ML19309D411
| ML19309D411 | |
| Person / Time | |
|---|---|
| Issue date: | 09/17/2019 |
| From: | Derek Widmayer Advisory Committee on Reactor Safeguards |
| To: | |
| Widmayer, D, ACRS | |
| References | |
| NRC-0569 | |
| Download: ML19309D411 (173) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Future Plant Designs Subcommittee Docket Number:
(n/a)
Location:
Rockville, Maryland Date:
Tuesday, September 17, 2019 Work Order No.:
NRC-0569 Pages 1-168 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005 (202) 234-4433
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com 1
1 2
3 DISCLAIMER 4
5 6
UNITED STATES NUCLEAR REGULATORY COMMISSIONS 7
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 8
9 10 The contents of this transcript of the 11 proceeding of the United States Nuclear Regulatory 12 Commission Advisory Committee on Reactor Safeguards, 13 as reported herein, is a record of the discussions 14 recorded at the meeting.
15 16 This transcript has not been reviewed, 17 corrected, and edited, and it may contain 18 inaccuracies.
19 20 21 22 23
1 UNITED STATES OF AMERICA 1
NUCLEAR REGULATORY COMMISSION 2
+ + + + +
3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
(ACRS) 5
+ + + + +
6 FUTURE PLANT DESIGNS SUBCOMMITTEE 7
+ + + + +
8 TUESDAY 9
SEPTEMBER 17, 2019 10
+ + + + +
11 ROCKVILLE, MARYLAND 12
+ + + + +
13 The Subcommittee met at the Nuclear 14 Regulatory Commission, Two White Flint North, Room 15 T2D10, 11545 Rockville Pike, at 8:30 a.m., Dennis 16 Bley, Chair, presiding.
17 18 COMMITTEE MEMBERS:
19 DENNIS BLEY, Chair 20 RONALD G. BALLINGER, Member 21 MICHAEL L. CORRADINI, Member 22 CHARLES H. BROWN, JR. Member 23 VESNA B. DIMITRIJEVIC, Member 24 WALTER L. KIRCHNER, Member 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
2 DAVID PETTI, Member*
1 HAROLD B. RAY, Member 2
JOY L. REMPE, Member 3
PETER RICCARDELLA, Member*
4 MATTHEW W. SUNSERI, Member*
5 6
DESIGNATED FEDERAL OFFICIAL:
7 WEIDONG WANG 8
9
- Present via telephone 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
3 P R O C E E D I N G S 1
8:30 a.m.
2 CHAIR BLEY: Good morning. The meeting 3
will now come to order. This is a meeting of the 4
Advisory Committee on Reactor Safeguards Subcommittee 5
on Future Plant Designs. I'm Dennis Bley, Chairman of 6
the Future Plan Designs Subcommittee.
7 ACRS members in attendance are Ron 8
Ballinger, Charlie Brown, Walt Kirchner, Harold Ray, 9
Joy Rempe, and Mike Corradini. And on the phone we 10 have three members, Dave Petti, Pete Riccardella, and 11 Matt Sunseri. Weidong Wang of the ACRS Staff is the 12 designated federal official for this meeting.
13 The purpose of today's meeting is to 14 review the draft report NRC Non-Light Water Reactor 15 Vision and Strategy, Volume 2, Fuel Performance 16 Analysis for Non-LWRs. And a couple of months ago we 17 had a precursor meeting where we discussed Volumes 1 18 and 3.
19 The Subcommittee will gather information, 20 analyze relevant issues and facts, and formulate for 21 post positions and actions, as appropriate.
22 This matter, along with those other two 23 volumes in this report series, is scheduled to be 24 addressed at the October full Committee meeting. We 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
4 expect to write a letter at that time.
1 The ACRS was established by statute and is 2
governed by the Federal Advisory Committee Act, FACA.
3 The Committee can only speak through its published 4
letter reports.
5 We hold meetings to gather information and 6
perform preparatory work that will support our 7
deliberations at a full Committee meeting.
8 The rules for participation in all ACRS 9
meetings, including today's, were announced in the 10 Federal Register on June 13, 2019.
11 The ACRS section of the US NRC public 12 website provides our charter, bylaws, agenda, letter 13 reports, and full transcripts of our meetings, 14 including the slides presented. The meeting notice 15 and agenda for this meeting were posted there.
16 As stated in the Federal Register notice 17 and in the public meeting notice posted to the 18 website, members of the public who desire to provide 19 written or oral input to the Subcommittee may do so, 20 and should contact the designated federal official 21 five days prior to the meeting.
22 Today's meeting is open to public 23 attendance and we've received no written statements or 24 requests to make an oral statement.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
5 We also set aside ten minutes in the 1
agenda for the spontaneous comments from members of 2
the public attending or listening to our meetings.
3 Today's meeting is being held with a 4
telephone bridge line, allowing for participation of 5
the public over the phone.
6 A transcript of today's meeting is being 7
kept. Therefore, we request that meeting participants 8
on the bridge line identify themselves when they 9
speak, and to speak with sufficient clarity and volume 10 so they can be readily heard.
11 Participants in the meeting should use the 12 microphones located throughout the meeting room when 13 addressing the Subcommittee. And we have one more 14 member I didn't mention earlier. Vesna Dimitrijevic 15 is with us.
16 At this time I ask that all attendees in 17 the room please silence all cellphones and other 18 devices that make noises, to minimize disruptions.
19 And I remind speakers at the front table 20 to turn on the microphone. I've been chastened to 21 tell all of the people in this room to be very careful 22 turning the microphones on and off because we will 23 have problems if we don't. So, be careful.
24 And for those of you who are new to this, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
6 the button to turn on the green light and turn it off 1
is nearest you, where it says, push, on the 2
microphone.
3 We will now proceed with the meeting. I 4
call on Kim Webber, Deputy Director with the Division 5
of Systems Analysis, Office of Research, to make 6
introductory remarks. Sorry, I can't talk. Kim.
7 MS. WEBBER: Yes, good morning to all of 8
you. And yeah, my name's Kim Webber. I'm the Deputy 9
Director of the Division of Systems Analysis in our 10 Office of Research.
11 Today we're here to present to you our co-12 development plans and strategies for Non-Light Water 13 Reactor Fuel Performance Analysis.
14 In addition, we're here to also answer 15 questions that you had for the May 1st and 16 September 4th Subcommittee meetings.
17 As part of the Office of Research biannual 18 ACRS review meeting held on September 4th, we 19 presented an overview of our division's activities.
20 And one the key messages that I was trying 21 to convey at that time, is that we're really in a 22 position where we have to be ready to enable the 23 regulatory offices to license advanced reactors.
24 And so, we believe that the introduction 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
7 and the Volume 1 through 3 go a long way to helping 1
the staff get ready to help them develop their 2
expertise to better understand the phenomenon that are 3
likely to be experienced in advanced reactor designs, 4
and so we think that these reports go a long way to 5
help achieving those goals. Let's see if I can do the 6
next slide.
7 Back in May 1st, we actually presented to 8
you information about three volumes, or three 9
documents, I should say. One was the introduction, 10 which is shown at the top of your slide on the left-11 hand side. And the introduction really laid out the 12 approach that we're taking is, some of the 13 considerations that we had to make in order to produce 14 the rest of the volumes.
15 And then Volume 1 is on the computer codes 16 for Non-Light Water reactor design basis analysis, and 17 Volume 3 is on severe accident, source term, and 18 accident progression.
19 At that time, Volume 2 and Volume 4 --
20 Volume 2 is on fuel performance -- that was a work-in-21 progress -- in addition to Volume 4, which, although 22 it says radiation protection on this slide, is really 23 licensing and citing dose assessment codes. And so, 24 there's a fourth volume.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
8 So, today we're here to talk about 1
Volume 2. The fourth volume -- actually, at the 2
September 4th meeting, I think you all requested to 3
see Volume 4, which is the licensing and citing dose 4
assessment codes report. And so, we're in progress, 5
in terms of setting up a meeting on that report.
6 MEMBER CORRADINI: And so, can I -- I'm 7
sorry.
8 MS. WEBBER: That' okay. You were first.
9 MEMBER CORRADINI: Since all of these 10 advanced reactors in theory will be using higher 11 enrichment, what about transport? What about front 12 and back.
13 MS.
WEBBER:
Yeah.
So, they're 14 actually -- and I'm channeling Richard, who always 15 reminds me -- there's going to be a Volume 5 that 16 covers front-and back-end code development activities.
17 We have not started on that yet, but there 18 will be one. And so, what we're looking for, as 19 Dennis mentioned, is after the fully Committee meeting 20 on October 3rd, we're looking for you to review the 21 introduction, Volume 1, Volume 2 and Volume 3, and get 22 a letter on that.
23 And then we'll come back with Volume 4 24 maybe in the spring. And then we will, at some later 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
9 date, come back with Volume 5, if you're interested in 1
that.
2 MEMBER CORRADINI: From the standpoint --
3 maybe this is the wrong time to ask it -- but from the 4
standpoint of readiness, wouldn't the back-end and the 5
front-end --
6 MS. WEBBER: Yes.
7 MEMBER CORRADINI: -- happen before the 8
reactors?
9 MS. WEBBER: Yes. Your point is very well 10 taken and it's something that we do think about and 11 we're going to start working on it. I think what we 12 wanted to do is try to get these volumes together 13 and -- actually, that fifth volume involves the same 14 people.
15 And so, we're trying to balance the 16 workload that we have too, but your point is well-17 taken.
18 MEMBER REMPE: Let's talk about Volume 4, 19 which we won't have before the October meeting. You 20 said it's in progress.
21 MS. WEBBER: It is.
22 MEMBER REMPE: Can you give us a clue of 23 what you're thinking of. I assume MACCS is involved?
24 MS. WEBBER: No, no. So, MACCS is covered 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
10 in Volume 3. So --
1 MEMBER REMPE: Okay, that's true. Yeah.
2 MS. WEBBER: Right. So --
3 MEMBER REMPE: But for citing -- what are 4
you thinking of?
5 MS. WEBBER: So, it codes like RASCAL, 6
like RADTRAD. I can't think of the other codes. I'm 7
looking to --
8 MEMBER REMPE: So, the reason that I'm 9
asking this question then, is that you were at the 10 research meeting where we talked about, think about 11 the future and combine -- maybe have a simplified 12 MACCS, instead of RADTRAN and RASCAL, to try and 13 reduce some costs in the long term. Too far out for 14 you to think about that?
15 MS. WEBBER: Well, I think that's 16 something that we have talked about and we will need 17 to have more discussions on. So, I think we met with 18 you on September 4th and that comment was raised.
19 I think we still need some work internally 20 to figure that one out, quite frankly.
21 MEMBER REMPE: Okay.
22 MS. WEBBER: And so, we'll take the 23 comment under consideration. And when we come back to 24 brief you on Volume 4, that will be something that we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
11 can talking about perhaps then.
1 MEMBER REMPE: Okay, thank you.
2 CHAIR BLEY: Just a reminder to everyone 3
here and that is listening, the ACRS only speaks 4
through its letters, so comments you've heard 5
discussed from the research meeting and our previous 6
meetings, really aren't suggestions of the Committee.
7 They're comments by individual members.
8 MS. WEBBER: Okay. Okay, and then I give 9
most of my -- oops, I'm pressing the wrong button. I 10 want to give most of my time to the other presenters.
11 So -- actually, just to follow up our Subcommittee 12 Chairman's comment that need to combine the codes with 13 RASCAL and MACCS was actually in our biennial letter.
14 So, it was a little higher than our off-the-wall 15 comments.
16 Okay, so I just want to introduce James 17 Corson and Lucas Kyriazidis. They're going to be the 18 presenters for Volume 2 and they'll discuss important 19 scenarios and phenomenology for Non-Light Water 20 Reactor fuels, code selection considerations, and 21 information gaps for the major types of fuels, 22 including TRISO metallic fuels.
23 And then after that, Steve Bajorek is 24 going to respond to some of the questions that you 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
12 raised at the May 1st meeting, and then also on the 1
September 4th meeting.
2 And so with that, I'd like to turn the 3
presentation over to James, unless you have any other 4
questions. Okay, and good. We know Ron's going to 5
take care of this.
6 DR. CORSON: All right. Good morning, 7
everyone, I'm James. With Lucas, I going to be 8
talking about our strategy for fuel performance for 9
Non-Light Water Reactors.
10 And before I begin, we also have Ken 11 Geelhood, Dr. Ken Geelhood, here from PNNL. Ken is 12 the lead developer for the FRAPCON, FRAPTRAN and FAST 13 codes. And he also has a lot of experience doing 14 technical reviews for licensing topical reports for 15 NRR. So, he'll be here to answer any more detailed 16 questions you might have. So, next slide.
17 Just a quick motivation of why we're here.
18 Kim already touched on all of this. There's a whole 19 lot of new reactor designs out there. We need to be 20 ready to license them. And in fact, Congress has 21 directed us to be ready to license them.
22 So, part of that in the Office of Research 23 is to work on the tools that can support licensing, 24 that can do confirmatory analysis. So, we're 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
13 specifically focused on fuel performance today. Next 1
slide.
2 You can see here, there's a lot of 3
different fuel designs out there, paired with 4
different coolants and geometries. So, there's quite 5
a number of different designs that have been proposed.
6 And this means that we need to be 7
judicious and narrow down what we're really focusing 8
on here, at least in the near term, based on industry 9
priorities. So, next slide.
10 This just is the same thing, but not it 11 shows that we're really prioritizing the TRISO fuel 12 and metallic fuel, because we expect those designs to 13 come in first.
14 So, if you've read the report, we do have 15 plans for the other fuel types. But we are not really 16 addressing them. They're not as high priority right 17 now. Of course, as we learn more from our regulatory 18 offices, that could change. But right now, this is 19 our plan.
20 MEMBER CORRADINI:
The plan looks 21 reasonable. Explain what the lines through the other 22 one means.
23 DR. CORSON: I was just getting there.
24 MEMBER CORRADINI: Oh, okay.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
14 DR. CORSON: So yeah, the last thing, we 1
had crossed out fuel salts simply because as far as 2
fuel performance goes, we are not covering that, 3
because we are covering fuel that's in a solid form.
4 I mean, that's really what our fuel performance codes 5
are designed to do.
6 So, fuel salts are being considered, but 7
they're handled in Volumes 1 and Volume 3, which 8
you've already heard about. So, those codes are going 9
to handle the important phenomena for fuel salts. So, 10 that's why it's crossed out here.
11 It's not that NRC is not considering them, 12 it's that we, for fuel performance, are not 13 considering them.
14 MEMBER PETTI: So, I have a question. How 15 are you going to handle the fission product release in 16 the fuel salts in Volume --
17 MEMBER REMPE: This is Dave Petti.
18 MEMBER PETTI: Yeah, sorry. Dave Petti.
19 DR. CORSON: So, it's covered in Volume 3.
20 And I think part of it is going to be MELCOR. I know 21 scale also, or -- sorry. Oak Ridge also has a 22 Thermochimica system which will be involved. Those 23 are both described in more detail in Volume 3.
24 MEMBER PETTI: Okay, thanks.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
15 DR. CORSON: Next slide.
1 CHAIR BLEY: Just a matter of form in your 2
report. They aren't crossed off in the report.
3 DR. CORSON: No.
4 CHAIR BLEY: Is there -- I don't remember 5
a discussion. Was there a discussion that points us 6
to Volume 3?
7 DR. CORSON: Yes. So, if you go to like 8
Section 3.8 or so.
9 (Background interruption.)
10 DR. CORSON: So, I think if go to like 11 Section 3.8, there is a brief section on molten salts 12 that specifically says it's covered in Volumes 1 13 and 3. And I believe it's also covered in the 14 Executive Summary, but I'm not positive about that.
15 So yeah, you do have to read through a bit 16 I think, but it does say in Section 3.8 that this is 17 covered elsewhere.
18 CHAIR BLEY: Okay, thanks.
19 DR. CORSON: So, just a quick overview of 20 what we're going to be talking about for the rest of 21 our presentation. We will begin with talking about 22 how we do fuel performance confirmatory analysis.
23 So yeah, we'll talk about what we do for 24 Light Water Reactors, and how that might apply to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
16 Non-Light Water Reactors going forward. After that, 1
we'll discuss the phenomenology that we need to be 2
able to model in our codes.
3 We'll talk about what criteria we use to 4
select which code we're going to use for Non-Light 5
Water Reactor confirmatory analysis. We'll talk about 6
its development status. Lucas will cover that. And 7
then I'll wrap things up by briefly discussing the 8
interface between fuel performance described in this 9
volume, and the other volumes that cover design-basis 10 and beyond-design-basis events.
11 So, for LWRs, again, just before I get 12 into this, I will often throughout this presentation 13 talk about Light Water Reactors first, because there 14 are a lot of similarities in the way we intend to do 15 things, and even in some of the phenomenology.
16 So, we understand there's different 17 properties, different coolants, and so on. But a lot 18 of the basic phenomena are similar. So, that's why 19 I'll constantly be talking about Light Water Reactors, 20 even though the subject of this presentation is Non-21 Light Water Reactors.
22 So, for LWRs, when NRC is performing a 23 review we use NUREG-0800, the standard review plan for 24 Light Water Reactors.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
17 Section 4.2 specifically deals with the 1
fuel design, and in there it calls out a number of the 2
general design criteria.
3 And so these general design criteria 4
guarantee that the fuel system is not damaged during 5
normal operations and anticipated operational 6
occurrences.
7 The fuel system is never damaged so 8
severely that you can't insert control rods when you 9
need to, to shut down the reactor, and that you don't 10 underestimate the number of fuel failures, and that 11 you always maintain coolability during accidents.
12 So, those are the criteria we use when 13 we're reviewing Light Water Reactor designs. Next 14 slide.
15 So, when staff are reviewing licensing 16 topical reports, we try to take more of a graded 17 approach, where we focus more on issues that are 18 really important.
19 So, things that are more complex or have 20 higher safety significance or uncertainties, that's 21 where we really want to focus our attention.
22 And one of the ways that we do that is by 23 performing confirmatory analysis. One portion of 24 confirmatory analysis is actually doing code 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
18 calculations -- independent code calculations. But 1
there really are a number of other tools that NRR and 2
NRO use to do these reviews, from simple back-of-the-3 envelope-type calculations to experimental results, 4
and so on.
5 So, this is just one piece, but it's still 6
very important a lot of times in our reviews.
7 MEMBER CORRADINI: Is the intent, from a 8
validation standpoint, to take what you have from 9
experiments to make sure that there's always a 10 benchmark checking? I assume that's the case.
11 DR. CORSON: Yes. In general, we try to 12 rely very heavily on experimental results, both when 13 we're doing our reviews, and just for our code 14 development as well. I mean, you know, it would be 15 nice to be able to predict from first principles all 16 the phenomena, and I think we're moving in that 17 direction.
18 But right now, we still have to rely very 19 heavily on experiments for all that we do.
20 MEMBER CORRADINI: Okay, thank you.
21 DR. CORSON: So, next slide. How does all 22 this apply to Non-Light Water Reactors?
23 So, NRO has put together Reg Guide 1.232 24 that proposes advanced reactor design criteria that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
19 adapt the general design criteria to Non-Light Water 1
Reactors. And there's a number of changes as well, 2
but it follows the same general structure.
3 So, advanced reactor design criteria 10, 4
26 and 35 are analogous to the fuel general design 5
criteria.
6 And so, this last bullet, advanced 7
reactors may use one of two concepts to guarantee that 8
their fuel's not going to fail during normal 9
operations and AOOs.
10 MEMBER CORRADINI: Can I stop you?
11 Because -- so why -- now I should remember this but I 12 don't. Why is 26 replacing 27? My impression was 26 13 is normal operation, 27 is beyond normal operation.
14 So, help me --
15 DR. CORSON: You're exactly right. So, in 16 the general design criteria, you're right. Twenty-six 17 deals with normal operation AOOs, 27 deals with 18 accidents. And they're all for being able to shut the 19 reactor down.
20 Twenty-six in the ARDC combines the two 21 together. So, there's just 26 covers now, both normal 22 operations and accidents. It's all one thing. But 23 the language is basically the same between the two.
24 MEMBER CORRADINI: I forgot.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
20 DR. CORSON: Yeah, it's -- I know. We got 1
that same question and I probably should have 2
clarified that point.
3 MEMBER KIRCHNER:
And it's more 4
comprehensive than the 26, 27 for the LWRs.
5 DR. CORSON: It's not identical.
6 MEMBER CORRADINI: It's not identical.
7 DR. CORSON: But the intent is similar.
8 Very similar.
9 MEMBER CORRADINI: Well, I had a feeling 10 he was going to say this. So, that's why I wanted to 11 be clear, because I thought it was similar, as you 12 said, but he's saying that it's -- I'll use the term 13 more restrictive.
14 MEMBER KIRCHNER: More complete.
15 DR. CORSON: More complete. So, from your 16 perspective, the staff help me. So, he'll check us.
17 MEMBER CORRADINI: I think NRO could 18 probably -- Boyce could probably handle this a lot 19 better than I can. Delegating responsibility.
20 MR. TRAVIS: Yeah, okay. So, this is 21 Boyce Travis from NRO. So, when the ARDC were 22 developed, this was happening around the same time as 23 the paper that was published on GDC 26 for NuScale.
24 PARTICIPANT: Twenty-seven.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
21 MR. TRAVIS: Twenty-seven. Excuse me, 27.
1 So, the thinking behind that SECY paper was 2
incorporated into the ARDC 26. And 26 is meant to 3
cover the full spectrum of basically reactivity 4
control for the full spectrum of fuel designs, 5
operation, and conditions for the plant.
6 And the goal is then -- and this is 7
enumerated on in the Reg Guide 1.232 -- is to avoid 8
the sort of considerations that resulted in different 9
safety significance or safety-related components being 10 applied for 26 versus 27.
11 Twenty-six in the ARDC requires that you 12 have two independent means of reactivity control, and 13 it requires that -- and it's four different 14 requirements that cover --
15 MEMBER CORRADINI: This is ARDC.
16 MR. TRAVIS: ARDC 26 -- has four the full 17 spectrum, from normal operations, AOOs, and postulated 18 accidents.
19 MEMBER CORRADINI: That kind of help --
20 MEMBER KIRCHNER: It addresses the problem 21 that we saw with the exemption request for GDC 27. It 22 precludes essentially a return to criticality.
23 MR. TRAVIS: Yeah, exactly. That's 24 exactly right. It's specifically called out in the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
22 Rationale and Reg Guide 1.232, that an extended return 1
to criticality is precluded by ARDC 26.
2 MEMBER CORRADINI: Okay. And under 3
current, it's unclear.
4 MR. TRAVIS: That is correct.
5 MEMBER BROWN: What do you mean by 6
extended?
7 MR. TRAVIS: So, I --
8 MEMBER BROWN: Is that three months? Is 9
it one week? Is it one day?
10 MR. TRAVIS: No, no, no. So, I guess I'm 11 referring to something like a steam generator, like an 12 over pooling event that results in a short power spike 13 following the trip.
14 MEMBER BROWN: We're talking about minutes 15 or so?
16 MR. TRAVIS: Yeah, yeah. Something on the 17 order of seconds to minutes after the event. And 18 then -- you have to end in a shutdown state basically, 19 using ARDC 26.
20 MEMBER BROWN: But within minutes, as 21 opposed to weeks.
22 MR. TRAVIS: Right. Yeah, during the 23 course of the transient would be on the order of 24 minutes.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
23 MEMBER BROWN: Okay. Thank you.
1 MR. TRAVIS: Mm-hmm.
2 CHAIR BLEY: James, I'm still stuck. Two 3
slides back when you were talking with Mike, and I 4
certainly think anchoring to experiments -- which 5
weren't on the slide -- you need to do. But 6
experiments occur under very particular conditions.
7 And the real world happens under a variety, so how do 8
you use the experiments? Two need-to experiments can 9
be really dangerous.
10 How do you use the experiments and 11 consider the uncertainty in the way you're modeling 12 things, and how does that get factored into the codes?
13 DR. CORSON: I mean, that's a really good 14 question. So, so you say, what you do in a small 15 scale, nice controlled experiment, may not exactly 16 represent reality.
17 CHAIR BLEY: Never.
18 DR. CORSON: So, we do like to use a lot 19 of sensitivity studies, where we -- so we do some 20 uncertainty studies as well based on what we think the 21 uncertain parameter distributions are, and we do our 22 uncertainty studies there.
23 But we also do some sensitivity studies to 24 look at maybe how close are you really to some of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
24 these limits? Or how much -- if things are just a 1
little bit different from what we expect, or maybe a 2
bit different from what we expect, what would happen?
3 So, we certainly have to be careful for 4
that. But we really have to rely a lot on our own 5
engineering judgment as well, and our own experience.
6 So, fortunately for -- I will say for some 7
of these advanced designs, there have been operating 8
reactors, or at least test reactors, in the past. So, 9
we do have some indication at least of how things 10 behave. So, we're not completely flying in the dark.
11 But it's certainly a very difficult problem that we 12 have to deal with.
13 MEMBER CORRADINI: Are you politely saying 14 that you look for cliffs, and as long as you're far 15 away from a cliff, a strange or very non-linear 16 behavior, you feel comfortable?
17 DR. CORSON: Yes.
18 MEMBER CORRADINI: Okay.
19 MEMBER REMPE: And along those lines, I 20 know like when we had the Volume 1 and 3 discussion, 21 I was curious about the term design basis, because 22 back in the old days when we started out, people 23 didn't have a lot of experimental data, so they made 24 conservative assumptions, or they did something like 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
25 SAFDLs. They put limits on it. But I think your 1
proposal is to look at best-estimate based on 2
experimental -- right? -- data and put uncertainties 3
with your fuel modeling, is your approach here?
4 You're not going to try -- like in the old 5
days, people put like factor 20 percent on the decay 6
heat curve, and they had like design basis codes. And 7
then we'd have best-estimate for other things. And 8
you're going with best-estimate, but your intent is to 9
have some uncertainties?
10 DR. CORSON: I mean, we're -- so I will 11 say we'll try to be as best-estimate as we can. But 12 at the same time, these are new designs. And our best 13 estimate may still have very large uncertainties 14 associated with them.
15 So, while yes, we may try to do like a 16 best-estimate-type thing, the uncertainty bounds are 17 going to be much larger than what you might expect for 18 best-estimate plus uncertainty for LOGA, perhaps for 19 example. Because we just don't have as much 20 experience and as many experiments at this point.
21 CHAIR BLEY: As you begin to put this 22 material together -- not just for you folks, and I 23 liked your answers so far on uncertainty but we need 24 to see later how it all works out -- but potential 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
26 applicants really need to be reminded of this because 1
it's real tempting to start with a few simple 2
experiments and say I know everything there is. And 3
you have to be careful about that.
4 DR. CORSON: Yeah. We do try to stress 5
that in our vendor interactions. Certainly, NRO is 6
much more in tune with the vendors than we are, but we 7
also have interactions. We meet with DOE, so we 8
certainly do try to stress that point as much as 9
possible.
10 MEMBER REMPE: So, when you provide input 11 or output that goes to the design basis codes or 12 the -- beyond the MELCOR codes, your plan is to give 13 them the same output results with uncertainties or 14 whatever? I mean, you're not planning to have a 15 conservative result you feed to the design basis codes 16 and best-estimate that you feed to MELCOR? You know 17 what I'm trying to get to?
18 DR. CORSON: Yeah. No, I mean, we would 19 give like the same sort of information to both.
20 MEMBER REMPE: Okay. Mm-hmm.
21 DR. CORSON: So, we are going to -- so 22 we're going to talk about this at the end. But since 23 you bring it up, I think for design basis, I think the 24 fuel performance aspect may be more tightly -- I guess 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
27 I could say coupled -- but it will certainly be much 1
more involved than with beyond-design basis aspects.
2 So, beyond-design basis, what MELCOR is 3
really looking for would be fuel failure rates. So, 4
for Light Water Reactors, it has some simplified 5
models that can handle that and we don't really use 6
FRAPCON right now to support MELCOR.
7 But going forward in the future, we could 8
provide some fuel failure rates for TRISO fuel, for 9
example. Things that MELCOR doesn't know needs to get 10 from somewhere, whether it's fuel performance code, or 11 an experiment, or what have you, that's one role that 12 a fuel performance could have in beyond-design basis.
13 For design basis, I think we'll do 14 something similar to what we do now for like a MELLLA+
15 analysis, where yeah, we provide our input to -- she's 16 itching to jump in here.
17 MEMBER REMPE: Well, finish what you're 18 saying.
19 DR. CORSON: Yeah. So, where we provide 20 input to like a neutronics code, or a thermohydraulic 21 code, and we might have some feedback. Or, going 22 forward, we could also have a more coupled analysis.
23 So, as it is right now for Light Water 24 Reactors, we do have some of these feedbacks, but it's 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
28 in a much more manual manner. That may be one 1
approach for Non-Light Water Reactors, or we're also 2
working to couple some of our codes together to take 3
some of that manual difficulty out of the picture.
4 MEMBER CORRADINI: But before -- well, I 5
have a feeling Dr. Petti may say the same thing I'm 6
about to say. I'll be quiet.
7 MEMBER PETTI: Can I ask the question the 8
opposite way? Why would you decide to do a more 9
coupled and a more detailed calculation, when some of 10 these systems have inherent safety attributes that 11 perhaps a simpler hand calculation can get you the 12 level of detail that you need to confirm what the 13 vendor is saying, particularly if you're two orders of 14 magnitude away from the dose consequence code, for 15 instance.
16 It seems like it may be too much of a 17 sledgehammer approach when you just need an elegant 18 hammer instead.
19 DR. CORSON: So, I would agree with that 20 statement. What I would say is that, while we 21 certainly expect much larger margins and all that, we 22 don't necessarily know at this point exactly what 23 vendors are going to propose. So, we have to be 24 ready.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
29 That way we don't get stuck in a situation 1
where vendors come in with maybe tighter margins than 2
we were expecting and we need a more detailed 3
analysis. So, this doesn't necessarily mean we will 4
do this fully coupled analysis. But it makes sense to 5
prepare for it at this point.
6 MEMBER PETTI: Well, I mean, it just seems 7
to me that you should have a toolbox. The toolbox 8
includes everything from something simple that you can 9
learn a lot about the physics side without running 10 detailed codes, up to the detail code, if and when you 11 need it.
12 You can probably build understanding and 13 engineering judgment pretty quickly with simpler 14 analytical solutions. For instance, the analytical 15 solutions for some of these fuel systems, you just got 16 to go back a little and find them, because they were 17 developed before we had big computers.
18 They capture 90 percent of what you need, 19 and you can come up to speed pretty quickly without 20 having the detailed code in front of you.
21 DR. CORSON: Yeah. Again, I fully agree.
22 And we'll actually touch on those points a little 23 later.
24 MEMBER KIRCHNER: May I ask more of a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
30 philosophical question? So, is the expectation -- it 1
fits in with Dave Petti's comment and Mike's -- that 2
the applicants will come in with an NQA-1 for design 3
basis, an AOO event?
4 In other words, a topical report on the 5
code that's been reviewed and approved by the agency?
6 DR.
CORSON:
I
- mean, that's my 7
understanding. If I'm wrong, NRO can contradict me.
8 MEMBER KIRCHNER: And then, to fit in with 9
the comments, then is there thoughts about something 10 like an Appendix K, which is a very conservative 11 approach to the ECCS LOCA issues for an LWR?
12 Is there a thought that you would use that 13 kind of approach, which then fits in with Dave's 14 comments about using bounding
- analysis, more 15 simplistic tools, to establish some kind of regulatory 16 certainty, or confidence that you've got considerable 17 margin for what in many cases will be an experimental, 18 or first-of-a-kind prototype design?
19 DR. CORSON: So, I would defer to NRO for 20 that.
21 MEMBER KIRCHNER: Ah. Good.
22 DR. CORSON: I mean, it's kind of a 23 policy -- so, it's difficult, because it's kind of 24 a --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
31 MEMBER KIRCHNER: Well, you're being 1
pretty honest. I think what my colleagues are asking 2
you -- what I think we're all kind of getting at is, 3
start as simple as possible, bound the calculation.
4 If that doesn't get you there, then -- and so I'm kind 5
of asking should you wait for the vendors to do this, 6
or is it more cost-effective for you to start doing 7
this now on the two fuel types that you're 8
identifying, and just broaden it on two candidate 9
designs for those fuel types, so that you actually get 10 a feeling when you need to be detailed, versus 11 assuming you have to be detailed.
12 DR. CORSON: Yeah. I think we -- so, 13 internally, we do plan to do that sort of stuff. I 14 can't speak to what would be required of an applicant, 15 but I will say internally, we don't need the same 16 level of detail as an applicant would in general.
17 We can do some of these simpler 18 calculations. I mean, we really -- so eventually, we 19 want to move towards as best estimate as we can, but 20 certainly initially, as you said, we can start with 21 simpler calculations, to give us a feel for some of 22 these things.
23 And we do have some initial assessment 24 plans to do those types of calculations.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
32 DR. BAJOREK: Mike, I think we are 1
thinking along those lines. This is Steve Bajorek 2
from Office of Research.
3 You know, you one of the ways we would 4
utilize something like this is a fuel performance code 5
might go and tell us what thermal conductivity 6
degradation is. And if we want to minimize or 7
maximize resistance to heat flow, that's the type of 8
information we would put in a design basis-type tool 9
to bound it in a certain direction.
10 And if that's good enough, we don't have 11 to go any further. The difficulty we have right now 12 is that many of the applicants say there's lots and 13 lots of margin.
14 We think that's right. Okay? That 15 probably is. But the staff doesn't know that yet.
16 And we're going to have to do some of those 17 calculations, which may involve some of these types of 18 details, in order to give us some of that direction.
19 So, we'll initially start off as simple as 20 we can, build in those details as we need to follow 21 those up.
22 MEMBER PETTI: I think it would have been 23 helpful to have some of this discussion in the 24 relevant chapters of the report. There's a lot of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
33 ways to attack this. Because it comes across as more, 1
we got to get these codes ready. We got to get these 2
codes ready. We got to put all this physics in.
3 And you lose a lot of this nuance that 4
we've had in this discussion here.
5 MEMBER CORRADINI: I guess I'd go a little 6
further, which is probably not in the volumes, but 7
some sort of preliminary introduction that says, 8
here's our attack philosophy for these things. We're 9
going to start simple.
10 Now, my bias is even further. I would 11 start and I'd reverse the problem. What is the NRC --
12 why do we even need NRC in all this? The answer is, 13 source term.
14 I would essentially say, what do I need to 15 do to estimate the source term? How much certainty do 16 I need in source term before I want to say it's 17 enough? And I can do a relatively straightforward 18 calculation, totally -- not totally, but I'd say 19 pretty close to totally -- based on experimental data 20 that I can point to.
21 And then, Dennis' concern would be, is 22 that data appropriately prototypical. Right? But 23 that's -- I'd be looking for that outside of your 24 three volumes. I'd be looking at some sort of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
34 introduction -- I'd attack it -- on how you're going 1
to deal with all of this. Because you're never sure 2
what's going to happen.
3 MS. WEBBER: And so there is the 4
introduction, which we presented on May 1st. And we 5
have talked about including some of this thought in 6
that introduction.
7 CHAIR BLEY: I think that's important. I 8
was about to say before Mike that the things we're 9
talking about aren't just about Volume 2. They're 10 probably everywhere.
11 MEMBER KIRCHNER: I think it would be good 12 in the introduction to cover this. And also, working 13 with NRO. Outline, in a general sense, what the 14 expectations are of the applicants, because they're 15 going to be looking at these documents as well.
16 I go back to my analogies. They're an 17 Appendix K kind of approach that you would use with a 18 novel new design to bound things, and then expect the 19 applicant comes in with a code that's been validated 20 within the experimental data range that they're going 21 to try and operate within, including their design 22 basis event.
23 DR. CORSON: Okay, so I'll touch on this 24 the second half of this slide, that advance reactors 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
35 may use either SAFDLs or SARRDLs. So, SAFDLs, 1
Specified Acceptable Fuel Design Limits, deal with 2
criteria to prevent fuel failure, which are used for 3
LWRs and may be used for advance reactors.
4 And the Specified Acceptable System Radio 5
Nuclide Release Design Limits, they limit how many 6
radio nuclides you can release. And this is proposed 7
for TRISO fuel.
8 So generally, how we do things -- again, 9
this is generally how we do it for LWRs, but it mostly 10 applies to non-LWRs. We're going to go over on the 11 next few slides kind of the two ways that we do fuel 12 performance confirmatory analysis.
13 So, even
- though, again, there's 14 differences between LWRs and non-LWRs. The same 15 approach, in general, should work. So, next slide.
16 So, the first way we it is single-element 17 analysis. So, this would be like hottest rod-type 18 calculations. And in this case, we'd use a standalone 19 fuel performance code FAST, which is just the 20 successor to the FRAPCON code.
21 We have some input from other codes for 22 things like radio power profile, or axial power 23 profile, inlet conditions, but for the most part we 24 just run our fuel performance code and determine 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
36 whether or not the limits are met. So, there's no 1
reason why that would necessarily change for non-LWRs.
2 So, next slide.
3 So, the other thing we do would be a full 4
core analysis. And this is where the fuel performance 5
code would fit into a design basis -- accident design 6
basis -- event-type analysis.
7 So, this is kind of how we use it for 8
MELLLA+, for example, for Light Water Reactors, where 9
we have some information that we're passing, in this 10 case, manually back and forth between core simulator, 11 like PARCS, and thermal hydraulics like TRACE, and 12 doing our fuel performance calculations.
13 So, as I was saying, we could take that 14 same approach where we're manually maybe passing some 15 simple information back and forth, or we're prepared 16 to also do a more coupled analysis. And we'll 17 determine what's necessary once we receive some 18 applications, as we discussed. So, next slide.
19 Now, I'm going to switch gears and talk a 20 little bit about the fuel phenomenology that we need 21 to consider. Once again, I'm starting with LWRs 22 because here we have the list of LWR fuel rod 23 failures. A lot of these also apply to some of the 24 non-LWR designs.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
37 Of course, you have to be concerned with 1
things like overheating, pellet-clad or kernel-layer 2
interactions, these types of things.
3 So, that's why I put this up. You know, 4
your fuel performance code, which handles these things 5
in LWR, you can use a lot of the same infrastructure 6
to do it for non-LWRs, of course with different 7
properties, correlations, and so on. Next slide.
8 So, I talked about some of the failure 9
modes. This just talks about the phenomena you need 10 to consider. Basic things like heat conduction or 11 fission gas release, stress strain, and so on.
12 Once again, these are generally applicable 13 to LWRs and to most non-LWR designs. So, a little 14 later on I'll go through TRISO and metallic, 15 specifically to talk about some of the differences.
16 But in general, these are the things we 17 need to consider. And for all these phenomena, we 18 need to consider temperature, burn-up, the radiation 19 effects on material properties, and also the initial 20 manufacturing defects, which can be very important for 21 certain fuel designs. So, next slide.
22 The way we determine which phenomena are 23 really important, is generally we use the PIRT 24 process. So, we've, way back in 2004, there was a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
38 PIRT on TRISO fuel, which Dr. Petti was a part 1
of -- yeah, long time ago but still applicable 2
today -- and that's the best information we have on 3
4 There aren't any specific liquid metal or 5
sort of FAST reactor metallic fuel, PIRTs, but Argonne 6
does have a series of reports on metallic fuel, which 7
is quite useful.
8 So, there are PIRTs. I should clarify.
9 There are PIRTs on SFRs, but they don't necessarily 10 focus on fuel performance. They cover different 11 aspects. Design-based access and so on.
12 MEMBER CORRADINI: So, I have two 13 questions. One is, maybe Dr. Petti can correct me.
14 But some of your references go back to this time. Are 15 there not newer documents to refer to on TRISO?
16 That's question one.
17 Question two is, on FAST reactors, I was 18 under the impression that Argonne, with Sandia, did a 19 series of five reports accumulative to one major 20 report on metal fueled gas reactor, in terms of gaps.
21 They didn't call it a PIRT, they called it a gap 22 analysis. And I'm assuming you're well aware of --
23 okay.
24 DR. CORSON: Yeah. Yeah, the gap 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
39 analysis, at least in May report, doesn't specifically 1
touch on fuel performance. I mean, aspects, of 2
course, are important to design-based accidents and so 3
on.
4 But it's not specifically a fuels PIRT 5
like the TRISO PIRT was. As far as newer, to your 6
first question, the best document I would say, which 7
I don't reference on slides but is referenced in the 8
report, is the project plan for the advanced gas 9
reactor program.
10 It talks about some of the missing 11 information. And then it talks about how the AGR 12 program is going to deal with that. So, that might be 13 a more up-to-date -- maybe not as comprehensive, but 14 more up-to-date document, if you want to look at it.
15 MEMBER CORRADINI: Okay.
16 MEMBER REMPE: So, I'm not an expert on 17 fuels, but I am aware that X-energy is trying to do 18 some sort of manufacturing capability for all the gas 19 reactors relying on TRISO fuel. And they realize that 20 they're coming in and they're different than what the 21 labs have done.
22 And they'll have to get a radiation data 23 to show their capability is as good as what's been 24 done in the DOE-funded HER program.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
40 With sodium fuel, I think the situation's 1
a bit different. In your report, you talk about how 2
that there's been this FAST reactor working group 3
report, as well as a report done by -- perhaps it was 4
Argonne or someone in the DOE thing.
5 But they talk about the metal fuel. And 6
you talk about in your report that there's data out 7
there. But one of those reports talks about how the 8
metal fuel has evolved over the years. And we've not 9
had any metal fuel made and run in EBR-II in a lot of 10 years now.
11 And I didn't see that clear indicator in 12 your report that yeah, there's a lot of historical 13 data, but whoever comes in with a new fuel capability 14 better have some data to show that it's still valid.
15 I mean, I don't know if you've had time to 16 get into the details and see if all this data, even 17 though the later fuel from EBR-II perform the same as 18 the earlier of the fuel, or the defects might change 19 with the manufacturer.
20 And I guess I'm interesting in, do we know 21 as much about metal fuel fabrication nowadays? I've 22 heard over the years some saying, oh, we had two 23 vendors make it. But jeepers, Aerojet hasn't been 24 making fuel also for a lot of years.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
41 And so I think there's more uncertainty 1
with metal fuel. What's your take on it?
2 DR. CORSON: I mean, there's certainly 3
some uncertainty with restarting the manufacturing 4
process. But I think Argonne hasn't looked at this 5
and they feel pretty good that what we've done in the 6
past is --
7 MEMBER REMPE: Well, I know Argonne, and 8
I know Gene feel very good about it. But I am 9
concerned. And I think, again, even the Argonne folks 10 at the MFC were not the technicians who made it years 11 ago for EBR-II. And is there not some concern by 12 anybody?
13 MEMBER PETTI: Yes, I think -- let me just 14 add that this was a subject of big discussions at INL 15 before I retired. And there are more than one 16 fabrication route now for metallic fuel.
17 The approach that was used for EBR-1, the 18 historic approach, is casting. And some of the newer 19 vendors think that's not an economical process to 20 scale for industry, with you guys not making a lot of 21 it.
22 And so, if you change the fabrication 23 route, the sessions, well, you have to re-irradiate, 24 right? Because we know that how you make it can 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
42 impact how it performs. So, I think some of the 1
vendors that have looked at it more seriously 2
understand that.
3 I don't know that it's a belief held 4
across all the concept developers in sodium systems, 5
but some of them understand that.
6 DR. BAJOREK: This is Steve Bajorek.
7 We've recently started some work with Department of 8
Energy to start looking at the experimental gaps that 9
are in existence in a number of these different areas.
10 With respect to metallic fuel, when we met 11 with them, they think that most of the phenomena are 12 covered with EBR-II. However, because of the new 13 manufacturing, swelling of these metallic fuels, 14 that's something that has very high uncertainties.
15 It's probably going to need more work.
16 The concern there is as you start to go a 17 little bit higher burn-up, you're going to start 18 having much more swelling in these metallic fuels and 19 the database may not have sufficient coverage.
20 MEMBER REMPE: Again, X-energy folks at a 21 meeting clearly said, even though Oak Ridge has been 22 making this fuel, we know we've got to radiate it and 23 we've got a plan to show we are as good as what the 24 folks within a couple of years ago, even though we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
43 have folks from Oak Ridge coming and help us, that we 1
could still do the same thing.
2 Now, I know, maybe it's harder to make 3
TRISO fuel than metal fuel, but the folks from 4
Argonne, they've got procedures from the old days.
5 And I really would like to see the regulators say, 6
well, I would like to see some radiation data to show 7
that you had the same type of fuel that they made a 8
decade ago, or whenever it was they last made it.
9 And that's what I would like to convey 10 here today, that I think you need to check it and make 11 sure they can still do it.
12 CHAIR BLEY: Amy?
13 MS. CUBBAGE: Do I just --
14 CHAIR BLEY: Introduce yourself.
15 MS. CUBBAGE: Hi. This is Amy Cubbage, 16 NRO. I'd just like to make kind of a broad statement 17 about the meeting here and the scope that we're 18 focusing on our independent confirmatory capabilities 19 and what data needs we see to support that. And 20 separately from that, NROs engaging with potential 21 applicants on what they would need to do.
22 So, our requirements -- they're not even 23 requirements, they're things that we're doing because 24 we want capability -- are separate from what 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
44 applicants would need to do to demonstrate their fuel.
1 And at this time, the linear term 2
applicant using metallic fuel would be the Oklo 3
vendor, and we're dealing with them separately, 4
outside the scope of these discussions.
5 MEMBER REMPE: I get that. But what I 6
guess is that, one, even when you refer to the FAST 7
reactor working group report as a source of data and 8
the other places, there's been a change in the metal 9
fuel over the years.
10 MS. CUBBAGE: Right.
11 MEMBER REMPE: And so, I'd like to see, as 12 you --
13 MS. CUBBAGE: So, to the extent that the 14 staff wants to rely on that data to validate its 15 codes, that's a factor we should consider. But I was 16 concerned that you were raising with regard to the 17 applicants. We'd like to keep those discussions 18 somewhat separate.
19 MEMBER REMPE: I get that. But I'd like 20 to see if there's been a difference. And then I'd 21 like to note that no matter what the applicant is, 22 that even their fuel may be different than what you've 23 had historically, is the point I'm trying to raise 24 today.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
45 MS. CUBBAGE: No, I understand. And I 1
guess from a regulatory office perspective, we don't 2
have a near-term applicant, other than Oklo, using the 3
fuel. So, PRISM will be authorized by Department of 4
Energy if it goes forward for the VTR reactor.
5 They're still working on that. And so, we'll engage 6
with those applicants as they come.
7 DR. CORSON: Okay, so now I'm going to go 8
a little bit into some of the phenomenology for TRISO 9
and metallic fuel. So, tri-isotropic fuel has been 10 proposed for not just high-temperature gas reactors, 11 but also for these fluoride cooled, high-temperature 12 reactors, which have a molten salt coolant but still 13 fixed fuel, or at least solid fuel, I should say.
14 So, the kernel and the codings provide 15 barriers to fission product release. I'll show a 16 picture of that on the next slide. And TRISO fuel has 17 operated a
number of years domestically and 18 internationally.
19 Of course, the German AVR and THTR are the 20 big examples internationally. And then, in the US we 21 had Fort St. Vrain. So, we do have some operating 22 experience with TRISO. Next slide.
23 So, this just shows what a TRISO particle 24 looks like. So, you have a kernel which contains your 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
46 uranium
- fuel, uranium
- dioxide, your uranium 1
oxycarbide, whatever it is.
2 And then you have a porous buffer layer 3
and a silicon carbide layer sandwiched between two 4
pyrolytic carbon layers. And on the next slide I'll 5
talk about what those different layers do.
6 These little particles are on the order of 7
1,000 microns, give or take a bit. And they're packed 8
in a graphite matrix. Here, I'm showing a pebble 9
design. These are some typical dimensions back when 10 we were looking at pebble bed reactor for NGNP.
11 And there's quite a few of these particles 12 per fuel compact, 1,000 to 10,000 or so. So, next 13 slide.
14 So, the way this works, the kernel 15 provides some fission product retention. It's just 16 like the pellet in U-02. It doesn't keep everything 17 in but retains quite a bit of the fission products.
18 The current plans are to use uranium 19 oxycarbide. In the past, the reactors have used 20 uranium dioxide, but the current irradiations that are 21 being done for the AGR program used uranium 22 oxycarbide.
23 The porous buffer layer accommodates the 24 kernel expansion, fission gas release, and fission 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
47 recoil energy, so it protects the structural layer, 1
essentially.
2 The pyrolytic carbon provide fission gas 3
retention, and also some structural support for the 4
silicon carbide layer, which is the main structural 5
support. It's essentially the pressure vessel for 6
this kernel design. So, next slide.
7 This just goes through some of the 8
phenomena that are important to safety. These are 9
just the things that we need to be aware of when we're 10 doing our fuel reviews.
11 So, you have to worry about fission 12 product migration through these layers, or attack of 13 the silicon carbide layer. So, historically, there 14 have been large silver releases from these particles, 15 and there's also some evidence that palladium can 16 attack and fail the silicon carbide layer. So, these 17 are just things we need to be aware of.
18 Another big thing historically that was an 19 issue for TRISO, is oxygen and carbon monoxide release 20 from the fuel kernel. It led to a phenomena of kernel 21 migration, where the kernel would actually move and 22 fail the structural layers.
23 But the uranium oxycarbide kernel should 24 minimize, or pretty much eliminate, this failure mode.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
48 So, that's one of the main reasons that we've gone to 1
that uranium oxycarbide.
2 And then, of course, we need to deal with 3
the usual suspects, like pressure buildup and 4
mechanical stresses, strains, and so on.
5 MEMBER CORRADINI: So, since I don't know 6
anything about this, I guess I can ask this question.
7 I didn't understand the first bullet. Does that mean 8
that it always attacks, or there's a level of burn-up 9
above which you have to worry about this? The way you 10 write it --
11 DR. CORSON: Yeah, there's -- so, for 12 silver it just a high diffusivity through the layers.
13 So, that's just something you have to deal with. For 14 palladium, it's like a certain amount you need.
15 The exact amount and so on is being 16 investigated. I mean, that's one of the things that 17 the AGR program is looking at. But yeah, it's just 18 something that has been identified as a concern. How 19 much of a concern hasn't been quantified yet. But 20 yeah, AGR is working on it.
21 MEMBER KIRCHNER: James, can I get you 22 back one slide previous?
23 DR. CORSON: Sure.
24 MEMBER KIRCHNER: I'll just quibble a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
49 little bit with your third bullet.
1 DR. CORSON: Okay.
2 MEMBER KIRCHNER: The pyrolytic carbon 3
really protects the silicon carbide. That's really 4
what it's there for.
5 DR. CORSON: Mm-hmm.
6 MEMBER KIRCHNER: But inner and outer.
7 DR. CORSON: Yeah.
8 MEMBER KIRCHNER: It doesn't provide a lot 9
of fission gas retention at all.
10 MEMBER PETTI: No, that's not true, Walt.
11 It provides a high degree of fission gas retention.
12 MEMBER KIRCHNER: Yes, but it's subject, 13 Dave, to cracking and all kinds of things under 14 irradiation and stress from heat. So, I mean, you 15 really hang your hat on the silicon carbide. The 16 performance.
17 MEMBER PETTI: But, for instance, if the 18 silicon carbide layer failed in the particle, what is 19 seen experimentally is cesium release but no noble gas 20 release, because the pyrolytic carbon layers are 21 intact and prevent the fission gas from coming out.
22 So, it's how to interpret what you see experimentally.
23 MEMBER KIRCHNER: Okay, I'll yield the 24 point, but not too much. And I would have preferred 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
50 the zirc carbide. But anyway, they didn't go there.
1 MEMBER PETTI: Well, we can have that 2
discussion some other time. I'm on the other side of 3
the fence on that.
4 MEMBER KIRCHNER: Thank you.
5 DR. CORSON: So yeah, just quickly to go 6
through metallic fuel, so next slide. So, it's 7
specifically looking at U-10 zirc, so ten percent 8
zirconium. That's what we expect most of -- or the 9
initial applicants to come in with.
10 We're also considering a mix of uranium 11 and plutonium, with tin-zirconium as well. But 12 certainly initially, we expect U-10 zirc. So, this 13 has been proposed for sodium FAST reactors and, as Amy 14 mentioned, for some heat pipe reactors.
15 This fuel is typically paired with high-16 temperature steel cladding, HT-9 was used towards the 17 end of the EBR program, and we expect that's what 18 applicants would use as well.
19 And this fuel has years of domestic and 20 international operating experience. Once again, 21 EBR-II is the best example that we have.
22 MEMBER CORRADINI: Again, something I'm 23 not familiar with, so I just thought I'd ask. So, 24 FFTF I thought ran with HT-9 also?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
51 DR. CORSON: FFTF had -- yes, but it had 1
oxide fuel. So, FFTF had oxide driver fuel. But they 2
did test metallic fuel. But the actual fuel in that 3
fuel reactor was an oxide. MOX fuel. So yeah, next 4
slide.
5 So, this just shows what a metallic fuel 6
rod looked like. This is back from PRISM, which might 7
be used -- similar design might be used for the BTR.
8 So, I think the big thing to note here is 9
a very large fission gas plenum. And also, there's a 10 very large initial gap between the fuel slugs and the 11 cladding, which is initially filled with sodium.
12 As Steve had mentioned, this metallic fuel 13 swells quite a bit. That's why you have this big gap.
14 And you have this big plenum because you release a 15 very large amount of fission gases. So --
16 MEMBER KIRCHNER: Since you brought it up, 17 or someone did, what would they do -- if I remember 18 it, the fuel was cast. So, is it proposed now to go 19 to some other alloy approach or extrusion, or would it 20 be an extrusion?
21 DR. CORSON: Yeah.
22 MEMBER KIRCHNER: Okay. And then, do we 23 have any feeling about general dynamic behavior with 24 regard to swelling? Because, as you point out, the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
52 large gap and the sodium is there to provide good 1
conductivity, and so on and so forth. Do we know --
2 do we have any preliminary evidence about how an 3
extruded slug would work versus the cast?
4 DR. CORSON: So, I have to say not that 5
I'm aware of. I mean, certainly I agree we're going 6
to have to deal with larger uncertainty until we have 7
actual evidence of manufactured fuel, how it performs.
8 But certainly, for our purposes in 9
research, we can get away a little bit more with 10 larger uncertainties, just because how we do our 11 confirmatory analysis.
12 MEMBER KIRCHNER: So, just sticking to 13 what you're doing in your co-development, so when it 14 comes to FAST, do you have confidence in the models, 15 that they'll deal with the change in the -- what would 16 you call it, the structural -- yeah, the fabrication 17 technique for those slugs?
18 DR. CORSON: I mean, certainly we're going 19 to rely on what our historical data. And if we have 20 new evidence, then we will update the codes as --
21 MEMBER KIRCHNER: Then, you're going to 22 have to put some uncertainty --
23 DR. CORSON: Yeah.
24 MEMBER KIRCHNER: -- kind of factor.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
53 DR. CORSON: And that's why we do things 1
like sensitivity studies, to see, okay, we say 2
swelling is this much, but what if it's really a 3
little bit greater? How would that affect things?
4 So yeah, that's why we really rely on 5
sensitivity studies, to look at these cliff effects.
6 MEMBER BALLINGER: My understanding is 7
that the swelling was actually intentional. In other 8
words, once you get above about 30 percent, or 20 or 9
30 percent, you get interconnected porosity. And then 10 the fission gas solutions gets easy. You got this 11 large plenum. So, it's pretty well understood, I 12 thought.
13 DR. CORSON: For the old manufacturing 14 process. So --
15 MEMBER BALLINGER: But I guess -- but 16 metallic fuel is metallic fuel. It's going to swell 17 like a dead fish.
18 (Laughter.)
19 DR. CORSON: Yeah, so you're certainly 20 right, it's going to swell.
21 MEMBER CORRADINI: I'm trying to visualize 22 that. Say that again.
23 PARTICIPANT: You've seen veils flow up.
24 DR. CORSON: Yeah, you're right. It's 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
54 going to swell no matter what. But is it going to 1
swell exactly the amount that we saw in the past, or 2
are there going to be differences?
3 MEMBER BALLINGER: Well, when you get to 4
30 percent, once you get interconnected porosity, 5
you're pretty much done. Right?
6 MEMBER CORRADINI: What I think Ron is 7
saying, is if it's more, it's of no consequence.
8 That's what I thought you were getting at.
9 But I know we're getting into the weeds, 10 but I thought I remembered that we had a -- well, 11 there was some meeting, I can't remember what 12 meeting -- and Terra Power already had similar 13 radiations of their extruded fuel. Am I mis-14 remembering?
15 DR. CORSON: Could be. Sorry, I'm not 16 aware of that.
17 MEMBER CORRADINI: Okay.
18 MEMBER PETTI: There are a plan to do some 19 irradiations on the extruded fuel, but there may --
20 MEMBER CORRADINI: Oh, I thought they had 21 done the irradiations in Russia.
22 PARTICIPANT: M-46.
23 MEMBER PETTI: Yeah. Well, okay, I'm not 24 sure at what scale that extrusion was done.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
55 MEMBER CORRADINI: Well, I'm sure there 1
were teeny tiny axial links. But I thought they had 2
done extruded fuel. That's the reason I was curious.
3 MEMBER PETTI: There's larger scale 4
extrusion work going on right now that's sort of at a 5
pilot scale, I think. And the plan is to irradiate 6
that as well.
7 MEMBER CORRADINI: I see. Okay.
8 MEMBER PETTI: I think Terra Power -- I 9
would personally put Terra Power in the same sort of 10 knowledge and approach as Joy described. They know 11 they have to irradiate, and so they're working that.
12 MEMBER CORRADINI: So, they're aware of 13 the need.
14 MEMBER PETTI: Yes.
15 MEMBER REMPE: In the data that you have 16 looked at -- because again, some of those other 17 reports emphasize how EBR-II fuel varied over its 18 operating years -- are you seeing much variability 19 when some of the -- I mean, have you started plotting 20 certain performance parameters as a function of -- for 21 different fuels, and seeing any -- or is it all just 22 the same no matter who made it, or not even started 23 looking at the data yet?
24 DR. CORSON: So, I would say EBR-II 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
56 actually used different metallic fuels throughout its 1
lifetime. So, that could be why the performance 2
changed. They used what, U-Fissium, I think, early 3
on, and then they eventually went to U-10 zirc.
4 But --
5 MEMBER BALLINGER: That's because of the 6
reprocessing part.
7 DR. CORSON: Yeah.
8 MEMBER BALLINGER: Yeah.
9 MEMBER REMPE: So, are you seeing 10 variation, is what I'm kind of -- I mean, have you 11 started looking at the data and your models, or you 12 just aren't that far yet? You just --
13 DR. CORSON: Yeah, not that far. I mean, 14 we've done preliminary assessments. But we've used 15 three test cases. I mean, there are certainly way 16 more test cases in the database that Argonne has. So, 17 we'll look at that going forward.
18 MEMBER REMPE: I'd be interested if you 19 can see some differences in the performance for over 20 the years of fuel.
21 DR. CORSON: Yeah. I mean, that's 22 certainly something we'll need to take into 23 consideration going forward.
24 So, metallic fuel performance, I already 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
57 touched on the last point, the fission gas release and 1
swelling. Of course, it's metal. It has very good 2
thermal conductivity. But then, it also has a lower 3
melting point. So, those are just things that you 4
have to be aware of when you're doing your analysis.
5 So, next slide.
6 Phenomena important to safety. One 7
interesting thing that happens with these metallic 8
fuels is that the zirconium and uranium tend to 9
redistribute inside this fuel slugs as you go along.
10 And this can have -- basically, they can form like 11 local
- phases, which are different melting 12 temperatures, or eutectic behavior, and so on.
13 So, it's something to be aware of, because 14 it can impact fuel performance. Of course, you always 15 have to be aware of manufacturing. And then, usual 16 suspects, fee conduction, and so on.
17 MEMBER PETTI: So, I have a question. Do 18 you guys actually plan to try to model the 19 redistribution in the FCCI?
20 DR. CORSON: Yeah, that's something where 21 it's a maybe right now. So, our plan right now is to 22 look at whether or not we think it's necessary. So, 23 we're going to do some preliminary assessments and 24 determine whether or not we need the zirconium 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
58 redistribution model.
1 I would also say that there are some 2
models out there that we can scavenge if we feel we 3
need to. So, to answer the question, maybe.
4 MEMBER PETTI: Yeah, because you'll find 5
a lot of stuff in the literature today on FCCI and 6
trying to understand it better. Lots of university 7
work going on and the like. So, it's still an active 8
area of research.
9 DR. CORSON: Yeah, I know. We did talk 10 about this a little bit with DOE, like the lanthanide 11 behavior and how that impacts fuel failure, and so on.
12 So yeah, we're aware of it. Whether we actually model 13 it or not is to be determined.
14 MEMBER PETTI: Okay.
15 DR. CORSON: So, right now we're going to 16 talk a little bit about how we selected which code 17 we're going to use going forward.
18 So, there's a lot more on this in the 19 report, but I'll go through this real quickly. So, 20 again, we need a tool that we can use.
21 We're going to focus on fuel performance 22 and we're going to focus on solid fuels, so this just 23 excludes our consideration of fuel salts for this 24 volume, or salt fuel reactors. And then, I'm going to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
59 discuss a few of the criteria we used.
1 There's really two big choices. It's 2
either FAST, NRC's code, which is a follow-on of the 3
FRAPCON code, and then there's INL's BISON code. So, 4
I'll just go through this fairly quickly. I talked 5
about --
6 MEMBER CORRADINI: So let's stop there for 7
a minute. But both of them rely on the same database.
8 DR. CORSON: Yeah. They do. So, there 9
are --
10 MEMBER CORRADINI: So, is it like 11 Chevrolet and Ford?
12 DR. CORSON: There are certainly a lot 13 of --
14 MEMBER CORRADINI: Why do I need a Ford 15 when I've got a Chevrolet?
16 DR. CORSON: So, there's a lot of 17 similarities. I will say BISON is more detailed and 18 it also has some interplay with the MARMOT code, which 19 has the mesoscale-type calculation. So, it goes into 20 a little more detail than FAST does. And it doesn't 21 have the same assessment database, at least for Light 22 Water Reactors, at this point.
23 MEMBER CORRADINI: So, you gave me some 24 pros. So, I sense the last one was a con.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
60 MR. GEELHOOD: Yeah, I know that's a major 1
con was --
2 MEMBER CORRADINI: I think you have help 3
from the peanut gallery. You have to identify 4
yourself and talk into the mike.
5 MR. GEELHOOD: Yeah, this is Ken Geelhood.
6 PARTICIPANT: Is this turned on or not?
7 MEMBER CORRADINI: It should be on. You 8
just got to get real close, Ken. Real close. Real 9
close.
10 PARTICIPANT: Aim it straight --
11 MR. GEELHOOD: So, I would say the 12 assessment database of FRAPCON and FAST is much 13 bigger. Probably almost a factor of ten greater than 14 what INL's used to assess BISON.
15 So, I would say that would be a pro of 16 FAST, is that we have such a large assessment 17 database, we can have higher confidence within a 18 larger area of power and burn-up levels.
19 MEMBER CORRADINI: But then, let me ask a 20 question back to James, but you have help. So, if I 21 now take myself out of the Light Water business and I 22 take myself into the other business of metallic fuel 23 or TRISO fuel, both will have to take the same 24 database and, excuse my English, tune itself to that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
61 database. Is that not true?
1 MR. GEELHOOD: I don't really think that 2
tuning to a database is really what we do. So, we 3
kind of have two levels of data. So, there's separate 4
effects data. When you irradiate something, then you 5
measure thermal connectivity, how much it swelled, 6
things like that.
7 So, we develop models based on that. And 8
then, we do assessments. So, when you irradiate 9
something, maybe you've put a lot of instrumentation 10 in, you can measure temperature in situ. You can 11 measure straining afterward.
12 Then, we just compare to that. It would 13 be tempting to go back and then try to change 14 something to get your predictions, but that's not 15 really our approach.
16 Our approach is to fit all the separate 17 pieces, and then see how well they do.
18 MEMBER CORRADINI: So, the separate 19 effects experiments lead to your model improvement, 20 and then the assessment database with what I'll call 21 a more integral measured test --
22 MR. GEELHOOD: Yes.
23 MEMBER CORRADINI: -- is what you compare 24 to.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
62 MR. GEELHOOD: Mm-hmm.
1 MEMBER CORRADINI: Okay, I got it.
2 MEMBER REMPE: But you need to say who you 3
are too. I'm sorry.
4 PARTICIPANT: He did. He did.
5 MEMBER REMPE: You did say?
6 MR. GEELHOOD: Yeah, I'm Ken Geelhood from 7
PNNL.
8 MEMBER REMPE: Okay, thank you. Then, I 9
had a question to -- I mean, you made the comment 10 about MARMOT. And in the interface between BISON and 11 MARMOT, the explanation you gave is good in that you 12 talk about we have separate effects for thermal 13 conductivity degradation, and then we have an integral 14 database.
15 My understand is this mesoscale MARMOT 16 database is very, very, very, very limited, if any, 17 especially when you get into FAST reactors. And so, 18 that's, I think, a difference.
19 My understanding is the fundamental 20 equations are a bit different in BISON. Now, you're 21 right that they're to the same integral effects, but 22 it's not clear to me that you have separate effects 23 for the BISON MARMOT type of interface. Is that a 24 true statement?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
63 MR. GEELHOOD: I would say that's a very 1
true statement. I would say that when we already know 2
the answer, so to speak, in LWRs, then the proof's in 3
the pudding. So, MARMOT could forget what we already 4
know, fission gas release, then maybe you could have 5
some confidence in moving it on.
6 But to date it hasn't been successful at 7
predicting what we already know based on fundamental 8
equations. And so, if it can't do that, I don't think 9
it's ready to go to what we don't know.
10 MEMBER REMPE: Thank you.
11 MEMBER CORRADINI: Can I just ask either 12 James or Ken just to be clear? As you would learn 13 from what people call mesoscale, I assume grain 14 boundary size stuff, is that what MARMOT is, grain 15 boundary size stuff?
16 MR. GEELHOOD: Probably even lower than 17 that.
18 MEMBER CORRADINI: Thousands of atoms 19 stuff-ish?
20 MR. GEELHOOD: Probably. They call it the 21 mesoscale. It's in between atoms --
22 MEMBER CORRADINI: It's not an atom and 23 it's not a grain. It's somewhere between.
24 MR. GEELHOOD: Yeah.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
64 MEMBER CORRADINI: Okay. But as you learn 1
that, there would have to be separate effects 2
experiments either currently in the Light Water 3
business, or for TRISO or metal, to better improve the 4
modeling within MARMOT.
5 MR. GEELHOOD: Certainly. I think that 6
someone could use their advanced microscope skills, 7
various spectroscopy things, to try to get some sort 8
of data that would be used as input to MARMOT. But I 9
think that's very difficult and I don't think it's 10 been done successfully to date.
11 MEMBER CORRADINI: If I take that away and 12 ask are there separate effects enough -- well, I guess 13 we're getting away from why you answer --
14 MR. GEELHOOD: I think I know what you're 15 answering. Like the inputs, what it means, what are 16 its material properties? I don't think they're very 17 easy to measure.
18 MEMBER CORRADINI: Oh, I was going to say 19 something more -- I'm back to my Chevrolet and Ford 20 question, which is, if I have an experimental 21 database, can I not then use those separate effects 22 data to either help in the performance of FAST or 23 BISON?
24 MR. GEELHOOD: I suppose you could. But 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
65 on some level, if that's what you're doing, then would 1
you rather have one knob to tune, or 50 knobs way down 2
in the weeds to tune? And if ultimately you're 3
getting the same answer, then we should go with the 4
most simple approach.
5 MEMBER CORRADINI: Okay. All right, 6
thanks. I got it.
7 MEMBER BALLINGER: Along the lines of what 8
Dave Petti has been saying, we need to be careful 9
about hitting a fly with a sledgehammer. When you 10 have a FAST reactor fuel where you have a sodium gap, 11 what happens to the fuel is decoupled to some extent, 12 except for the fuel-clad chemical interaction part 13 from the cladding. And so, from the standpoint of a 14 source term determination --
15 MR. GEELHOOD: So, they're not totally 16 decoupled. So, we --
17 MEMBER BALLINGER: Well, I didn't say 18 totally decoupled. I didn't use the word totally. I 19 said, with the exception of.
20 MR. GEELHOOD: Well, but we see kind of 21 two swelling regimes. We see an unrestrained 22 swelling, maybe of like 20 percent.
23 MEMBER BALLINGER: Yep.
24 MR. GEELHOOD: But then, once it comes in 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
66 contact, certainly the swelling rate goes down quite 1
a bit, because the cladding is pretty strong and 2
provides quite a bit of restraint to that swelling.
3 And so, but when you get that restrained 4
swelling, that is now straining the cladding, and so 5
that is leading to damage of the cladding.
6 MEMBER BALLINGER: But when you have --
7 again, when you have 30 percent swelling, you got 8
interconnected porosity, when you get interaction 9
between that and the clad, the clad doesn't go out 10 exclusively. The fuel goes in.
11 In other words, the fuel is able to 12 accommodate the plastic deformation.
13 MR. GEELHOOD: To some degree. But it 14 does strain somewhat --
15 MEMBER BALLINGER: Yeah, yeah. It does 16 strain the cladding.
17 MR. GEELHOOD: -- maybe like a couple of 18 percent. And so, that is now a damage mechanism that 19 you want to worry about.
20 MS. WEBBER: Can I just make one comment?
21 So, notwithstanding the staff's approach to selecting 22 one code versus another, I think the other thing I 23 think about is that some of these advanced reactor --
24 potential applicants, I'll call them -- may use BISON.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
67 And so, there's active development on BISON.
1 And so, at the very least, the staff needs 2
to have some level of familiarity with BISON, to be 3
able to ask the appropriate questions during a 4
licensing review.
5 And so, I think while they're considering 6
the use of BISON, there still may be a lot of 7
development work that needs to be undertaken. But 8
having a familiarity with BISON will help if we have 9
applicants who use it in the future.
10 MEMBER CORRADINI: So, I'm sure this is a 11 terrible analogy. However, I look upon this as RELAP5 12 and TRACE.
13 MS. WEBBER: I don't know. I don't know 14 that much. Personally, I don't know that much about 15 the process.
16 MEMBER CORRADINI: I don't either, but I 17 look upon these as two things that then demand an 18 experimental database that must be fed, both, as Ken 19 said, as separate effects and --
20 MS. WEBBER: I don't think we would 21 disagree.
22 MR. GEELHOOD: That's a reasonable 23 analogy. And the way we're looking at FAST and BISON 24 is, we want the capability of using either one. Okay?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
68 And we'll factor that in. And eventually, we'll find 1
out which one's working better.
2 DR. CORSON: Next slide.
3 MS. WEBBER: Okay.
4 DR. CORSON: So yeah, we've already talked 5
about all the phenomena we need to model. We've also 6
touched on multi-physics, do we need to couple, or can 7
we take a more serial approach.
8 Certainly, we would like a code that could 9
do either, ideally. Next slide.
10 So, of course, we need to consider code 11 development costs. We don't want to spend unnecessary 12 money. And that means we need to consider full life 13 cycle costs: validation, code maintenance, because we 14 expect if applicants do come in and reactors do get 15 built, we're going to need these codes for the long-16 term, because there's training, and so on. So, that's 17 just something to consider.
18 Whether or not we're going to be ready to 19 meet industry schedule, as I said, we really need to 20 be ready and we want a code that's going to be ready 21 on time.
22 Computational research requirements. So, 23 right now we basically run on Windows, although this 24 is changing, I will say.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
69 While it's nice to have something that can 1
run on Windows, we are moving to the cloud. We have 2
a Linux instance that we've been working on. So, this 3
isn't so prohibitive necessarily if we don't have 4
Windows.
5 Impacts on staff. So, certainly, we've 6
touched on this a little bit. But as much as 7
possible, we would like something simpler, just if 8
possible, because it's easier to understand.
9 It's not to say that we can't understand 10 something that's complex, but certainly we have 11 limited time, it's easier for us if we have a simpler 12 model. And it makes sense too, because it might allow 13 us to do our reviews much more quickly if we have 14 something simple that we can understand, than going 15 into a very complex code.
16 And then lastly, this issue of regulatory 17 independence that comes up. We really need to 18 understand all the assumptions that go into whatever 19 code we use. It certainly helps if we're actively 20 involved with code development, to really understand 21 what the code is doing, so that we can better perform 22 our reviews. So, based on all these criteria --
23 CHAIR BLEY: I like what I've read and 24 what you've been saying here. As we went through 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
70 Volume 1 especially, I think things were a little more 1
up in the air.
2 It strike me there's -- and you guys have 3
said this a couple of times today in another way --
4 there's a difference between what you choose to do for 5
confirmatory analysis, what you choose to require, or 6
at least suggest as a good option for you to review, 7
is very different from what a vendor decides to do.
8 But I think some of the concern we heard 9
earlier at other meetings was, if we require vendors 10 to use more complicated tools than they really need, 11 that kind of doubles the expense of dealing with all 12 of these issues.
13 So, I like the way you're talking today 14 about what you need in-house, and I think that'll help 15 the vendors as well.
16 MS. WEBBER: I mean, I just want to 17 comment on that. So, we talk about this quite a lot, 18 that the vendors have to design these reactors and 19 make sure that they are designed, constructed, and 20 operate safely.
21 So, they may have to have computational 22 capabilities well beyond what our needs are. And so, 23 that's something that we think about and talk about 24 quite a bit. And so, the capability of those codes 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
71 may be much more -- I don't know what the word is --
1 developed and more extensive than perhaps what our 2
needs are. So, I just wanted to make that comment.
3 DR. CORSON: So yeah, based on all these 4
criteria, we're going to work on FAST going forward.
5 I mean, we've been actively involved with FRAPCON, 6
FRAPTRAN, FAST code for decades. So, it makes sense 7
to continue with this.
8 But as Ken said, we are still following 9
what the BISON team are working on. And it certainly 10 helps us to be familiar with the BISON models. So --
11 (Off-mic comment.)
12 MS. WEBBER: Protesting the question.
13 MEMBER CORRADINI: So, no, we're just 14 educating ourselves.
15 DR. CORSON: Okay. Just making sure you 16 didn't have a question.
17 MEMBER CORRADINI: No no.
18 DR. CORSON: So, okay.
19 MEMBER CORRADINI: We're not shy.
20 DR. CORSON: Okay. So now, Lucas is going 21 to talk about what we're doing to get FAST ready, and 22 what our plans are then for the future.
23 MS. WEBBER: Are we moving off of this?
24 DR. CORSON: Yeah.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
72 MEMBER PETTI: So, just a question. Are 1
you aware of some of the latest results in TRISO-2 modeling that the AGR program has done where they've 3
done a really detailed look at every thermal 4
mechanical material property that's needed and varied 5
them, and figured out which are the ones that really 6
make a difference and which of the ones don't make a 7
difference? That might be quite helpful for you guys 8
as you think about what you're doing for FAST.
9 DR. CORSON: So, yeah. That would 10 certainly be helpful. Is that a published INL report?
11 MEMBER PETTI: Yeah. I was the reviewer 12 before I retired. So, it was done in 2018.
13 DR. CORSON: Okay.
14 MS. WEBBER: Can we ask you for an action 15 item, to give us that report?
16 DR. CORSON: We may have it. But I have 17 so many INL reports that it may have gotten lost.
18 MEMBER PETTI: Yeah. I mean, it can 19 really help you leapfrog. Similarly, are you aware of 20 the four volume, comprehensive nuclear material 21 publication that goes through all the Gen-4 systems, 22 and there's chapters on fuels and there's chapters on 23 the modeling of those fuels?
24 DR. CORSON: Yes. So, we are aware of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
73 that.
1 MEMBER PETTI: Good. Good. That's 2
scheduled for updating, actually --
3 DR. CORSON: Okay.
4 MEMBER PETTI: -- in the next year or so.
5 So, be on the lookout.
6 DR. CORSON: Yeah, that would be good. If 7
you could give us a heads-up, that would certainly be 8
helpful as well.
9 MEMBER CORRADINI: So, Dave, just to close 10 the loop, maybe you want to -- when you send it, send 11 it to Weidong and Derrick so that they can pass it on 12 to Kim, just so we --
13 MEMBER PETTI: Okay.
14 MS. WEBBER: Yeah, pass it along to James 15 Corson as well.
16 DR. CORSON: Yeah, and thanks a lot for 17 that. That would be really helpful.
18 MS. WEBBER: He'll get to it a lot faster 19 than I will.
20 DR. CORSON: Maybe.
21 MS. WEBBER: Are we moving on?
22 MR. KYRIAZIDIS: All right, so good 23 morning. So, my name is Lucas Kyriazidis. I'll be 24 reviewing the development plans for FAST for non-LWRs.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
74 So, as James has highlighted, the NRC has 1
selected to pursue using FAST as its fuel performance 2
code to further develop and use it during confirmatory 3
calculations for non-LWRs.
4 Highlighting some of the reasons why we've 5
made that selection, is just that we've been actively 6
involved with the development of FAST, in addition to 7
FRAPCON and FRAPTRAN, throughout its creation.
8 Also, as Ken has highlighted, has an 9
extensive assessment database for LWRs. So, if you 10 want to go on to the next slide.
11 I'll go over the generic non-LWR 12 development needs, and what its current capabilities 13 are, and then what we plan on doing to the code.
14 So, in its current state, FAST is limited 15 to modeling 1.5-D cylindrical fuel rods. In here, 16 1.5-D cylindrical fuel rods, or 1.5-D analysis, simply 17 refers to the fact that each radial node is 18 calculated, and then -- well, let me -- sorry.
19 1.5-D simply refers to the fact that the 20 transport equations are solved in the radial direction 21 for each axial slice. And then, simple relations are 22 used to handle the actual transport.
23 MEMBER CORRADINI: Hang on. So, that 24 seems reasonable. So, where does that break down?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
75 Because what you're telling me is in the world of 1
reactor physics, that was what life was before some 2
magical 3-D slice connection. Right?
3 I mean, all reactor physics are basically 4
just 2-D analysis. And then, I connect actually by 5
some sort of simple connection algorithm. So, what 6
you're telling me here is I do a radial, and then I 7
don't have any circumferential variation, but I then 8
connect axially over some lane scale.
9 MR. KYRIAZIDIS: Correct.
10 MEMBER CORRADINI: Okay. So, why is 11 that -- where does that break down?
12 MR. KYRIAZIDIS: So, the only reason 13 that's okay is because the fuel rods are volume 14 skinny. And so if you had a fuel rod --
15 MEMBER CORRADINI: A fat fuel rod.
16 MR. KYRIAZIDIS: A fat fuel rod, you could 17 have some axial heat transfer.
18 MEMBER CORRADINI: Okay, fine.
19 MR. KYRIAZIDIS: But when you have a 20 pencil, you don't really get significant heat transfer 21 up and down.
22 MEMBER CORRADINI: That's what I guessed 23 you were going to say. So, pellets, or TRISO fuel, 24 are the issue? I'm still trying to understand where 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
76 the issue lies that I might -- where I'm going with 1
this is -- I'm sorry, I'm jumping -- I'm trying to not 2
do 2-and 3-D because all I can see is just 3
computers --
4 DR. BAJOREK: Like, I think --
5 MEMBER CORRADINI: Churning.
6 DR. BAJOREK: -- part of this might be we 7
have seen a couple of designs. The AHTR, one of the 8
molten assault, the fuel looks like a manhole cover.
9 MEMBER CORRADINI: Oh.
10 DR. BAJOREK: Okay, with swats through it.
11 MEMBER CORRADINI: Oh, I see.
12 DR. BAJOREK: Okay, very 3-D. And it's 13 geometry. We've also -- we would also be using FAST 14 for some accident-tolerant fuels. If you've seen, I 15 think it's the light bridge design, looks kind of like 16 a licorice stick. Okay? That one would be very 17 difficult to handle with this, but almost everything 18 else would be a one-and-a-half.
19 MEMBER CORRADINI: Okay. All right.
20 Thank you. I appreciate it. Thank you again.
21 MR. KYRIAZIDIS: So, based on the 22 currently proposed non-LWR fuel forms, FAST would be 23 extremely limited. So, we're going to overview what 24 we need to do to get the code ready. So, a couple of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
77 needs that we've identified is to expand the 1
geometries that FAST can handle.
2 So, we want to incorporate new solvers 3
that can handle spherical geometries, in addition to 4
irregular geometries, and also expand the 1.5-D 5
assumption to handle 1-D spheres, 2-D and 3-D 6
dimensions.
7 And this will happen when we incorporate 8
new solvers, so there are contracts in place and works 9
ongoing to incorporate new finite difference and 10 finite volume options.
11 And I think it was stated by Steve that 12 1-D will be sufficient for spherical geometries, but 13 the 2-D and 3-D capabilities will really handle the 14 heat pipe reactors.
15 So, I think James touched a little bit 16 earlier in the presentation, but FAST is also being 17 coupled to other neutronics and thermal hydraulic 18 codes. So, there are ongoing work to couple FAST to 19 TRACE through the MOOSE framework.
20 And this will really expand the thermal 21 hydraulic capabilities of FAST, and possibly allow for 22 feedback effects to be accounted for.
23 But another benefit of this is to allow 24 FAST to communicate to other codes in the MOOSE 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
78 framework. And then, coupling SCALE and EPIC to FAST 1
would allow for more redefined power profile, or radio 2
power profile.
3 MEMBER CORRADINI: EPIC is part of the 4
SCALE?
5 MR. KYRIAZIDIS: It is. It is. And I'll 6
give you just a brief highlight of it. But yes, SCALE 7
and EPIC would be used to calculate the intra pen 8
power distribution, in addition to the isotropic 9
distribution.
10 And EPIC is just a simplified, faster 11 running version of SCALE, designed for use in FAST.
12 And yeah, like I said, it would be used to just 13 calculate the radial power profiles. Next slide.
14 So, moving from the generic FAST 15 development needs, I'm going to start focusing on 16 TRISO, then follow that with metallic fuels.
17 So, in its current state, FAST can't be 18 used to model any spherical fuel form, such as TRISO.
19 It's restricted to 1.5-D cylindrical fuel rods. But 20 there are features in FAST that we can recycle to 21 model TRISO. Some of these features are calculating 22 fission gas pressure.
23 And also, we have the basic infrastructure 24 in place to incorporate new material properties, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
79 properties such as UCO, pyrolytic carbon, and silicon 1
carbide.
2 And the bottom half of the slide just 3
highlights specific development needs for modeling 4
TRISO in FAST.
5 MEMBER CORRADINI: I was going to ask 6
about PARFUME. So, that's the basis by which --
7 MR. KYRIAZIDIS: That is going to be our 8
starting point for material properties. We'll also do 9
a code-to-code comparison between FAST and BISON, 10 BISON AND PARFUME. Yeah, I'll get into that a little 11 bit.
12 MEMBER CORRADINI: So, maybe --
13 (Simultaneous speaking.)
14 MEMBER CORRADINI: We're letting our 15 Chairman lead the way in example. So, you're doing 16 a -- I wanted you to repeat that so I understood.
17 So, PARFUME is strictly a spherical 18 geometry modeling of TRISO fuel? Am I understanding 19 that correctly? I know that --
20 MR. KYRIAZIDIS: You got to do TRISO.
21 MEMBER CORRADINI: -- somebody online 22 knows this, but --
23 MR. KYRIAZIDIS: Yeah.
24 MEMBER CORRADINI: -- I wanted to task 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
80 you.
1 MR. KYRIAZIDIS: It does do TRISO fuel 2
particles, but I think I read that it also does clay-3 type fuel. But I'm not familiar with that. But yes, 4
it was used to do TRISO fuel particles.
5 MEMBER CORRADINI: And so that's why you 6
want to do the code-to-code --
7 MR. KYRIAZIDIS: Yes. Yep, yep.
8 MEMBER PETTI: Yeah. So, Mike, the 9
particles are spherical, but the fuel elements come in 10 different geometries. So, it does spherical at the 11 particle level and it does cylindrical plate at the 12 fuel element level.
13 MEMBER CORRADINI: But to be politely 14 kidding, you're on conflict since you were one of the 15 authors of PARFUME.
16 MEMBER PETTI: Yeah. I'm just providing 17 technical background.
18 MEMBER CORRADINI: Thank you. That's all 19 we were -- I figured that's where we were going with 20 this.
21 MR. KYRIAZIDIS: So, I just want to 22 highlight some of the specific development needs for 23 FAST to model TRISO. So, as we stated, the FAST needs 24 to currently be modified to allow the model of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
81 spherical geometries, but also be able to solve the 1
governing equation, such as heat transportation and 2
mechanical stress through spheres. And this is 3
ongoing right now.
4 But also need to expand FAST's material 5
property library. Currently, we don't have properties 6
that relate to TRISO in FAST. And our starting point 7
would be just to take what PARFUME has in its material 8
properties, and work that into FAST.
9 And then lastly, once these above-10 mentioned features are developed, we would require to 11 do validation and assessment of whether or not we 12 incorporated these features correctly. And this will 13 consist of open literature benchmarks against AGR, in 14 addition to code-to-code comparisons between BISON and 15 PARFUME.
16 And here, just a note of clarification 17 from the slides. Underdevelopment means that we are 18 currently working on this task right now. And then 19 under contract means that we have a contract in place 20 but work hasn't begun yet. Next slide please.
21 So, moving forward, the FAST development 22 team has identified specific data needs.
23 MEMBER CORRADINI: Can I stop you?
24 MR. KYRIAZIDIS: Yep.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
82 MEMBER CORRADINI: So -- or maybe this 1
slide is where I was going to go. I was going to ask, 2
when does the use of FAST stop and the use of MELCOR 3
begin? Or is that coming?
4 MR. KYRIAZIDIS: That comes towards the 5
end of the presentation.
6 MEMBER CORRADINI: Okay, fine.
7 MR. KYRIAZIDIS: And James will cover 8
that.
9 MEMBER CORRADINI: Because when we had the 10 Volume 3 presentation, I seem to remember the MELCOR 11 types told us that they were also using a TRISO model 12 within MELCOR. So, I assume there's some sort of 13 pass-off.
14 MR. KYRIAZIDIS: Yes.
15 MEMBER CORRADINI: Okay.
16 MEMBER KIRCHNER: So, this slide really 17 just wants to highlight some of the specific data 18 needs that we need to model TRISO. So, we've 19 identified or highlighted three areas, the first being 20 coding material properties, the second being UCO 21 properties, and lastly, in a broader one, it's study 22 state and transient fuel performance data.
23 So, coding material properties have 24 referred to the pyrolytic carbon layers. And the AGR 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
83 program, it has been cited that there are large 1
uncertainties. This is okay, but if we want to bring 2
down those uncertainties, we would need more data.
3 The second being a bigger area as UCO 4
properties. Currently, in PARFUME it assumes that UCO 5
properties are the same as UO2. This will be adjusted 6
with AGR. But as for the time being, we are missing 7
UCO properties.
8 And the last being study state and 9
transient fuel performance data. This would be data 10 that would be used to assess FAST. This also includes 11 like the effects and interactions between the coolant 12 and particle graphite matrix, and effects of air/water 13 ingress.
14 We have used this data to determine 15 whether or not we would need to use FAST to calculate 16 the effects. But yeah, next slide please.
17 FAST assessment data for TRISO fuel. So, 18 as stated, FAST will require integral data to perform 19 validation. So, the current plan is to utilize data 20 from DOE's AGR program. And based on our current 21 understanding of AGR, AGR will provide the industry 22 TRISO data for the development of material property 23 models and integral assessment data.
24 It should be noted that AGR will focus on 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
84 UCO fuel kernels. So, that addresses the data need on 1
the previous slide. AGR will also produce a variety 2
of PIE data that will be used to develop material 3
property models.
4 And it also performs safety tests. And 5
this will be where we get our assessment data and code 6
assessment -- code assessment and validation data.
7 MEMBER KIRCHNER: Curiosity. The AGR, is 8
the program being run to support the BISON code at a 9
mesoscale, or is the data being generated at a more 10 macro level?
11 MR. KYRIAZIDIS: My understanding is macro 12 scale. But if anyone wants to add to the --
13 MEMBER KIRCHNER: That's what I would 14 expect.
15 MEMBER PETTI: That's what it is, Walt.
16 MEMBER KIRCHNER: Yeah. Okay.
17 MEMBER PETTI: Just a point that there's 18 another benchmark you guys might be interested in 19 besides the IAEA that Gen 4 benchmark, that focuses 20 just on the fission particle release from the fuel.
21 And we can get you that. I think it's 22 just completed. So, it'll be another one for you to 23 compare against.
24 MS. WEBBER: That would be great.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
85 MR. KYRIAZIDIS: And the last point I want 1
to touch on this slide is, the NRC will coordinate 2
with INL to come up with a similar and same VNV test 3
suite between BISON and FAST. And this will be where 4
we do our code-to-code comparisons.
5 So now, moving on to metallic fuel.
6 Metallic fuel is much more developed in FAST when 7
compared to TRISO. Modeling metallic fuel can be 8
done, but only limited to study state and 1.5-D 9
cylindrical fuel rods.
10 FAST has been implemented with material 11 properties for U-Zr and U-Pu-Zr fuel, with H29 12 cladding. But as stated, it's only limited to study 13 state calculations.
14 Material properties that need to be 15 implemented for transient analysis includes heat 16 capacities, entropies, and H29 cladding the old stress 17 models.
18 I think Ken mentioned that we have done 19 preliminary assessment on EBR-II fuel. This was a 20 paper presented at top fuels in 2018. We'll highlight 21 some of the results in the next couple of slides, but 22 I'll just give you an overview of what the paper 23 discussed.
24 It discussed the modeling process, the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
86 limitations of FAST, and additional future work that 1
FAST needs to be changed to perform, I guess, refined 2
assessments.
3 Two areas of future work were noted, the 4
first being a more refined fission gas release model, 5
especially in the low burn-up areas. And the second 6
was to evaluate the need to implement a zirconium 7
redistribution model.
8 We'll get into the details of the results 9
in the next couple of slides. But I guess I just 10 wanted to highlight the development needs for metallic 11
- fuels, the first
- being, incorporate material 12 properties for transient analysis, the second being, 13 for redefined fission gas release model, and the third 14 being, incorporate a zirconium redistribution model, 15 fourth being to expand allowable geometries -- this 16 would be specifically to address non-cylindrical 17 geometries, such as the heat pipe reactor -- and 18 lastly, just perform more assessment, because of the 19 available data, such as EBR-II data.
20 MEMBER CORRADINI: I don't know enough 21 about these designs. So, in the heat pipe-cooled 22 reactor, the fuel is not cylindrical? I thought it 23 was like a block of stuff with cylindrical pencils 24 interspersed with heat pipes.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
87 MR. KYRIAZIDIS: So, I don't know how much 1
of this I can go into.
2 MEMBER CORRADINI: Well then fine. Never 3
mind.
4 MR. KYRIAZIDIS: So, yeah. It's not 5
necessarily --
6 MEMBER CORRADINI: If there's no more that 7
you can say about it, that's fine. But the answer 8
is --
9 MR. KYRIAZIDIS: Yeah. I don't want to --
10 MEMBER CORRADINI: -- it's complicated.
11 MR. KYRIAZIDIS: Yeah.
12 MEMBER CORRADINI: Okay, fine.
13 MR. KYRIAZIDIS: Yeah, I don't want to say 14 more than that.
15 MEMBER CORRADINI: Okay, fine.
16 MEMBER PETTI: Could you guys model 17 annular fuel?
18 MR. KYRIAZIDIS: With the new solvers, we 19 will be able to.
20 MEMBER PETTI: Okay, great.
21 MR. GEELHOOD: We can currently model 22 annular pellets. So, like VBR pellets, or pellets in 23 the blanket region.
24 MEMBER PETTI: Good.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
88 MR. KYRIAZIDIS: So, this slide highlights 1
some of the metallic fuel data that we have identified 2
that we require. This would really expand what we can 3
model in FAST, but also give us the ability to model 4
transients in FAST.
5 Some of the data needs here identified are 6
H29 cladding, strain capability and fatigue data, and 7
this really would give us an insight of how much 8
strain the cladding can withstand before failure.
9 Another data need identified is rod 10 internal pressure limits and end-of-life internal 11 pressure. This would be used to assess FAST against.
12 And the last being study state and transient data.
13 And this includes transient fission gas release data, 14 cladding strain and centerline fuel temperature.
15 And again, this would be used to validate 16 fission swelling models in FAST, in addition to 17 fission gas release. And the data would likely come 18 from EBR-II, FFTF, and TREAT-M.
19 So, next point I want to talk about is the 20 assessment data for metallic fuel. And this really --
21 so, in addition to thermal mechanical data, we need 22 integral assessment data. And Argonne National Lab 23 has been developing a database to house a large 24 portion of the EBR-II data. And we believe that this 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
89 should provide the essential data needed to validate 1
FAST against.
2 So, recently, DOE hosted a DOE/NRC non-LWR 3
data need meeting. This happened in September of this 4
year. And during this meeting, an overview of what 5
data was available was presented to us with respect to 6
metallic fuels.
7 Examples included were EBR-II and FFTF 8
transient testing data. So, there is ongoing work for 9
us to get access to this database and begin reviewing 10 what's in this database.
11 It is believe that there is enough data in 12 there to be sufficient for initial licensing efforts.
13 And then, similarly to the TRISO statement, NRC will 14 coordinate with the BISON team to establish a common 15 validation of verification test cases for metallic 16 fuels. And this includes like a code-to-code 17 comparison.
18 MEMBER CORRADINI: So, to get back to 19 Joy's point, Dr. Rempe's point, this is all cast 20 versus what some of the potential --
21 MR. KYRIAZIDIS: Correct.
22 MEMBER CORRADINI: -- designs would have 23 as extruded. So, that has yet to be -- that 24 difference has yet to be determined.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
90 MR. KYRIAZIDIS: Correct.
1 MEMBER CORRADINI: Okay.
2 MEMBER REMPE: But my point's even a bit 3
broader. I'm not sure the cast from EBR-II was always 4
the same, because if you look at -- like look up a 5
couple of the reports you cited in this document and 6
they were explicit about what's changed over the 7
years.
8 And I'm just curious what kind of 9
differences one sees in the change over the years, 10 because you may get -- since they haven't done it for 11 a long time, some differences, just because somebody's 12 coming in new with cast fuel. And you talked about 13 manufacturing defects. I'm real curious on if it's 14 the same.
15 MR. KYRIAZIDIS: I mean, that's just 16 something that we'd handle with our sensitivity and 17 uncertainty-type analysis. So, if the database does 18 show differences in performance historically from 19 EBR-II, at least we'll know -- we'll have some idea of 20 what the range of differences are.
21 So, we could perhaps bound our 22 calculations, or even maybe push them a little further 23 in different directions, to explore any cliff edges 24 that we might have.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
91 So, yeah, that's certainly something that 1
we'll need to do once we start digging into this 2
database.
3 So, the next slide gets into the 4
preliminary assessment that we've done. So, as I 5
mentioned, in 2018 NRC staff and PNNL lead the efforts 6
of doing a preliminary assessment on EBR-II fuel in 7
FAST.
8 But we have copies of the report, or it's 9
out there publicly on the top fuel website. But in 10 this assessment, EBR-II fuel forms of U-Zr and U-Pu-Zr 11 fuel bonded with liquid sodium clad in H29, were 12 modeled. Coolant was liquid sodium.
13 And here, we just showed the outlet 14 temperature and fission gas release. Just wanted to 15 touch -- we do have like a simplified fission gas 16 release model in FAST and that's just assumed to be a 17 constant 70 percent independent of burn-up.
18 And I previously mentioned that we do want 19 to have a more redefined fission gas release model, to 20 really address the lower burn-up regions, outlet 21 temperature for sodium, values agree fairly well, so 22 it really showed that we incorporated our hue transfer 23 correlations correctly in FAST.
24 MEMBER CORRADINI: So, help me. What's 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
92 16 percent burn-up atom percent to megawatt days per 1
ton?
2 MR. GEELHOOD: So, one atom percent is 3
close to ten gigawatt days per ton.
4 MEMBER CORRADINI: Okay, fine. Okay, 5
thank you.
6 MR. KYRIAZIDIS: So, these results only 7
highlight the X-425 case. And this was the U-Pu-Zr 8
case. But we also looked at U-Zr and that report 9
highlighted three cases. We'll go on to the next 10 slide.
11 And this really just shows cladding 12 strains as a function of axial height, but also burn-13 up. And this just shows that FAST is able to predict 14 the trend and, I guess, the magnitude, fairly well.
15 We're within.3 percent strain.
16 Ken was one of the leading authors of this 17 paper, so if there's specific questions, we could have 18 him speak to them. But this is all I really want to 19 touch on, the preliminary assessment.
20 So, in Volume 2, we covered other fuel 21 forms that weren't metallic, and charged those, such 22 as oxide-based fuels, such as U02 and MOX, but also 23 ceramic-based, such as carbides and nitrides.
24 So here, I just want to touch a little bit 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
93 on what the methodology would be required to get these 1
into FAST.
2 MEMBER REMPE: Could I slow you down for 3
just a second?
4 MR. KYRIAZIDIS: Yeah. Yeah, yeah.
5 MEMBER REMPE: I'm thinking about this 6
response about the correlation between atom percent 7
and ten gigawatt-based for metric ton uranium. What 8
data point? Was that like 170 gigawatt days from 9
that --
10 MR. KYRIAZIDIS: Yes, that reactor fuel 11 was burned to very high burn-ups.
12 MEMBER REMPE: Which reactor was that in?
13 MR. KYRIAZIDIS: This was EBR-II.
14 MEMBER REMPE: Really? Okay, thank you.
15 I just was curious. That's a lot. Okay, thanks.
16 Yeah, that's a lot. Thank you. Sorry to interrupt.
17 MR. KYRIAZIDIS: Oh, no problem. No 18 problem. So, there are other non-LWR fuel forms being 19 investigated. As I highlighted, there are U02, MOX, 20 ceramic-based fuels, such as carbides and nitrides.
21 These fuel forms are of lower priority, 22 based on our current understanding of industry plans.
23 We've laid out the methodology in the TRISO section 24 and metallic fuel sections of what we need to do to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
94 get these fuel forms into FAST.
1 And then historically speaking, we have 2
seen U02 MOX fuel being used in the FFTF, but also 3
nitrides and ceramics have been used in EBR-II. So, 4
we believe that the data is out there, but we just --
5 there's no current near-term plans for it.
6 MEMBER KIRCHNER: Lucas, I've just --
7 seeing the UN there reminds me that at Los Alamos we 8
made fuel for the SP-100. That was a space reactor.
9 And it was now a FAST system.
10 I know the fuel was made. I don't know if 11 there are any detailed irradiation data. But the pin 12 size and such was very similar to the EBR-II.
13 MR. GEELHOOD: We do have a lot of the 14 SP-100 data at PNNL. It's not publicly available and 15 it's marked expert control right now. But we do have 16 access to that if need be.
17 MR. KYRIAZIDIS: So, this last slide, at 18 least the last slide in this section, just deals with 19 the methodology that we would take to get these fuel 20 forms into FAST.
21 So first, we would identify important 22 phenomenon on UO2, the carbides and nitrides-based 23 fuels. We did a little bit of that in Volume 2. The 24 second step would be to gather material properties and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
95 determine whether or not we would have to model these 1
important phenomena in FAST.
2 So, an example of this is, we identified 3
zirconium redistribution from metallic fuels. So, now 4
we're doing the valuation of whether or not we need to 5
incorporate that into FAST. That's just one example 6
for metallic fuels. And the last point is to perform 7
assessment.
8 So, like I said before, we believe that 9
there is data out there for uranium carbide and 10 uranium nitrides, and UO2.
11 So, that covers my section. I'm going to 12 pass it back to James to do the discussion between 13 fuel performance and DBs, and beyond DBs.
14 MEMBER REMPE: Before you go there, I've 15 been thinking from the very beginning of this meeting 16 the discussion about where you crossed off the flowing 17 metal fuel in your comment. And you're right.
18 Section 3.8 says, hey, we're going to do that in 19 Volumes 1 and 3 type of tools.
20 I'm just kind of thinking about, well, how 21 would you do that? And I guess I can understand in 22 MELCOR how you might be able to do it, although the 23 devil will be in the details.
24 But I'm just kind of wondering what you 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
96 would do in Volume 1 with a flowing metal fuel? Would 1
you have some sort of heat source in there? Because 2
you won't be dealing with any sort of fission product 3
release or transport, but it -- what would you do in 4
Volume 1 type of.
5 DR. BAJOREK: The molten salts are very 6
interesting and they kind of go into two areas. What 7
we would be doing is we would be tracking the 8
precursors through the system. So, you have to add 9
those to the codes.
10 And you have to have feedback between your 11 reactor kinetics -- the point kinetics, or something 12 more advanced -- in order to get your power 13 distribution within the system.
14 It also becomes a little bit of a 15 chemistry problem, because in some of these molten 16 salt reactors, the fuel salt in particular, the 17 fission process will be building up with time.
18 That's going to be changing your thermal 19 physics properties. It's going to change some of your 20 cross-sections in there? And as we go along with the 21 molten fuel salts, we're going to have to understand 22 the chemistry at the time of the expected scenario, 23 whatever those might be, and be able to have a 24 feedback on the kinetics that keep track of where the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
97 precursors are.
1 Because if you have, let's say a loss of 2
flow, okay? Now, instead of the precursor is dumping 3
their neutron over by the heat transfer mechanism, it 4
might be back in the core and that might be a 5
reactivity increase.
6 In terms of fuel performance, okay, we 7
can't really think of anything right now that fits 8
into that realm of your traditional fuel performance.
9 What changes with burn-up, apart from the chemistry as 10 you use salt.
11 MEMBER REMPE: So, with the first example 12 you cited, affirming data to validate this precursor 13 tracking thing that you're talking about?
14 DR. BAJOREK: MSRE is probably -- well, 15 it's the best and only information for the fuel salt 16 reactor. That's something that, as we go on, we'll be 17 pointing out as a data need, that we are going to need 18 better information on that.
19 MEMBER REMPE: But MSRE didn't have some 20 sort of test where you have data. So, there's no data 21 in them. Folks are going to have to think about that.
22 DR. BAJOREK: That has to be developed.
23 MEMBER REMPE: Okay, thank you.
24 DR. CORSON: Okay, yeah. So, to wrap 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
98 things up, I'm going to talk about how FAST fuel 1
performance interfaces with design-basis accidents and 2
beyond-design basis accidents.
3 So, a lot of what FAST is used for for 4
LWRs is normal operations and AOOs. So, it runs by 5
itself, does its own thing with minimal inputs from 6
other codes.
7 But for LWRs, we do use FAST for some 8
design-basis accident calculations, in conjunction 9
with TRACE and with some neutronics codes. So, the 10 MELLLA+ calculations are really good examples of where 11 we use FAST. And we feed initial conditions and take 12 input from TRACE and from the PARCS. So, that's just 13 one example of how we use it for design-based 14 accidents.
15 For beyond-design basis accidents in LWRs, 16 as I mentioned earlier, we really don't use FAST to 17 support that. So, next slide.
18 Now, going forward, how do we do this for 19 a design basis event and beyond design basis events 20 for non-LWRs? For design-basis events, I think it 21 would be a similar sort of approach to what we 22 currently do now, where we pass some information back 23 and forth between codes.
24 We've touched on this already. Whether 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
99 it's a tight coupling or something more manual is to 1
be determined. But it's certainly a very similar 2
approach we would use.
3 Now, for beyond-design basis events, we do 4
think FAST would have some role. And Dr. Corradini 5
asking about this. But it's especially important I 6
think for TRISO fuel. MELCOR does a lot of the 7
fission product tracking and it does have some models.
8 But it doesn't know what the fuel failure criteria is.
9 Right now, it would take fuel failure as 10 a function of temperature burn-up and so on. And it 11 doesn't know what that table looks like. So, FAST 12 could be a code that could provide that.
13 Again, we could also use experiments or 14 something else, or we could use FAST to try to extend 15 the experimental database as much as we can. But 16 that's where FAST could have a role for beyond-design 17 basis events.
18 MEMBER CORRADINI: But if there's -- I 19 mean, just to use Dr. Petti's analogy, if there is a 20 weak coupling, FAST could develop a table which then 21 can be used within MELCOR. I mean --
22 DR. CORSON: Yeah.
23 MEMBER CORRADINI: Okay.
24 DR. CORSON: Yeah, exactly. So, MELCOR 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
100 needs a table, whether it's fast or experiment --
1 MEMBER CORRADINI: Okay, I got it.
2 DR. CORSON: Yeah.
3 MEMBER CORRADINI: I got it. Okay.
4 DR. CORSON: So, next slide. So, this 5
goes --
6 MEMBER PETTI: Just a point. Back in the 7
NGNP days, INL provided NRC -- Stu Ruben in 8
particular -- of a spot surface analysis of exactly 9
that. So, it's probably somewhere in the archives of 10 the NRC, so you get a fact of what was done.
11 DR. CORSON: Yeah, we're aware of that.
12 It's just with the new fuel, oxycarbide fuel, there 13 would be some differences. So, MELCOR has something 14 right now. But we would like to be able to update it 15 with the latest information, if possible.
16 So, this is just a TRISO example that I 17 just talked about. So, we can skip over this and just 18 go to the summary.
19 MEMBER REMPE: Before you go to the 20 summary --
21 DR. CORSON: Sure.
22 MEMBER REMPE: -- I brought this up when 23 we had the Volume 1 and 3 discussion, and I guess its 24 caused some consternation from folks. But instead of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
101 calling things design basis and beyond-design basis, 1
to me, I guess in the SRP they talk about design basis 2
accidents.
3 And when we think about fission product 4
release, it covers AOOs, the whole spectrum. And what 5
I really see is something about a more integrated 6
evaluation in Volume 3 that could be used for the 7
whole spectrum in a simpler way, and it does provide 8
input to a source term.
9 And then I think Volume 1's more detailed 10 evaluations, where you have to have this coupling, and 11 there's a need for that because you have to be ready.
12 But why don't we try and think of it that way, instead 13 of design basis versus beyond-design basis.
14 Because AOOs aren't exclusively covered if 15 you go with this division you have. And so, it's a 16 simple terminology to it. And so, I was trying to 17 point out and people got hung up on a different --
18 DR. BAJOREK: We're rethinking the 19 terminology. It's the design basis, beyond-design 20 basis, that's part of our dialect that we've gotten 21 used to for the last 40 years.
22 As we start to use for some units LMP, 23 okay, there's going to be a gray area between those 24 two. And I think I'm going to go into it --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
102 MS. WEBBER: Yeah, Steve's going to talk 1
about that.
2 DR. BAJOREK: -- when I talk later on.
3 It's addressing the same problem from two different 4
angles of attack.
5 MEMBER REMPE: And historically, they're 6
not going with the old way of -- and that's what I was 7
trying to bring up at the beginning about this. We 8
don't put conservatisms in this. No, we're going to 9
do best-estimate and put uncertainties. And I think 10 that's what you're going to do, from what I'm hearing.
11 DR. BAJOREK: Yeah. And I think 12 Dr. Petti's point is very well taken. Our initial 13 approaches are trying to be as simple as possible. In 14 the last couple of slides we talked about the coupling 15 of FAST and other codes, and using it for thermal 16 hydraulics.
17 That's one of those things that we would 18 likely use as more of a side calculation, as opposed 19 to the everyday type of evaluation.
20 I like to think of it a little bit more 21 as, do I need all of the detail in my reactor core 22 when I -- the more I look with TRACE? Now, even 23 though we can go through and model it like a model, 24 every assembly in there, we don't really care about 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
103 that most of the time.
1 We're after the hot assembly. Okay? And 2
that might be the parallel over here, where the 3
coupling is necessary now for me to take my more 4
simple calculation, and now look at some of those 5
details that might become of interest in the review.
6 MEMBER REMPE: So, I think we're 7
approaching this -- a similar conclusion here, that I 8
think, again, changing the volume names, and then your 9
introduction document that you gave us a while back to 10 talk about that approach would really make this hang 11 together a lot better.
12 But I know you don't have a lot of time 13 before you come back to us and all that. But boy, I 14 sure would like to see that. Because I think it makes 15 a much better story.
16 MS. WEBBER: Well, what I would like to do 17 is, so Steve's going to touch on a lot of those 18 comments. And so, we're just ending with this 19 presentation. I don't know if you -- did you finish 20 this slide.
21 MEMBER REMPE: I interrupted him before he 22 did, because he was going the route design basis and 23 beyond-design basis, so I thought it was a good time 24 to interrupt him.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
104 MS. WEBBER: So, it might be good to 1
continue the dialogue when Steve's presentation --
2 MEMBER REMPE: Sure. Okay.
3 DR. CORSON: I mean, this is just a 4
summary though, of what we talked about. We think 5
FAST is the preferred path. We're working on 6
developing it for non-LWRs for metallic and TRISO 7
fuel. And we expect we'll have the data that we need 8
to support licensing.
9 Again, we have a very different level of 10 what we need than their's might need. We don't 11 necessarily need quite as much detail.
12 So, I think, while we say we might be in 13 really good shape, the vendors may have a slightly 14 different perspective on what data they need. So, 15 just to clarify that point and to conclude. So, I'd 16 be happy to answer any other questions you have right 17 now.
18 CHAIR BLEY: I think for the members, 19 we're going to have a talk by Steve. We're just 20 trying to address a number of things we've brought up 21 in various meetings. So, questions of that general 22 nature will wait until then.
23 At this time we're going to recess for 24 15 minutes. Be back here at quarter till.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
105 (Whereupon the above-entitled matter went 1
off the record at 10:30 a.m. and resumed 2
at 10:48 a.m.)
3 MS. WEBBER: Yes, let's turn it over to 4
Steve.
5 DR. BAJOREK: Thank you very much. Good 6
morning, everyone. I am Steve Bajorek from the Office 7
of Research. And what we wanted to follow-up this 8
morning on is a more -- continuing discussion on 9
Volumes I -- Volume 2 and Volume 3 and how they all 10 fit together. We've had a number of questions over 11 the last couple of months. Just to bring everybody --
12
-- everybody back, we had a subcommittee meeting on 13 May the 1st. We went through Volumes 1 and 3. A 14 couple of weeks ago, September 4, we had a -- I guess 15 it was a ACRS review of research activities. We 16 talked about what the work the we're doing in DSA.
17 And there were some additional questions on non-LWR 18 analysis because that falls -- the code development, 19 that falls in the DSA scope.
20 Now, just to kind of go back and, you 21 know, set the stage in where we're at with Volumes 1, 22 2 and 3 is the whole idea behind the IAPs. You know, 23 and the strategy, too, that we've been talking about 24 is readiness. Our mission has been to be ready. We 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
106 started this work about two years ago. It's -- it's 1
evolving as we go on. Now, and part of readiness 2
means -- doesn't -- don't be prepared for a particular 3
unit, or one or two particular units. Our goal here 4
is to be ready for all plant designs. Okay, we're 5
assuming that they are all equally likely to come in 6
for a design certification.
7 Licensing approach has not been fully 8
defined yet. Okay, now over the last couple --
9 CHAIR BLEY: I read that somewhere else.
10 They're all equally likely -- well maybe they are 11 eventually. But you know, of course -- you know, some 12 are coming in sooner than others. And your -- in the 13 last report, Volume 2, you have a big table that shows 14 what you're doing next year and what you're doing the 15 year after. And that -- you didn't tell us how you 16 came up with what you're doing when. But I -- I 17 assume it has mostly to do with what you expect to see 18 first?
19 DR. BAJOREK: Well, up until now, most of 20 our work has been generic. We have to do that, one, 21 to avoid costs to any particular potential applicant.
22 And we have been able to do that because there has 23 been a number of different physical phenomena --
24 things that you need to do that is there regardless 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
107 of, you know, the design. Putting different solvers 1
in a code and that type of -- type of thing. Doesn't 2
matter who it's for. We are getting to the point now 3
where, and -- by working with NRO we are starting to 4
see some of the applicants maybe taking a lead. And 5
our expectation, there are going to be a couple of 6
them that are going to be front-runners in this review 7
process. So we are starting to shift gears and put a 8
little bit more emphasis on those now.
9 CHAIR BLEY: So you're still expecting 10 applications late this year? Maybe next year?
11 DR. BAJOREK: The NRO will correct me, but 12 I believe that Oklo is still targeting a submittal in 13 December of this year. We have been talking with 14 another applicant, and they are regular visitors here 15 to talk to us on their applications. I don't know the 16 schedule, but I think a fluoride high-temperature 17 reactor is -- is one of those. And -- and thank you 18 very much, because that's what I wanted to do is, you 19 know, talk just a little briefly about the advanced 20 reactor landscape. It does continue to evolve and 21 change. This is the -- the current schematic that we 22 use of it. If you go back and look maybe a year ago, 23 microreactors were not on there. They were sort of 24 buried in here somewhere else. But you can see the --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
108 the lay of the land here. And the green that we see 1
-- that's supposed to be a kind of green on the -- on 2
the -- the schematic here for the micro-reactors.
3 It's pretty much apparent to the staff and maybe 4
everybody that these are getting an awful lot of 5
attention right now. We've been talking with Oklo.
6 They look like they are very serious at coming in on 7
the near term. There are others out there. Only two 8
of them have given us a RIS. And, you know, to talk 9
about when they are going to do submittals. But we 10 attended a micro-reactor workshop back in, I think it 11 was June. And it was surprising that there were --
12 there must have been about 10 or 12 different 13 organizations which are all targeting micro-reactors 14 and some -- some various flavor to come in. And 15 they're talking about having these up and going within 16 the next couple -- three years. I mean, very, very 17 short -- short time scales.
18 And the complication for the staff is --
19 is both with the -- the technical analysis and with 20 policy. The microreactors -- you can see, some are 21 stationary. Some are mobile. And that's a brand new 22 thing for -- for the staff to have to -- to really 23 deal with. But the other point on this landscape is, 24 if we look at some of the microreactors, we're looking 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
109 at things which are down maybe a megawatt, two 1
megawatts. That's almost more on a test reactors type 2
of scale -- up to some of these which I think are at 3
least several hundred. I think a couple of them are 4
several thousand thermal megawatts. So as we are 5
preparing for these, we are still a little bit leery 6
about saying, hey, one size of review is going to fit 7
all. And we think that is going to have to be 8
addressed as we move on. So our bottom line is to be 9
ready for all of these, and for the various types of 10 reviews that might -- that might go on. The next one.
11 MEMBER KIRCHNER: Before you go on, how --
12 you've colored also the liquid salt fueled -- two of 13 them in your green. Am I reading this chart 14 correctly? Terrestrial and terra power?
15 DR. BAJOREK: Terrestrial and terra power 16
-- I think the color is not coming -- coming through 17 here. That's --
18 MEMBER KIRCHNER: Okay, so -- so that's 19 not one of the leading --
20 DR. BAJOREK: That's not -- those aren't 21 one of the leads. I would say it's --
22 (Simultaneous speaking.)
23 MEMBER KIRCHNER: That would complicate 24 your life greatly.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
110 PARTICIPANT: You're asking about the 1
applicants?
2 DR. BAJOREK: Which applicants are most 3
likely to come in --
4 (Simultaneous speaking.)
5 MS. CUBBAGE: Yes, so we're not going to 6
speculate on that right now. The only thing I can say 7
publicly is that Oklo is planning to submit a combined 8
licensed application in December. And we are engaged 9
with those other applicants, and they have varying 10 plans. And as Steve mentioned, Kairos, we've had very 11 extensive pre-application engagement over the last 12 year, which is indicative of their progress.
13 MEMBER CORRADINI: So Amy, since you're 14 there.
15 MS. CUBBAGE: Yes.
16 MEMBER CORRADINI: Help me -- remind me --
17 so, the first expectation is directly to a combined 18 license application without a
without a
19 certification?
20 MS. CUBBAGE: That's right. So the 21 combined licensed application would include all of the 22 siting, environmental and design information in the 23 combined license and not reference to certify design.
24 MEMBER CORRADINI: Under 10 CFR --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
111 MS. CUBBAGE: Part 52.
1 MEMBER CORRADINI: Okay. Under 52.
2 MS. CUBBAGE: So there's an option for a 3
combined license that they could reference and certify 4
design, an ESP, neither or both. And they're opting 5
to reference neither and add all the information in 6
one application.
7 CHAIR BLEY: What makes that different 8
from Part 50, then?
9 MS. CUBBAGE: It's -- it's -- Part 50 10 would be the construction permit and then operating 11 license, whereas this would be a combined operating 12 license.
13 CHAIR BLEY: All at once? Okay.
14 MS. CUBBAGE: All at once.
15 CHAIR BLEY: But it wouldn't then be a 16 certified design?
17 MS. CUBBAGE: It would not be a certified 18 design. Of course, with the one issue -- one review 19
-- you know, we certainly could leverage this review 20 to -- to provide a license for a future reactor.
21 MEMBER CORRADINI: So I know we don't care 22 about this, but I -- in this committee -- but I guess 23 I am kind of curious. This strikes me as a -- well, 24 so maybe I am not understanding. So under Par 52 and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
112
-- but they would be using the -- I will get it wrong.
1 I want to call it the LMP. They'll be using the LMP 2
approach? Or it's not clear?
3 MS. CUBBAGE: I can speak to that as well.
4 We had some early engagement with them on a pilot for 5
using an LMP approach. And they have not yet formally 6
communicated how they're going to do their analysis.
7 And it may involve a maximum credible, or a maximum 8
hypothetical accident approach with some underpinnings 9
at LMP. But we're waiting to see that in their 10 application.
11 MEMBER CORRADINI: So -- so it could be a 12 light water reactor-like approach where -- or, a non-13 power reactor approach.
14 MS. CUBBAGE: That's possible.
15 MEMBER CORRADINI: Because most research 16 reactors essentially have an MCA. Okay.
17 MEMBER REMPE: So if they were to use an 18 LMP, how does that -- or, excuse me, a -- if they were 19 to use the maximum credible accident, why would we 20 need a detailed code? I mean, aren't you just going 21 to blow the core out?
22 MS. CUBBAGE: I think we would be looking 23 more at a MELCOR/MACCS type of an approach if we were 24 to do confirmatory calculations against something like 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
113 that event.
1 MEMBER REMPE: Okay.
2 DR. BAJOREK: Can we move on?
3 MEMBER CORRADINI: So -- I am sorry. I --
4 I know we're talking about this, but -- where'd she 5
go?
6 (Laughter.)
7 MEMBER CORRADINI: She's fast. She 8
doesn't want to be --
9 (Laughter.)
10 PARTICIPANT: I am sorry, what?
11 MEMBER CORRADINI: What I guess -- I mean, 12 Joy -- Joy asked a question about the tool. I want to 13 ask a higher-level question. If I take an MCA 14 approach like a non-power reactor, the licensing, if 15 audited and analyzed appropriately, could be 16 significantly simpler.
17 MS. CUBBAGE: There's a potential that we 18 could have a -- a simpler review if you have a -- a 19 scenario that we can all agree is credible, or -- is 20 bounding of credible events and significant 21 demonstrated margin. So those are things that we 22 would consider in the review -- how much margin is 23 there available --
24 MEMBER CORRADINI: Okay.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
114 MS.
CUBBAGE:
and what other 1
certainties exist?
2 MEMBER CORRADINI: Okay, thank you very 3
much. Appreciate it.
4 DR. BAJOREK: The approach that we've 5
taken in Volumes 1, 2 and 3 have been primarily to try 6
to identify the gaps in the needed capabilities. It's 7
not to try to tell how we're going to review any of 8
these, or even say what scheme we're going to be 9
using. But it's, what gaps are we going to need to 10 satisfy in order to have the tools that will do this 11 broad range -- okay -- of different -- different --
12 different types of designs. As we see it, we have two 13 distinct but complimentary and coordinated sets of 14 analysis tools that are going to look at this range of 15 anticipated staff review questions. Volume 1 is more 16 looking at -- I'm not even -- I'm not even word -- to 17 use -- I'm no longer use design basis anymore.
18 MEMBER REMPE: That's why it's in red, 19 right? That title has been put in red, and I wondered 20 if that's why you're thinking of changing the title?
21 DR. BAJOREK: Actually, we forgot to 22 change it in the first place.
23 MEMBER REMPE: Oh, okay.
24 (Laughter.)
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
115 DR. BAJOREK: Change it to black. Okay?
1 But the goal behind the types of codes that we get in 2
Volume 1 are to answer staff review questions -- are 3
the safety functions and the systems acceptable?
4 Okay. Basically like you would in a light water 5
reactor world, is the ECCS system adequate? We expect 6
to see those types of questions. When that plant, or 7
that system is operated, is it operating within the 8
operational limits that we feel are safe? And 9
finally, I think one of the best things that you 10 really get out of those types of tools is you -- is 11 you get the -- that's where you develop your staff 12 expertise. How does the machine work? Okay? By you 13 exercising those tools, you start to understand what 14 are the physics? What are the phenomena which are 15 important to the operation of that design? What are 16 the sensitivities to those phenomena? Okay? And what 17 are the uncertainties that you should be considering 18 as part of that review to make sure you don't head 19 yourself over some type of a cliff?
20 On the opposite end of things, when you're 21 looking at the -- the analyses from Volume 3, you'll 22 start off, well what is that fission product 23 inventory? Where's the -- what is the source term?
24 And where do those fission products go? We can look 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
116 at it from that opposite end. And even if you do get 1
a -- a smaller-type of reactor -- more on a research 2
end -- you could still have a fairly sizable fission 3
product inventory. Remember lifting rods in a trigger 4
reactor. You made sure there was nobody in the bay 5
when you lifted one out of the water because you could 6
get a very sizable dose very quickly. So even though 7
it's a small power megawatt size, you may still have 8
a fission product inventory that you want to make sure 9
is contained.
10 So together, as we exercise both of these 11
-- these types of tools, with Volume 2 feeding into 12 either one of these, okay, this is where the staff is 13 going to learn how these new machines operate. That 14 is going to be important to us because in many cases 15
-- or in some cases, we are dealing with systems that 16 do not have a lot of operating experience. We have 17 not operated or developed these in the past. And 18 they're being developed by organizations that, I 19 think, have been characterized as mom and pop shops 20 where they do not have a lot of experience in 21 licensing reactors, licensing fuel, or licensing 22 evaluation models in order by which to -- to evaluate 23 them.
24 A few questions on the role of NRC codes 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
117 with respect to the LMP. In development of the tools, 1
this remains to be defined. Okay, it's going to 2
depend on what the applicant submits to it, which 3
route they want to go. They may or may not use the 4
LMP. They may decide to go to something -- something 5
different. So our goal? Be ready because we don't 6
know what that approach is going to be. And we don't 7
know what the user office is going to need as part of 8
that review at this point. They may need those 9
details at some point, they may not. But we need to 10 be ready on both ends of the spectrum.
11 Now we've gotten a few of the LMP pilot 12 studies in. And -- and I think that they do provide 13 a little bit of insight on where review questions 14 might be generated. Can I just put the -- the LMP up 15 there, with just a few stars up there on where things 16 are -- what you see. I would characterize the ones 17 that I've looked at -- most of the stars are on that 18 y-axis.
They're on that frequency curve.
19 Essentially, they're saying there is no source term.
20 There is not going to be release. The defense-in-21 depth barriers work, and there's lots of margin. That 22 is where we would expect our reviewers to start having 23 question. Do you have that safety margin? Do you 24 have that defense-in-depth? If I fail that next 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
118 barrier -- might be the fuel, it might be the vessel, 1
okay? Top or bottom. I mean, we've got very thin-2 walled vessel. It might need a heat exchanger.
3 Something along those. Does that point on that left 4
axis and it's uncertainty with those phenomena pop you 5
to the opposite side here?
6 Given that the vast -- almost the majority 7
of points are lining up on there, we would expect some 8
of those review questions to start focusing on those 9
types of questions. And our belief right now is that 10 is where you are going to need some detail in your --
11 in your analysis in order to justify that margin, or 12 show it doesn't exist. You've got other cases out 13 there where, yes, you're going to have a source term.
14 There's going to be a non-zero dose. And I think 15 that's very apparent in gas-cooled reactors where you 16 have circulating activity, you have graphite dust, you 17 have air ingress, you have water ingress. You have a 18 high pressure system to begin with. And there's going 19 to be some mobility of those -- those fission 20 products. That is probably a case where I am hoping 21 when it comes to the tools I am dealing with, go away.
22 Okay?
23 CHAIR BLEY: That would be really nice.
24 Can you tell us anything about what the pilot studies 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
119 have been?
1 (No audible response.)
2 CHAIR BLEY: Are they -- the pilot study 3
4 (Simultaneous speaking.)
5 CHAIR BLEY: -- the LMP pilot studies?
6 DR. BAJOREK: There's been, I think, four 7
or five of them --
8 CHAIR BLEY: Are they public? Are they 9
available?
10 DR. BAJOREK: I --
11 MEMBER KIRCHNER: Are these -- are these 12 the desktop exercises?
13 (Simultaneous speaking.)
14 PARTICIPANT: I believe so.
15 DR. BAJOREK: Several companies --
16 PARTICIPANT: So they provided us several 17 references.
18 MEMBER KIRCHNER: Oh, it's the same ones?
19 PARTICIPANT: Yes.
20 MEMBER KIRCHNER: That's good.
21 DR. BAJOREK: There's been four or five --
22 I'm not sure if they're public or not. That's why I 23 kind of made up the --
24 (Simultaneous speaking.)
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
120 PARTICIPANT: Okay, yes.
1 MEMBER CORRADINI: I think staff last year 2
provided -- I am trying to think which two.
3 PARTICIPANT: This one, I know -- one.
4 PARTICIPANT: There was PRISM, too.
5 DR. BAJOREK: PRISM -- PRISM was out 6
early. There's -- there's a --
7 PARTICIPANT: PRISM and MHGTR.
8 MEMBER CORRADINI: That's right. That's 9
the one.
10 DR. BAJOREK: We've also seen a couple 11 recent ones from a molten salt and at least one of the 12
-- the microreactors. At least conceptual. So we're 13
-- we're starting to get more information out there.
14 I am just not sure whether it's publicly available or 15 distributable. But anyway, the bottom line is, we're 16 starting to see a couple of ranges -- a couple of 17 different types of situations in -- you know, our goal 18 then is to be prepared for this range of questions, 19 okay, that we just don't know because we don't have 20 the application yet.
21 MS. WEBBER: I think John Segala is at the 22 microphone.
23 MR. SEGALA: Yes, hello, this is John 24 Segala, Chief of the Advanced Reactor Policy Branch in 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
121 the Office of New Reactors. We have six pilot 1
reports. They're all on our public website under 2
advanced reactors. You can click on all of them and 3
take a look through them.
4 PARTICIPANT: Thank you.
5 MR. SEGALA: They're all limited scope.
6 They don't do the entire LMP process.
7 MEMBER CORRADINI: Thank you, John.
8 DR. BAJOREK: Okay. In one of the emails 9
we received there were a number of questions. So I 10 tried to organized --
11 CHAIR BLEY: Yes, I -- I kind of picked 12 those out of the transcript and from note people fed 13 me. So those are individual people's questions or 14 comments.
15 DR. BAJOREK: Thank you --
16 MEMBER CORRADINI: These aren't the same 17 questions that Mike asked of us 18 CHAIR BLEY: That Mike asked of us?
19 MEMBER CORRADINI: Mike Case asked us some 20 questions as we had --
21 (Simultaneous speaking.)
22 CHAIR BLEY: No. Well, unless he parroted 23 them back.
24 MEMBER CORRADINI: Okay.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
122 CHAIR BLEY: They came from us first.
1 MEMBER CORRADINI: Oh, okay.
2 DR. BAJOREK: These are the ones that came 3
from -- from your email. And I wanted to cover some 4
of these and I don't think we have time to give to 5
each and every one. So certainly ask if you have 6
questions. One of those, you know, what are the most 7
significant modeling and simulation gaps for non-LWRs 8
in the reports submitted? And, you know, we -- you 9
know, an ethic and prioritize. You know, we -- we 10 feel that the Volumes 1, 2 and 3 do point out the gaps 11 that are needed both in what you need to do to the 12 codes, okay, and where some of the experimental data 13 is needed and lacking and what the validation you're 14 going to need. We do agree that, yes, there needs to 15 be some prioritization. As we move forward it's 16 certainly in our best interest to start to focus on 17 one or two of these designs and try to make more 18 progress in those rather than spreading the wealth 19 among everybody at this point. And that's going to be 20 especially the case as we start to do more validation 21 work -- which is fairly time intensive.
22 Right now I would say that microreactors 23 and the Kairos design look like they're maybe a little 24 ahead of the pact. But that's -- that can change 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
123 quickly. Data needs something that's very important.
1 And now that we're starting to identify the gaps and 2
what validation is, kind of one of our next steps is 3
to continue with that validation and identify data 4
needs. We had a recent meeting with Department of 5
Energy and the labs a couple of weeks ago. And we're 6
characterizing the data as data that preexists. Okay, 7
maybe like EBR -- EBR-II. We need to get it. We need 8
to understand it. We have to bring it in-house. We 9
have to be assured of the quality of that data. And 10 I think the question is on completeness. You know, 11 they -- extruding versus the casting. Those types of 12 questions. But first we have to get those data.
13 There's a lot of other data out there. We 14 don't have ready access to it, so we're trying to 15 identify where it's at. Is it in a format that we can 16 still use? Because a lot of these data are quite old.
17 Analysts now want to have this in electronic format so 18 they can go ahead and use and start to get this in --
19 in the case where there are gaps in the experimental 20 data, nobody has it. We're trying to point that out 21 and at least identify to Department of Energy in the 22 labs that for us to be convinced of the safety case, 23 there are uncertainties that are going to have be 24 addressed in the data. We've started that process.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
124 And I think this is the -- the next step in 1
understanding where we need to develop these codes, 2
and what's important for some of these -- these 3
reactor designs.
4 Does this strategy represent the best 5
course of action with -- to develop sufficient 6
expertise? What changes? We've hit on this a few 7
times, and we think that the introduction report -- I 8
called it the overview here -- we need to build on 9
that one because that's -- it's too vague and 10 confusing at this point -- is what the approach. You 11 know, I think the -- the idea is that, hey, we have a 12 set of tools that is going to be looking at adequacy 13 of the safety functions. And I use that term as 14 rather than an active system because everyone wants to 15 go to passive systems, or a microreactor where it's 16 natural forces that remove the energy from the system.
17 Okay, they may not even want to say that they have a 18 safety-significant component as part of all this. But 19
-- you know, it's -- so we needed to get the adequacy 20 of those safety functions and also be able to make 21 sure we understand the source term and its 22 distribution for both internal and external events, if 23 that's what results in an MHA.
24 We think that both capabilities are going 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
125 to be needed as the NRC moves ahead. If we're going 1
to be doing our independent analysis, and we're going 2
to do it in an efficient, effective manner, okay, we 3
expect that the reviews to be much more expedited than 4
they have been in the past. I think that the 5
applicants need that, and that might be justified 6
because of the margins which are out there. But that 7
means for us, we have to get those capabilities 8
developed before those applications come in the door.
9 I am sorry, Mike?
10 MEMBER CORRADINI: No, no, no. I'm just 11 listening. I'm trying to understand the third bullet.
12 So are you saying the staff has a -- a consensus view 13 on how to re -- I will use the rewrite, or re-describe 14 the overview of the attack of this?
15 MS. WEBBER: Well the -- I think the 16 reference to the overview report is really a reference 17 to the introduction, which we presented on May 1st.
18 Is that right, Steve?
19 DR.
BAJOREK:
- Yes, that's the 20 introduction.
21 CHAIR BLEY: Are you in the process of 22 revising that? Might we see a revision before you 23 come back in a few weeks?
24 MS. WEBBER: We hadn't start revising it 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
126 yet, but --
1 CHAIR BLEY: Well, then you aren't going 2
to give us any --
3 (Simultaneous speaking.)
4 MEMBER CORRADINI: Okay, that's fine. But 5
-- so let me a little bit about it. The -- the first 6
thing in quotes tells me that you would need a set of 7
tools to make sure that I'm within operational limits 8
for AOOs? I'm still -- I'm struggling to understand 9
-- I understand you need a tool for the source term.
10 That I get. I don't understand the first thing in 11 quotes -- what are -- your meaning there?
12 (Off-mic comments.)
13 MEMBER CORRADINI: Yes, I know that, but 14 I --
15 (Simultaneous speaking.)
16 MEMBER CORRADINI: Can I show that without 17 a tool? That's what -- what's going through my mind 18 is --
19 DR. BAJOREK: Some systems will look at a 20 decay heat removal system. Okay? And they may be 21 dependent on a shutdown to remove the energy, okay?
22 Is the staff convinced that that system works 23 adequately, which -- within the -- the operating 24 limits of the reactor? Some of the -- some of these 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
127 designs say we're so safe, we can do a unanticipated 1
loss of heat sink, or loss of flow -- multiple 2
failures. And we don't need an SSCs to do that.
3 Okay, well if you don't take credit for that and you 4
have to remove the energy by either natural forces or 5
your remaining systems, we expect those types of 6
things to be -- come up as questions in the review.
7 Now those -- now those are the types of 8
things that you don't have a source term.
9 MEMBER CORRADINI: No, I understand.
10 DR. BAJOREK: -- everything within the 11 system. But the staff -- we would expect that as we 12 look at -- you know, maybe an example -- and, you 13 know, this is -- there is a large break LOCA. Okay, 14 where you have a tool that basic function is to help 15 define the power limit for the reactor --
16 (Simultaneous speaking.)
17 MEMBER CORRADINI: No, I mean -- I mean, 18 let's just -- let's pick on something since I don't 19 know even what this thing looks like. The Oklo 20 design, it strikes me, if I have a heat pipe and I 21 understand -- based on experiments, the performance of 22 the heat pipe, now I have to understand how the heat 23 pipe couples to the core, how it couples to the power 24 conversion system, how it also couples to a -- some 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
128 sort of redundant or a high, reliable decay heat 1
removal system when it's shutdown. And then I have to 2
ask myself, how many of these things don't work and I 3
still am okay? So I could do that with a lot of 4
relatively simple calculations once I have the data on 5
the heat pipe performance. But without that data --
6 MEMBER KIRCHNER: Mike, since I worked in 7
that area --
8 (Simultaneous speaking.)
9 MEMBER KIRCHNER: Well, I -- probably what 10 they'll do is design it so whether the heat pipe 11 functions or not, doesn't matter. They'll just have 12 passive decay heat removal out of the -- by conduction 13
-- out of the body of the reactor core. And I 14 shouldn't pre-guess what they're design approach would 15 be. But that would be one approach.
16 (Simultaneous speaking.)
17 MEMBER KIRCHNER: And then you can 18 demonstrate with simple calculations that you could 19 reject all of the decay heat.
20 MEMBER CORRADINI: Okay.
21 MEMBER REMPE: So I guess to -- there was 22 an example brought up during the Volume 1 discussions 23 where I could see what Steve's saying make sense. The 24 gas reactor and the hot deck. Thermal stratification.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
129 That might be something that needs a tool. Now, you 1
could also just put a hole in MELCOR and blow it out 2
and see what happens with sensitivity studies. But if 3
a vendor were to come in and really need to reduce the 4
margin for some reason, you might need it. But I 5
guess -- and so I like what you're coming through here 6
and saying.
7 Associated with it, I think, is the 8
implied assumption that the resources devoted to it 9
may not be as much at first until you -- you know, 10 you're going to be ready, but you realize that these 11 detailed tools may probably not be needed as often as 12 the workhorse to get you source term. Is that a good 13 extrapolation of what you're saying too?
14 DR. BAJOREK: I think it is, and you know, 15 I mean, I'll throw out one example. You know, because 16 we are interested in microreactors, we've already 17 completed a -- a small reference model that looks and 18 should operate much like one of the microreactors.
19 And this is a way that we will be able to examine, you 20 know, the performance -- the operation. What happens 21 if you fail one heat pipe? The parts have said, hey, 22 scenario you've got to look at is cascading effects.
23 No, that's -- that -- you might get the -- with a 24 simple calculation -- might need to be more detailed 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
130 as you look at some of these structures, which are 1
kind of complex. And some scenarios. So you -- you 2
explore things on a simple basis. When we did this 3
one, we don't want a all-inclusive model of a heat 4
pipe and the sonic velocities inside the pipe. It's 5
a super conductor. Dial-in a large thermal 6
conductivity. See how it operates. Are you close to 7
a limit? Are you less than a limit? If I increase 8
the power, you know, a little bit higher, do I run 9
into a eutectic temperature in the metallic fuel?
10 Okay, when you go through some of those transients.
11 Once we get our feet wet on that, we're going to be 12 able to tell I think pretty quickly whether we 13 actually have to go to a -- a lot more detail, or hey 14 that's -- that's what you -- that's enough for the 15 staff to ask intelligent questions of the applicant 16 because it's their analysis that's going to be the --
17 the analysis of record.
18 CHAIR BLEY: I wanted to jump in. We --
19 I talked earlier and said, could you get us this 20 report ahead of time? I don't think that's 21 reasonable, even if Steve says it's not a big deal.
22 But if you had some slides at the full committee 23 meeting that outlined what you were going to include 24 in that revision, I think that would make our letter 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
131 writing much easier.
1 MS. WEBBER: Yes, I took that as a no 2
earlier in the presentation.
3 CHAIR BLEY: Okay. I think that could be 4
really important for us.
5 DR. BAJOREK: Final point on here, I think 6
one of the comments is a pilot study. You know, I --
7 just in talking amongst -- yes, we think a pilot study 8
on the staff's end is a good idea. Pick a design.
9 Stay away from proprietary issues. Address it from 10 both tools. How would we calculate something with the 11 LMP? It will at least be informative as we move 12 forward and we do get a real -- a real application in.
13 So -- it's a matter of resources and what would be a 14 design that we would be able to use in a -- more of a 15 public format?
16 Should the NRC consider developer and 17 applicant codes for confirmatory or sensitivity 18 analyses? I think, you know, our view is that we want 19 to -- we -- I think we agreed with one of the comments 20 we got there. We really don't want to just pick up 21 the applicant's tool and repeat some calculations.
22 And I think the idea here is that you have to 23 understand these codes. You just can't pick these 24 things up and use them as a black box. You have to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
132 learn how to use them. You have to understand what's 1
inside them. You've got to look under the hood before 2
you really understand how they really are. And many 3
applicants -- you know, the applicants are all over 4
the place in what they've talked about as potential 5
codes they're going to use. So we'd rather, you know, 6
focus on maybe a smaller set, learn those and use 7
those rather than try to address anything that those 8
50 or 60 designs might come in with.
9 We have found -- you know, our -- looking 10 at this from the safety function analysis point of 11 view is that using the -- the NIM tools in combination 12 with some of the NRC tools gives us something that 13 allows us to use the details if we want to go there.
14 But it also represents a very large cost savings to 15 the NRC. In working with DOE, we're expecting them to 16 do the code development -- the verification of those 17
-- of those codes. We -- yes?
18 CHAIR BLEY: What do you have to do to 19 have confidence in the codes once the DOE says, V&V is 20 done, these are great?
21 DR. BAJOREK: Well first of all, we work 22 with them to do V&V. And we say, hey, if you look --
23 (Simultaneous speaking.)
24 CHAIR BLEY: Okay, so it's not just --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
133 DR. BAJOREK: It's not just what -- now, 1
we're pointing out -- and that was part of the data in 2
the NIM, hey we're going to point out what validation 3
that we think is important for us to have the 4
confidence in the codes. And one way that's not real 5
difficult -- because as you look at the non-LWR --
6 there's not a whole lot of data out there. So you're 7
-- we're all kind of -- we all want to know which life 8
preserver we're going after. So that -- I don't think 9
that's a -- that's really -- that's not a -- a point 10 of contention. I think the developers realize, yes, 11 they have to validate those types of tools. Well, 12 they're doing that. And that's -- that's a big 13 expense. And I am not even sure we have the total 14 number of staff to go ahead and do that.
15 But, as part of the learning process, we 16 are doing some of that -- that validation ourselves.
17 We're engaged in an IAEA -- or we think we are going 18 to be able to do this one using a Chinese fast reactor 19 data. That's a separate issue. You know, that will 20 help us benchmark some of the fast reactor analysis 21 tools. We have a staff member that's setting up a 22 model for FFTF. We're going to do this. We're doing 23 this in-house. I am hoping to actually maybe talk 24 with Dr. Corradini. I am looking for the Wisconsin 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
134 RCCS data, and the NSTF RCCS data. We will be taking 1
some of these things and we'll be doing our own 2
validation because that's how we're going to learn and 3
we're going to train ourselves to make sure we 4
understand those tools, look under the hood to make 5
sure they're doing the right things for the right 6
reasons. And if they aren't working properly, we're 7
going to get on the phone to the developers and have 8
that corrected.
9 CHAIR BLEY: Let me -- you made the --
10 kind of a big point in Volume 2 that in some cases the 11 costs work out kind of equally no matter which way you 12 go. But also that, you know, you have a computer base 13 here that's pretty limited. Someone mentioned 14 earlier, I think it was James, that there's a way to 15 go --
16 DR. BAJOREK: Using the --
17 CHAIR BLEY: In the cloud and run --
18 DR. BAJOREK: We -- we've -- we think 19 we've mastered that. We've been doing that with TRACE 20 and --
21 CHAIR BLEY: So you don't think you'd be 22 computer limited should you need to use these more 23 than you expect you'll have to?
24 DR. BAJOREK: No, we're looking at it in 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
135 three different ways. One, we think most of these 1
things will run on desktop-type systems with multiple 2
processors. Those are our -- those aren't very 3
expensive anymore. Using the cloud gives us access to 4
a lot. But we've also been talking and working with 5
the Department of Energy, and they've given us access 6
to their high performance computing system. So, you 7
know, some of us have accounts on those. We're 8
starting to use those. So we've got three different 9
areas. But we think, for the most part, running on 10 desktop-type systems of 16 or 32 CPUs is going to 11 handle -- handle these types of things. And that's --
12 we're pretty confident that the high performance 13 system is not going to -- not going to come bite us.
14 CHAIR BLEY: And issue? Okay.
15 MEMBER REMPE: So could you go to slide 12 16 where you have BlueCRAB. And I'd like to understand 17 this last bullet a little bit more. You've got -- yes 18 because I think sometimes it's good to look at the 19 actual slides here -- or, the -- the codes.
20 PARTICIPANT: Got it?
21 (Simultaneous speaking.)
22 MEMBER REMPE: Yes, that one.
23 DR. BAJOREK: Let's -- let's --
24 MEMBER REMPE: So you're saying -- you 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
136 said the DOE codes is going to save us money. Which 1
of those codes that are now white -- because you --
2 you did listen to a prior suggestion of putting all 3
the NRC ones on the left. Which of the codes is going 4
to save you money there?
5 DR. BAJOREK: I think it's ultimately 6
MOOSE and SAM that will do most -- do a lot of this.
7 MEMBER REMPE: And help NRC in saving 8
money?
9 DR. BAJOREK: As I mentioned, MOOSE allows 10 you to do data transfers. Okay? And I think I said 11 on one of the other slides, you know, you see all the 12 stuff in coupling. For the Fortran lovers out there, 13 think of these as sub-routines. Okay? It's much 14 easier than it had been 20 or 30 years ago. Okay, so 15
-- but, you know, something like MOOSE allows you to 16 do these data transfers. It also allows you to do 17 tensor mechanics. So if I am looking at a -- let's 18 say a microreactor that is heating up, the big 19 feedback there is as it grows radially and axially, 20 that's your -- that's your source of negative 21 reactivity. That's what shuts it down. And that's 22 what you've got to predict. We don't have that in our 23 NRC codes, okay?
24 You have the same problems in a sodium 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
137 fast reactor, VTR. Now they -- they get a little bit 1
easier because the core plate is all in a constant 2
temperature bath and you can kind of do a side 3
calculation. But this allows you to attack those 4
types of things. SAM has some attribute in there that 5
allows you to get at thermal striping stratification 6
where systems codes don't do a very good job. And 7
just to point out, a couple of things that we've tried 8
to clean up on this is, you know, the solid arrows --
9 those are things which are done. We've completed 10 those. It's operating And we're ready to move ahead 11 on those. The dashed are things that we're working 12 on. And that's coupling fast either to trace directly 13 or through MOOSE. MOOSE might actually be the easiest 14 way. And just to kind of pick up on, you know, this 15
-- this coupling And why sometimes it's good -- right 16 now, if you go to FAST to TRACE, you've got that solid 17 line over there on the left. What that basically 18 means -- I go run FAST And FRAPCON. I take the input.
19 I give it to the analysts. Now, for those 30-some 20 heat structures, you type in the arrays. It may take 21 you a couple of weeks, but you'll have something to do 22 for those couple of weeks.
23 So doing that type of a coupling is 24 actually a big cost and time saver, okay? But anyway, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
138 let's assert a point here. Now, if you go to the next 1
one --
2 MEMBER REMPE: No, before you leave here.
3 You said MOOSE and I got the data transfer and there 4
might be even some numerics that you didn't -- or 5
numerical solutions that you might like -- and all 6
that, the SAM -- that's with the sodium reactor? Or 7
what does it do? Help me remember.
8 DR. BAJOREK: That -- oh, I am sorry.
9 That does all liquid metal reactors -- lead-bismuth, 10 sodium, also does molten salts.
11 MEMBER REMPE: Okay, why can't you put --
12 I thought you'd already put some liquid metal 13 properties in TRACE? And what does SAM do you don't 14 have the capability with your current --
15 (Simultaneous speaking.)
16 DR. BAJOREK: We could put them in the 17 TRACE, but SAM is where the validation is going on.
18 MEMBER REMPE: So --
19 DR. BAJOREK: I can --
20 MEMBER REMPE: It can -- because --
21 DR. BAJOREK: You know, I can go and 22 validate TRACE against all the liquid metal type of 23 tests out there, but I will be asking for my colleague 24 at the end of the table for quite a pot of money in 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
139 order to go ahead and do that.
1 MEMBER CORRADINI: But I think what Joy is 2
asking -- I am not -- I am not sure where she is 3
going, but I think at least on this question what she 4
is asking is SAM just simply a liquid metal RELAP?
5 (No audible response.)
6 MEMBER CORRADINI: Isn't it?
7 DR. BAJOREK: Yes, it's basically -- you 8
know, in one way it's RELAP that you can do a sodium 9
-- you know, you can do the liquid metals, the sodium, 10 molten salts.
11 MEMBER CORRADINI: I mean, it was 12 developed in a -- in a different group, but in -- for 13 all intents and purposes, a tube and tank model.
14 (Simultaneous speaking.)
15 MEMBER CORRADINI: It's essentially 16 orifice --
17 DR. BAJOREK: Yes, it's a systems code.
18 MEMBER CORRADINI: Okay, fine.
19 DR. BAJOREK: There are -- there is --
20 there is -- there are some models in there that kind 21 of gives you a -- a CFD light that you -- if you 22 choose to turn it on. So it has some of those 23 capabilities.
24 MEMBER CORRADINI: Okay.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
140 (Simultaneous speaking.)
1 DR. BAJOREK: But, anyway --
2 MEMBER REMPE: So -- okay, so then I would 3
be careful with that bullet that was earlier in this 4
presentation. It's not all the DOE codes, it's some 5
codes. And it may save you money is where you -- I 6
wouldn't be over-stating it. But there's selected 7
ones. It's not the -- everything in your --
8 (Simultaneous speaking.)
9 DR. BAJOREK: We've done an evaluation.
10 It will save us money.
11 MEMBER REMPE: Okay, but for all of these 12 codes? PRONGHORN and MAMMOTH, too?
13 DR. BAJOREK: I do it by application.
14 MEMBER REMPE: So all of these codes that 15 are in white and green?
16 DR. BAJOREK: Applying -- yes, yes.
17 MEMBER REMPE: Are going to save you money 18 if you use them? And you do want to do all of them 19 because the DOE is going to validate them to your 20 standards?
21 DR. BAJOREK: Let me -- let me just also 22 explain on this. This -- this figure, which we call 23
-- all that stuff on there BlueCRAB -- is sort of the 24
-- the artists' pallet. Okay? These are all the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
141 tools that we're interested in. But when it comes to 1
an application, okay -- modeling a sodium fast, a 2
microreactor -- you don't use all of these. Let's go 3
to the next one.
4 In our microreactor model, these are the 5
only ones that get turned on. Okay, you get cross-6 sections from SERPENT, okay? And you have your 7
kinetics passing information as the model expands.
8 And getting its heat-pipe information from SAM through 9
a very simplified component. So when it comes to 10 validating, you know, something for a microreactor, 11 you're looking at a very -- a small population out of 12 all of those tools.
13 MEMBER CORRADINI: I guess -- I guess --
14 I understand how you're explaining to Joy. I -- where 15 I am personally -- I have a hard time is, since I 16 don't know it, is if you have invested all of this 17 time and effort in SCALE and Parks, to switch to 18 SERPENT and MAMMOTH strikes me as a big training 19 exercise. I understand SAM because you either got to 20 put it there, or you got to put it in TRACE. But the 21 reactor physics part of this strikes me as an odd sort 22 of choice. Just me alone. I'm -- knowing -- because 23 I then drag your reactor physics people in and ask 24 them quite --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
142 DR. BAJOREK: This might be the blind 1
leading the blind, then, but one of the things that we 2
kind of have to keep in mind when we're dealing with 3
fast reactors -- and for that matter, you know, some 4
of the gas-cooled reactors -- the mean free path of 5
the neutrons are much, much larger than they are --
6 MEMBER CORRADINI: Sure.
7 DR. BAJOREK: -- in a light water reactor.
8 PARCS is diffusion-based, okay? MAMMOTH can do 9
diffusion, but it also has transport capabilities.
10 And it has a mesh that I can put the detail where I 11 want it, or I can ignore it. Okay? So I can -- I can 12 make it run much like a PARCS or a diffusion-based 13 code, but where I have to look at a situation where my 14 control rod over here is -- is affecting my assembly 15 over here, I can do that. I can't really do that with 16 PARCS. Joe can probably explain this much better.
17 MR. KELLY: This is Joe Kelly from Office 18 of Research. In this particular example for the 19 microreactor --
20 PARTICIPANT: Turn the mic on.
21 MR. KELLY: Okay, Joe Kelly from the 22 Office of Research. This particular microreactor 23 example is a good one to show why you would choose the 24 advanced tools. As Steve stated earlier, the two 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
143 primary negative activity feedbacks have to do with 1
radio expansion of the core support plate, and then 2
axio-thermal expansion of the fuel slugs. So we do 3
that, you know, with the thermal analysis in SAM and 4
then the tensor mechanics module in MOOSE. That gives 5
us the displacements. But how do you factor those 6
displacements into the reactivity feedbacks? In the 7
case of MAMMOTH, because this is all an unstructured 8
finite element mesh, we actually distort the mesh and 9
let the reactivity feedbacks calculate themselves.
10 And I verified that for uniform dilations by checking 11 with Monte Carlo solution versus the MAMMOTH solution 12 for the microreactor model and it worked very, very 13 well. And there's no way we can do that with our 14 legacy tools.
15 DR. BAJOREK: Thank you, Joe. The other 16
-- the other point that, you know, I might want to 17 make -- and if Dr. Petti is still there -- is this is 18 also a bit of an example on how we can take some 19 simplifications. And SAM, for example, we have a very 20 simple model in there to represent the heat pipe. And 21 we're not convinced we need all the detail. That's 22 one place. But if you notice FAST and BISON are shut 23 off on this. Okay? The initial approach is just to 24 use models for thermal conductivity, specific heat --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
144 you know, those material properties -- those go in 1
either through SAM or for MOOSE. And just use it like 2
that. Okay? We can get most of the information out.
3 If we have to go to the detail, then we 4
can go back and start turning on something like FAST 5
or BISON if you need more information for the fuel.
6 Or if there's a reactor cavity cooling system that's 7
associated with this, a water-cooled system -- now 8
that we already have the link over to TRACE and we can 9
model helical coil tubes and all sorts of geometries 10 where we have boiling and sub-cooled boiling and stuff 11 that we really like to spend our time on -- okay, 12 because that's fun too -- we can go ahead and use 13 TRACE for that or any of the secondary systems.
14 So we do make it a little bit simpler for 15 the analyst that if he has to do something like model 16 the secondary system in RCCS, there are things that we 17 have some familiarity with. And we think that for 18 looking at boiling in tubes or boiling anywhere in a 19
-- in a bundle, we think TRACE is extremely well 20 validated for that type of thing. So it is taking 21 advantage of things that we've already validated for.
22 Taking advantage of tools that have largely been 23 developed for these types of tasks -- and taking 24 advantage of working together with Department of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
145 Energy where they are responsible for validation and 1
development of these tools. We take on the -- the 2
need to understand them, learn them effectively, and 3
use these to help address the questions that we get 4
from our friends at NRR or NRO. And I don't know 5
where we're at in terms of slides.
6 MEMBER PETTI: Steve, I had a question.
7 What's the strategy in the event DOE changes its 8
direction and decides not to support these at the 9
level that they decided to, you know, a year ago.
10 This is not uncommon in the DOE space.
11 (Simultaneous speaking.)
12 DR. BAJOREK: How would you know that?
13 (Laughter.)
14 DR. BAJOREK: That's -- that's certainly 15 an -- is a question mark. The funding for these types 16 of tools in advanced modeling simulations has been 17 consistent over the last ten years. I think in the 18 most recent bill, actually the number has gone up. I 19 am -- I am sorry, I can't predict the future, 20 especially with our funding. But --
21 (Simultaneous speaking.)
22 MEMBER CORRADINI: But, I --
23 DR. BAJOREK: -- we have the same problem 24 with our -- our funding is for those types of things.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
146 MEMBER CORRADINI: I guess what -- what --
1 there are still kind of underlying assumption, which 2
I am not sure you're the right one to answer this, but 3
I am going to put Kim on the hot seat. The assumption 4
is that what you get delivered from DOE is validated.
5 I am real concerned that you're going to actually have 6
enough data to take -- for these complicated 7
calculations and actually have the data you need.
8 That's the one thing I thought we started with that 9
Joy asked.
10 DR. BAJOREK: That's -- that's a bit of a 11
-- that's a separate question.
12 MEMBER CORRADINI: Right.
13 DR. BAJOREK: It's one, taking the data 14 you have and validating those tools. The other 15 question -- and that's one of the reasons we had this 16 data need meeting -- do we have all the data that we 17 need out there to validate it? And one of the things, 18 I mean, since we had microreactors up here -- we can 19 model these things. There's some information for heat 20 pipes. I don't think there's a whole lot of 21 information when it comes to taking one of these 22 monolith structures and ensuring that you can get 23 convection, radiation and conduction through a -- a 24 system where you may have a lot of contact resistance.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
147 Physically, you can do it.
1 PARTICIPANT: Right.
2 DR. BAJOREK: But the way you have 3
something that the whole system of codes can handle --
4 that's a shortcoming. And --
5 MS. WEBBER: But I also think that, you 6
know, with the notion of developing these reference 7
plant models ahead of the application, you know, the 8
purpose for that is to sort of evaluate where the 9
safety issues are, and then focus data needs in that 10 regard.
11 DR. BAJOREK: Oh, yes, and it was brought 12 up earlier -- the idea of uncertainties, that's key.
13 I mean, when we get these reference models, one of the 14 things we want to do are sensitivity studies to 15 explore and understand the uncertainties. Now you 16 kind of -- it's kind of code, telling you that. But 17 what are the phenomena which tend to be more 18 problematic in your -- your evaluation? At least 19 tells you where you can sharpen the pencil, or where 20 you might want to ask and push for more data. And the 21 other thing with uncertainties -- we don't want to put 22 any more codes on top of this. But the goal is to use 23 our SNAP tool. Symbolic Nuclear Analysis Programs.
24 It's a graphical -- a graphical process, say --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
148 processors that guys of this age like to use. You 1
know, I -- I kind of like card decks. But, you know, 2
that's a way of processing the input and output. But 3
we have -- we have linked that. We have merged that 4
with the DAKOTA statistical package so that when we 5
run something from SNAP, we can take a model, run it 6
in times -- whatever you need to get the -- the 7
statistics -- bring that stuff back and then you get 8
a statistical evaluation of your figure of merit and 9
some of the coefficients that you get you out of it 10 that tell you which ones were more dominant.
11 So you know, our goal as we go along is 12 certainly to integrate uncertainty in the uncertain 13 capabilities that we would use with light water 14 reactors. So that's -- you know, not talked -- talked 15 up much in the report, but that is part of the -- part 16 of the goal.
17 MEMBER REMPE: So this story hangs better.
18 A couple of questions. One, you are using these codes 19 to identify data needs, and the codes aren't 20 validated, so you might have a pitfall there. The 21 other thing is, I think I asked this a long time ago, 22 and I think the last time I asked it, I was told yes, 23 DOE said they would give you the source coding. Is 24 that still true?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
149 DR. BAJOREK: Yes. We have access to 1
that.
2 MEMBER REMPE: Okay. And you'll be able 3
to archive it? I mean, if you invest any time on it, 4
that way you better make sure because, like Dave 5
mentioned, DOE does change its mind sometimes.
6 DR. BAJOREK: If somebody comes along and 7
disbands Department of Energy, yes we would be able to 8
get the source code and move on. But you know, I 9
don't think that's going to happen.
10 CHAIR BLEY: Steve, just to give you an 11 anchor, you had just finished slide 8. So if Kim 12 types number 9 and return, she'll jump right to slide 13 9.
14 DR. BAJOREK: Go back one more please.
15 Yes, I think we covered that one.
16 MEMBER CORRADINI: I think -- I think --
17 but if -- but since you did it -- I think Dr. Rempe's 18 point about significant cost savings is -- without --
19 without putting words in her mouth, strikes me as a 20 stretch. I can see where there's cost savings in very 21 specific instances, but it's almost a case-by-case 22 sort of discussion.
23 DR. BAJOREK: Well, put together a model 24 for EBR-II, okay, FFTF, MSRE -- DOE is off doing 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
150 those. Now I -- I've kind of gone under the rule of 1
thumb that if you set up a plant model or a very large 2
integral tests, it can take you a good six months to 3
put those -- that model together with any kind of --
4 any kind of, simple -- you know -- you start adding up 5
things which are going to take, you know, a dozen or 6
so of these tests -- and separate effects tests -- you 7
start adding those up, you get to a large number.
8 Okay, we're talking about things that's already on 9
Department of Energy's plan, okay? We're going to 10 reinforce the need for those. They're going to do 11 them. They're going to try to push the -- but the 12 schedule --
13 MEMBER CORRADINI: But where I am pushing 14 back, though Steve, is where you pointed to SAM versus 15 TRACE in installing a liquid metal, I can see it. But 16 I would have to almost look at each one of these 17 individually to decide if I make the decision. That's 18 where the -- where I think Joy and I kind of reacted 19 to the bold and the underlying.
20 DR. BAJOREK: Yes, I think it's on an 21 application basis --
22 MEMBER CORRADINI: Okay, fine.
23 DR. BAJOREK: You know, you're closer in 24 one place to the other. Microreactors -- we aren't 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
151 going to get there with our -- our tools.
1 MEMBER REMPE: But you may not need such 2
a sophisticated tool for the microreactors. They may 3
not be pushing the margin. You might be able to do 4
something a lot simpler with MELCOR MACCS as mentioned 5
earlier by NRR.
6 DR. BAJOREK: Well like you said, we're --
7 we're becoming ready. I -- if we can approach it with 8
MELCOR and MACCS, okay that's good. In other reviews 9
that, you know, I've been involved with, the staff 10 wants to get into the details. Okay? We're --
11 (Simultaneous speaking.)
12 MEMBER KIRCHNER: It's not like that. I 13 mean, you're always going to have a safety function.
14 I was going to save this comment, but maybe this is a 15 good time to make it. There always will be a safety 16 function, even with the microreactors. Whether it's 17 passive or active, there's going to be -- for example, 18 advanced reactor design criteria in 26 is going to 19 have to be satisfied.
20 DR. BAJOREK: Sure.
21 MEMBER KIRCHNER: So you are going to have 22 to verify that. So you always have a safety function 23 like evaluation to complete. That's what sets a 24 microreactor apart from a spent nuclear fuel cap. You 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
152 have to ensure that it shuts down. So I don't think 1
you do that in MELCOR.
2 MEMBER REMPE: You have to make sure that 3
you don't go critical in a spent fuel path too.
4 (Laughter.)
5 (Simultaneous speaking.)
6 DR. BAJOREK: You know, kind of follow on 7
that, if -- if they say that the reactor will shut 8
down passively, we're going to want to be convinced 9
and do our calc and show, yes, it will shut down. And 10 if it shuts down, will it go re-critical under some 11 situation? And I -- you know, I'd like to -- you 12 know, I love to do hand calculations. They're more 13 fun. But it's -- you know, there are going to be 14 situations where we're going to have to delve in to 15 some of those details. And that's what we're -- where 16 we're getting ready for those.
17 CHAIR BLEY: To the comments from my 18 colleagues here. You've demonstrated -- at least to 19 my satisfaction -- that there are situations in which 20 there will be substantial cost savings. If we get no 21 applications, you know, it's -- there's no cost 22 savings anywhere. But you are prepared to take any 23 application and use the tools that -- or you're 24 becoming prepared to use the tools that will address 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
153 whatever the application is. And that to me is the --
1 makes sense.
2 (Simultaneous speaking.)
3 DR. BAJOREK: Ready and -- the one thing 4
about, you know, using some of the legacy tools -- we 5
saw that landscape of dozens of different reactor 6
types. I don't know what that landscape will evolve to 7
in five years. Are we going to have ten gas-cooled 8
reactors? Thirty microreactors? A couple of lead-9 bismuth and some molten salt? I don't think anyone 10 things we're going to see that. It's probably going 11 to wind up to a design or two. Okay? So at this 12 point, okay, we aren't taking all of this new 13 infrastructure and putting it in my TRACE or my NRC 14 codes, which I am going to have to hang on to. The 15 more I put into those codes, the more complex I make 16 those and their maintenance as I make any -- any other 17 change.
18 So at this point, it's not a bad -- we 19 don't think it's a bad idea to go ahead and use 20 separate set of tools, okay, and if the day comes that 21 we're building 50 microreactors, okay, and then we can 22 start doing some more consolidation and refining them 23 where we're at. We're not there yet. Where are we 24 at?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
154 Oh, last question -- we talked about it 1
earlier on the you
- know, the licensing 2
modernization and, you know, some of the -- the 3
various cases and sort of in conclusion, the -- you 4
know, the scope and type of independent calculations 5
-- they'll likely depend on what that design is, what 6
the margin is, what the perceived margin is. It may 7
or may not use the LMP. But because of the -- you 8
know, the lack of experimental and operational data 9
for non-LWRs, we feel there are going to be -- we 10 would expect there's going to be some technical issues 11 that are going to have high analytical uncertainty and 12 we may need to use the -- the details of these codes, 13 or we may be able to get around it. Time will tell.
14 And that's -- we have -- we've kind of covered, I 15 think the -- those parts there. So -- that's -- yes, 16 that was just a -- that's just extra. Anyway, I 17 wanted -- I do want to thank everyone's questions.
18 It's -- it's very engaging. And we appreciate it.
19 Thank you.
20 CHAIR BLEY: Thank you. Anything else 21 from the members? We'll be going around the table in 22 a few minutes. We're going to get the phone line 23 opened up. But while we wait for that, is there 24 anybody in the room who would like to make a comment?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
155 If so, please come to the microphone and identify 1
yourself.
2 MEMBER CORRADINI: They look happy back 3
there. They look happy.
4 CHAIR BLEY: It's open. It's magically 5
quiet. Anyone on the phone line who would like to 6
make a comment, please identify yourself and make your 7
comment. There's only one on the phone line. Say yes 8
or no. I guess that's it.
9 Well, at this point, I would like to go 10 around the table and I'd like to ask people to think 11 of all three volumes and the introduction when you 12 make your comments and thinking about things that we 13 need to flag in the letter and things we'd like to see 14 the staff focus on at the full committee meeting. Why 15 don't we start on the phone. Who's on the phone?
16 Matt Sunseri. Matt, do you have anything?
17 MEMBER SUNSERI: Yes. Thanks, Dennis. I 18 thought as far as the staff has gone to meeting their 19 obligation to Congress to develop codes and be ready 20 for the next generation of this work is a big step, I 21 guess, in that direction. And a lot of effort is 22 being put in.
23 As far as things for us to think about, I 24 agree with the comments that other committee members 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
156 have made regarding, and I'll characterize, the 1
framing the application of this suite of information.
2 So there's the various sections, if you will, that 3
apply to different functional areas and then some to 4
come and then the introduction.
5 But as it was mentioned, I think putting 6
something together like a framework of how applicants 7
will tie this thing, the beginning to end, would be 8
useful, I think. We can talk about that.
9 CHAIR BLEY: Thanks, Matt.
10 MEMBER SUNSERI: Thank you for all the 11 thorough presentations.
12 CHAIR BLEY: Pete Riccardella, are you on 13 the phone still?
14 MEMBER RICCARDELLA: I am. I am. I have 15 to say I found the presentations very education, worth 16 getting up early for. I don't have any comments.
17 CHAIR BLEY: Well, you can go back to bed 18 now. Dave Petti?
19 MEMBER PETTI: Yeah. I appreciated the 20 presentations. I just think that it's going to be 21 really important in that introduction to set the 22 context a little bit better. I think a lot of what 23 we've heard wasn't adequately reflected yet in that 24 introduction, the need for flexibility and different 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
157 approaches depending on when you need the sharpest 1
pencil, you get the sharpest pencil. But sometimes 2
you don't need that. Some of that, I think, in the 3
introduction would help it a lot going forward.
4 CHAIR BLEY: Thank you, Dave.
5 MEMBER PETTI: No problem.
6 MEMBER REMPE: Oh, yeah. I want to thank 7
everyone for their presentations and their work. I 8
agree that the intro but also the volumes will need to 9
have some tweaking to get -- it's not just the intro.
10 I really am glad the decision or discussion of saying 11 maybe we won't talk about design basis anymore and 12 we'll talk about more detailed tools or safety 13 function tools.
14 But I think writing the letter would be 15 simplified, as you've said, if you commit to all the 16 plan changes in the intro. As you do that, I'd recall 17 the opening statement or a question Mike had about 18 what's not still covered. If you have a transportable 19 reactor, that's going to be a big thing that the NRC 20 has still got to deal with in high enrichment and 21 things like that.
22 So it would make the story even hang 23 better if you identify what you have done and haven't 24 done. And again, I appreciate you considering our 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
158 comments and trying to address them because it helps 1
to better understand what you really had in mind.
2 Because I think many members have said it wasn't 3
communicated in the reports. And so a lot of the 4
questions came because we were concerned.
5 MS. WEBER: I think on just your latter 6
point, there is a paragraph in the introduction that 7
describes Volume 1, 2, and 3 and alludes to Volume 4 8
and 5. It's just not clearly identified as Volume --
9 Volume 4 is identified. Volume 5 is a little less 10 clear. But we can enhance that.
11 MEMBER REMPE: Okay. Thank you.
12 CHAIR BLEY: Mike Corradini?
13 MEMBER CORRADINI: Oh, you're going in 14 some different order.
15 CHAIR BLEY: Just random order.
16 MEMBER CORRADINI: Okay. All right. So 17 thanks to the staff. I appreciate that. I think 18 we've been doing this now for at least the third or 19 fourth meeting. I can't remember how long we've been 20 doing this. I think it all started because one of the 21 Commissioners asked a question and I won't put the 22 context other than that.
23 I guess I'm looking -- I think the way the 24 discussion has evolved since the last May 1st meeting 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
159 to this one, I think -- I sense that there's a coming 1
together of the staff in how you want to explain this.
2 But I think there's four attributes that have got to 3
be there.
4 Dave already said one which was the need 5
for flexibility. I think Walt and Joy mentioned it a 6
couple times, a need for simplicity. I think Steve's 7
point about -- I think -- I can't remember what slide.
8 It was Slide 7 where he pointed out the need for --
9 show the adequacy of safety function operational 10 limits.
11 There's a need for completeness. You 12 don't necessarily stay on one tool for everything.
13 But on the other hand, you want to make sure you're 14 complete. And then the one that I guess I thought I 15 said in May but I'll just repeat it here is you've got 16 to work the problem backwards.
17 The only reason -- for all intents and 18 purposes, the only reason we care about this is the 19 source term. So I'd work it backwards from there and 20 ask, what do I need to verify? I know what the source 21 term is. I essentially feel confident that the safety 22 functions per the advanced reactor design criteria are 23 met and then I stop.
24 I will say something that maybe we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
160 shouldn't talk about here but I worry about is if the 1
cost of licensing is the same cost and the size of the 2
machine decreases, all of a sudden, the cost of 3
licensing becomes predominant. That's a bad thing.
4 And if I were the reactor vendors, I'd sure make that 5
apparent to some audience and it may not be you guys.
6 So it seems to me I've got to make sure 7
this kind of fits together in some sort of overall 8
package. But those four attributes I think are 9
important. And I assume, as Ken was saying and Dennis 10 suggested, if you guys -- you can't rewrite it. But 11 if you can at least enunciate the key points of it 12 come the October full committee, I think that'd be 13 very beneficial. That's it.
14 CHAIR BLEY: Harold Ray is next.
15 MEMBER RAY: I agree with the comments 16 that have been made, Dennis. And I have nothing to 17 add.
18 CHAIR BLEY: Ron?
19 MEMBER BALLINGER: Yeah, I can't add -- I 20 was going to say pretty much the same thing about the 21 introduction and things like that but can't add much.
22 MEMBER DIMITRIJEVIC: I'm late to the 23 party, so I have nothing to add to a lot of writing or 24 anything. I just want to say that in general this 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
161 principle of simplicity and flexibility will be the 1
ones which will help most on expedite the review 2
process.
3 You said that you are training expertise 4
to expedite that process. But actually what would 5
expedite that process is finishing and developing 6
licensing approach which will allow for the lowest 7
facility to have a streamlined process which may not 8
require very complex calculations or uncertainties.
9 CHAIR BLEY: Thank you, Vesna. Charlie?
10 MEMBER BROWN: Not being a code person, an 11 electrical I&C guy, I actually understood some of what 12 James and Steve were talking about as well as Lucas on 13 the generalities. And I guess I appreciated that 14 presentation in terms of your interactions with the 15 codes and stuff like that.
16 On Steve's presentation on one of the 17 slides, I did not ask this because everybody else was 18 on a roll and it would've interrupted the flow on the 19 technical side. But on your Slide 5, you had a 20 heading, a role of NRC codes with LMP and then stated 21 in the first bullet that that remains to be --
22 CHAIR BLEY: Whoever is on the phone, put 23 yourself on mute, please.
24 MEMBER BROWN: The first bullet said, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
162 remains to be defined and will be based on applicant 1
submittals and user needs. And so I didn't ask it 2
because it was kind of a, what's the overall context?
3 I thought codes were codes and you needed codes. And 4
it made it -- it sounded like you didn't need any 5
codes based on somebody's submittal.
6 And so didn't quite understand the context 7
of the bullet relative to all the rest of the 8
presentation. I don't think that adds anything to the 9
letter. But I, quite frankly, did not understand the 10 idea that LMP is just thrown out the window based on 11 what the applicants may submit. So that was -- if you 12 have an answer to that, that would --
13 MS. CUBBAGE: I'd like to answer --
14 MEMBER BROWN: -- be appreciated.
15 MS. CUBBAGE: -- that. This is Amy 16 Cubbage, general staff. I actually reacted to that 17 bullet as well with the same thoughts you had. And I 18 think regardless of whether it's an LMP approach or a 19 more traditional approach, there will be events.
20 There'll be events. They'll need to be analyzed, and 21 you'll have to have codes to do it that are validated.
22 So the LMP are not -- it's a little bit of a red 23 herring in this context.
24 MEMBER BROWN: Okay. Codes are codes.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
163 You will be using codes. That's --
1 MS. CUBBAGE: Yes.
2 MEMBER BROWN: -- the way I read your 3
answer.
4 MS. CUBBAGE: Yes.
5 (Simultaneous speaking.)
6 DR. BAJOREK: Yeah. My point was to the 7
extent we have to apply. You have to use --
8 MEMBER BROWN: Whether simple or complex.
9 DR. BAJOREK: Simple or complex.
10 (Simultaneous speaking.)
11 MEMBER BROWN: I got it. Then I 12 understand. Okay. Thank you very much. Sorry to 13 throw that little --
14 CHAIR BLEY: Thanks, Charlie.
15 MEMBER BROWN: -- fish in the water here.
16 CHAIR BLEY: Walt?
17 MEMBER KIRCHNER: Thank you for the 18 presentations. I will be repeating myself, I guess.
19 But I see that the staff has a lot of flexibility and 20 choices. Their analyses are confirmatory, so they can 21 pick the tool they think they need to match the 22 requirements to make their assessment on safety 23 functions or whatever. It's different for the 24 applicant because the applicant has to make proof.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
164 I think FRAPTRAN which was most of the 1
presentation this morning. That is, I think, a good 2
choice, a prudent choice for the NRC. So that's one 3
member's opinion.
4 What I still remain unclear about, and 5
this is more for NRO, not for you and NRR. It's just 6
not clear to me what the expectations are for the 7
applicants. If the applicants pick up your document 8
and think that applies to them, I think that would be 9
very misleading.
10 So I
think in the introduction 11 strengthening what this is about and who it's for, 12 it's not for the applicants. It's for the staff --
13 MS. WEBER: I think that's part of the --
14 MEMBER KIRCHNER: -- to be ready.
15 MS. WEBER: -- intro.
16 MEMBER KIRCHNER: It's part of the intro, 17 but I think it could stand out and make it clear that 18 if they pick the same thing you pick, that doesn't 19 mean that they have verified qualified codes for their 20 application. And so I remain a little bit -- again, 21 it's a question more for a different time. But it's 22 not clear to me what the expectations of the 23 applicants are going to be for these non-LWRs.
24 I would expect that their codes would be 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
165 NQA-1 and DNV. And the burden is on the applicant 1
through a topical report or some other mechanism to 2
make that case.
3 DR. BAJOREK: In meetings --
4 MEMBER KIRCHNER:
And that's much 5
different than what is required of the staff.
6 DR. BAJOREK: In meetings with them, we've 7
tried to point out that, hey, this is a path the NRC 8
is doing. And it's because of the flexibility of the 9
number of types out there.
10 MEMBER KIRCHNER: Yeah.
11 DR. BAJOREK: If an applicant wants to go 12 a separate way, that's certainly their choice. And if 13 you only have one design to have to design and 14 license, it's going to look a lot different than what 15 we're --
16 MEMBER KIRCHNER: Yeah.
17 DR. BAJOREK: -- presenting here.
18 MEMBER KIRCHNER: So with that, thank you, 19 Dennis.
20 CHAIR BLEY: Yeah, thanks. I do want to 21 compliment all of you on really good presentations and 22 discussions and open discussions. I appreciate that.
23 I appreciate also what feels to me like a coming 24 together of thoughts since our first meeting. Well, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
166 it was kind of we got all these options and we've 1
evaluated some. Here, we're starting to think about 2
here's how we'll actually try to use them when things 3
come up.
4 And again, you won't have the document.
5 But if you can give us some slides on what that 6
revised overview is going to look like and the things 7
it's covering, I think that's great. I like Mike's 8
principles and I especially like Walt's mention.
9 Yeah, I was looking at these and I wrote a note to 10 myself.
11 These are probably going to be NUREGs, 12 although they don't say so on the cover. I'm not sure 13 what they're going to be, but they're going to be 14 looked at. And I think that first document has got to 15 make clear what you've said here today, that this is 16 for your to decide how to be prepared to review 17 things. And applicants are free to come up with their 18 own approaches.
19 And you haven't -- I don't think you've 20 put out guidance that tells them anything about what 21 codes are okay to use, if you use these, you're okay.
22 And that we're expecting -- and you've said it twice 23 today -- that these are going to be QA'ed and come in 24 with a -- I think the document itself, at least Volume 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
167 2 talked about having topical reports come in to get 1
approved on the codes people use. So making all that 2
stuff clear I think is important.
3 MS. WEBER: I think that --
4 CHAIR BLEY: I'm sorry. Go ahead.
5 MS. WEBER: -- to that last point, I mean, 6
I don't think in the history of NRC we've ever said 7
that applicant has to use Code XYZ.
8 CHAIR BLEY: No, but you've often said if 9
you use Code X, your life is going to be a whole lot 10 easier. And we're going to be looking for that at 11 some point. I was just hanging up I didn't talk to 12 Derek and I forget how much time we have on the 13 agenda. So talk with Weidong and Derek in between 14 times to see how much time. I'm not quite sure how 15 you --
16 MS. WEBER: It's two hours as I understand 17 it.
18 CHAIR BLEY: You squished this up to cover 19 all the volumes. So you got to think on that. Kind 20 of emphasize --
21 MS. WEBER: Yeah, the intro.
22 CHAIR BLEY: -- the intro and the kind of 23 things we were talking about that everybody has talked 24 about in our meetings and give a summary of the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
168 possible codes in each of the three areas and how 1
you're addressing that. So lots less than we had in 2
the two meetings, three if you count the research 3
meeting. But a real focus in on how they're going to 4
be used and what you're going to be expecting.
5 So thanks very much. At this point, we 6
are adjourned.
7 (Whereupon, the above-entitled matter went 8
off the record at 12:03 p.m.)
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
Advanced Reactor Code Development Status Fuel Performance Analysis for Non-Light Water Reactors Kimberly A. Webber Office of Nuclear Regulatory Research Presented to:
ACRS Future Plant Design Subcommittee Meeting September 17, 2019
NRCs Implementation Action Plan, Strategy 2 - Computer Codes Introduction Volume 1 Volume 3 Volume 2 = Fuel Performance - Subject of September 17th meeting Volume 4 = Radiation Protection - Work in Progress ML19093B266 ML19093B322 ML19093B404 2
Agenda
- Overview - Kim Webber
- Volume 2 - Fuel Performance Analysis for Non-LWRs - James Corson, Lucas Kyriazidis
- Response to May 1st and September 4th Questions on IAP Strategy 2 Codes -
Steve Bajorek Add pic of Volume 2 w ML #
3