areas, these 17 mechanistic transport models, in the appendices 18 specifically understood.
19 MR. CAMPBELL: Can we INSPECTOR BOTH: back 20 to that final slide? No, that was it. That was it.
21 MEMBER BROWN: Sorry to disrupt.
22 MR. CAMPBELL: No, thank you, no. So I 23 think the second bullet here is pretty obvious.
24 There's a lot of retention in the suppression pool.
25 Third bullet, preliminary investigation of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
159 1 the fission product inventory showed limited effect 2 for high burnup and high enriched uranium. So this 3 was that whole inventory aspect that you have marginal 4 increase in your actual activity. So inventory 5 matters.
6 And then finally, this is something that 7 we're investigating right now, we're looking into.
8 There's potential to apply MELCOR to better inform 9 some of those removal mechanisms that I was talking 10 about in that first slide, those lambdas and to inform 11 those for the simplified tools. And I just wanted to 12 point out here we're in the process of drafting a 13 document that summarizes this work that we've been 14 doing on investigating concentrations within the steam 15 line. So this is preliminary in that we're looking 16 for early feedback from ACRS on our process and so on 17 as we're drafting this report.
18 MEMBER BROWN: Well, don't take my 19 questions as being negative or critical. It's nice to 20 see we're not sitting on our past accomplishments, but 21 yet we're trying to make sure we're doing it right for 22 both today and the future. So don't take my comments 23 as being majority that are critical because it's good 24 to see somebody doing good research that comes up with 25 some results we can deal with, at least reduce it to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
160 1 the understanding of the common man.
2 MR. CAMPBELL: We take all feedback as 3 constructive.
4 MEMBER REMPE: Okay. What are you going 5 to do about the existing source term and the 6 differences between? I mean, you've got this new 7 model of higher release fractions because of the way 8 that the sequences modeled. And can't do back fits 9 probably. And so what will you do with this --
10 MR. DICKSON: We're still in the review 11 process. We're just now kicking off increased 12 enrichment. And with that is the work in developing 13 Reg Guide, Rev. 2.
14 MEMBER REMPE: It's a different problem 15 because it's --
16 MR. DICKSON: It is.
17 MEMBER REMPE: -- a penalty for the high 18 burnup, high enriched fuels because of our increase in 19 knowledge. And how does one deal with that?
20 MR. DICKSON: This is the beginning of 21 that process looking at this report and doing these 22 additional analyses. At this point, I can only share 23 so much and how we want to look at this and what we 24 want to be doing in Rev. 2, Reg Guide 1.183.
25 MEMBER REMPE: This is not going to come 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
161 1 up in the next month or two.
2 MR. DICKSON: Timeline, it's matching up 3 with increased enrichment. So this is under the 4 umbrella of increased enrichment. So I think we said 5 beginning or end of the school year, calendar year 6 2024, beginning of 2025 we'll be seeing something.
7 MEMBER ROBERTS: I'm not sure you answered 8 Vicki's questions. If you can go back to Slide 8.
9 No, the one after that, the one with the colors. That 10 one. No. Thank you.
11 Yeah, I think what I've got out of these 12 last couple hours is there's really no difference in 13 ultimate consequence between the blue, the orange, and 14 the green because they're driven by changes in 15 modeling. And the fission products are going to come 16 out from the core. At some point, the progression, 17 just those somewhat arbitrary division of lower vessel 18 head rupture.
19 That defines when you stop making that 20 plot. But what Vicki asked I think is interesting.
21 If you have a Level 3 PRA with these different models, 22 would the end result be significantly different?
23 Because the timing, right, is not all that different.
24 It's the timing of the events that are 25 happening, not so much the timing of the degradation 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
162 1 of the fission product release. So if you did a Level 2 3 PRA with these different models, would you have a 3 different answer? Or would it just all come out in 4 the wash? I mean, there's nobody here who can answer 5 that question.
6 DR. ESMAILI: And this is Hossein Esmaili.
7 Can I just say so we're not going to be using this or 8 any Level 3 PRA. Remember what Shawn showed in the 9 first or second slide that a Level 3 PRA, that's the 10 thing that we are actually doing right now is that we 11 are going to be doing a very mechanistic accident 12 progression source term calculation, going all the way 13 to lower head failure, MCCI, et cetera.
14 So whatever is going to come out of that, 15 whether it's a containment failure, what is the 16 release to the containment. So we are not looking at 17 the release to the containment. And don't forget, 18 this release to the containment, these bars that you 19 see, it's everything.
20 It's airborne, deposit that in the 21 suppression pool, et cetera. In a consequence 22 analysis, we are not going to be looking at what's in 23 the deposit, et cetera. So for a Level 3 PRA type of 24 analysis, we are just going to go that path, the one 25 that Shawn showed at the beginning.
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
163 1 We're going to go all the way, do a very 2 integrated analysis and look at what's going to come 3 out of the containment, whenever it's leaking or 4 containment failure, et cetera. This one is just for 5 the purpose of doing that bottom one which means that 6 we are doing a simplified approach right? I mean, 7 that simplified approach, somebody is going to take 8 care of the position, et cetera, those lambdas in the 9 real -- up one in the integrated analysis, that's all 10 part of the calculation.
11 MELCOR will calculate the position and the 12 structure, the position on the pool, in the pool 13 surface itself. So in that respect, you know, we are 14 just being very mechanistic. What we come down to is 15 that we specify what that lambda is from this 16 calculation, right?
17 MEMBER ROBERTS: Okay. Thanks, Hossein.
18 I think what I got out of that is the blue that's 19 there would be NUREG-1150. So whatever models produce 20 the blue also produce results at NUREG-1150. The 21 green is whatever is coming out of your current Level 22 3 PRA project.
23 And so you would see if there's an effect 24 between the blue and the green if you can find it with 25 all the other different things that have changed over 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
164 1 the last 30 years. But yeah, thanks for the answer.
2 I think that answered my question. So there's no --
3 in the severe accident world or Level 3 PRA world, 4 there's really no distinction to be drawn by this 5 blue-yellow or orange-green.
6 (Simultaneous speaking.)
7 DR. ESMAILI: Yes, in Level 3 PRA, we 8 would be looking at the accident sequences and just 9 combining them, whatever the plant damage say, what is 10 the release category, et cetera. This thing is only 11 when we come up with that green box that Shawn showed.
12 And then we do that path of a simplified model.
13 MEMBER ROBERTS: I didn't get an answer to 14 Charlie's question then. Is the blue versus orange 15 versus green have a real meaning in the reactor space?
16 MR. CAMPBELL: Yeah, that's sort of what 17 I want to say.
18 MEMBER BROWN: I walked away with that.
19 I transitioned back to the regular stuff. There's 20 still the differences you have to deal with.
21 MR. CAMPBELL: Yeah, so some of these are 22 artifacts because it's these regulatory applications.
23 Like, Level 3 PRA, you'd actually -- you want to 24 consider taking out the effects of the suppression 25 pool. You have to do that for the downstream 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
165 1 licensing calcs only. So that's --
2 CHAIR PETTI: I just think it shows how 3 difficult it is to take what's done and let's call it 4 the state of the art. And to try to boil something 5 down that you can put into a licensing approach that's 6 simple because you don't want something complex 7 because there's so much artificiality. And you can 8 get misled as much as you can get an answer that's 9 good. So good luck, Elijah.
10 MEMBER REMPE: Okay. So back in the 11 NUREG-1465 days, they thought an hour and a half was 12 enough for a source term, right? If you did an hour 13 and a half with your 2023 source term, you'd probably 14 get a lower amount, not a higher amount, right, or 15 something comparable? If you want to be consistent, 16 why not just say an hour and a half and cut it off 17 because vessel failure is irrelevant.
18 DR. ESMAILI: So this is Hossein Esmaili.
19 So Joy, I'm just going to defer to NRR, right, because 20 we are -- no, I'm just trying to be very careful 21 because what we are doing is that we are showing 22 everything. We are showing you everything.
23 This is what happens in the containment.
24 This is what happens if you just take out the 25 suppression pool. You can see in terms of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
166 1 concentration. And concentration is what's really 2 driving this. You go from that green, that big green.
3 You go to that red which is --
4 (Simultaneous speaking.)
5 DR. ESMAILI: And we are going to have 6 meetings. We're going to have public workshops in the 7 next few months starting in January going to April.
8 And whatever regulatory decision they're going to 9 make, it's going to come at the end of that.
10 MEMBER REMPE: So the question really 11 wasn't for you or for --
12 (Simultaneous speaking.)
13 MEMBER HALNON: Okay.
14 MEMBER REMPE: It's more for Elijah.
15 Elijah, that's what was done back in the days of 1465.
16 And that was enough in order to be a consistent 17 predictable regulator. It seems to me that that was 18 enough back then to say, okay.
19 DR. ESMAILI: But talking about the vessel 20 and lower head failure, we have learned over the years 21 that there was the calculations we used to do, like, 22 20, 30 years ago, as soon as the debris would melt and 23 come down. It would reach with the lower head, 24 whether it's a drain plug, whether it's an instrument 25 tube, and fail it and go out. So that was that.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
167 1 Nowadays --
2 MEMBER REMPE: We used to worry about high 3 pressure injection too and a lot of things we don't 4 worry about anymore. But anyway, it's just a 5 suggestion that might be --
6 DR. ESMAILI: Yeah, yeah. No, that's a 7 very good suggestion. All I'm suggesting is that some 8 of the -- this durations, and this came out of SOARCA, 9 is because we're doing a better modeling of heat 10 transfer to the water, you know, that it has to boil.
11 There's a massive amount of structures in the lower 12 plenum of a BWR against the duration of the (audio 13 interference). So all of those things factor into the 14 fact that now it's in vessel phase. Actually, this 15 was a SOARCA insight.
16 It's going to take a long time, right?
17 And I hear you. I'm not going to make any judgments.
18 Leave it up to NRR, what they want to do. But we have 19 the data. We have the analysis. We can go data 20 mining. We have other choices. It's just that 21 decision has not been --
22 MEMBER REMPE: I get it. I just was 23 trying to figure out how one would get out of this 24 mess. And so it's not a question.
25 DR. ESMAILI: Yeah, the other thing is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
168 1 that you can see that we talk about the suppression 2 pool. It's really not -- we should not be comparing 3 the source term. We should really comparing the 4 concentrations because when you compare, like, in a 5 BWR, we have the suppression pool.
6 So you automatically see that the 7 concentration goes down with the different Mark I or 8 Mark III design. But in the PWR case, you don't have 9 a suppression pool. But you have a huge containment 10 volume, right?
11 So the concentration you can see, it's 12 just like comparable even to a -- the Mark III. So 13 that was the purpose of showing this. Sorry, thank 14 you.
15 CHAIR PETTI: I guess we should probably 16 go out for public comment. Any members of the public 17 that would wish to make a comment, please unmute 18 yourself, name, organization, and comment.
19 Hearing nobody online, we do have someone 20 in the room. Please go ahead.
21 MR. CSONTOS: So I just wanted to say 22 thank you very much for the presentations. The first 23 time -- oh, Al Csontos, NEI, Director of Fuels. Thank 24 you very much for the presentation and discussion. I 25 think it's very helpful. We had a lot of comments and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
169 1 questions and looking forward to having those 2 workshops that you talked about.
3 I think that will be really important to 4 count as dialogue on how the impact of the new 5 modeling that's here goes out to the actual 6 implementation steps. Charlie, to your question, a 7 comment you made. EPRI did do a scoping study looking 8 at the impact of high enrichment, higher burnup on the 9 back end. Okay. And --
10 MEMBER BROWN: Without severe accident.
11 MR. CSONTOS: Yes, this is more -- you had 12 mentioned dry cask spent fuel, things like that. That 13 report number is 3002027535. I'll pass it on to Larry 14 and folks to provide you.
15 But the bottom line there was that there's 16 very little impact to the back end. There's a small 17 -- longer time that you might have to put it in the 18 pool, but it's manageable. And actually dose rates 19 for the workers actually goes down because you're 20 loading less canisters. So that's the -- but you can 21 read the study.
22 MEMBER BROWN: The good news is you've 23 looked at it.
24 MR. CSONTOS: Yes, we've looked at it.
25 But thank you very much and more on this later. And NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
170 1 also during the rulemaking reg basis comments that the 2 industries would like.
3 CHAIR PETTI: Thank you. With that, I 4 don't see any more comments. So I think we -- oh, I 5 want to -- yes.
6 MEMBER HALNON: Yeah, you guys did a great 7 job. I mean, you kept energy through this whole 8 thing, and that kept us going. For those of us like 9 my young friend here who are not modelers, it was 10 understandable. Appreciate it.
11 MEMBER BROWN: I couldn't have asked the 12 question without absorbing a little bit of what you 13 said for the last two and a half hours, three hours, 14 whatever it is now.
15 MEMBER REMPE: It's good to hear things.
16 I mean, one, we were curious about this because of the 17 Reg Guide 1.183 but also the new calculations and 18 understanding this will help others when they're 19 trying to deal with Reg Guide 1.183, Rev. 2.
20 CHAIR PETTI: Very important when we hear 21 the next go around of (audio interference). Technical 22 basis here was just very informative. Okay. Thank 23 you everyone, and we are adjourned.
24 (Whereupon, the above-entitled matter went 25 off the record at 4:44 p.m.)
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
Securing the future of Nuclear Energy High Burnup Fuel Accident Source Terms ACRS Briefing Nov 16, 2023 Presented by Lucas I. Albright and David L. Luxat Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-NA0003525.
Contents
- Motivation and Background
- Key Messages
- Deep Dive
- Summary
- Independent Peer Review
- Upcoming Work 2 of 70
Motivation and Background 3 of 70
High Burnup Fuel Source Term Analysis Motivation
- Develop alternative source term applicable to LWR cores with HBU/HALEU fuel
- Different burnup levels and enrichments considered
- Extends NUREG-1465 and SAND2011-0128 SAND2023-01313 alternative source terms 4 of 70
Historically Relevant Studies
- TID-14844: Calculation of Distance Factors for Power and Test Reactors, -
USAEC 1962
- NUREG-1465 - Accident Source Terms for Light-Water Nuclear Power Plants,
- USNRC 1995 (code: STCP)
- SAND2011-0128 - Accident Source Terms for Light- Water Nuclear Power Plants Using High-Burnup or MOX Fuel (code: MELCOR 1.8.5) 5 of 70
Source Term Timeline Fukushima HBU/HALEU/
Phébus FP Daiichi ATF SA PIRT BSAF RG 1.183 ATF Major SOARCA V2 Source Term Developments NUREG-1560 Surry SOARCA SOARCA V1 UAs TID-14844 MELCOR Peach Bottom 95 11 23 NUREG-1465 SAND2011-0128 SAND2023-01313 Accident Source Terms Participation in OECD/NEA International Programs 6 of 70
Severe Accident Modeling Advancements
- Heterogeneous, integrated reactor core modeling tends to promote to progressive and extended core degradation.
- 2D discretization of the reactor core
- No more distinct gap release phase
- Prolonged core damage progression
- Longer times to lower head failure
- Prevalence of accident-induced low-pressure scenarios - SOARCA
- Hot leg creep rupture for majority of PWR sequences 7 of 70
Impact of Early Depressurization BWR: Thermally induced SRV seizure PWR: Hot Leg Creep Rupture Early loss of the primary pressure boundary induces depressurization of the reactor coolant system and opens a release pathway for radionuclides to transport directly to containment during early in-vessel core degradation
- Diagrams are for illustration purposes only 8 of 70
Selected Severe Accident Datasets More recent severe accident datasets have improved characterization of core damage progression and subsequent radionuclide releases since NUREG-1465
- Severe accident experiments used to validate severe accident codes
- Phébus FP
- Early fuel failure
- Hypothesized CsMoO4 as the dominant chemical form of Cs
- VERCORS
- Early fuel failure
- High burnup fission product release rates
- Severe accidents are a primary data source for severe accident code validation
- Fukushima Daiichi
- Existing data confirms that CsMoO4 is the dominant chemical form of Cs 9 of 70
Severe Accident Knowledge Advancements
- Chemical form of iodine:
- NUREG-1465 assumed 95% of iodine in the form of CsI
- Current practice assumes all Iodine to be CsI
- Still assume 5% of the total iodine inventory is present in the gap inventory
- Chemical form of cesium:
- NUREG-1465 assumed Cs predominantly in the form of volatile CsOH
- Current best-practice assumes 5% of cesium present in the gap inventory as both CsI and CsOH
- All remaining cesium assumed to react with Mo to form Cs2MoO4
- Mo release:
- Mo releases are now higher than other metallic fission products such as Ru and Pd.
- Te release:
- Current best practice is more extensive Te release than reported in NUREG-1465
- Due to change in chemical form with more efficient transport of Te to containment 10 of 70
HBU/HALEU/ATF PIRT
- HBU/HALEU fuel severe accident behavior
- No significant differences between HBU and HBU/HALEU fuels
- Thermophysical property differences expected
- Fuel fragmentation and sintering can impact core degradation
- Fission product chemistry may change
- Possibility of cladding embrittlement
- Greater potential for recriticality during reflood using unborated water for HALEU 11 of 70
Key Findings 12 of 70
Study Highlights Key Finding 1: Increased burnup and enrichment does not strongly impact in-containment source term
- Most significant variation in source term arises due to differences between accident scenarios Key Finding 2: Larger early releases to containment result from early primary pressure boundary failure
- Set of accident scenarios dominated by low pressure accident sequences
- NUREG-1465 prescribed a larger number of high pressure scenarios Key Finding 3: Releases to containment significantly reduced if primary pressure boundary remains intact
- Low pressure scenarios lead to more significant releases to containment
- Evolution of severe accident modeling state-of-art since NUREG-1465 (e.g., SOARCA) 13 of 70
High Burnup and Extended Enrichment Impact on Source Term Core Types:
ORNL/TM-2023/1833 (1) 60 GWd/MTU LEU (2) 80 GWd/MTU LEU ML210888336 (3) 60 GWd/MTU HALEU (4) 80 GWd/MTU HALEU Time region of interest Burnup and enrichment do not significantly Increased burnup and enrichment does not change decay heat after reactor shutdown strongly impact in-containment source term 14 of 70
Impact of Accident Scenarios on In-containment Source Term Reference Hot leg creep rupture enabled No HLCR Hot leg creep rupture disabled Accident progression and in- Primary pressure boundary failure during critical containment source terms different accident phases is a significant factor in accident across accident sequences progression and in-containment source term 15 of 70
In-Containment Source Term Differences Gap Release Early In-vessel Late In-vessel Ex-vessel Report 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 Phase Duration 0.70 0.50 6.7 1.5 44.6 10.0 3.1 3.0 BWR Noble Gases 0.016 0.050 0.95 0.95 0.005 0.0 0.011 0.0 Halogens 0.005 0.050 0.71 0.25 0.16 0.010 0.017 0.30 Alkali Metals 0.005 0.050 0.32 0.20 0.021 The NRC has 0.010 determined 0.009(SECY 0.35 302, December 19, 1994) that Te Group 0.003 0.0 0.56 0.050 0.19 0.005 0.003 0.25 design basis source terms will not Gap Release Early In-vessel Late In-vessel Ex-vessel include the ex-vessel and late in-Phase vessel phases.
Duration 1.3 0.50 4.0 1.3 24.0 10.0 1.9 2.0 PWR Noble Gases 0.026 0.050 0.93 0.95 0.010 0.0 0.018 0.0 Halogens 0.007 0.050 0.58 0.35 0.031 0.10 0.020 0.25 Alkali Metals 0.003 0.050 0.50 0.25 0.013 0.10 0.015 0.35 Te Group 0.006 0.0 0.55 0.050 0.019 0.005 0.005 0.25
- Longer in-vessel phase durations due to progressive core degradation 16 of 70
In-Containment Source Term Differences Gap Release Early In-vessel Late In-vessel Ex-vessel Report 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 Phase Duration 0.70 0.50 6.7 1.5 44.6 10.0 3.1 3.0 BWR Noble Gases 0.016 0.050 0.95 0.95 0.005 0.0 0.011 0.0 Halogens 0.005 0.050 0.71 0.25 0.16 0.010 0.017 0.30 Alkali Metals 0.005 0.050 0.32 0.20 0.021 The NRC has 0.010 determined 0.009(SECY 0.35 302, December 19, 1994) that Te Group 0.003 0.0 0.56 0.050 0.19 0.005 0.003 0.25 design basis source terms will not Gap Release Early In-vessel Late In-vessel Ex-vessel include the ex-vessel and late in-Phase vessel phases.
Duration 1.3 0.50 4.0 1.3 24.0 10.0 1.9 2.0 PWR Noble Gases 0.026 0.050 0.93 0.95 0.010 0.0 0.018 0.0 Halogens 0.007 0.050 0.58 0.35 0.031 0.10 0.020 0.25 Alkali Metals 0.003 0.050 0.50 0.25 0.013 0.10 0.015 0.35 Te Group 0.006 0.0 0.55 0.050 0.019 0.005 0.005 0.25
- Longer in-vessel phase durations due to progressive core degradation
- Progressive releases to containment due to enhanced reactor coolant system modeling 17 of 70
In-Containment Source Term Differences Gap Release Early In-vessel Late In-vessel Ex-vessel Report 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 Phase Duration 0.70 0.50 6.7 1.5 44.6 10.0 3.1 3.0 BWR Noble Gases 0.016 0.050 0.95 0.95 0.005 0.0 0.011 0.0 Halogens 0.005 0.050 0.71 0.25 0.16 0.010 0.017 0.30 Alkali Metals 0.005 0.050 0.32 0.20 0.021 The NRC has 0.010 determined 0.009(SECY 0.35 302, December 19, 1994) that Te Group 0.003 0.0 0.56 0.050 0.19 0.005 0.003 0.25 design basis source terms will not Gap Release Early In-vessel Late In-vessel Ex-vessel include the ex-vessel and late in-Phase vessel phases.
Duration 1.3 0.50 4.0 1.3 24.0 10.0 1.9 2.0 PWR Noble Gases 0.026 0.050 0.93 0.95 0.010 0.0 0.018 0.0 Halogens 0.007 0.050 0.58 0.35 0.031 0.10 0.020 0.25 Alkali Metals 0.003 0.050 0.50 0.25 0.013 0.10 0.015 0.35 Te Group 0.006 0.0 0.55 0.050 0.019 0.005 0.005 0.25
- Longer in-vessel phase durations due to progressive core degradation
- Progressive releases to containment due to enhanced reactor coolant system modeling
- Larger release magnitudes prior to lower head failure due to early loss of the primary pressure boundary (by safety relief valve seizure and hot leg creep rupture) 18 of 70
In-Containment Source Term Release Rates Gap Release Early In-vessel Late In-vessel Ex-vessel Report 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 Phase Duration 0.70 0.50 6.7 1.5 44.6 10.0 3.1 3.0 BWR Noble Gases 0.023 0.10 0.14 0.63 0.0001 0.0 0.003 0.0 Halogens 0.007 0.10 0.11 0.17 0.004 0.001 0.006 0.100 Alkali Metals 0.007 0.10 0.047 0.13 0.0006 The NRC has determined 0.001 0.003(SECY 0.12 Te Group 0.005 0.0 0.091 0.033 0.005 302, December 19,0.001 0.0005 1994) that 0.083 Gap Release Early In-vessel design basis source terms Ex-vessel Late In-vessel will not Phase include the ex-vessel and late in-Duration 1.3 0.50 4.0 1.3 24.0 10.0vessel phases.
1.9 2.0 PWR Noble Gases 0.019 0.10 0.21 0.73 0.0008 0.0 0.009 0.0 Halogens 0.003 0.10 0.16 0.27 0.001 0.010 0.009 0.12 Alkali Metals 0.001 0.10 0.15 0.19 0.0005 0.010 0.008 0.17 Te Group 0.003 0.0 0.15 0.038 0.0008 0.0005 0.002 0.12
- Assumes uniform release rate across the entire phase duration 19 of 70
In-Containment Source Term Release Rates Gap Release Early In-vessel Late In-vessel Ex-vessel Report 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 Phase Duration 0.70 0.50 6.7 1.5 44.6 10.0 3.1 3.0 BWR Noble Gases 0.023 0.10 0.14 0.63 0.0001 0.0 0.003 0.0 Halogens 0.007 0.10 0.11 0.17 0.004 0.001 0.006 0.100 Alkali Metals 0.007 0.10 0.047 0.13 0.0006 The NRC has determined 0.001 0.003(SECY 0.12 Te Group 0.005 0.0 0.091 0.033 0.005 302, December 19,0.001 0.0005 1994) that 0.083 Gap Release Early In-vessel design basis source terms Ex-vessel Late In-vessel will not Phase include the ex-vessel and late in-Duration 1.3 0.50 4.0 1.3 24.0 10.0vessel phases.
1.9 2.0 PWR Noble Gases 0.019 0.10 0.21 0.73 0.0008 0.0 0.009 0.0 Halogens 0.003 0.10 0.16 0.27 0.001 0.010 0.009 0.12 Alkali Metals 0.001 0.10 0.15 0.19 0.0005 0.010 0.008 0.17 Te Group 0.003 0.0 0.15 0.038 0.0008 0.0005 0.002 0.12
- Assumes uniform release rate across the entire phase duration
- Release rates (release fraction/hour) are generally smaller 20 of 70
In-Containment Source Term Release Rates Gap Release Early In-vessel Late In-vessel Ex-vessel Report 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 2023 NUREG-1465 Phase Duration 0.70 0.50 6.7 1.5 44.6 10.0 3.1 3.0 BWR Noble Gases 0.023 0.10 0.14 0.63 0.0001 0.0 0.003 0.0 Halogens 0.007 0.10 0.11 0.17 0.004 0.001 0.006 0.100 Alkali Metals 0.007 0.10 0.047 0.13 0.0006 The NRC has determined 0.001 0.003(SECY 0.12 Te Group 0.005 0.0 0.091 0.033 0.005 302, December 19,0.001 0.0005 1994) that 0.083 Gap Release Early In-vessel design basis source terms Ex-vessel Late In-vessel will not Phase include the ex-vessel and late in-Duration 1.3 0.50 4.0 1.3 24.0 10.0vessel phases.
1.9 2.0 PWR Noble Gases 0.019 0.10 0.21 0.73 0.0008 0.0 0.009 0.0 Halogens 0.003 0.10 0.16 0.27 0.001 0.010 0.009 0.12 Alkali Metals 0.001 0.10 0.15 0.19 0.0005 0.010 0.008 0.17 Te Group 0.003 0.0 0.15 0.038 0.0008 0.0005 0.002 0.12
- Assumes uniform release rate across the entire phase duration
- Release rates (release fraction/hour) are generally smaller
- Larger Te group release magnitude prior to lower head failure 21 of 70
Deep Dive 22 of 70
In-containment Source Term
- In-containment source term characterizes total radioactive inventory in containment
- In-containment source term combines deposited, airborne, and escaped radionuclide inventories
- MELCOR simulations can track deposited and airborne masses separately
- This additional information not used in determining in-containment source term
- Radionuclide removal mechanisms accounted for in downstream calculations with RADTRAD 10 CFR 50.2 - Source term refers to the magnitude and mix of the radionuclides released from the fuel, expressed as fractions of the fission product inventory in the fuel, as well as their physical and chemical form, and the timing of their release 23 of 70
Alternative Source Term "Alternative Radiological Source Terms for Evaluating Design Basis Accidents at Nuclear Power Reactors," Regulatory Guide 1.183 Alternative Source Term (AST) must be based on major accidents involving a substantial meltdown of the Fulfills Criteria core AST must be represented in terms of the quantities, times, rates, chemical speciation for fission product SAND2023-01313 Peer Review Fulfills Criteria release into containment AST must not based on a single accident scenario but characterizes a spectrum of credible severe accident Fulfills Criteria events Assessment (ERI/NRC 23-201)
Fulfills AST must have a defensible technical basis Criteria Fulfills AST must be peer reviewed Criteria 24 of 70
Process for Source Term Development BWR and PWR core damage accident scenario identification Develop radionuclide inventory and decay heat using the SCALE code package Perform accident progression and source term analyses using MELCOR Develop statistically representative source term across all accident scenarios and BWR/PWR plants 25 of 70
Evolution from SAND2011-0128
- Overall SAND2023-01313 methodology is consistent with SAND2011-0128
- Focus on assessing impact of HBU/HALEU fuel on alternative source term
- Key areas of consistency between the studies are
- Nuclear power plants modeled
- Accident scenarios simulated
- Radionuclide chemical classes represented
- Radiological release phases first identified in NUREG-1465 are defined using SAND2011-0128 criteria
- Representative release phase source terms and timings are statistical median values 26 of 70
Extending SAND2011-0128 Source Terms
- Plants analyzed - from SAND2011-0128
- BWR: Mark I containment (Peach Bottom) and Mark III containment (Grand Gulf)
- PWR: Ice Condenser containment (Sequoyah) and large-dry containment (Surry)
- Accident scenarios analyzed - from SAND2011-0128
- PWR: SBLOCA, LBLOCA, STSBO Phase Onset Criteria - from SAND2011-0128 End Criteria - from SAND2011-0128 Gap Release RPV water level below top of active fuel Release of 5% of initial, total Xe inventory from fuel Early In-Vessel Release of 5% of initial, total Xe inventory from fuel Lower Head Failure Ex-Vessel Lower Head Failure 95% of total ex-vessel Cs releases Late In-Vessel Lower Head Failure 95% of total late in-vessel Cs releases Peer Review Findings
- Ex-vessel and late in-vessel phase criteria have limited technical justification
- NRC determined (SECY-94-302, December 19, 1994) design basis source terms will not include ex-vessel and late in-vessel phases 27 of 70
SAND2023-01313 Accident Selection
- NUREG-1560 - Individual Plant Examination Program
- Based on SAND2011-0128 accident selection
- Consistent with NUREG 1560 IPE results
- Representative accident sequences similar to those selected for NUREG-1465
- Provides coverage of all major sequences
- Incorporates SBO, LOCA and ATWS scenarios and range of mitigating system operation Peer Review Findings
- More recent PRA studies may potentially show different core damage contributors
- For the intended applications the scenarios used in the current [SAND2023-01313]
appropriate with regards to the progression of severe accidents, radionuclide release and transport.
28 of 70
Peach Bottom Accident Scenarios 7 SBOs RCIC operation RPV at low Early failures Containment Status Initiating Events RPV Status Coolant Injection
- 4 immediate loss of
- 3 scenarios credited pressure prior to
- Drywell liner melt-DC power RCIC lower head failure through
- 3 with prolonged DC No coolant
- 8 low pressure
- Torus overpressure power
- 6 scenarios had no credit for any coolant pressure until Late Failure injection system lower head failure
- High-temperature penetration failure
- 1 high pressure scenario Containment failures occurred at or after lower head failure 29 of 70
Grand Gulf Accident Scenarios 5 SBOs RCIC operation RPV at low Early failures Containment Status Initiating Events RPV Status Coolant Injection 1 ATWS
- 3 scenarios credited pressure prior to
- High containment
- 6 low pressure pressure (ATWS)
- Recirculation injection scenarios line break RPV at high Late Failure
- 4 scenarios had no credit for any coolant pressure until
- High containment injection system lower head failure pressure
- 1 high pressure scenario Containment failures generally occurred at or after lower head failure 30 of 70
Surry Accident Scenarios 2 SBOs 1 scenario RPV at low Early failures Containment Status Initiating Events RPV Status Coolant Injection 3 LOCAs crediting coolant pressure prior to
- Hydrogen injection lower head failure deflagration 4 scenarios with
- 3 low pressure Late Failure scenarios no coolant
- All SBOs exhibit hot-leg creep rupture prior to lower head failure Containment failures occurred at or after lower head failure 31 of 70
Sequoyah Accident Scenarios 2 SBOs 5 scenarios RPV at low Early failures Containment Status Initiating Events RPV Status Coolant Injection 5 LOCAs crediting coolant pressure prior to
- Hydrogen injection lower head failure deflagration 2 scenarios with
- 2 low pressure Late Failure scenarios no coolant
- All SBOs and RCP seal LOCAs exhibit hot-leg creep rupture prior to lower head failure Containment failures occurred at or after lower head failure 32 of 70
BWR Radionuclide Inventories 60 GWd/MTU - 80 GWd/MTU - 60 GWd/MTU - 80 GWd/MTU -
Class (kg) 5 wt% Enrichment 5 wt% Enrichment 10 wt% Enrichment 10 wt% Enrichment BWR Mark I - Peach Bottom Noble Gases 1323.99 1848.13 (+40%) 1280.34 (-3%) 1790.12 (+35%)
Halogens 52.83 73.70 (+40%) 49.41 (-6%) 69.53 (+32%)
Alkali Metals 748.78 980.11 (+31%) 817.97 (+9%) 1082.33 (+45%)
Te Group 142.94 195.01 (+36%) 139.99 (-2%) 190.51 (+33%)
Ba/Sr Group 551.99 763.09 (+38%) 586.41 (+6%) 814.05 (+47%)
Ru Group 1058.01 1598.56 (+51%) 919.02 (-13%) 1374.61 (+30%)
Mo Group 973.05 1305.64 (+34%) 1007.92 (+4%) 1364.59 (+40%)
Lanthanides 2943.70 3702.34 (+26%) 2922.84 (-1%) 3686.46 (+25%)
Ce Group 2469.33 2916.84 (18%) 2559.90 (+4%) 3107.02 (+26%)
- percent differences shown relative to reference core (60 GWd/MTU - 5 wt% enrichment)
- all fuel bundles assumed to reach reported burnup Peer Review Findings
- Radionuclide class mass differences are not equal to radionuclide class activity differences for the considered enrichments and burnups
- Unlikely that siting calculations would be significantly impact by burnup 33 of 70
PWR Radionuclide Inventories 60 GWd/MTU - 80 GWd/MTU - 60 GWd/MTU - 80 GWd/MTU -
Class (kg) 5 wt% Enrichment 5 wt% Enrichment 8 wt% Enrichment 8 wt% Enrichment PWR with Large-Dry Containment - Surry Noble Gases 740.20 987.15 (+33%) 717.66 (-3%) 959.00 (+30%)
Halogens 29.31 39.35 (+34%) 27.44 (-6%) 37.06 (+26%)
Alkali Metals 421.27 537.41 (+28%) 455.26 (+8%) 584.21 (+39%)
Te Group 74.62 99.01 (+33%) 73.02 (-2%) 96.81 (+30%)
Ba/Sr Group 305.28 401.76 (+32%) 323.92 (+6%) 428.01 (+40%)
Ru Group 559.35 807.23 (+44%) 487.92 (-13%) 701.18 (+25%)
Mo Group 530.59 689.06 (+30%) 546.71 (+3%) 714.95 (+35%)
Lanthanides 1035.01 1396.16 (+35%) 1048.46 (+1%) 1409.24 (+36%)
Ce Group 1535.14 1780.67 (+16%) 1599.41 (+4%) 1903.19 (+24%)
- percent differences shown relative to reference core (60 GWd/MTU - 5 wt% enrichment)
- all fuel bundles assumed to reach reported burnup Peer Review Findings
- Radionuclide class mass differences are not equal to radionuclide class activity differences for the considered enrichments and burnups
- Unlikely that siting calculations would be significantly impact by burnup 34 of 70
Iodine and Cesium Chemical Form
- 5% Iodine inventory is gaseous (I2 and other organic iodides)
- 95% Iodine inventory is CsI
- Remaining Cs inventory assumed volatile (CsOH)
- SAND2023-01313 - consistent with SOARCA
- Of Cesium not forming CsI
- 5% assumed to form CsOH
- 95% assumed to form Cs2MoO4 Peer Review Findings
- Uncertainty in Iodine speciation persists despite experimental studies (FPT3, DF-4, and BECARRE)
- Fukushima Daiichi post-accident analyses confirm assumption that Cs2MoO4 is dominant chemical form of Cs
- Recommended consideration of/validation against French CEA HBU VERDON tests 35 of 70
Other Analysis Assumptions
- In-containment source term does not consider impact of
- Variation in the gap inventory at the start of the accident
- Fraction of aerosolized iodine in containment
- Radionuclide removal and retention in containment
- Source term analyses based on current state-of-the-art
- Latest major code version - MELCOR 2.2
- Majority of modeling best-practices established under SOARCA
- Some modeling best-practices have evolved since SOARCA
- Time-at-temperature fuel rod failure model uses default time-at-temperature fuel rod lifetime curve
- UO2 and ZrO2 liquefaction temperatures reduced to 2479 K to account for material interactions
- Failure temperature of oxidized fuel rods have been reduced to 2479 K 36 of 70
Other Analysis Assumptions
- Relative contribution of accident sequences to total BWR/PWR CDF not changed by cores with extended enrichment HBU
- Dominant uncertainty from range of possible accidents that could be realized (i.e.,
aleatory uncertainty)
- Phenomenological (or epistemic) uncertainty not incorporated into BWR/PWR in-containment source terms
- Impact of phenomenological uncertainties considered through sensitivity calculations
- Key phenomena identified in a PIRT study are investigated through sensitivity studies
- Containment removal mechanisms not credited
- Some removal mechanisms, such as containment sprays, are incorporated in downstream RADTRAD calculations
- Suppression pool scrubbing not credited
- Release fractions (source terms) below 1x10-6 considered negligibly small and truncated 37 of 70
Non-Parametric Statistical Analysis
- Non-parametric bootstrap methodology used to determine statistically representative source term across accident scenarios
- Can be applied to data that follow any distribution
- Utilizes repeated re-sampling (bootstrapping) of data
- Estimates empirical cumulative distribution function (ECDF) of a given quantity of interest (QoI) 50th Percentile
- Representative source term is the median (50th percentile) estimate from the ECDF
- Equally weights all simulations *Dashed colored lines illustrate confidence intervals spanning +/- standard deviation () at each percentile Peer Review Finding
- Representative source term based on median value appropriate to avoid introducing bias from potential outliers 38 of 70
Bootstrap Procedure Calculate Compute Interpolate Quantity of sample / of each to obtain Generate N Interest percentiles percentile QoIs ECDF samples of (QoI) from k size k simulations
- Incorporates variability due to different plants and accident scenarios in representative source term
- Bounds on empirical cumulative distribution function (ECDF) characterize sampling uncertainty 39 of 70
Results and Discussion 40 of 70
Restating Key Aspects of the Analysis
- Objective
- Extend the NUREG-1465 alternative source term to address LWRs with cores designed to utilize HBU fuel with varied fuel enrichments
- Plants analyzed
- BWR: Mark I containment (Peach Bottom) and Mark III containment (Grand Gulf)
- PWR: Ice Condenser containment (Sequoyah) and Large-dry containment (Surry)
- Reactor cores analyzed
- 1. Core average burnup of 60GWd/MTU for enrichment of 5 wt%
- 2. Core average burnup of 80GWd/MTU for enrichment of 5 wt%
- 3. Core average burnup of 60GWd/MTU for enrichment of 8 wt% (peak 10 wt% for BWRs)
- 4. Core average burnup of 80GWd/MTU for enrichment of 8 wt% (peak 10 wt% for BWRs)
- Accident scenarios analyzed
- PWR: SBLOCA, LBLOCA, STSBO Phase Onset Criteria End Criteria Gap Release RPV water level below top of active fuel Release of 5% of initial, total Xe inventory from fuel Early In-Vessel Release of 5% of initial, total Xe inventory from fuel Lower Head Failure Ex-Vessel Lower Head Failure 95% of total ex-vessel Cs releases Late In-Vessel Lower Head Failure 95% of total late in-vessel Cs releases 41 of 70
Revisiting the Impact of Reactor Core on In-containment Source Term Early In-vessel Early In-vessel Core Type (1) (2) (3) (4) Core Type (1) (2) (3) (4)
Phase Duration 4.0 3.8 4.2 3.8 Phase Duration 6.7 6.3 6.5 6.3 BWR Noble Gases 0.94 0.96 0.94 0.94 PWR Noble Gases 0.93 0.92 0.91 0.92 Halogens 0.71 0.71 0.76 0.71 Halogens 0.57 0.56 0.57 0.58 Alkali Metals 0.31 0.31 0.31 0.26 Alkali Metals 0.5 0.5 0.5 0.51 Core Types:
(1) 60 GWd/MTU LEU, (2) 80 GWd/MTU LEU (3) 60 GWd/MTU HALEU (4) 80 GWd/MTU HALEU An increase in burnup and enrichment does not strongly impact the in-containment source term 42 of 70
BWR In-containment Source Term Evolution Gap Release Early In-vessel Study 2023 2011 NUREG-1465 2023 2011 NUREG-1465 Phase Duration (hr) 0.70 0.16 0.50 6.7 8.0 1.5 Noble Gases 0.016 0.008 0.050 0.95 0.96 0.95 Halogens 0.005 0.002 0.050 0.71 0.47 0.25 Alkali Metals 0.005 0.002 0.050 0.32 0.13 0.20 Te Group 0.003 0.002 0.0 0.56 0.39 0.050 Ba/Sr Group 0.0006 0.0 0.0 0.005 0.005 0.020 Ru Group <1.0e-6 0.0 0.0 0.006 0.003 0.003 Mo Group 1.9E-05 0.0 0.0 0.12 0.020 0.003 Lanthanides <1.0e-6 0.0 0.0 <1.0e-6 <1.0e-6 0.0002 Ce Group <1.0e-6 0.0 0.0 <1.0e-6 <1.0e-6 0.0005
- SAND2023-01313 and SAND2011-0128 utilized MELCOR
- Accident scenarios and modeling best-practices lead to tendency for increased early in-vessel halogen releases
- Peach Bottom and Grand Gulf modeling best-practices in SAND2023-01313 represent improvements due to SOARCA 43 of 70
BWR In-Containment Source Terms Consistent with SOARCA
- SOARCA found limited in-vessel halogen retention during early-in vessel phase PB SOARCA halogen releases (STSBO without RCIC blackstart)
- In-containment source terms reported in SAND2023-01313 characterize total radioactive inventory in containment 44 of 70
PWR In-containment Source Term Evolution Gap Release Early In-vessel Study 2023 2011 NUREG-1465 2023 2011 NUREG-1465 1.3 0.22 0.50 4.0 4.5 1.3 Phase Duration 0.026 0.017 0.050 0.93 0.94 0.95 Noble Gases 0.007 0.004 0.050 0.58 0.37 0.35 Halogens 0.003 0.003 0.050 0.50 0.23 0.25 Alkali Metals 0.006 0.004 0.0 0.55 0.30 0.050 Te Group 0.001 0.0006 0.0 0.002 0.004 0.020 Ba/Sr Group
<1.0e-6 0.0 0.0 0.008 0.006 0.003 Ru Group 2.0E-05 0.0 0.0 0.15 0.080 0.003 Mo Group
<1.0e-6 0.0 0.0 <1.0e-6 <1.0e-6 0.0002 Lanthanides
<1.0e-6 0.0 0.0 <1.0e-6 <1.0e-6 0.0005 Ce Group
- SAND2023-01313 and SAND2011-0128 utilized MELCOR
- Accident scenarios and modeling best-practices lead to tendency for increased early in-vessel halogen releases
- Surry and Sequoyah modeling best-practices in SAND2023-01313 represent improvements due to SOARCA 45 of 70
PWR In-Containment Source Terms Consistent with SOARCA
- SOARCA found limited halogen in-vessel retention after hot leg creep rupture SQN SOARCA halogen releases (LTSBO)
- In-containment source terms reported in SAND2023-01313 characterize total radioactive inventory in containment 46 of 70
In-containment Release Rate Evolution BWR PWR Early In-vessel Early In-vessel NUREG- NUREG-2023 2011 2023 2011 Study 1465 1465 Noble Gases 0.14 0.12 0.63 0.21 0.21 0.73 Halogens 0.11 0.059 0.17 0.16 0.082 0.27 Alkali Metals 0.047 0.016 0.13 0.15 0.051 0.19 Te Group 0.091 0.049 0.033 0.15 0.067 0.038 Ba/Sr Group 0.0009 0.0006 0.013 0.0007 0.0009 0.015 Ru Group 0.0009 0.0003 0.002 0.002 0.001 0.002 Mo Group 0.017 0.003 0.002 0.045 0.018 0.002 Lanthanides <1.0e-6 <1.0e-6 0.0001 <1.0e-6 <1.0e-6 0.0002 Ce Group <1.0e-6 <1.0e-6 0.0003 <1.0e-6 <1.0e-6 0.0004 Reported as [release fraction/hour]
47 of 70
Fuel Thermal Conductivity Sensitivity Increased burnup leads to decrease of fuel thermal conductivity Sensitivity Fuel Thermal Results shown for Case Conductivity [W/m-K] Surry Reference 4.92 Reduced 2.02 Low 0.2 No impact from variation of fuel thermal conductivity 48 of 70
In-vessel Particulate Debris Porosity Very high burnups have been postulated to promote disintegration of the fuel material Three sensitivity cases to assess impact on in-containment source term Sensitivity In-Vessel Results shown for Case Particulate Debris Peach Bottom Porosity Reference 0.4 High 0.6 Low 0.2 No impact from variation of in-vessel particulate debris porosity 49 of 70
Diameter of In-vessel Particulate Debris Sensitivity Higher burnups result in a greater degree of fuel breakup Sensitivity In-core Particulate Lower Plenum Debris Diameter Particulate Debris
[cm] Diameter [cm]
Reference 1.0 0.2 High 1.5 0.5 Results shown for Surry Low 0.5 0.1 Variation in particulate debris diameter impacts in-containment source term Impact smaller than changes across accident scenarios 50 of 70
Particulate Debris Falling Velocity Sensitivity Particulate debris sizes could impact particulate debris fall velocity into lower plenum Sensitivity In-Vessel In-Vessel Results shown for Particulate Particulate Surry Debris Fall Debris Fall Velocity [m/s] Velocity [m/s]
Peach Bottom Surry Reference 0.94 0.094 Low 0.094 0.064 No impact on source term due to variation in particulate debris fall velocity 51 of 70
Fuel Relocation Temperature Sensitivity Material interactions can cause early failure of fuel assemblies and other core components
- MELCOR uses either the interactive materials model or eutectics model to represent material interactions Sensitivity Fuel Relocation Results shown for Temperature [K] Surry Reference 2479 High 2728 Low 2230 Eutectics Eutectics model Material interactions that cause early fuel failure and can impact accident progression timings and in-containment source terms based on SOARCA uncertainty studies 52 of 70
Fuel Rod Lifetime Sensitivity Fuel assemblies at high temperatures exhibit early failures
- Early failures captured in MELCOR simulations using a lifetime function Sensitivity Fuel Rod Lifetime Model Results shown for Reference Default time-at-temperature model Peach Bottom Increased Lifetime function that accrues damage from 22.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> to 20 Lifetime minutes at temperatures from 2100K - 2600K Reduced Lifetime Lifetime function that accrues damage from 1.67 hours7.75463e-4 days <br />0.0186 hours <br />1.107804e-4 weeks <br />2.54935e-5 months <br /> to 3.3 minutes at temperatures from 2100K - 2600K SOARCA Lifetime Lifetime function that accrues damage from 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> to 5 minutes at temperatures from 2100K - 2600K No impact due to variation of the fuel rod lifetime modeling on source term Oxidized fuel assembly temperature failure model generally dominates 53 of 70
Hot Leg Creep Rupture Sensitivity Key insight from SOARCA is potential for induced RPV pressure boundary failures
- Severe accident conditions lead to high pressure and temperature conditions at RPV boundary Results shown for
- Thermally-induced hot leg creep rupture found likely for Surry PWRs
- BWRs exhibited thermally-induced seizure of cycling SRVs Sensitivity RPV Induced Pressure Boundary Failure Modeling Reference Hot leg creep rupture enabled No HLCR Hot leg creep rupture disabled Significant increase in early in-vessel source term for induced RPV failure for SBOs 54 of 70
In-containment Source Term Variability Peer Review Finding
- Potential for combined effects of various sensitivity studies to be larger than separate effects Nonlinear processes in severe accidents tend to limit amplification of response variability in multi-parameter sensitivity studies such that single scenario variability is less than variation across scenarios In-containment source term variation dominated by variation across sequences 55 of 70
Summary
- Increased burnup or extended enrichment does not significantly impact source term
- Most significant variation in source term arises due to differences between accident scenarios
- Status of RPV has significant impact on early in-vessel releases
- Low pressure scenarios exhibit more significant releases to containment
- NUREG-1465 prescribed larger number of high SAND2023-01313 pressure scenarios than SAND2023-01313
- Early in-vessel source term greatly reduced if RPV pressure boundary intact 56 of 70
Independent Peer Review 57 of 70
Focus of SAND2023-01313 Peer Review
- Review technical basis of SAND2023-01313 ERI/NRC 23-201
- Recommend improvements to SAND2023-01313
- Assess suitability of SAND2023-01313 source terms for regulatory applications 58 of 70
Peer Review Organization Panel Membership Panel Objectives
- Dr. Mohsen Khatib-Rahbar - Panel Chair
- Assess technical adequacy with respect to:
- Energy Research, Inc. (ERI)
- Overall analysis approach
- Specific applications of MELCOR to development
- Dr. Richard S. Denning of in-containment source terms
- Consultant
- Assess appropriateness of severe accident
- Mr. Jeff Gabor sequences selected
- Jensen Hughes
- Assess applied models and assumptions in terms of
- Dr. Didier Jacquemain
- Current understanding of severe accidents and
- Organization for Economic Co-operation and source terms Development/Nuclear Energy Agency (OECD/NEA)
- Adequacy considering available experimental data, and observations
- Dr. Luis E. Herranz
- Centro de Investigaciones Energéticas,
- Assess that source terms are Medioambientales y Tecnológicas (CIEMAT) representative, rather than conservative or bounding
- Dr. Yu Maruyama
- Japan Atomic Energy Agency (JAEA)
- Assess adequacy of documentation against
- Completeness of technical bases specification
- Approach to analysis of uncertainties 59 of 70
Peer Review Process
- Draft High Burnup Fuel Source Term Accident Sequence Analysis (Completed 2021)
- Virtual Meetings (began in 2022)
- 1. Briefing on the peer review objectives and the draft report by NRC and SNL
- Panelist review reports delivered to SNL
- Preliminary resolution of comments by SNL
- Preparation of the draft peer review report
- 2. Discussion of draft peer review report, comment resolution, and summary of unresolved comments
- Final resolution of comments by SNL
- Revision of High Burnup Fuel Source Term Accident Sequence Analysis report
- 3. Discussion of revised report, peer review panel findings, and conclusions
- Final High Burnup Fuel Source Term Accident Sequence Analysis report released (2023)
- Final peer review report released (2023) 60 of 70
Acceptability of the SAND2023-01313 Source Term
- [The peer review panel] endorses the approach taken in [SAND2023-01313]
- [SAND2023-01313] provides a defendable technical basis for the proposed source terms
- The peer review panel finds that the four nuclear power plants considered in the [SAND2023-01313] reasonably represent the U.S. nuclear fleet
- The spectrum of accidents is sufficient to satisfy the following stated attributes of an acceptable alternative accident source term (RG 1.183):
The accident source term must be expressed in terms of times and rates of appearance of radioactive fission products released into containment, the types and quantities of the radioactive species released, and the chemical forms of iodine released.
61 of 70
Qualities of the SAND2023-01313 Source Term
- Study is a significant technical improvement using state-of-the-art methods implemented in latest version of MELCOR
- In-containment source terms for HBU/HALEU fuels are representative MELCOR estimates, rather than conservative or bounding estimates
- No bias in the approach identified that could overestimate in-containment source terms
- Sensitivity studies documented in SAND2023-01313 valuable in supporting applications
- Sensitivities explored limitations in understanding of HBU/HALEU fuel response under severe accident conditions
- Results demonstrated impact of thermally induced (creep) depressurization of RCS for PWRs on in-containment source terms 62 of 70
Peer Review Report Recommendations
- Gap release phase incorporated into the early in-vessel phase
- The panel considers the current approach of separating the gap and early in-vessel release phases, a product of the simplified single channel treatment of the STCP models of circa 1980s that is reflected in the NUREG-1465 source terms, outdated. During severe accidents, the gap and in-vessel releases from the fuel overlap to the extent that it is not possible to truly separate the two as distinct phases. Therefore, it is recommended that the gap release be incorporated into the early in-vessel release phase.
- More appropriate to represent impact of burnup using core inventories for HBU expressed in terms of radiological activities
- The implication of comparison of mass inventories in kilogram [SAND2023-01313] is to incorrectly conclude that at higher fuel burnups, off-site doses would likely be substantially higher for high burnup fuels as the direct result of larger core mass inventories of radionuclides.
In fact, when compared on the basis of integrated radiological activity, there would not be any significant differences for the two levels of fuel burnup.
- Examples shown in the next presentation: Follow-on Calculations 63 of 70
Other Comments And Recommendations
- Panelists requested additional clarification (reflected in final report) that
- Containment bypass scenarios and air ingression not considered in development of tabular source terms
- Fission product removal mechanisms in containment not included in tabular source terms
- Captured in MELCOR simulations, but post-processed out of reported MELCOR source terms
- Peer reviewers acknowledged more recent PRA studies could have different contributors to core damage
- For the intended applications the scenarios used in the current [SAND2023-01313] appropriate with regards to the progression of severe accidents, radionuclide release and transport
- Panelists noted for most radionuclides no increase in activity with burnup sufficient to impact siting calculations
- Peer reviewers noted the uncertainty in Iodine speciation based on experiments (FPT3, DF-4, and BECARRE)
- Peer review noted that current Fukushima Daiichi post-accident analyses confirm the assumption that Cs2MoO4 is dominant chemical form of Cs
- Peer review panel considered the use of median estimates appropriate to avoid bias due to potential outliers 64 of 70
Other Comments And Recommendations Finally, even though tabular severe accident in-containment source terms provide a simplified tool for regulatory applications and analyses, it is important to recognize their limitations and the panel encourages the direct application of a state-of-the art severe accident code to specific issues when appropriate.
65 of 70
Fission Product Retention in Suppression Pools Gap Release Early In-vessel Including Suppression Pool Excluding Suppression Pool Including Suppression Pool Excluding Suppression Pool Release Category Inventory Inventory Inventory Inventory Noble Gases 0.016 0.016 0.95 0.95 Halogens 0.005 1.30E-06 0.71 0.06 Alkali Metals 0.005 1.20E-06 0.32 0.006 Te Group 0.003 <1.0e-6 0.56 0.038 Ba/Sr Group 0.0006 <1.0e-6 0.005 0.0003 Ru Group <1.0e-6 <1.0e-6 0.006 7.40E-06 Mo Group 1.90E-05 <1.0e-6 0.12 0.0001 Lanthanides <1.0e-6 <1.0e-6 <1.0e-6 <1.0e-6 Ce Group <1.0e-6 <1.0e-6 <1.0e-6 <1.0e-6 Peer Review Findings
- In-containment source terms should consider the impact of retention in suppression pools, especially for SBO scenarios that discharge directly into the suppression pool
- Estimates of retention in suppression pools provided in SAND2023-01313 could be used in regulatory guidance to establish suppression pool decontamination factors Significant effect of retention in suppression pool on key radionuclide groups 66 of 70
Upcoming Work 67 of 70
Cr-Coated ATF Concept
- Cr-coated ATF concept most similar to conventional fuels
- Thin, protective chromium coating on Zircaloy fuel cladding delays exothermic Zircaloy oxidation onset DRAFT
- Cr-coated analysis informed by ATF severe accident PIRT (NUREG/CR-7283) findings 68 of 70
FeCrAl ATF Concept
- FeCrAl ATF concept utilizes a novel fuel cladding material
- Substitution of Zr-based alloy with an FeCrAl alloy
- Intended to reduce both oxidation in the core and associated hydrogen production DRAFT
- FeCrAl analysis informed by ATF severe accident PIRT (NUREG/CR-7283) findings
- Sensitivity analyses deployed to interrogate FeCrAl cladding knowledge uncertainties 69 of 70
Thank you for your attention!
70 of 70
Backup Slides 71 of 70
Acronyms Acronym Definition Acronym Definition AC Alternating current NRC Nuclear Regulatory Commission ADS Automatic depressurization system ORNL Oak Ridge National Laboratories AFW Auxiliary Feedwater PB Peach Bottom AST Alternative source term PIRT Phenomena Identification and Ranking Table ATF Accident tolerant fuel PORV Pilot-operated relief valve ATWS Anticipated transient without scram PRA Probabilistic risk assessment BWR Boiling water reactor PRT Pressurizer relief tank CCFL Counter current flow PWR Pressurized water reactor CDF Core damage frequency QoI Quantity of interest DC Direct current RCIC Reactor core isolation cooling system ECCS Emergency core cooling system RCP reactor coolant pump ECDF Empirical cumulative distribution function RCS Reactor coolant system GG Grand Gulf RHR Residual heat removal HALEU High-assay low-enriched uranium SBLOCA Small-break loss of coolant accident HBU High burnup SBO Station blackout HLCR Hot leg creep rupture SOARCA State-of-the-Art Reactor Consequence Analyses HPCI High pressure coolant injection system SQN Sequoyah HPSI High-pressure safety injection SRV Safety relief valve LBLOCA Large-break loss of coolant accident STCP Source Term Code Package LEU Low-enriched uranium STSBO Short-term station blackout LOCA Loss of coolant accident SU Surry LPCI Low-pressure coolant injection TDAFW Turbine-driven auxiliary feedwater LPSI Low-pressure safety injection TID Technical information document LTSBO Long-term station blackout TMI-2 Three Mile Island Unit-2 LWR Light water reactor 72 of 70
Cs and I Releases
- SAND2011-0128 considered deposition of radionuclides on the lower head, leading to significantly decreased in-vessel phase releases.
- This consideration delays a significant fraction of radionuclide release to containment until after lower head failure during the ex-vessel phase (employed for Peach Bottom and Sequoyah)
- This practice is no longer considered appropriate, and was not employed in SAND2023-01313
- CsI (all original I inventory and ~10% original Cs inventory) transports readily from the primary system to containment during core damage due to the relatively large CsI vapor pressures at elevated primary system temperatures
- Consistent with Peach Bottom SOARCA results 73 of 70
NUREG-1465 Accident Selection
- Dominant sequences were chosen based on impact on source term
- BWRs are predominantly SBO/ATWS accidents PWR Plants Sequence Description BWR Plants Sequence Description Surry AG LOCA (hot leg), no containment hear removal systems Peach Bottom TC1 ATWS w/ reactor depressurized TMLB LOOP, no PCS and no AFWS TC2 ATWS w/ reactor pressurized V Intefacing system LOCA TC3 TC2 with wetwell venting S3B SBO with RCP seal LOCA TB1 SBO with battery depletion S2D-d SBLOCA, no ECCS and H2 combustion TB2 TB1 with containment failure at vessel failure S2D-b SBLOCA w/ 6" hole in containment S2E1 LOCA (2"), no ECCS and no ADS Oconee 3 TMLB SBO, no active ESF systems S2E2 S2E1 with basaltic concrete S1DCF LOCA (3"), no ESF systems V RHR pipe failure outside containment Sequoyah S3HF1 LOCA RCP, no ECCS, no CSRS w/ reactor cavity flooded TBUX SBO with loss of all DC power S3HF2 S3HF1 w/ hot leg induced LOCA LaSalle TB SBO with late containment failure 3HF2 S3HF1 w/ dry reactor cavity Grand Gulf TC ATWS early containment failure fails ECCS S3B LOCA (1/2") w/ SBO TB1 SBO with battery depletion TBA SBO induces hot leg LOCA - H2 burn fails containment TB2 TB1 w/ H2 burn fails containment ACD LOCA (hot leg), no ECCS no CS TBS SBO, no ECCS but reactor depressurized S3B1 SBO delayed 4 RCP seal failures, only steam driven AFW operates TBR TBS with AC recovery after vessel failure S3HF LOCA (RCP seal), no ECCS no CSRS S3H LOCA (RCP seal) no ECCS recirculation 74 of 70
High Burnup Fuel Source Term Accident Analysis Boiling-Water Reactor Follow-On Calculations ACRS Radiation Protection And Nuclear Materials Subcommittee Briefing November 16, 2023 Shawn Campbell and Michael Salay Fuel & Source Term Code Development Branch Division of Systems Analysis Office of Nuclear Regulatory Research 1
Background and Motivation
- The High Burnup (HBU) Peer Review panelists commented on the potential impact of the suppression pool on the containment source term.
- Table 5-16 of SAND2023-01313 provides the boiling-water reactor (BWR) containment release fractions including and excluding the suppression pool.
- Supplemental investigations following the peer review in BWRs:
- Investigate fission product concentration variation between different regions of the reactor system and containment since some scenarios and pathways bypass the suppression pool (e.g., main steam line).
- Modified the two (Peach Bottom, Grand Gulf) full-scale BWR input decks to better capture aerosol behavior in the containment and steam line.
- Performed a set of BWR source term calculations.
2
Source Term Methodology Early Vessel Late Fuel heat up Core FP Release and Containment MCCI/FP containment Breach containment Integrated Clad oxidation relocation Transport Leakage Release failure? failure?
Analysis (e.g., L3PRA, Containment SOARCA, In-Vessel Source Term Ex-Vessel (ST)
Fukushima)
Mechanistic Modeling FP Inventory FP removal mechanisms () Leak Rate ()
() e.g., Sprays/natural deposition User Specified User Specified Simplified Modeling Simplified Modeling Regulatory Source Containment Term C0 =ST/ Vol C t = C0 exp() FP release = C Dose Calculation Source Term (ST)
Analysis (for DBA) 3
Illustration of BWR Modeling Practices Area with refined modeling Peach Bottom 4
New BWR Main Steam Line (MSL) Modeling RPV steam dome For each BWR, the Main Steam Lines were broken up into finer nodalization Containment 2 to 4 SRVs 3 RCIC (MSL A) boundary per MSL 10 HPCI (MSL B)
MSIV #2 TSV to TCVs and turbine MSIV #1 open Vented to the The reported source term Condenser environment fractions in the steam line are averaged airborne fission All 4xMSLs modeled Vented to Condenser separately environment products in the green portion.
5
BWR Source Term (ST) Inventory Fractions - Early In-Vessel Pool Containment Steam Line Radionuclide RG1.183 (rev0) RG1.183 (rev1) SAND2023 (SAND2023 (SAND2023 (Preliminary Group Table 5-16) Table 5-16) Follow-on Calcs)
Noble Gases 9.50E-01 9.60E-01 9.50E-01 0.00E+00 9.50E-01 1.1E-03 Halogens 2.50E-01 5.40E-01 7.10E-01 6.50E-01 6.00E-02 5.1E-05 Alkali Metals 2.00E-01 1.40E-01 3.20E-01 3.10E-01 6.00E-03 1.3E-05 Te Group 5.00E-02 3.90E-01 5.60E-01 5.20E-01 3.80E-02 2.7E-05 Ba/Sr Group 2.00E-02 5.00E-03 5.00E-03 4.70E-03 3.00E-04 2.4E-07 Ru Group 3.00E-03 2.70E-03 6.00E-03 6.00E-03 7.40E-06 2.4E-07 Mo Group 3.00E-03 3.00E-02 1.20E-01 1.20E-01 1.00E-04 3.0E-06 Lanthanides 2.00E-04 <1.0e-6 <1.0e-6 <1.0e-6 <1.0e-6 1.0E-11 Ce Group 5.00E-04 <1.0e-6 <1.0e-6 <1.0e-6 <1.0e-6 8.4E-12 6
BWR Example Fission Product (FP) Concentrations (C0)
C0 =ST/ Vol FP Concentration (x 10-5) FP Concentration (x 10-5)
- 2023 Follow-on *2023 Follow-on calculations do not include calculations do not include FPs retained in the FPs retained in the suppression pool suppression pool 7
BWR/PWR Example Containment Concentrations Halogen (Iodine) x 1E-5 C0 =ST/ Vol 2023 Follow-on calculations do not include FPs retained in the BWR BWR PWR suppression pool Alkali Metals (Cesium) x 1E-5 Typical containment volumes from Figure 4.1-1 in NUREG/CR-6042, Rev. 2 8
Example HBU Inventories Radionuclide Group BWR (Bq) BWR (%) -> HBU PWR (Bq) PWR (%) -> HBU Halogens (I) 3.54E19 <1% 2.53E19 <1%
Alkali Metals (Cs) 4.46E18 +7% 3.09E18 +5%
Chalcogen (Te) 1.16E19 <1% 8.35E18 <1%
GE14 10x10 GE14 10x10 W 17x17 W 17x17 Core Avg. end of cycle BU 36.2 41.4 43.5 48.3 (MWd/MTU)
Avg. Assembly discharge BU 52.6 58.0 60.7 71.6 (MWd/MTU)
Initial Enrichment (%) 4.45 5.30 4.65 5.25 Power (MWt) 4016 4016 2893 2893 Cycle Length (months) 24 24 18 24 9
Conclusions and Next Steps
- Refined modeling provides better estimation of fission product distribution in the steamline.
- Concentration in the steam line is distinct from that of containment.
- Significant retention of fission products were predicted in the suppression pool.
- Preliminary investigation of fission product inventories show limited effect for high burnup/high-assay low-enriched uranium (HBU/HALEU)fuels.
- Potential application of MELCOR to inform better estimates of fission product removal mechanisms in the simplified tools for regulatory applications and analysis where appropriate.
10
Backup Slides 11
Acronyms Bq Becquerel MWt Megawatt thermal BWR boiling-water reactor PWR pressurized water reactor DBA design-basis accident RCIC reactor core isolation cooling FP fission product RG (NRC) regulatory guide GE General Electric RPV reactor pressure vessel HALEU high-assay low-enriched uranium SOARCA State-of-the-Art Reactor HBU high burnup Consequence Analyses HPCI high pressure coolant injection SRV safety relief valve MSIV main steam line isolation valve ST source term MSL main steam line TCV turbine control valve GWd/MTU gigawatt-days per metric ton of TSV turbine stop valve uranium W Westinghouse 12