ML21313A064
| ML21313A064 | |
| Person / Time | |
|---|---|
| Issue date: | 10/05/2021 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| Brown, C, ACRS | |
| References | |
| NRC-1707 | |
| Download: ML21313A064 (109) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Subcommittees on Metallurgy & Reactor Fuels Docket Number:
(n/a)
Location:
teleconference Date:
Tuesday, October 5, 2021 Work Order No.:
NRC-1707 Pages 1-82 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1716 14th Street, N.W., Suite 200 Washington, D.C. 20009 (202) 234-4433
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com 1
1 2
3 DISCLAIMER 4
5 6
UNITED STATES NUCLEAR REGULATORY COMMISSIONS 7
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 8
9 10 The contents of this transcript of the 11 proceeding of the United States Nuclear Regulatory 12 Commission Advisory Committee on Reactor Safeguards, 13 as reported herein, is a record of the discussions 14 recorded at the meeting.
15 16 This transcript has not been reviewed, 17 corrected, and edited, and it may contain 18 inaccuracies.
19 20 21 22 23
1 UNITED STATES OF AMERICA 1
NUCLEAR REGULATORY COMMISSION 2
+ + + + +
3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
(ACRS) 5
+ + + + +
6 SUBCOMMITTEES ON METALLURGY & REACTOR FUELS 7
+ + + + +
8 TUESDAY 9
OCTOBER 5, 2021 10
+ + + + +
11 The Subcommittee met via Video 12 Teleconference, at 9:00 a.m. EDT, David Petti, 13 Subcommittee Chair, presiding.
14 COMMITTEE MEMBERS:
15 DAVID A. PETTI, Chair 16 RONALD G. BALLINGER, Member 17 CHARLES H. BROWN, JR. Member 18 GREGORY H. HALNON, Member 19 JOSE MARCH-LEUBA, Member 20 JOY L. REMPE, Member 21 MATTHEW W. SUNSERI, Member 22 23 ACRS CONSULTANT:
24 STEPHEN SCHULTZ 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
2 DESIGNATED FEDERAL OFFICIAL:
ALSO PRESENT:
3 ZENA ABDULLAHI 4
LARRY BURKHART 5
DAVID HOLCOMB 7
HOSSEIN NOURBAKHSH 10 MICHAEL ORENAK 11 WENDY REED 12 JANET RINER 13 RICHARD RIVERA 14 MOHAMED SHAMS 15 TAMMY SKOV 16 CHRISTOPHER VAN WERT 17 SHANDETH WALTON 18 WEIDONG WANG 19 DEREK WIDMAYER 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
3 TABLE OF CONTENTS 1
Opening Remarks and Objectives 4
2 Staff Opening Remarks..............
6 3
Discussion of Report 9
4 Public Comments................. 81 5
Adjourn..................... 82 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
4 P R O C E E D I N G S 1
9:00 a.m.
2 CHAIR PETTI: This is a meeting of the 3
Metallurgy and Reactor Fuels Subcommittee of the 4
Advisory Committee on Reactor Safeguards. I'm Dave 5
Petti, chairman of today's subcommittee meeting.
6 ACRS members in attendance are Ron 7
Ballinger, Charles Brown, Joy Rempe, Jose March-Leuba, 8
Greg Halnon, Matt Sunseri, and our consultant, Steve 9
Schultz, is also with us. Christopher Brown of the 10 ACRS staff is the Designated Federal Official for this 11 meeting.
12 During today's meeting, the subcommittee is 13 going to hear about the Draft NUREG report entitled 14 ORNL Molten Salt Reactor Fuel Qualification. The 15 subcommittee will hear presentations by and hold 16 discussions with the NRC staff, staff from ORNL, and 17 other interested persons regarding this matter.
18 The rules for participation in all ACRS 19 meetings, including today, were announced in the 20 Federal Register on June 13th, 2019. The ACRS section 21 of the U.S. NRC public website provides our charter, 22 bylaws, agendas, letter reports, and full transcripts 23 of all full and subcommittee meetings including slides 24 presented there.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
5 The meeting notice and agenda for this 1
meeting were posted there. We received no written 2
statements or requests to make an oral statement from 3
the public.
4 The subcommittee will gather information, 5
analyze relevant issues and facts, and formulate 6
proposed positions and actions as appropriate for 7
deliberations by the full committee anticipated in 8
November.
9 The rules for participation in today's 10 meeting have been announced as part of this meeting 11 previously published in the Federal Register.
12 A transcript of the meeting is being kept 13 and will be made available as stated in the Federal 14 Register notice.
15 Due to the COVID pandemic, today's meeting 16 is being held over Microsoft Teams for ACRS and NRC 17 staff. There's also a call-in number with a pass code 18 to allow participation of the public over the phone 19 using Microsoft Teams. Refer to the bottom of the 20 published agenda for this number.
21 When addressing the subcommittee, the 22 participants should first identify themselves and 23 speak with sufficient clarity and volume so that they 24 may be readily heard. When not speaking, we request 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
6 the participants mute your computer, microphone, or 1
phone.
2 We will now proceed with the meeting and I'd 3
like to start by calling on Mohamed Shams, Division 4
Director of NRR, Division of Advanced Reactors and 5
NPUF for opening remarks.
6 Mohamed.
7 MR. VAN WERT: I believe Mohamed is actually 8
on travel right now. So I will be giving the opening 9
remarks.
10 CHAIR PETTI: Okay, so ahead, sir.
11 MR. VAN WERT: This is Chris Van Wert and I 12 am Acting Branch Chief for the Technical Branch No. 2 13 in the Division of Advanced Reactors and Non-Power 14 Utilization Facilities.
15 So just a couple of opening remarks here to 16 give you an idea of the framework in which we're going 17 to be hearing this presentation.
18 So first of all, thank you very much for 19 having us today and the kind of setting for this 20 presentation is that we've had these on-going efforts 21 related to fuel qualification guidance. This is in 22 support of NEIMA. And you recently heard a 23 presentation by Tim Gestwicki regarding NUREG-2246 24 which was the staff's fuel qualification guidance.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
7 This was, in general, a technology neutral approach to 1
fuel qualification, although it was noted at the time 2
that certain aspects of molten salt fuel reactors make 3
it a little bit difficult to apply the guidance to it.
4 So with that in mind, we have been in the middle of 5
ongoing efforts with Oak Ridge to develop guidance a 6
little bit more specific for molten salt 7
qualification.
8 And so we still are using a similar 9
framework, as you heard before with NUREG-2246, in 10 that we're trying to focus on fundamental safety 11 functions, but we recognize that there are differences 12 between manufactured fuel, solid fuel that we're more 13 familiar with, as compared with synthesized molten 14 salt fuel.
15 So this is also an ongoing effort with the 16 molten salt reactor fuel qualification. We have had 17 previous efforts with Oak Ridge and documents have 18 been presented related to this topic, but this has 19 culminated in this draft NUREG that we'll be 20 presenting today.
21 This is also something that will be 22 continuing after we complete this NUREG. There is on-23 going talks of a Reg. Guide that the staff will 24 develop in coordination with or with assistance from 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
8 Oak Ridge to endorse this NUREG. At this point, as 1
I'm clear, we will do a single new, single Reg. Guide 2
to endorse both NUREG-2246, as well as this NUREG, or 3
if we will use separate Reg. Guides to endorse these 4
documents.
5 And also, just to give you an idea of the 6
time line, this NUREG has gone out for public comments 7
and so we are expecting to get comments and hoping to 8
get comments, as well as incorporating any internal 9
generated comments here. And we will come back with 10 a final version of this report at that point.
11 So with that in mind, I just want to thank 12 you again and I will leave it up to Dave Holcomb from 13 Oak Ridge to present the technical meat of his 14 presentation, but if you have any questions for me, 15 I'd be more than happy to answer as well. Thank you 16 very much.
17 CHAIR PETTI: Thanks, Chris.
18 MEMBER REMPE: Dave, this is Joy. Just so 19 I'm on the right page, you've stated in your opening 20 remarks that you're planning to have a letter in 21 November, but yet, they're going to be coming back 22 with comments later.
23 Is it appropriate to have a letter now or is 24 staff requesting a letter now? Or why are we thinking 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
9 about a letter now versus after they come back with 1
the updated document? And I was just wondering about 2
that.
3 CHAIR PETTI: We had a meeting with staff 4
and felt that there probably would not be a lot of 5
significant comments that will come back from the 6
public and so we just decided that we would go ahead 7
with these comments because we've these comments, real 8
important for them to look at.
9 MEMBER REMPE: Okay.
10 CHAIR PETTI: David.
11 DR. HOLCOMB: So good morning, folks. My 12 name is David Holcomb. I work for Oak Ridge National 13 Laboratory. As Chris has indicated, we've been 14 supporting the NRC for the past few years looking at 15 what it would mean to qualify or develop qualification 16 guidance. And that's the purpose of the presentation 17 here. This is intended as a summary of the Draft 18 NUREG CR that has been shared with you and so I'm 19 trying to provide highlights from that. If you have 20 detailed questions on a specific text on there, please 21 feel free to raise those as well.
22 I'm representing here my colleagues, George 23 Flanagan and Mike Poore, as well, as we worked as a 24 team on developing this.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
10 Please, next slide.
1 MEMBER MARCH-LEUBA: Dave, this is Jose.
2 Before you move forward, just so I can focus on what 3
you are saying. In my mind there are two types of 4
molten salt reactors, two big pluses. One is where 5
the fuel or some material is dissolved in the coolant 6
and it moves freely through the reactor; another one 7
where you segregate the fuel into some, what do you 8
call it, cladding of some type.
9 We are going to talk only about the first 10 class today?
11 DR. HOLCOMB: We are talking about reactors 12 in which the fuel is a salt. In some cases, that 13 would be where fuel is in bins and then the salt 14 circulates in the bins. And those are small natural 15 circulation loops. They do not require to be pumped.
16 On the other hand, we are not talking about 17 reactors for which the fuel is not a salt. For 18 example, if it is a TRISO which is cooled by a salt 19 that would be a solid fuel. So this is intended to be 20 a generic capability of looking at fuel salt. And so 21 it is not that if it's segregated by cladding, if it's 22 still in a circulating form and natural circulation in 23 the fuel bins does indeed qualify as that.
24 MEMBER MARCH-LEUBA: Okay. Thank you.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
11 DR. HOLCOMB: I've got somebody waiting in 1
the lobby. I'm going to admit them.
2 Okay, well, we're starting as you've seen in 3
2246 they have a definition for fuel qualification 4
which involving behaving according to a manufacturing 5
spec. Well, as fuel salt is not a manufactured 6
product, it's a synthesized product, we are relying 7
upon a different definition that we're supporting.
8 And this comes from an NRC presentation from a few 9
years ago by Joe Williams.
10 This is about indicating that fuel 11 qualifications is a process that provides high 12 confidence. The physical and chemical behavior of 13 fuel is sufficiently understood that can be adequately 14 modeled in both normal and accident conditions, 15 reflecting the role of the fuel design and the overall 16 safety of the facility.
17 Uncertainties are defined so that the 18 calculated fission product releases include 19 appropriate margins to ensure conservative 20 calculations of the radiological dose consequences.
21 We'll keep coming back to this as this is our touch 22 point of what we're trying to mean -- to show that the 23 fuel is -- what it takes to be a qualified fuel. Sort 24 of pictorially on this is that fuel qualification 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
12 allows you to do the fuel performance measurement --
1 to do the calculations of fuel performance under 2
accident scenarios, under normal and AOO conditions.
3 We really upon the traditional functional 4
containment, defense-in-depth, and the barriers to 5
prevent the generation of the source term and provide 6
reasonable assurance of adequate protection.
7 Next slide, please.
8 Say fuel -- qualification we're building 9
from the NUREG-2246. We've got a slightly different 10 take on things because of the different 11 characteristics of liquids as compared to solids, but 12 it is still looking at the goal decomposition process 13 where you start at top level goal of the fuel is 14 qualified and break it down into individual sub-goals.
15 And we're really focusing heavily on the fundamental 16 safety functions as to determine what are the success 17 criteria for whether you have -- you do have 18 adequately performing fuel.
19 Well, part of the reason we have to do this 20 right now is that there are literally dozens of MSR 21 design variants under consideration and nearly all of 22 them have been developed in the past decade. And 23 they're emerging, I mean within the past month or two 24 we've had another U.S. company finally put out 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
13 publicly what its design is going to look like. Even 1
though they're not yet credibly going to be going to 2
the Commission I would not be surprised if Alpha Tech 3
Research at some point in the future says that a U.S.
4 company presents its innovative design for regulation 5
and it's very different from other designs. So which 6
designs are going to be presented for regulatory 7
adequately safety review, it remains uncertain.
8 So we're really looking at the process based 9
on the salt chemistry and physics which really is 10 largely independent of the reactor designer's 11 configuration.
12 Next slide, please.
13 So the fuel qualification, when you're 14 looking at the fundamental safety function, a function 15 that really is independent. You can support a 16 performance-based or prescriptive process. So 17 understanding how the fuel salt property support 18 achievement of fundamental safety function is truly a 19 performance-based evaluation. We evaluate the 20 acceptable range of fuel salt properties to maintain 21 appropriate margin from design limits under normal and 22 AOO conditions and limit the damage to safety-related 23 SSCs to ensure proper functioning during accidents.
24 Essentially, the fuel salt can't result in 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
14 so much damage that safety-related SSC ceases to 1
perform their functions. And the fuel salt is a 2
safety-related SSC, so it in part performs -- it 3
performs its safety-related functions, but it is not 4
the only safety-related SSC.
5 And also, the prescriptive -- we determine 6
the range of fuel salt properties that result in 7
compliance with specific requirements, both fuel and 8
coolant requirements. And that's really the major 9
difference with the solid is that you also have to be 10 in compliance with the coolant requirements.
11 Next slide, please.
12 So liquid fuel really does have substantial 13 fundamental differences from solid fuel. It is both 14 the nuclear fuel and the primary heat transfer media 15 and it must meet the requirements of both processes.
16 So liquid fuel is chemically damageable, but the 17 chemistry can be -- may be repaired during use. It 18 depends upon whether there's access to the salt.
19 For example, in a bin-type fuel, it's not 20 really accessible, but in a circulating loop there may 21 be accessibility to add or remove components during 22 use. Solid fuel is mechanically damageable. The 23 composition of liquid fuel again may be adjustable 24 during use, but solid fuel is not adjustable. It's 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
15 set prior to use.
1 The properties of liquid fuel depend upon 2
its composition and state. The properties of solid 3
fuel depend upon its fabrication processes as well as, 4
of course, its composition which is part of the 5
fabrication spec.
6 And another thing is because the fuel is a 7
liquid, fluids are formed to fit their container. If 8
you poke a hole in one container, the next container 9
becomes its container, and it could release nearly all 10 of the radionuclides from a breach. Unlike a solid 11 which doesn't form to fit its containers, so if you 12 crack a fuel bin all of the fuel comes out. You have 13 to really, massively rupture to cause a true loss of 14 all solid fuel.
15 Next slide, please.
16 Fuel salt qualification applies while it's 17 a regulated product at the reactor site. So from 18 receipt of regulated material until transfer to 19 independent storage. However, the properties that 20 you're concerned about depend upon where in the fuel 21 salt life cycle. For example, if you're fueling a 22 reactor for the first time, for example, if you've 23 transported a reactor with the 99 percent of the fuel 24 component in a transportable reactor, but it doesn't 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
16 have any of the fissile material and you're adding 1
fuel salt concentrate on that, you're certainly going 2
to be interested in the reactivity impact of the fuel 3
salt concentrate.
4 The fluid properties, on the other hand, are 5
largely immaterial once it's frozen as it's a used 6
fuel, once it's in a long-term storage. Materials 7
that leave the fuel salt and lack a reasonable means 8
cease to be part of the fuel salt and have to meet 9
other safety regulations. Things like insoluble 10 materials, fission gases, vapors, and aerosols because 11 part of the radioactive waste stream, once they no 12 longer have the ability returning to the bulk of the 13 fuel salt. On the other hand, plated-out materials 14 that couldn't re-dissolve or re-suspend in the liquid 15 remain part of the fuel salt.
16 Next slide, please.
17 So this means as one of the key issues is 18 what is fuel salt? What is and what isn't? Because 19 in solid fuel, we've traditionally cladding and fuel 20 assemblies are qualified as part of the solid fuel.
21 Other liquid fuel doesn't come in discrete elements.
22 It may be provided in barrels. It may be provided 23 where you get all of the non-fissile parts and then 24 you add fissile material slowly to it or you take 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
17 things out of it during service. So it doesn't really 1
come in rods or assemblies. And also it moves 2
independently of its container during normal 3
operations. And the container could even be replaced 4
as part of normal operations. You could drain it 5
into a drain tank with a new heat exchanger or a new 6
reactor vessel in and then return the fuel salt to 7
use. So it doesn't really -- it's not really an 8
integral system.
9 So fuel salt therefore includes all the 10 materials containing fissionable materials or 11 radionuclides that remain in hydraulic communication, 12 but does not include the surrounding systems, 13 structures, or components. Salt vapors and aerosols 14 remain part of the fuel salt system until they become 15 adequately trapped in which case -- until then they 16 become part of the radioactive waste stream.
17 The container corrosion products, however, 18 become part of the fuel salt. Fresh and used fuel 19 salt in on-site storage is within the scope of this 20 qualification process.
21 Next slide, please.
22 The common properties and the common plant 23 functions enable a general fuel salt evaluation 24 method. Although the specifics are very different 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
18 depending on the plant design and the specific 1
accident sequences are very much design dependent, the 2
basic operational and safety functions are common to 3
any nuclear power plant. That's -- the fundamental 4
safety functions pertain to all plants.
5 And the halide salt characteristics are 6
common to any MSR. They all have high boiling points 7
which result in low operating pressures. It's low 8
Gibbs free energy in the salt. Basically, it's a low 9
chemical potential energy. You're not going to get 10 vigorous chemical reactions with anything.
11 Natural circulation heat transfer properties 12 are also part -- they're just an inherent part of the 13 fuel salt. It's not a really good high thermal 14 conductivity material. In many ways, it's useful to 15 think of the heat transfer properties of liquid water, 16 water are very similar to those of -- with salt, just 17 at much higher temperatures.
18 The fuel salt interacts with its containers 19 by a common chemical and physical mechanism. For 20 example, by a thermal energy transfer, chemical 21 reactions, and mechanical processes, just the 22 hydraulic load, for example.
23 Next slide, please.
24 MEMBER REMPE: David, this is Joy. I should 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
19 have jumped in, I guess, on the prior slide, but this 1
is more relevant to the report than what's in your 2
slides.
3 Your report emphasizes the difference 4
between used and spent fuel because I guess with the 5
molten salt you'll be processing the molten salt and 6
then sticking it back in the reactor in some of these 7
designs. And if that -- first of all, it's kind of 8
interesting because for years -- we tried to 9
distinguish between spent and used fuel to say oh, we 10 just have used fuel and that didn't go very well as 11 emphasized by a recent GAO report.
12 But my question is more in how you're going 13 to have to pick an EPC for these reactors because 14 you're going to have to consider all the sources of 15 radionuclides on a site which means as you accumulate 16 the used fuel and you're processing it, you need to 17 consider that as a potential source for radionuclide 18 release as well as the spent fuel when you just can't 19 use it any more, right?
20 And that's something that the -- maybe this 21 is a question for the staff more than you, but it just 22 seems like that's something we're going to have to 23 think about, right?
24 DR. HOLCOMB: Well, it's certainly true that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
20 all radionuclides on site are of interest. And it may 1
be during the processing that you have the highest 2
potential for release because once you're no longer 3
processing and it's been used for a while, the natural 4
circulation heat transfer on there will go to decay 5
heat. So it's not a really exciting system once it's 6
been out of core for quite a while.
7 And it's a good question as to what choices 8
the designers are going to make because there are a 9
number of options about how much processing to do.
10 Some of the processing is just inherent. I mean 11 things that are insoluble are going to plate out or 12 you're going to filter them out. The melt of the 13 fission gases have really low solubility in the 14 materials and particularly in the fast spectrum 15 systems where they won't even go into -- they don't 16 have anything like graphite to go into. They 17 essentially all come out.
18 And yes, that rad waste stream is a 19 substantial source of the radionuclides there and that 20 may be stored on site. That may not be stored on site 21 and that has to be considered in the safety analysis.
22 MEMBER REMPE: So are the designers aware of 23 this because until there's a place -- again, this is 24 all design, conceptual design work, I suppose at this 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
21 time. But as the design developers are coming up with 1
their design, is this being emphasized to them that 2
they can't go around claiming that they can survive 3
with a low ECP until they actually figure out what 4
they're going to do with either shipping it off site 5
or accumulating it on site. These are big questions 6
in order to work with NEIMA which requires the whole 7
fuel cycle to be developed. And as we think about how 8
to pick a site boundary, those kind of issues need to 9
be really emphasized with the design developers, 10 right?
11 DR. HOLCOMB: Yes, certainly Terrestrial is 12 the furthest along on this and they are leading 13 through the Canadian process. But the announced 14 Terrestrial plans about going ahead and having seven 15 sets of cores on site is indicating they are thinking 16 about how they are going to be storing multiple 17 generations of the cores and indicating that 18 currently, they're planning on the once through fuel 19 cycle because it is a much simpler system, but they 20 are very amenable to if they can get both the 21 technical and the regulatory processes, essentially 22 continuously reusing the fuel is an option to them.
23 But their announced plans were seven cores 24 on site and they will have a net life time storage of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
22 their fuel for the entire life of the plants and 1
that's what they have indicated. Those are in public 2
statements. They're not in regulatory filings, at 3
least not public regulatory filings and you can see 4
where they are with the Canadian process, but they are 5
not nearly as advanced in the U.S. yet, but I'll let 6
Chris if he wants to comment further on that.
7 MR. VAN WERT: I don't have too much more on 8
that, in general, other than I want to point out that 9
that's kind of separate from the fuel qualification 10 NUREG that we're developing here. The intent of this 11 is to lead them down the path such that when they go 12 through their fuel qualification process they will 13 understand their fuel well enough to understand their 14 source term and be able to calculate the subsequent 15 analyses.
16 But yes, you are absolutely correct that 17 they do need to consider how they operate in the plant 18 in its entirety, whether or not they're reprocessing 19 on site, or just storing it, and then shipping it off.
20 I believe we have some efforts and one that 21 I'm thinking right now Oak Ridge has been developing 22 some considerations as well, and as I recall 23 correctly, some of the environmental concerns are 24 brought up in there and -- but as far as how far along 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
23 we are in communicating regulatory concerns, I think 1
the fuel qualification is the furthest along we are in 2
any of these topics, but yes, we do need to do more 3
regarding EPC and other topics as well.
4 MEMBER REMPE: I understand separate from 5
what the fuel qualification is, but on the other hand, 6
we are also considering emergency planning right now 7
and the Agency needs to make sure that we consider all 8
of these aspects as we move forward in Part 53 and 9
other types of activities. So thank you.
10 MR. VAN WERT: Agreed. Thanks.
11 DR. HOLCOMB: Okay. I think we've finished 12 with this slide.
13 Again, the method tailors the solid fuel 14 qualification process to the characteristics and 15 functions of liquid salt fuel. Modifications, both 16 add and remove issues from the solid fuel 17 qualification process. So we've got an example here.
18 Fuel salt is not a manufactured process in the sense 19 that NUREG-226 describes the solid fuel as 20 biomanufacturing specifications.
21 Liquids can't be mechanically damaged, just 22 a basic function of liquids. And fuel salt also 23 serves as the primary reactor coolant. So the fuel 24 salt properties determine its capability to adequately 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
24 support achievement to the fundamental safety 1
functions. And the fuel salt regulatory basis derives 2
from the role of fuel salt in establishing compliance 3
with existing regulations.
4 Next slide, please.
5 Liquid fuel salt does not have a
6 mechanically determined life time. The identification 7
of life-limiting failure and property degradation 8
mechanisms that occur as a result of the radiation 9
during reactor operation remains a key focus. We 10 don't really care that it's a radiation, so you can 11 say core's operations of fuel salt circulation, if 12 corrosion happens for the temperature differences and 13 stuff with use in core.
14 The fuel salt lifetime is the period during 15 which it contains an adequate quantity of fissile 16 material so it still works as fuel. Does not include 17 too many neutron absorbers. Again, still works as 18 fuel. And maintains acceptable thermophysical and 19 thermochemical properties.
20 The composition of the fuel salt may be 21 adjustable during operations to correct degrading 22 conditions. Again, it depends upon the design.
23 Next slide, please.
24 MEMBER REMPE: David, this is Joy again.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
25 Okay, so this slide in your report implies that there 1
is some sort of degradation associated with fluence, 2
then I thought some of your words on this slide 3
indicate well, there isn't any degradation associated 4
with fluence. Is there or isn't there?
5 And then what data do we have to say that 6
there isn't some sort of degradation with fluence? I 7
mean how far out? How many gigawatt days of burnup 8
have you ever seen with this molten salt?
9 DR. HOLCOMB: I'm trying to see --
10 radiation, of course, causes some degradation, burn 11 out to fissile materials. That is a degradation 12 process on this.
13 You also change the redox condition.
14 Fission is an oxidative process and that changes.
15 Redox makes the salt more corrosive on this.
16 As you add more materials to compensate for 17 burnup, you may reach a solubility limit on there and 18 then begin to plate out materials. Some of those 19 could be fissile materials that were played at places 20 you don't want them to. So there are a number of 21 adverse things that happen as a result of operation, 22 but as far as ionic liquid and the characteristics of 23 ionic liquids, they've had enormous amounts of 24 radiation applied to them. We've put them in beam 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
26 lines before at very high powered densities and you 1
realistically don't see anything because ionic liquids 2
are the recombination rates are multiple orders of 3
magnitude greater than anything on radiolytic damage.
4 It's just not -- there's not true radiolysis 5
that's going in these types of temperature range where 6
you're forming things, but on the other hand you're 7
getting a number of things which cause property 8
degradation over time which -- and that has things 9
like where you're accumulating fission products in the 10 salt. Or the fission products can form solids. These 11 are insoluble. They might form larger solids in 12 there. They agglomerate.
13 There just are things that happen, but it's 14 not to the basic salt. It is there are other things 15 that are accumulate. Again, we put those in, 16 quantities of fissile materials, amounts of neutron 17 absorbers, and some are physical and some are chemical 18 properties.
19 MEMBER REMPE: Has there much fluence --
20 what's the -- how much operating experience do you 21 have other than the MSRE and how long did it operate?
22 How much fluence did they see?
23 DR. HOLCOMB: Well, that's not where we got 24
-- tried to maximize fluence on there. If you wanted 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
27 to see, there were capsule tests done on an MTR, but 1
mostly if you were looking at peak things, that was 2
done to the Aircraft Reactor Experiment where they 3
were putting things into beam lines where things are 4
-- where you -- and most of those were like high-5 energy proton or radiations and they were putting peak 6
flux. And they were mostly looking at peak fluxes to 7
see whether you could do anything, rather than looking 8
at fluence.
9 And it's tough to say when you're taking 10 things out of the salt and putting more things back 11 into the salt, what the total fluence is on any 12 particular part of the salt.
13 CHAIR PETTI: David, just a question on --
14 you said that on these ionic liquids that 15 recombination is very quick and I certainly knew that 16 for the fluoride system, but there was some discussion 17 I knew among some salt researchers that that wasn't 18 necessarily known for all the different salts that 19 were under consideration, that the chloride system had 20 not experimentally established that. Is that true or 21 do we now believe that, you know, it's sort of like 22 the --
23 DR. HOLCOMB: Halide salt, is it -- we are 24 orders of magnitude from where you would be of concern 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
28 because people have used them where you put high 1
fluxes of other things like electrons or other things 2
into there.
3 You are correct that no one has run 4
experimentally -- them in a reactor. We have enough 5
analogies on there that recombination is not what 6
we're expecting to be a problem. That is just -- it's 7
such a fundamental property of ionic liquids, liquids 8
that if you arc and spark and you put protein beams 9
and electron beams and the like in them and you get 10 recombination. It is very much energetically 11 favorable.
12 CHAIR PETTI: Okay.
13 DR. HOLCOMB: But you're right, has anybody 14 run this in a reactor before? No, there's been very 15 little done in chloride salts.
16 CHAIR PETTI: Right.
17 MEMBER HALNON: David, one question from me, 18 Greg Halnon. I'm not super familiar with molten salt, 19 but are voids of concern during operation at all?
20 Does that change the composition?
21 DR. HOLCOMB: Bubbles matter, but it depends 22 upon how much your power density is on this or whether 23 your pump has done something like entrained bubbles on 24 there. And so it does things like well, gee, the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
29 reactivity, the other the reactivity 1
coefficients for the bubbles also changes things like 2
reactivity feedback because some of the expansion of 3
this is limited by if it's a bubbly material, so it is 4
of interest and it's something you have to be able to 5
model, but we're not expecting to have any real damage 6
as a result of this.
7 So you'll be getting things like fission gas 8
bubbles will be forming in this. Certainly, the MSREs 9
saw little burps in reactivity because at one point 10 they were ingesting gases which were part of their gas 11 drippings where fission gas dripping system and 12 pumping it through the core and they saw little 13 reactivity burps.
14 And so yes, that is indeed something which 15 does impact reactivity. Do we think it has any safety 16 significance? Probably not, not unless you've got 17 enough that you really are exhibiting adequate 18 negative reactivity feedback from the formal 19 expansion.
20 MEMBER HALNON: I guess in the morning I 21 made an analogy to the mechanical deformation of the 22 fuel pin and I guess when you mention there's no 23 mechanical deformation or effect on the fuel because 24 it's liquid, I guess that's where my head went. I 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
30 assume that that's all taken into consideration during 1
the fuel qualification process then?
2 DR. HOLCOMB: Yes, certainly, bubbles form 3
on this and one of the reasons no one is considering 4
hydrocarbons and lubricants on there, as it turns out 5
hydrocarbons, when they break down and go to the 6
surface of this there are fun surfactants (phonetic) 7 and it results in foam. And foam is not -- then foam 8
overflows and it does a number of things that he would 9
prefer it not to do. They don't seem to be really 10 substantial issues, but it's still having a foaming 11 system, I don't think is in anybody's desirable 12 design.
13 Otherwise, bubbles, you just have to have 14 enough free space up above it so that the gas can 15 escape and then -- in a sense, there's a lot of energy 16 in the fission gases. You have to adequately cool the 17 fission gases and all the bubbles coming off of the 18 system. But bubbles just reach the surface and 19 collapse and they inherently just migrate out to the 20 density difference in buoyancy.
21 MEMBER HALNON: Okay. Thanks.
22 DR. HOLCOMB: Any more questions on this 23 slide? Comments? We'll go on to the next one.
24 Functional containment is really important 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
31 to how MSRs provide adequately radionuclide retention.
1 The barrier performance must be degraded to release 2
the radionuclides to the environment and performance 3
degradation can occur through failure or for bypass.
4 Bypass is very important on MSRs. Anything from yes, 5
it's a low-pressure system, somebody accidentally 6
leaving a hatch open, to the fact that you have to 7
reject the decay heat out to the environment and 8
you've got lines that therefore go through the 9
containment. And you have to make sure that you don't 10 have rupture on any of the lines.
11 And fuel salt properties that stress 12 barriers cause them to be -- are more likely to 13 release radionuclides. For example, you increase the 14 temperature, be it higher vapor pressure in the cover 15 gas and you decrease the strength of the container and 16 we're operating much closer to the formal stress 17 limits because of we're at higher temperatures and you 18 get more money, you get the higher temperatures you 19 can operate it at. That's just higher thermal 20 efficiency.
21 And the performance requirements for 22 materials normally in contact with the salt versus 23 those that only need to withstand accidents are 24 different. For example, if you are not crediting the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
32 normally salt-wetted layer for accident performance on 1
there is only your guard vessel. Then that really 2
needs to withstand the salt -- the safety-related 3
component and you might get -- and that's for a 4
limited duration.
5 Next slide, please.
6 Fuel salt properties have substantial impact 7
on containment performance. Normal operation, the 8
salt-seeking radionuclides are chemically retained.
9 Most of the cesium, strontium, the traditional things 10 that you worry about in light water reactors really 11 tend to go ahead and stay in salt because they form 12 very high -- low vapor pressure, high boiling point, 13 fluorides or chlorides.
14 The fission gases, those with more than a 15 few seconds of half-life (audio interference). The 16 noble -- our definition of noble means not soluble in 17 the salt. Largely, they plate out or escape into the 18 vapor because they form nanoparticles and they're 19 salt-wetted so they're not in equilibrium there, so 20 they can just migrate out into the cover gas.
21 The fuel salt causes stress on potentially 22 resulting in damage to container materials, core 23 erosion, erosion, creep. The tritium behavior depends 24 upon operations, composition, and redox state. The 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
33 redox, if it actually is oxidated, we tend to 1
chemically bind the tritium and it doesn't tend to go 2
out. If you are in a reducing state which is good for 3
the corrosion, the tritium tends to free and then it 4
tends to go -- diffuse out through where it's got 5
large surface areas and thin walls.
6 MEMBER MARCH-LEUBA: Hey, Dave, this is 7
Jose.
8 MR. HOLCOMB: Yes.
9 MEMBER MARCH-LEUBA: When you say fission 10 gasses largely escape, you mean they escape the salt 11 12 MR. HOLCOMB: Yes.
13 MEMBER MARCH-LEUBA: -- they're still 14 retained by the vessel and containment, right?
15 MR. HOLCOMB: Yeah, yes. You have -- you're 16 required to have a containment around the fission 17 gasses. And you would have multiple layers of 18 containment around the fission gasses.
19 MEMBER MARCH-LEUBA: And in some reactors, 20 you may have a significant production of tritium. So 21 that's a concern?
22 MR. HOLCOMB: Yes. Tritium, tritium 23 probably isn't -- most of it is probably not going to 24 come out in that manner because the diffusion 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
34 coefficients of tritium are really small. So what you 1
tend to have is tritium is probably going to come out 2
through the power cycle mostly.
3 And that is a very significant concern 4
because it could be, you know, during normal 5
operations is when you have your largest potential for 6
radioactive material release into the power cycle.
7 And most of the designers so far have just 8
accepted the fact that we're going to have to have a 9
chemical trap in the coolant -- in the secondary 10 coolant. And that was indeed what the molten salt 11 breeder reactor program in the 1970s had intended.
12 And so the nitrate salts are very good 13 chemical traps for the tritium. And, but that is very 14 much a concern, mostly through the power cycle.
15 Though you're correct that there will be amounts that 16 (audio interference), there will be amounts that go 17 through the reactor vessel.
18 And then so you have to have a cleanup of 19 the tritium out of the containment gas base, if indeed 20 we're using a containment (audio interference),
21 because a number of people are keeping the fuels, the 22 entire vessel and cover gas system immersed in a 23 secondary coolant salt, so it's a liquid containment.
24 MEMBER MARCH-LEUBA: So there is very little 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
35 1
MEMBER BLEY: This is --
2 MEMBER MARCH-LEUBA: There is very little 3
hydrogen? Go ahead, Ron.
4 MEMBER BLEY: This is especially true for 5
FLiBe, right?
6 MR. HOLCOMB: FLiBe is worst, yep. Anything 7
which -- with lots of lithium or lots of beryllium.
8 You essentially think of it as quantities very similar 9
to what a heavy water reactor produces, except for the 10 fact that because the structural boundaries are above 11 300 C, and that's rough -- and it goes exponentially 12 above that. It's -- there-- it penetrates through.
13 But where the thin walls and high surface 14 area are, the heat exchanger tubes. And they have to 15 be thin to function, so most of it goes through the 16 power cycle.
17 MEMBER MARCH-LEUBA: But what I hear -- what 18 I hear you say -- what I hear you say is that there is 19 very little hydrogen in the system, so you can afford 20 to trap the tritium chemically.
21 MR. HOLCOMB: Oh, yeah, there is. Yeah, no 22 one wants to have water anywhere near these reactors.
23 Because water gives you the possibility of steam 24 explosions, gives you the possibilities of getting 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
36 high pressurization.
1 So even things like pump cooling and stuff 2
is probably going to be done by refrigerating the 3
atmosphere, you know, something other that -- or using 4
a liquid metal. Some other coolant that does not have 5
a phase change. We really will not be using --
6 MEMBER MARCH-LEUBA: Yeah, so --
7 MR. HOLCOMB: And it's just a -- and that is 8
one of the learning things out of MSRE is there, you 9
know, most, you know, their maximum credible accident 10 was leaking there -- they actually their shielding was 11 a pebble structure of -- that used water cooling.
12 And that intermixing with the fuel salt in 13 a double spill was their maximum credible accident.
14 So I don't think anybody that I've seen in any of the 15 designs has water near the system. So --
16 MEMBER MARCH-LEUBA: In my mind -- in my 17 mind tritium is a problem, you know, tritium reactors 18 because it hides in the tons and tons of water that 19 are there.
20 MR. HOLCOMB: Yep, and we really do not have 21 water. It's a bad idea to have water near an MSR.
22 MEMBER MARCH-LEUBA:
So it's really 23 conceivable to have a very good hydrogen filter 24 chemically which will grab -- 90% of what it will grab 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
37 will be tritium. Okay, thank you.
1 MR. HOLCOMB: Yes.
2 MEMBER BLEY: But it's, in the case of 3
FLiBe, the thing is it's very -- the diffusivity's a 4
very strong function of temperature. Ten degrees C 5
it'll go by a factor of 2. So if you have temperature 6
gradients in the system in the interface between the 7
-- well, primary and secondary, whatever you want to 8
call the secondary, you could have big differences in 9
migration rate of tritium depending on the 10 temperature.
11 MR. HOLCOMB: Yes. The tritium -- and T 12 hot, they'll be more mobile at the upper end of the --
13 in the heat exchanger. Again, it's the heat exchanger 14 tubing where more of the stuff will go through.
15 I think one of the leading techniques that 16 people are -- other folks are considering is to 17 saturate the secondary (audio interference) side with 18 hydrogen and then back -- then try to back-diffuse the 19 hydrogen into the primary system and essentially try 20 to block a diffusion out of the system by overwhelming 21 it within diffusion.
22 MEMBER BLEY: Yeah, that runs a little bit 23 counter to potential control, but anyway.
24 MEMBER REMPE:
So what kind of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
38 instrumentation is used to monitor these devices?
1 What was used at MSRE and what's being proposed?
2 MR. HOLCOMB: Do you mean -- what aspects of 3
these devices? I mean, thermocouples work at 4
temperature on air. And you know, PWRs traditionally 5
don't -- you know, do the heat balance on the 6
secondary side. And so you can do heat balance on 7
these things on the secondary side, much like you 8
would in other reactors.
9 You know, flow monitors on here, you have 10 differential pressure gauges on there, they are 11 traditionally used. They had not used any flow 12 monitoring on the fuel salt at MSRE because you shift 13 the redox -- because essentially the liquid in the 14 impulse lines was, traditionally is NaK.
15 And if you broke the -- had a leak of the 16 NaK into the fuel salt, you make it much more 17 reducing. And it turns out the uranium would then 18 attack the carbon and form uranium carbide in core.
19 And so they just didn't monitor it, monitor the flow.
20 There are a lot of different ways that you 21 might do this. On their activation flow meters, 22 certainly you can watch things going around. On their 23
-- probably, again, I think most of the measurements 24 are going to be made on the secondary side, on --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
39 MEMBER REMPE: What about flux? And again, 1
what about the interactions of the fuel salt with any 2
sort of sheath on these, on the instrumentation? And 3
again, you know, flux impurities, are they just 4
sampling to make sure their cleanup's working, or do 5
they actually monitor anything real time?
6 MR. HOLCOMB: Well, some of the things 7
people talk about are looking at the cover gas on the 8
-- here. But again, if you're saying what instruments 9
are they using to look at cleanup, generally there's 10 a filter line.
11 And so they grab what's on the filters, and 12 the filters are being replaced. If it's, you know, 13 and those are things like metal mesh filters, nickel.
14 Or they're, you know, you can grab sample in the 15 salts, and that's just a sort of bucket and well 16 that's, and then, you know, offline sampling.
17 You know, it depends. There are people who 18 look at using probes on this for electrochemical 19 probes here. But things like uranium concentration, 20 it turns out it's much more sensitive to look at its 21 impact on reactivity than anything else. And 22 reactivity you monitor with neutron measurements, 23 which don't touch the salt.
24 There are very few things that you actually 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
40 need to touch the salt to measure.
1 MEMBER REMPE: Okay, thanks.
2 CHAIR PETTI: But there probably will be 3
active redox measurements in most of these systems, 4
David, you think?
5 MR. HOLCOMB: I'm not sure, because you 6
could look at U3, U4 grab samples. What we did at 7
MSRE was look at it every few days. Because it's not 8
something we're expecting to evolve very fast. And we 9
expect it to evolve under a normal process. It's kind 10 of like confirming that your models are working.
11 So I'm not really expecting very much active 12 instrumentation on this. You know, it's -- the 13 chemical processes, we know sort of what's being 14 generated when fission happens. And okay, it's 15 oxidative. And then sometimes when the decay process 16 can't be reducing on there. Some of the chloride 17 salts will actually become more reducing over time 18 during the decay.
19 And but we have models of this. And what 20 you'll be doing is confirming that your models are 21 working. And none of these things, the reason you 22 care about them is the damage to the materials. Most 23 of those things are corrosion and those are long --
24 those are longer time period things.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
41 And you're not expecting, you know, sort of 1
these cliff edge, anything where I've got to know this 2
this minute. If I know it tomorrow or next week, 3
generally that's pretty good.
4 So as I say, most of the operational stuff 5
I expect to be done on the secondary side, just as 6
traditional reactor -- you know, PWRs are, just 7
because it's easier. And you get good information 8
there.
9 Again, there are high potential consequences 10 for large -- a large early gas releases. We're 11 pointing out that mechanistic analysis of radioactive 12 material depends upon the fuel salt properties.
13 One of the things you'll note is that 14 there's an awful of the energy, you know, like 40% of 15 the fission products have a noble gas someplace in 16 their decay chain. And if that noble gas has more 17 than a few seconds of lifetime, they will then, you 18 know, that'll escape. And so a fair amount of the 19 cesium-137 production comes out of the xenon-137 20 production.
21 And so you get a lot of cesium-137 22 production in the cover gas because it's got a 3.8 23 minute half-life. And sort of exactly where that's 24 produced depends upon what your design is.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
42 Are you trying to keep all the -- this in 1
your reactor vessel, or do you want to actively purge 2
this and put that into a different location so it's 3
not available for an accident later. Or it's gone 4
into some rad waste protection system as opposed to 5
there. And so there's a lot of design variance on 6
this.
7 CHAIR PETTI: So David, what about iodine 8
and tellurium? I've seen --
9 MR. HOLCOMB: Iodine, they -- a lot of these 10 things depend upon the redox on there. The tellurium, 11 for example, is, well if it goes onto -- if you keep 12 it reducing, the tellurium largely stays in the fluid 13 and then doesn't do very much.
14 On the other hand, if you let the things 15 become oxidative, in the nickel -- in the nickel 16 alloys that we have experience with, the tellurium 17 then plates onto the surface and goes into this --
18 goes into the grain boundaries and tends to form 19 surface cracks.
20 That could become important on things that 21 are thin, particularly things like heat exchanger 22 tubing over time. So that becomes why redox control 23 is so important, because we run enough experiments to 24 know well, if you keep the tellurium in solution, it 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
43 doesn't do that. It's got to get played out.
1 Iodine, there will be some iodine that will 2
come into the cover gas -- gas space, much like the --
3 cesium and such. And you may be an iodine and 4
fluoride things. And so that amount of stuff which is 5
in the cover gas either needs to be trapped or well-6 confined. There's no question about that.
7 You also have real problems with things like 8
if you end up with the zirc fluoride on there.
9 Because the zirconium fluoride might plug the cover 10 gas line on there because it's higher vapor pressure, 11 and then tends to solidify.
12 And then it doesn't have a nice, low melting 13 point. So you'd have to use something like a screw 14 augur in your cover gas line to send it back into the 15 fuel salt or send it for treatment.
16 Same thing with uranium tetrachloride.
17 Chloride. So there are a number of things that go on 18 there. Exactly where the iodine goes, it depends upon 19 the redox, depends upon the operations. And we don't 20 have a -- that's one of the ones that we don't have a 21 perfect map of where things are. And so we need to 22 get more information if we want to have detailed this 23 is where this radionuclide goes.
24 DOE is currently engaged in a number of salt 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
44 spill evaluation programs. On the other hand, we're 1
quite confident it doesn't generate things like 2
pressure waves. So your major issue is is just making 3
sure that you have a sealed containment or multiple 4
layers of sealed containment so that -- because 5
basically you cannot -- a large, early cover gas 6
release would be a catastrophic event.
7 So these things have to be designed so that 8
essentially you cannot go ahead and release all the 9
cover gas immediately. And so shearing off the whole 10 reactor vessel or some method for releasing most of 11 the cover gas early would -- you just -- that's --
12 that's must be a beyond-design-basis event well beyond 13 credible probably.
14 And we'll also pointing out, the MSBR, if 15 you're not familiar with that, was the eventual 16 project for, in the 1970s, for developing breeder 17 reactors in the large one. And they just had done 18 most of the calculations. And realistically the --
19 you know, 99.9% reduction in the radiotoxicity in the 20 cover gas in 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />.
21 So really, it's dropping very fast initially 22 on this. And so it is that early release has got the 23 most significance on there. It is not to say that 24 remaining parts, which are largely krypton and xenon, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
45 xenon, are not important to contain. But they're not 1
the same large land denial accidents after that 48 2
hours. Next slide please.
3 Fuel salt boundary breech accident 4
progression is part of performance-based and 5
deterministic fuel qualification. You know, multiple 6
locates in 10 CFR require evaluations of a possible, 7
postulated fission product release in core into 8
containments. So essentially, salt spill accidents.
9 Fuel salt or cover gas can't directly 10 express exterior containment layers without first 11 breeching the inner containment layer. So a lot of 12 the evaluation is how do you breech the inner 13 containment layer.
14 So high radiation and high temperatures 15 immediately outside of the fuel salt boundaries 16 substantially circumscribe the characteristics and the 17 materials adjacent to the fuel salt container.
18 So that's a lot of those instrument issues 19 and wiring issues and the like. You can't really have 20 electronics or, you know, or organic materials right, 21 even right outside the reactor vessel. The 22 temperature and the radiation environment is simply 23 too harsh.
24 The focus is on the fuel salt properties 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
46 that must be known to adequately model accident 1
progression interaction characteristic with materials 2
and containment. So you have to look at okay, what 3
might be in containment and what could you spill the 4
salt on, on what's credible about that. Next slide 5
please.
6 The fuel-related advanced reactor 7
requirements and similar for liquid and solid fuel 8
because both of them are fuel. So, and there's the 10 9
CFR 50.43(e)(1)(i) recommends that performance of each 10 safety feature -- requires that the performance safety 11 feature of the design has been demonstrated either 12 through
- analysis, appropriate test
- programs, 13 experience, or a combination thereof.
14 Fuel salt thermochemical and thermophysical 15 properties provide the information necessary to model 16 its role in the plant safety features and when 17 performing their plant safety functions. The fuel 18 salt properties vary with composition and temperature.
19 20 So we need to determine the fuel salt 21 properties across the range in temperatures and 22 composition that span the potential operational and 23 accent conditions.
And the quality of the 24 measurements need to be sufficient to enable modeling 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
47 the role of fuel salt in achieving the plant 1
fundamental safety functions.
2 So if you've got a variation with viscosity 3
with composition, well, we need to know the 4
composition well enough. Because some things don't 5
vary a great deal with composition. Some things vary 6
a great deal. Things like the reactivity. If you 7
change the amount of uranium in the salt, well, you 8
get a fair amount of reactivity change.
9 On the other hand, even with lots of fission 10 product buildup, there really wasn't much viscosity 11 change at MSRE. Next slide, please.
12 And the liquid fuel salt must meet the 13 safety intents of the coolant-related GDCs or ARDCs as 14 appropriate. So if we look at the Reg Guide 1.232 15 approach to this and we're saying well, that's, you 16 know, in Part 50 that's how you establish the minimum 17 principal design criteria. There is compliance with 18 the appropriate set of design criteria.
19 So GDC 15 requires the coolant system be 20 designed so the design conditions -- the reactor 21 coolant pressure boundary are not exceeded during 22 normal operations or AOO. ARDC just removes the 23 pressure from the -- from the coolant boundary 24 functioning.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
48 So the key choice of MSRs that employ 1
functional containment is which set of layers comprise 2
the reactor coolant boundary. And the layers that are 3
credited to achieve the safety function must meet the 4
requirements. So if a normally salt-wetted layer, 5
such as the reactor coolant boundary, is credited, the 6
fuel salt conditions that would damage the layer must 7
be tested and corrected before causing significant 8
damage.
9 So this is a real challenge, and that's what 10 Joy may have been getting to in the instrumentation.
11 Because it is not easy to check on the thin wall 12 materials in the heat exchanger tubes, for example.
13 And this is why I think a number of companies are 14 likely not to credit the normally salt-wetted layer.
15 It may be possible on something like the 16 reactor vessel, where you've got -- or you just make 17 it thick enough that you watch the chromium 18 concentration build up over time -- time. And then 19 you just sample and watch for the chromium, you know, 20 and say that if you don't have enough chromium I 21 haven't gotten deep enough into my system.
22 However, if I'm concerned about something 23 like getting a hole in my heat exchanger tubing, and 24 that's accredited safety function on there, then 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
49 you're going to have to be able to detect that under 1
normal operating conditions. And we don't really have 2
the technology to show small leaks in the heat 3
exchanger tubes on there.
4 Normally, you're going to be looking at 5
things like, well, you'll run the reactor, the fuel 6
salt at a lower pressure and your secondary coolant 7
would then be at a higher pressure. And so you'd look 8
at things like, well, what's the volume in my 9
secondary coolant. So I know I've got inflow into my 10
-- into the primary coolant, into the fuel salt.
11 And I think that generally you'll often be 12 moving the accredited boundary to the next layer, 13 which isn't nearly as radioactive on there, to be able 14 to monitor for damage under normal operating 15 conditions.
16 So watching things like degradation of the 17 heat exchanger tube boundary is a fairly, you know, 18 it's technically difficult.
19 MEMBER BROWN: It's Charlie Brown, can I ask 20 you a question?
21 MR. HOLCOMB: Sure.
22 MEMBER BROWN: The Naval Nuclear Program had 23 a substantial amount of experience with sodium-cooled 24 reactor in the early days, the Sea Wolf and its 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
50 prototype in West Milton. And the corrosion and the 1
leaks in the steam generator tubes was a very 2
significant issue, very difficult, and that's why the 3
whole project was scrapped and rebuilt.
4 It seems -- you're saying it's tough.
5 MR. HOLCOMB: I'm saying detecting it is 6
difficult. We have a number of materials which turn 7
out if you keep the salt in a reducing it condition, 8
and so you keep the redox there. Well, the materials 9
are in the alloy are in the most reduced stated. And 10 you can drive the corrosion down very effectively by 11 keeping the redox where it should be.
12 If you let the redox go, you're having --
13 you will have problems on this. But the monitoring of 14 this, and then because Joy was asking about 15 instruments that you would put in to do this, you are 16 not expecting it to actually damage.
17 However, having an instrument to check 18 whether it has, is -- because it is a very harsh 19 environment. And trying to get an instrument that'll 20 look at tube wall thinning in, you know, right next to 21 the fuel, that's not something we can really do well.
22 But we do expect to be able to -- I mean by 23 far, the dominant corrosion mechanism in a halide salt 24 is oxidative damage. And so you just drive it into a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
51 reducing condition so it won't -- so that -- so you 1
can make that very small. There were a number of test 2
loops run for up to a decade on there, and they look 3
pristine.
4 That is not to say that other modes of 5
corrosion do not occur. However, they're either 6
really fast, where you've done something like you put 7
a dissimilar material into this and then, well, you 8
find out that's not a good idea. In 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> you've 9
corroded through a stainless steel part when you put 10 it into a nickel alloy loop. Or they're very slow, 11 non-oxidative corrosion.
12 So yes, these things do exist, but if you --
13 so we are very interested in maintaining the proper 14 chemistry in the fuel salt loop. Much like water 15 reactors, maintaining the proper chemistry is very 16 important.
17 MEMBER BROWN: Okay, how do you keep this 18 stuff liquid all the time, as opposed to going solid?
19 MR. HOLCOMB: Well, normally --
20 MEMBER BROWN: I haven't seen a reactor like 21 this --
22 (Simultaneous speaking.)
23 MR. HOLCOMB: Your challenge is after it's 24 been run for a while is removing -- is continuing to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
52 cool it on there, because you've got buildup of 1
fission products within it. And so it naturally would 2
tend to overheat as opposed to cool.
3 Initially, before it's built that up, you're 4
going to need heaters. Heaters, you know, when you 5
put in fresh salt, you're going to have to keep it 6
hot. And so you'll have either external heaters, or 7
you may put in gas heaters for the cover gas. It's a 8
number of different ways.
9 But you know, initially until you build up 10 the fission -- enough fission products. After that, 11 it's used fuel. Used fuel overheats rather than 12 underheats.
13 MEMBER BROWN: Okay, thank you.
14 MR. HOLCOMB: Next slide, please. So the 15 liquid fuel salt assessment frame where it follows the 16 template developed for the solid fuel advanced 17 reactors, where we would take the top-down approach, 18 decomposing the top-level goal of the fuel is 19 qualified to lower-level supporting goals.
20 You
- know, qualifying fuel develops 21 confidence the fuel will adequately perform its role 22 in enabling the safety -- the facility to achieve its 23 safety objectives. Lower level supporting goals are 24 further decomposed until the clear objective goals are 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
53 identified that can be satisfied with direct evidence.
1 So if you look at this, that's a goal is the 2
fuel is qualified for use. That's a sub -- that's the 3
top goal in the solid fuel. And the safety criteria 4
can be satisfied with high confidence. Margin to 5
design limits during normal operations in AOO would be 6
met with high confidence.
7 The margin rating applied to release limits 8
under accent conditions, and the ability to achieve 9
and maintain safe shutdown can be assured. So it is 10 breaking things down into pieces. Next slide, please.
11 However, one of the (audio interference) you 12 know, liquid fuel is not a -- is a synthesized not a 13 fabricated process. So one of the major branches in 14 the solid fuel goal decomposition simply does not 15 apply to liquid fuel -- fuel. You know, both 16 properties can be determined by measuring, you know, 17 properties of product samples.
18 So if we just grab a sample of this, and 19 fortunately it's flowing well, so you just -- you get 20 a pretty homogenized system by grabbing. So neither 21 the fabrication-based definition of fuel qualification 22 nor a manufacturing specification branch is -- applies 23 to liquid fuel. So essentially truncate much of the 24 solid fuel program. Next slide, please.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
54 Qualifications based upon understanding the 1
chemical and physical properties of representative 2
fuel samples. A liquid state significantly changes 3
the behavior of fuel. They don't accumulate internal 4
- stresses, so we don't have history-dependent 5
properties. So I don't need to have accumulated a 6
particular stress state.
7 The flow homogenizes the fluid properties, 8
so you don't get position-dependent property or size-9 dependent properties. So I don't have the major 10 issues about leak test assemblies and bowing and such 11 that are all inherent characteristics of things being 12 solids.
13 The chemical and physical properties are set 14 by the elemental composition in temperature. This is 15 actually quite important because that means I don't 16 care about the isotopic content for the chemical and 17 physical properties, which means that I can take -- so 18 one I have determined what the elemental composition 19 is, I can use small, minimally radioactive liquid fuel 20 salt samples to get representative physical and 21 chemical properties.
22 I don't need to use cesium-137 in my salt to 23 determine its thermal conductivity. I can get that 24 from, you know, a stable cesium on this. And that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
55 really is quite helpful that it's an elemental 1
- property, not an isotopic property for the 2
thermophysical and thermochemical. Not of course the 3
reactivity. Next slide, please.
4 Liquid fuel salt qualification establishes 5
acceptable salt composition ranges that maintain the 6
safety functions. Liquid fuel salt's a Newtonian 7
fluid. Classic fluid that's not thixotropic or 8
anything exotic.
9 The heat transfer and flow behave in well-10 known manners, their continuous variation in the 11 physical properties with composition. Yet reasonable 12 assurance of adequate protection derived from a 13 combination of measured composition and knowledge of 14 the resulting chemical and physical properties.
15 A liquid fuel salt property database would 16 capture the relationship between fuel salt composition 17 and properties. Fortunately, the major properties on 18 there are not strongly impacted by things which are 19 very small fractions of it, things that are 0.01% of 20 things are not dominant issues in this. And it's 21 just, it's a nice feature of ionic liquids. Next 22 slide, please.
23 CHAIR PETTI: So David, just a question. I 24 agree with this, you know, the statements for the, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
56 let's call them the heat transfer normal operation and 1
accident properties. But solubilities, how close do 2
you get to solubility limits for some of the fission 3
products on these -- in these systems?
4 MR. HOLCOMB: Largely it depends. And 5
unfortunately that's the answer on there. For 6
example, the thermal, the -- if you're using a 7
fluoride salt and it's more complex because the 8
solubility of the fission products are, you know, 9
impacts the solubility of the fissile materials. So 10 they compete for solubility.
11 So if I get a lot of lanthanides in there, 12 I can plate out my actinides. And so you really do 13 need to have a map of where you are. The nice thing 14 is if it's just this fission product and it becomes 15 insoluble, initially it comes out as a particle, and 16 you filter on this.
17 So the things that are -- we call them the 18 noble fission products, well, some of them have very 19 low solubilities, approximately none on there. It's, 20 the real issue on this is that when I get something 21 like a, you know, where I've not caused my plutonium 22 to plate out at my low temperature because I've got so 23 much lanthanide buildup. But in reality, things like 24 cesium, strontium, most of these things form really 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
57 good fluorides.
1 And what you do is as a whole you adjust 2
things like what's the melting point. And they're not 3
-- they're really -- they're well mistable (phonetic).
4 So you're not -- yes, phase separation on particularly 5
on the fast spectrum reactors where you've got very 6
large fractions of fissile materials, phase separation 7
is a possibility of something which can happen.
8 And you will need to have a map of where you 9
are and where you're headed. That's why that property 10 database is important. And part of this is how fast 11 do you get there and what type of a margin do I need 12 to have from these types of phase separation problems.
13 CHAIR PETTI: So, but the chemistry here 14 sounds somewhat complex in the liquid phase. So you 15 have some, hopefully some pretty good models to help.
16 MR. HOLCOMB: Yes, fortunately ionic 17 liquids, halite salts are some of the best, well-18 studies types of materials that -- in existence. I 19 think we provided the roadmap prior to this that the 20 Campaign and NEAMS -- combination Campaign and the 21 NEAMS program have generated about how we're going to 22 get the thermophysical and thermochemical properties, 23 you know, adequately measured. Or -- and adequately 24 modeled.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
58 But some of the relationships are 1
complicated, but mostly we're not -- in most instances 2
we're not near the boundaries. But that is not a 3
universal statement. In some instances people are 4
running much closer to the boundaries, and that will 5
be a design concern. And it will certainly be 6
something that they will need to be able to 7
demonstrate that we have adequate knowledge of.
8 And fortunately because we have continuous 9
variations in the properties, what they're probably 10 going to do is take extremal points and say well, 11 we're not going to get further than this in 12 composition and show that we still haven't had a phase 13 separation.
14 CHAIR PETTI: So are these salts ideal, or 15 do they also exhibit some non-ideal behavior in a 16 chemical sense?
17 MR. HOLCOMB: Halite salts are pretty, you 18 know, are -- are very distinctive ionic liquids. They 19 have very normal, you know, Newtonian heat transfer.
20 You know, very well-known. But things like phase 21 separation, well, that is certainly a non-ideal 22 property. And phase separation is a possibility. You 23 can get solids that form in this with composition 24 changes.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
59 So in most instances and in most respects, 1
if you are well away from boundaries, when you are 2
well away from boundaries, they look pretty ideal.
3 But if you get near composition boundaries 4
where you're -- where gee, you add a little bit more 5
to this, and I would ten get a phase separation where 6
I might get a solidification out of something, then I 7
worry about my accidents. Where, gee, I get an 8
overcooling transient, and now I'm plating out my 9
fissile material at the lower end of my heat 10 exchanger.
11 CHAIR PETTI: Right, right.
12 MR. HOLCOMB: So.
13 CHAIR PETTI: This is just a comment that 14 some of this discussion is really good, and it didn't 15 jump out to me in the report itself, the chemical 16 ideality and non-ideality and how important that is in 17 some of the defense, like you're talking about, so.
18 MR. HOLCOMB: We'll take comments on the 19 report, certainly. I would appreciate if we can add 20 things which will add clarity to that. We are trying 21 to make this a useful document, not just a regulatory 22 limits thing, so that we can both help the designers 23 and the regulators. So if you've got suggestions on 24 how to improve, we're -- I'm interested, my ears are 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
60 open.
1 CHAIR PETTI: Yeah, no, we've got -- those 2
are -- my comments are things like that, this needs a 3
little more here a little more there sort of, to help.
4 MR. HOLCOMB: Because unfortunately I'm 5
awful close to this.
6 CHAIR PETTI: Right.
7 MR. HOLCOMB: And so my presumption of 8
knowledge on things is probably different than most 9
others. And so I need extra eyes. Okay.
10 CHAIR PETTI: Keep going.
11 MR. HOLCOMB: Next slide, please. So a 12 liquid fuel property database relates composition to 13 physical and chemical properties to add developers.
14 The database development is currently underway, 15 sponsored by DOE-NE.
16 It's also, it turns out that folks like the 17 Europeans through the Joint Research
- Centre, 18 Karlsruhe, have created another -- the database. And 19 they -- we've finally been working through the GIF to 20 get them to use the appropriate IEC quality standards 21 to give an equivalent data.
22 And so people can contribute data on the 23 phases, on the properties of the -- of salts. Because 24 there's a fair amount of complexity in the number of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
61 potential compositions that are there. And we -- and 1
the database is becoming more and more populated, but 2
it's got a few years.
3 And so what I expect is that individual 4
vendors will pay for data on their particular salt and 5
may or may not then elect to add it to the database.
6 So we may end up not being very efficient and having 7
to do things several times.
8 But for example, I was quite pleased to hear 9
that TerraPower in their work with Los Alamos on 10 sodium chloride-plutonium chloride data are agreeing 11 to take that data, which they own through -- and add 12 it to the common database. And so we just are adding 13 more information to the database over time.
14 Eventually, you just going to have salt 15 composition and just look up what the salt composition 16 temperature in this, look up the properties in the 17 database. Chances are, you know, it depends upon how 18 much an applicant is willing to do beforehand to say 19 do I have to do more thermophysical and thermochemical 20 property measurements to know where do I have to 21 maneuver things to set my boundaries.
22 But if I, you know, and but again, this is 23 one of the largest efforts that is under -- currently 24 underway, supported jointly by DOE, the Europeans, and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
62 the -- and DOE both through its MSR Campaign and 1
through NEAMS. Because largely they're using some 2
fairly fancy modeling tools to give predictions of 3
what the properties are.
4 And then they're measuring them at select 5
points and using the combination of modeling and 6
simulation and validation of the models -- models to 7
get the overall database. Next slide, please.
8 MEMBER REMPE: Does the database also 9
consider materials interactions?
10 MR. HOLCOMB: Yes, you have to -- there are 11 a couple of other programs that are just being 12 published now that NEAMS is doing on a significant 13 amount of interactions with the material. We do not 14 have truly first principle understanding of some of 15 the corrosion processes and we're doing engineering --
16 engineering scale approximations to number of things.
17 But as far as the salt composition, yeah, or 18 we stick more chromium in this because we corroded it.
19 That is one of the composition elements in the -- in 20 the salt, as far as the salt properties. Now, as far 21 as the accident progression modeling, we are having to 22 do things like well, except -- well, gee if it leaks 23 or if it does break, what happens.
24 Not -- and we are using some of the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
63 engineering development tools to try to say, well, 1
let's select these conditions or these things to make 2
that less likely. But we don't have a full first 3
principles understanding of everything which is going 4
on in, particularly in some of the corrosion parts on 5
here, because that would certainly help us to design 6
new alloys.
7 And so we're looking at developing, 8
continuously improving the development of the -- of 9
the -- of our understanding on there. But what we're 10 having to do is, currently, is mitigate the impacts of 11 if something does fail, can we still prevent, you 12 know, a large accident, accidents?
13 And that's kind of why we have the advantage 14 that the salt's only touching that first layer. And 15 CFR already tells us we have to accommodate an 16 accident where that first layer fails.
17 And things like if you have, say for example 18 you have a 304 stainless guard vessel around this 19 thing, and even if your vessel ruptures, well, the 304 20 stainless is going to last for indefinitely, a long 21 time. It's really radiation-damage tolerant and under 22 the condition that we're getting, so the neutrons 23 aren't going to have to done very much to it.
24 So you've got a pretty good secondary 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
64 containment around this, even though you didn't have 1
very good evidence, perhaps, that your first -- that 2
you have a fundamental understanding of your -- of the 3
damage to your first material.
4 MEMBER REMPE: So your response -- thank you 5
for it, but your response emphasizes this first 6
principles, which now I'm starting to wonder about 7
your earlier discussion of the other data that's being 8
put into this database.
9 I understand that data are limited and it's 10 nice to have some sort of tool that shows the curve 11 that best matches the data. But hopefully it will be 12 obvious to the regulator where the data stop and where 13 those tools are used to extrapolate the data.
14 Is the data going to be presented in a way 15 that's easy for the regulatory to say, okay, we have 16
-- we understand based on data points between X1 and 17 X2, but any data beyond X2 is just some model 18 someone's created and extrapolated?
19 MR. HOLCOMB: Typically we tend to take the 20 data at the boundary points on here, and then not have 21 to do extrapolation and just doing interpolations.
22 And so we take it at a higher temperature than would 23 be acceptable and at maximum, you know, accident 24 conditions. So we don't try to do extrapolations 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
65 because we don't really have good abilities of doing 1
that.
2 And so that means that what you're doing is 3
interpolating the data and relying upon the fact that 4
the fuel salt is generally well-behaved with 5
continuous property variations. And yes, we're using 6
models to say well, we don't have a really dense data 7
field. And so it might be a bit off.
8 But most of the -- but the ways that they're 9
trying to accommodate this is by taking data that it 10 would -- beyond the acceptable range. Taking data 11 where gee, this is at hotter than, well, we know that 12 this nickel-based alloy is going to turn butter by 13 these temperatures.
14 And we take the data at these temperatures 15 because we don't have to use a nickel-based alloy when 16 I'm doing this. It's probably in some refractory 17 material when I'm taking the original data set.
18 MEMBER REMPE: Thank you, that's what I 19 wanted to hear. Years ago there was this idea that 20 the DOE codes could be used to extrapolate because we 21 had first principles knowledge, and I'd still like to 22 be interpolating rather than extrapolating.
23 MR. HOLCOMB: Yeah, we don't really -- we 24 don't have -- we haven't validated that we have the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
66 knowledge on there. And until we do, we can't really 1
use them that way.
2 I see Stephen Schultz has got his hand up.
3 DR. SCHULTZ: Hi, Dave, thank you. A 4
question on the previous slide you mention that 5
appropriate quality assurance needs to be applied in 6
the database development. Is there sufficient 7
guidance and programmatic information that assures 8
that in terms of quality assurance?
9 MR. HOLCOMB: They are doing, in terms of 10 setting the boundaries of the properties. So we'll 11 know, gee, is the heat capacity off ten percent or 12 something. And that is being done by following good 13 scientific methods.
14 On the other hand, the implications of being 15 off on that are different because that involves you 16 having adequate models of the accident progression 17 sequences. And so you have both a concern about how 18 good do you need the data to be. And we can do a 19 number of things to say this is how well we know the 20 data and what the boundaries are.
21 But the question about how good do we know 22
-- need to know -- to really need to know the data is 23 also going to be very design-dependent. Some designs 24 may have a great deal more margin in certain 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
67 properties than others will. And that's going to mean 1
that, well, we wouldn't need to get it down to a, you 2
know, a five percent knowledge, but maybe I could have 3
-- take a factor of 2 on this.
4 So we've got the scientific basis for 5
- saying, well, what's my uncertainty in the 6
measurement. But what's my -- but what's the 7
implications of that uncertainty is going to be very 8
design-dependent.
9 DR. SCHULTZ: Thank you, that helps. And 10 retrospectively, the database as it has been developed 11 in the past, do you have a sufficient quality 12 assurance capability identified in that past data that 13 14 MR. HOLCOMB: We are unfortunately having to 15 regenerate on there. There is not an -- now, the 16 things that have been done since 2002 by the JRC, 17 because they did not initially -- it was -- that's one 18 of the advantages of working together in the GIF 19 program, is I'm the US GIF representative and we 20 really worked with the JRC to make sure they 21 understood people were going to make design and safety 22 decisions based upon this.
23 And they have fully implemented an 24 equivalent to, you know, an ASME NQA-1 type thing, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
68 quality assurance. But largely this has been, you 1
know, the database has been originally developed with 2
this as its purpose, with the understanding that we 3
need to know this adequately.
4 On the other hand, the database has not yet 5
been publicly released. It is DOE continues to 6
consider the export control implications of releasing 7
the data. However, right now I believe the decision 8
is this is fundamental scientific data, which by US 9
policy must be in the public domain. But we're --
10 we'll see what we get, you know, is the basic message.
11 But it is in the original, you know, design, 12 the original goals that this is to be a -- acquired 13 under an NQA-1 type approach.
14 (Simultaneous speaking.)
15 MR. HOLCOMB: And so I don't have a real 16 answer for you because the database isn't out there 17 yet.
18 DR. SCHULTZ: Understood.
19 MR. HOLCOMB: On there. But every piece 20 that we've seen so far, you know, that is all the way 21 through the acquisition program, round robin testing 22 is there -- is to apply an NQA-1 type mindset from day 23 one.
24 DR. SCHULTZ: Thank you.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
69 MR. HOLCOMB: So the fuel properties support 1
modeling the reactor performance under normal and the 2
accident condition. So the heat transfer of Newtonian 3
fluid is primarily determined by classics, density, 4
viscosity, heat capacity.
5 So thermal conductivity and radiative heat 6
transfer parameters can become important and 7
specialized situations, things like narrow tube walls 8
or very high temperatures or stagnant conditions. But 9
mostly it's density it's largely (audio 10 interference).
11 Fuel, you know, fuel salt database then has 12 the fuel salt composition, which gives you the fuel 13 salt properties. That is a function of temperatures, 14 which then tells you the fuel performance. Then you 15 take normal conditions and accident scenarios and you 16 determine whether you achieved the fundamental safety 17 functions.
18 And so there's a lot of work that's 19 currently going on to sticking the fuel salt 20 properties and the scenarios and the capabilities into 21 MELCOR and the accident progression codes, as well as 22 things like SAM to go ahead and say where are the 23 radionuclides to start with.
24 Because that's one of our biggest challenges 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
70 is if you have your accidents is where is everything 1
when the accident starts. And that is -- it's not 2
necessarily so much fuel qualification, but it is 3
certainly part of the DOE-NE program to develop tools 4
that support the NRC so people can at least validate 5
the predictions that the designers are making.
6 CHAIR PETTI: David, I didn't see any 7
discussion of surface tension. I would think that if 8
you're going to do melt spreading, that you might want 9
to know what the surface tension of these salts are.
10 MR. HOLCOMB: It is indeed correct. It 11 turns out that the gravity -- there have been a couple 12 of reports recently, and it is the -- you know, it's 13 a gravity type, you know, that's the leading type of 14 things. And the amount of wetting is not the biggest, 15 you know, impact on there. But you're correct, 16 surface tension is one of the properties.
17 If you look in the details of the report, 18 particularly the one before, and one are the things 19 that are in there, surface tension is one of the 20 columns, columns.
21 And it turns out most people on the -- when 22 it flows, you've channelized it. You've put a sloped 23 pan so it flows down into a tank. And it flows pretty 24 well like water, you know, water.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
71 And so yes, that is one of the things we'll 1
need to have. No, we don't think it's one of the 2
things that's really a dominant issue. But it is one 3
of the columns -- columns that's listed as yeah, you 4
need to know that. Next slide please.
5 So, periodic fuel salt property assessment 6
will be an element of reactor operations. You know, 7
I sort of think of this as analogous to materials 8
surveillance coupons.
You pull the coupons 9
periodically, look at gee, are the -- is -- are the 10 properties evolving according to you models.
11 If they're according to your models, you go 12 forward. If not, you got to update your models or 13 show that you maintain -- your continuing to be within 14 acceptable boundaries.
15 You'll see the frequency of how often you 16 measure it, depends upon the rate of change and how 17 close you are to an allowable limit. That was part of 18 the questions about the phase change there is if 19 you're close, you're going to have to measure more 20 frequently and have a better model of how your 21 composition is changing.
22 So for example on here, when we were at the 23 MSRE, they measure the chromium composition weekly on 24 there. And largely that was because they were 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
72 concerned about oxygen ingress.
1 Any time you open things, you always get a 2
little bit of oxygen, and that made it more oxidative, 3
which would get the chromium out. And you'd see it 4
and you'd keep adding beryllium to it in small 5
quantities until you stopped seeing the chromium 6
content go up.
7 The uranium content was inferred from the 8
reactivity impact. They tried to measure it, but they 9
found it was a thousand times more sensitive to look 10 at the reactivity impact because they just kept adding 11 small amounts of uranium to keep the reactivity where 12 they wanted it to be.
13 But the MSRE really didn't accumulate 14 sufficient fission products to require reassessing 15 most properties, things like density, viscosity, etc.
16 After years, they really didn't have enough to do very 17 much.
18 (Simultaneous speaking.)
19 MR. HOLCOMB: Go ahead.
20 CHAIR PETTI: Just to clarify, I hear people 21 talk about the, you know, MSRE operated for years, but 22 I also understand it was really only one effective 23 full power year of operation. Is that --
24 MR. HOLCOMB: Pretty close to two, but yeah, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
73 to 13,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />. But yeah, but yes, the -- yeah, 1
there is a limited duration experience, you are 2
correct.
3 CHAIR PETTI: Yeah.
4 MR. HOLCOMB: Next slide, please. Fuel salt 5
supports plant SSCs in achieving the fundamental 6
safety functions and regulatory requirements. So the 7
qualification focuses on identification and 8
understanding of fuel salt property degradation 9
mechanisms that occur as a result of irradiation 10 during reactor operations.
11 So property repair, composition adjustment 12 may be incorporated into normal operation so that 13 things like, you know, filtering out the solids is 14 just, I envision most people are going to want to 15 filter out the solids.
16 But again, in something like a Moltex design 17 where the fuel is in a pin, well, you can't do that.
18 So you're going to run it until you've got -- you go 19 outside of the range of acceptable properties. During 20 normal operations, an AOO fuel salt properties must 21 not result in sufficient damage -- must result in 22 sufficient margin from damage to safety-related SSCs.
23 Essentially, if you said something like your 24 heat exchanger tubes are safety-related, you have to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
74 have some evidence that you don't -- you got 1
sufficient margin from damage. Which is why I think 2
that in most cases people are probably going to say 3
those are not safety-related and my -- it's going to 4
be my next boundary that I'm going to credit.
5 Under accident condition, the fuel salt 6
properties must not result in sufficient damage to 7
safety-related SSCs to prevent them from achieving 8
their function. So it's not required if other things 9
are happening during the accident to prevent the other 10 things, but they -- but it cannot itself cause, you 11 know, damage to other safety-related SSCs. Next 12 slide, please.
13 So our fuel qualification draft's available 14 for review and comment, there's the ML number. You 15 know, we're certainly welcome for suggestions for 16 improvements. The approach, you know, send it to Stu 17 while he's still employed there or to Chris or to me, 18 it will certainly be welcome. Welcome this, we're 19 hoping to, you know, accept comments, comments from 20 the ACRS and get a draft, you know, get a candidate 21 draft as soon, you know, in a timely manner.
22 I think the next one's just the last slide 23 then.
24 CHAIR PETTI: Okay, so David, you didn't go 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
75 into some of the issues. I do have one important 1
issue I wanted to raise with you in the reactivity 2
control. What's there is good, but I'm worried about 3
actual reactor control. Not so much in thermal 4
systems and fluoride salts, but in fast systems.
5 There's the effective delayed neutron 6
fraction in a FAS system where some of the fuel is 7
leaving. So the delayed neutron fraction keeps 8
getting small. You know, it's smaller in a FAS system 9
than a thermal system. Then I'm moving some of the 10 fuel so it's smaller still.
11 Some designs that might let the actinides in 12 the FAC system sit there for a long time and fission 13 them as an in situ actinide burner brings more 14 uncertainty.
15 I think an experimental assessment is needed 16 that you know what that delayed neutron fraction is.
17 I thinking particularly of the chloride FAS systems 18 where, if you're going to use them for a long times, 19 they have a lot of buildup of material. You're 20 fissioning the higher order actinides. I think life 21 gets really complicated.
22 MR. HOLCOMB: You got a really robust set of 23 reactivity feedback mechanisms on this. Unlike, you 24 know, a sodium reactors or anything like this is that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
76 that Doppler feedback is really quite strong on this.
1 And you're yanking out the things, you know, 2
when you're removing the delayed, that are more 3
effective because they're -- some of them are born at 4
a lower energy. So you are getting rid of some of 5
those neutrons that have more reactivity impact on 6
this.
7 But the Doppler broadening in the FAS 8
systems, because the problem on some of the thermal 9
systems is the potential for a positive reactivity 10 from the competition between the U233 and the thorium 11 because of the -- so you can get some warnings.
12 But in a FAS system, you got -- you have two 13 very strong negative reactivity impacts, one of which 14 being the Doppler broadening. The other one is just 15 the volumetric expansion that you've got a critical 16 geometry on this. And you are at the speed of sound 17 driving fuel out of the reactor.
18 And recall that your reactivity limit is not 19 even necessarily prompt criticality. It is so far 20 into prompt criticality that you caused a pressure 21 wave that damages the reactor vessel.
22 Because recall one of the original designs, 23 and they were trying to do this as a competition to 24 the TRIGA initially, was to use MSRs as prompt first 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
77 reactors. Because they've got such really good, 1
negative fuel-based reactivity characteristics.
2 And so yeah, you might be able to drive, it 3
could be conceivable to get these very, you know, some 4
form of a prompt first. But if you -- if you've got 5
adequate evidence of the very strong negative 6
reactivity feedbacks, okay, you're designing a new 7
TRIGA. So I'm not --
8 MEMBER BLEY: This is Ron again. Most of 9
these designs have more -- have some form of 10 continuous purification, am I correct?
11 MR. HOLCOMB: Well, some of it's inherent, 12 you don't have a choice. The fission gasses are 13 coming out.
14 MEMBER BLEY: Right.
15 MR. HOLCOMB: And the solids, you probably 16 got a -- you're going to probably have to filter them.
17 I don't know whether I'd say anything beyond most on 18 those. But the ones that you have to do, yeah, 19 everybody's going to do the ones you have to.
20 MEMBER BLEY: Yeah. And you can --
21 MR. HOLCOMB: Whether they're going to be 22 other ones, look at Moltex, they're not doing it. But 23 Terrestrial probably will.
24 MEMBER BLEY: And you can more or less treat 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
78 this as a homogenous system, am I right also there?
1 MR. HOLCOMB: Yeah, it's pretty close to 2
that. There's temperature driven, so you have to be 3
concerned about composition, you know, from the top to 4
the bottom because of thermally driven things. But 5
basically it, the fluid has to flow very rapidly to 6
transfer the heat so it's well mixed.
7 MEMBER BLEY: So this can be treated, and 8
Dave can correct me if I'm wrong, more or less as a 9
computational, what do you want to call it, easy 10 system?
11 MR. HOLCOMB: It's got a lot of inherent, 12 you know, as a pretty ideal system on many ways. But 13 again, what you have to be concerned with is that you 14 know the reactivity impacts on this and that you have 15 adequate confidence that you have strong and timely 16 negative reactivity feedback coefficients.
17 So that if you do get something which is --
18 then you get a reactivity transient, that it does not 19 rupture the system. And I think where we will have 20 pretty darn good confidence that we've got strong and 21 effective negative reactivity coefficients on there.
22 Thermal expansion is a pretty well-known property, and 23 so you're -- it's -- I'm not that concerned about the 24 controllability --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
79 CHAIR PETTI: Dave, I talked to some folks 1
interested in the chloride system and I've raised this 2
issue. Because you know, I mean, in the early days of 3
nuclear there were concerns about controllability of 4
FAS systems, that it could be -- I mean, right, if 5
there's no delayed neutrons, you can't have a reactor, 6
right. We know that boundary, right?
7 MR. HOLCOMB: Yep.
8 CHAIR PETTI: And so in talking to some of 9
the chloride guys, again, this is not my area of 10 expertise, but they thought it was worth them looking 11 at, you know, a small, very small system to 12 demonstrate it before jumping to something big.
13 MR. HOLCOMB: Well, I think there's a whole 14 bunch of things. That molten chloride reactor 15 experiment is really important. I think there's a 16 whole bunch of stuff you're going to get with that, 17 and I would not be supportive, and I don't think 18 anybody is supportive, of going beyond the small scale 19 reactor experiment. And they will do reactor physics 20 right at the beginning of that reactor.
21 CHAIR PETTI: Right.
22 MR. HOLCOMB: That's not going to be a big 23 system on there. For much like the same reason, you 24 know, you started MSRE as a small system in this. It 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
80 is just no question there's a lot to learn about this.
1 I am concerned about things like the phase separations 2
and such and really understanding that
- well, 3
understanding the corrosion.
4 There are a lot of things that you will 5
learn from that first test reactor. And going to a 6
power reactor with a fast chloride salt system, I just 7
don't see anybody doing that. It seems to be 8
extremely risky on our level of knowledge, knowledge 9
we don't have the MSRE to provide us real confidence 10 that yeah, we built this, it runs this way.
11 So I don't think anybody's going to present 12 that as the, well, let's go do this. I do, you know, 13 certainly you will have startup reactor physics 14 experiments that are going to be very important on the 15 MCRE. And you know, that's the first one. You know, 16 DOE and you know, Southern Services, you know, with 17 TerraPower are going to be doing as a test -- as a 18 test scale reactor, as a non-power reactor.
19 You are going to be doing zero power 20 criticals. I think that's important. And then 21 they'll do (audio interference) to criticals and the 22 full set of reactivity experiments. I don't disagree 23 that you, you know, that that is the correct thing to 24 do and it should be done.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
81 CHAIR PETTI: Right. Any other comments, 1
members? Okay, I'll guess we'll turn to public 2
comment then. Can we open the public line?
3 MR. BROWN: Dave, we don't use the public 4
phone line, we use the Microsoft calling line.
5 CHAIR PETTI: Okay, so yeah. So anybody 6
from the public listening in that would like to make 7
a comment, please unmute now.
8 Okay, with that, I guess we'd like to thank 9
you, David. I actually really liked the report. I 10 thought it was a very good report. You'll see in our 11 letter just some minor things. You wanted some 12 feedback, you'll get those in our letter.
13 But I thought it was a good report. I 14 really liked the approach, starting with the safety 15 functions, exactly how I would have approached it. So 16 it was well done and well-reasoned, so.
17 So members, I do recommend that we write a 18 letter, just given the importance of the topic. The 19 fact that next month we're also going to be briefed on 20 the more generic framework in NUREG 2246. Getting 21 them both out I think will just help the staff as they 22 move forward here.
23 Any other comments from anybody?
24 MEMBER SUNSERI: This is just Matt, Dave.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
82 I do appreciate you pulling this together, and 1
presentation. I found it very informative, so thank 2
you.
3 CHAIR PETTI: Okay, so with that, then I 4
guess we adjourn the meeting, and we'll see everybody 5
at the full committee here this afternoon.
6 (Whereupon, the above-entitled matter went 7
off the record at 10:45 a.m.)
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
ORNL is managed by UT-Battelle, LLC for the US Department of Energy MSR Fuel Salt Qualification Guidance Development ACRS Subcommittees on Metallurgy & Reactor Fuels David Holcomb, George Flanagan, and Mike Poore October 5th, 2021
22 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Fuel Qualification is an Element in Achieving Sufficient Understanding of Fuel Behavior Fuel qualification is a process which provides high confidence that physical and chemical behavior of fuel is sufficiently understood so that it can be adequately modeled for both normal and accident conditions, reflecting the role of the fuel design in the overall safety of the facility. Uncertainties are defined so that calculated fission product releases include the appropriate margins to ensure conservative calculation of radiological dose consequences.
NRC Presentation on Possible Regulatory Process Improvements for Advanced Reactor Designs, August 3rd, 2017 (ML17220A315)
33 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Liquid Fuel Salt Qualification Builds From Advanced Reactor Assessment Framework
- Liquid salt fuel assessment framework employs goal decomposition process parallel to NUREG-2246
- Success criteria derive from fundamental safety functions
- Dozens of MSR design variants under consideration
- Nearly all developed over the past decade
- Which designs will be presented for regulatory safety adequacy review over the next decade is uncertain
- Fuel salt qualification process based on fuel salt chemistry and physics and is largely independent of reactor configuration
44 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Fuel Qualification Supports Both Performance-Based and Prescriptive Safety Adequacy Evaluation Processes
- Understand how fuel salt properties support plant achievement of fundamental safety functions
- Evaluate acceptable range of fuel salt properties to Maintain appropriate margin from design limits under normal and AOO conditions Limit damage to safety-related SSCs to ensure proper functioning during accidents Performance-Based
- Determine the range of fuel salt properties that results in compliance with specific requirements Both fuel and coolant requirements Prescriptive
55 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Liquid Fuel Has Substantial, Fundamental Differences From Solid Fuel Liquid Fuel Chemically damageable -
may be reparable during use Composition may be adjustable during use Properties depend on composition and state Container breach could release nearly all radionuclides Solid Fuel Mechanically damageable Composition set prior to use Properties depend on fabrication process
- Liquid salt fuel
- Serves as nuclear fuel and primary heat transfer media
- Must meet requirements for both purposes
66 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Fuel Salt Qualification Applies From Receipt of Regulated Materials Until Transfer to Independent Storage
- Properties evaluated will be different at different points of fuel salt lifecycle - for example
- Reactivity impact of fuel salt concentrate is a concern during fueling
- Fluid properties immaterial to frozen, used fuel salt
- Materials that leave the fuel salt and lack a reasonable means of return cease to be part of the fuel salt
- Insoluble materials, fission gases, vapors, and aerosols become part of radioactive waste stream
- Plated out materials that could redissolve or resuspend in the liquid remain part of fuel salt
77 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Key Issue is What Constitutes Fuel Salt?
- Fuel salt does not come in discrete elements (rods or assemblies) and moves independently of its container during normal operations
- Cladding and fuel assembly structures are qualified as part of solid fuel
- Fuel salt includes all of the material containing fissionable elements or radionuclides that remain in hydraulic communication, but does not include the surrounding systems, structures, or components
- Salt vapors and aerosols remain part of the fuel salt system until they become adequately trapped
- Container corrosion products become part of the fuel salt
- Fresh and used fuel salt in on-site storage are within scope
88 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Common Salt Properties and Plant Functions Enable a General Liquid Fuel Salt Evaluation Method
- Specific accident sequences are design dependent
- Basic operational and safety functions are common to any nuclear power plant
- Halide salt characteristics are common to any MSR
- High boiling points (low pressure)
- Low Gibbs free energy (low chemical potential energy)
- Natural circulation heat transfer properties
- Fuel salt interacts with its container layers via common chemical and physical mechanisms - for example via thermal energy transfer, chemical reactions, and mechanical processes
99 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Method Tailors Solid Fuel Qualification Process To Characteristics and Functions of Liquid Salt Fuel
- Modifications both add and remove issues from solid fuel qualification process - for example
- Fuel salt is not a manufactured product (e.g., not in the sense that NUREG-2246 describes solid fuel via a manufacturing specification)
- Liquids cannot be mechanically damaged
- Fuel salt also serves as the primary reactor coolant
- Fuel salt properties determine its capability to adequately support achievement of fundamental safety functions (FSFs)
- Fuel salt regulatory basis derives from the role of the fuel salt in establishing compliance with existing regulations
10 10 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Liquid Fuel Salt Does Not Have a Mechanically Determined Lifetime
- Identification of life-limiting failure and property degradation mechanisms that occur as a result of irradiation during reactor operation remains key focus
- Fuel salt lifetime is the period during which it
- 1. Contains adequate quantities of fissile materials,
- 2. Does not include too many neutron absorbers, and
- 3. Maintains acceptable thermophysical and thermochemical properties
- Composition may be adjustable during operation to correct degrading conditions
11 11 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Functional Containment is Important to How MSRs Provide Adequate Radionuclide Retention
- Barrier performance must be degraded to release radionuclides into the environment
- Performance degradation can occur through failure or bypass
- Fuel salt properties that stress barriers cause them to be more likely to release radionuclides - for example
- Increased temperature increases radionuclide vapor pressure in cover gas and well as decreasing strength of container
- Different performance requirements for materials normally in contact with salt versus those that only need to withstand accidents
12 12 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Fuel Salt Properties Have Substantial Impact on Containment Performance
- Normal operations
- Salt seeking radionuclides are chemically retained
- Fission gases (with half lives more than a few seconds) largely escape
- Noble (insoluble) metals largely either plate out or escape into vapor
- Fuel salt causes stress on, potentially resulting in damage to, container materials (e.g.,
via corrosion, erosion, creep, etc.)
- Tritium behavior depends on operations and redox state
- Chemical configuration (free or bound) depends on redox
- Accident conditions
- Mechanistic analysis of radioactive material transport depends on fuel salt properties
- Substantial reduction in cover gas heat content and radiotoxicity over first couple of days - MSBR predicted 99.9% reduction after 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> (ORNL-4396)
- High potential consequences of large, early cover gas releases
13 13 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Fuel Salt Boundary Breach Accident Progression Part of Performance Based and Deterministic Fuel Qualification
- Multiple locations in 10 CFR require evaluation of a postulated fission product release from core into containment
- Fuel salt or cover gas cannot directly stress exterior containment layers without first breaching an inner containment layer
- High radiation and high temperatures immediately outside fuel salt boundary substantially circumscribes characteristics of materials adjacent to fuel salt container
- Focus is on fuel salt properties that must be known to adequately model accident progression and interaction characteristics with materials within containment
14 14 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Fuel Related Advanced Reactor Requirements Are Similar for Liquid and Solid Fuel
- Example
- 10 CFR 50.43(e)(1)(i) requires that the performance of each safety feature of the design has been demonstrated through either analysis, appropriate test programs, experience, or a combination thereof
- Fuel salt thermophysical and thermochemical properties provide the information necessary to model its role in enabling plant safety features to perform safety functions
- Fuel salt properties vary with both composition and temperature
- Fuel salt properties need to be determined across the range of temperatures and compositions that span potential operational and accident conditions
- Quality of the fuel salt property data needs to be sufficient to enable modeling the role of the fuel salt in achieving the plant FSFs
15 15 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Liquid Fuel Salt Must Meet the Safety Intent of the Coolant Related GDC or ARDC (as appropriate)
- Example
- GDC 15 requires that the coolant system be designed so that the design conditions of the reactor coolant pressure boundary are not exceeded during either normal operations or AOOs
- ARDC 15 removes pressure from the reactor coolant boundary function
- Key design choice of MSRs that employ functional containment is which layer or set of layers comprise the reactor coolant boundary
- Layer(s) credited to achieve the safety function must meet the requirement
- If a normally salt-wetted layer (e.g., reactor coolant boundary) is credited, fuel salt conditions that would damage the layer must be detected and corrected before causing significant damage
16 16 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Liquid Salt Fuel Assessment Framework Follows Template Developed for Solid Fueled Advanced Reactors
- Top-down approach used to decompose top level goal of fuel is qualified to lower level supporting goals
- Qualifying fuel develops high confidence that the fuel will adequately perform its role in enabling the facility to achieve its safety objectives
- Lower level supporting goals are further decomposed until clear objective goals are identified that can be satisfied with direct evidence
17 17 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Liquid Fuel is a Synthesized Not a Fabricated Product
- Bulk properties can be determined by measuring properties of product samples
- Neither the fabrication based definition of fuel qualification nor manufacturing specification branch of fuel assessment employed for solid fuel advanced reactors applies to liquid fuel Solid Fuel Goal Decomposition
18 18 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Qualification is Based Upon Understanding the Chemical and Physical Properties of Representative Fuel Samples
- Liquid state significantly changes the physical behavior of fuel
- Liquids do not accumulate internal stresses
- No history dependent properties
- Flow homogenizes fluid properties
- No position dependent properties
- No size dependent properties
- Chemical and physical properties are set by elemental composition and temperature
- Independent of isotopic content Small minimally-radioactive liquid fuel salt samples provide representative physical and chemical properties
19 19 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Liquid Fuel Salt Qualification Establishes Acceptable Salt Composition Range That Maintains Safety Functions
- Liquid fuel salt is a Newtonian fluid
- Heat transfer and fluid flow behave in well known manners
- Continuous variance in physical properties with composition
- Reasonable assurance of adequate protection derives from a combination of measured salt composition and knowledge of resulting chemical and physical properties
- A liquid fuel salt property database would capture the relationship between fuel salt composition and properties
20 20 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Liquid Fuel Salt Property Database Relates Composition to Physical and Chemical Properties to Aid Developers
- Database development underway sponsored by DOE-NE
- Salt property measurement program in progress
- Not currently including minor constituent transuranic elements (Am, Cm)
- Requires appropriate quality assurance for both new and existing data
- Database initially sparsely populated
- Safety evaluations / accident models performed with bounding values to establish acceptable performance range
- Additional data added to database over time
- Goal is to eventually only require salt composition and temperature measurement at operating plants and look up properties from database
21 21 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Fuel Salt Properties Support Modeling Reactor Performance Under Normal and Accident Conditions
- Heat transfer in Newtonian fluids is determined primarily by density, viscosity, and heat capacity
- Thermal conductivity and radiative heat transfer parameters can become important in specialized situations
22 22 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Periodic Fuel Salt Property Assessment Will Be an Element of Reactor Operations
- Analogous to material surveillance coupons
- Compare measurement to prediction
- Frequency of property measurement depends on potential rate of change and how close salt composition is to allowable limits
- Chromium composition was measured weekly at MSRE
- Uranium content was inferred from reactivity impact
- MSRE did not accumulate sufficient fission products to require reassessing most properties: density, viscosity, etc.
23 23 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Fuel Salt Supports the Plant SSCs in Achieving the FSFs and Regulatory Requirements
- Qualification focuses on identification and understanding of fuel salt property degradation mechanisms that occur as a result of irradiation during reactor operation
- Property repair (composition adjustment) may be incorporated into normal operation
- During normal operations and AOOs fuel salt properties must result in sufficient margin from damage to safety-related SSCs
- Under accident conditions the fuel salt properties must not result in sufficient damage to safety-related SSCs to prevent them from achieving their function
24 24 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Fuel Qualification Draft NUREG/CR (ML21245A493) is Available for Review and Comment
- Suggestions for improvements to the approach can be provided at any time
- Comments and suggestions can be provided to the NRC or the authors
25 25 October 5, 2021 ACRS Metallurgy & Reactor Fuels Subcommittees Meeting Discussion