Protecting People and the EnvironmentUNITED STATES NUCLEAR REGULATORY COMMISSION
1
1 UNITED STATES OF AMERICA
2 NUCLEAR REGULATORY COMMISSION
3 ***
4 MEETING WITH
5 ADVISORY COMMITTEE ON NUCLEAR WASTE (ACNW)
6 ***
7 PUBLIC MEETING
8 ***
9 Nuclear Regulatory Commission
10 Room 1F-16
11 One White Flint North
12 11555 Rockville Pike
13 Rockville, Maryland
14
15 Tuesday, July 21, 1998
16
17 The Commission met in open session, pursuant to
18 notice, at 1:33 p.m., the Honorable SHIRLEY A. JACKSON,
19 Chairman, presiding.
20
21 COMMISSIONERS PRESENT:
22 SHIRLEY A. JACKSON, Chairman of the Commission
23 NILS J. DIAZ, Member of the Commission
24 EDWARD McGAFFIGAN, JR., Member of the Commission
25
2
1 STAFF AND PRESENTERS SEATED AT COMMISSION TABLE:
2 B. JOHN GARRICK, Chairman, ACNW
3 CHARLES FAIRHURST, Member, ACNW
4 GEORGE HORNBERGER, Member, ACNW
5 R.G. WYMER, Member ACNW
6 JOHN LARKINS
7 JOHN C. HOYLE, Secretary
8 KAREN D. CYR, General Counsel
9 ANNETTE C. VIETTI-COOK, Assistant Secretary
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
3
1 P R O C E E D I N G S
2 [1:33 p.m.]
3 CHAIRMAN JACKSON: Good afternoon, ladies and
4 gentlemen. Commissioner Diaz is running a little late and
5 he asked that we begin.
6 Today the Commission will be briefed by the
7 Advisory Committee on Nuclear Waste on several technical
8 issues related to the management and disposal of radioactive
9 waste. The Commission looks to the ACNW, as it is called,
10 to provide it with technical advice to ensure the safe
11 management and disposal of this country's radioactive waste.
12 The Commission was last briefed by the ACNW on
13 December 18th of last year. We seem to have a long time
14 period between these briefings.
15 Today's briefing will include discussions on four
16 topics that are of great interest to the Commission. These
17 include, first, the ACNW's views on risk-informed,
18 performance-based regulation. Second, the interim guidance
19 in support of the final rule on radiological criteria for
20 license termination. Third, the NRC's waste-related
21 research program. And, fourth, the near-field environment
22 and performance of engineered barriers in a high-level waste
23 geologic repository.
24 In addition to these discussions, the ACNW will
25 also address its plans, priorities and accomplishments for
4
1 fiscal year 1998 and its plans and priorities for fiscal
2 year 1999. The Commission looks forward to interacting with
3 you on all of these topics.
4 And unless my colleague has any comments, please
5 begin, Dr. Garrick.
6 DR. GARRICK: Thank you. I agree with you it has
7 been a little bit too long since we have had a chance to
8 meet, and I think the resources in here are adequate to do
9 something about that, so we will try to do that.
10 We are going to first talk to you, as you
11 indicated, about the positions of the Advisory Committee on
12 risk-informed, performance-based regulation, and we have
13 been pretty direct and outspoken on those positions in a
14 number of letters.
15 CHAIRMAN JACKSON: Good.
16 DR. GARRICK: And what I want to do is just kind
17 of reiterate our views on some of the key points having to
18 do with this approach. So in my first exhibit, I point out
19 that we as a Committee strongly support whatever we can do
20 to enhance the language of this discipline, and important to
21 that is moving towards a common terminology. And we have
22 been very encouraged by the Commission's view on wanting to
23 do that as well. So I think that will help the process a
24 lot.
25 We have also expressed our position several times
5
1 that we believe that as we move towards a risk-informed,
2 performance-based method of operation, that it will give us
3 a basis for making our regulations more efficient and moving
4 in the direction of some form of optimization of the
5 regulations.
6 As we have said on a number of occasions, and as
7 you have also said, it is very important that if the agency
8 is going to move in this direction, that we do it in such a
9 way that the language applies to everything essentially that
10 the agency does. So even though this activity had its birth
11 in and has emerged primarily from the reactor business, the
12 underlying principles are sufficiently basic that they can
13 apply to, we believe at least, all of the activities of the
14 agency.
15 CHAIRMAN JACKSON: Please.
16 DR. GARRICK: Yes.
17 COMMISSIONER McGAFFIGAN: Dr. Garrick, we had a
18 stakeholders meeting last week, and I don't know -- I saw
19 some ACRS members there. I am not sure whether you all were
20 there. But we asked about risk-informing Part 50, and I
21 think the answer that we got was that there are some --
22 let's finish what we are doing now, get these various Reg.
23 Guides out and working. Dr. Remick said there may be an
24 opportunity in Appendix B to strip some stuff out that the
25 maintenance rule may be now adequately dealing with.
6
1 But there wasn't a lot of enthusiasm for a
2 comprehensive rewrite. And it strikes me that what I am
3 learning, and you said earlier that it came out of the
4 reactor side, but in some sense, the waste side has gotten
5 ahead of that reactor side because it is a new area.
6 DR. GARRICK: Yes.
7 COMMISSIONER McGAFFIGAN: It is easier to bring
8 this new framework into an area where you are starting from
9 scratch than it is where you have a large body of work
10 already there and the stability of the regulatory framework,
11 however deterministic and prescriptive, and whatever it may
12 be. Better the devil you know than the devil you don't
13 tends to become a counter-wait.
14 Do you have any thoughts about that? As I say,
15 you'll have to take my word that that was the general
16 consensus of some of the industry folks. And you might not
17 have concurred in it if you had been present, but whatever.
18 DR. GARRICK: Well, I think that you are correct
19 in that the waste field has some advantage, particularly on
20 the performance side, because the standards are basically
21 performance-based, and the primary activity has been in the
22 high-level waste arena and that is where most of the
23 attention has been given with respect to establishing a
24 performance-based standard. So I think there is an
25 advantage.
7
1 On the probabilistic side, I think that the waste
2 side has had some catching up to do, and I believe they have
3 done a very good job of that. And I think they have been
4 sometimes frustrated by not being able to capture as much of
5 the methods that come out of the reactor business and
6 transfer those to the waste business, as some would like.
7 But, certainly, some of the fundamental principles, they
8 have been able to do that.
9 As far as the question of how fast we should move,
10 I think that it is very difficult when you have got a system
11 that seems to be working, that people are well-skilled in,
12 trained. It is difficult to talk about change, and I think
13 there will be a natural resistance to that. On the other
14 hand, you would certainly expect that from me.
15 I think the change is justified. The benefits for
16 doing so are there. I think we are in a time of metrics and
17 measurements. I think the risk-based process gives us a
18 much better basis for measuring our performance and being
19 focused in terms of having reasonable confidence that we are
20 dealing with the right priorities. So I expect that. I
21 expect there will be a resistance and people saying that
22 maybe we shouldn't go make substantial change.
23 And I think we have to be very selective where we
24 make the change and what-have-you. And I would hope that
25 one of the areas where there would be rather quick change
8
1 would be in the analysis activities that are ongoing. There
2 is no reason that all of our analysis activities shouldn't
3 be risk-informed right now, regardless of the regulations.
4 I would like to think that a comprehensive, risk-oriented
5 analysis contains within it all that is required for the
6 existing regulations.
7 But I hope, as you will see in a moment, that we
8 move in a direction where maybe some of the existing
9 regulations can either be simplified or even eliminated.
10 CHAIRMAN JACKSON: So let me make sure I
11 understand your point. There are really two. One is that
12 you believe that even within the existing framework, that
13 essentially all of the analysis can be made risk-informed.
14 DR. GARRICK: Yes.
15 CHAIRMAN JACKSON: And that the second point you
16 make is that there are some selected regulations that should
17 be or could be made risk-informed.
18 DR. GARRICK: Yes.
19 CHAIRMAN JACKSON: Even if we don't do a
20 comprehensive rewrite of Part 50.
21 DR. GARRICK: Right.
22 CHAIRMAN JACKSON: Could you speak to where you
23 think some of the opportunities are?
24 DR. GARRICK: Well, certainly, we heard a lot on
25 that reactor side about Part 50 and about trying to embrace
9
1 the safety goals more directly into the regulatory process.
2 There has been lots of talk and discussion, and even work
3 towards elevation of the core damage frequency as a
4 surrogate of risk, and all of that is related in one way or
5 another to Part 50.
6 In the waste side, I think the differences that
7 are probably going to manifest themselves between the
8 existing Part 60, for example, and what we expect in the new
9 regulation, Part 63. Some of those are clearly going to be
10 driven by risk-informed interests and performance-based.
11 I think the idea of moving away from the
12 allocation of performance requirements to subsystem levels
13 is another direct indicator that we are moving in the
14 direction of a more performance-based and risk-informed
15 approach. So I think we are beginning to see things happen
16 and those are a couple of the regulations that I think would
17 be most -- most directly impacted.
18 CHAIRMAN JACKSON: You also speak to the fact that
19 you feel that the concepts need to be sufficiently general
20 to accommodate all NRC activities. Do you feel that, and I
21 know you have had some interaction at an earlier
22 incarnation, but do you believe that the concepts and
23 definitions embodied in now the staff white paper on the
24 risk-informed, do you think they are general enough to
25 accommodate those?
10
1 DR. GARRICK: I think -- yes, I think that is very
2 much in the right direction. The version that I have seen,
3 I am very encouraged. I think it clearly has a stronger
4 orientation to risk than -- and performance than any similar
5 paper that I have seen.
6 CHAIRMAN JACKSON: And the last question, we did
7 have a Commission meeting a couple of weeks ago on PRA and
8 the propagation of it into materials-related areas in
9 particular, waste management areas. Do you agree that -- or
10 do you believe that the staff has a comprehensive plan or a
11 comprehensive framework for using risk-informed approaches
12 to optimize our regulations and regulatory approaches,
13 including analyses, in these areas?
14 DR. GARRICK: Well, being sometimes accused of
15 being a zealot in this discipline, obviously, I am never
16 satisfied. And I think that, you know, there is a desire
17 always to see progress and more progress. But I have
18 followed what has been going on, and both facilities, the
19 nuclear waste facility side and then the reactor side, and
20 have been very encouraged that -- with most of what is being
21 done.
22 I have also been encouraged by the fact that, for
23 example, the ACRS has capability in this arena that they
24 haven't had in the past, and I think that is very, very
25 helpful.
11
1 So I think there is a lot of work to be done, but
2 I see some of the fundamental building blocks being put in
3 place, and the white paper is clearly one of those.
4 COMMISSIONER McGAFFIGAN: One of the points you
5 made a few minutes ago was that you think there could be a
6 quick change in the ongoing analysis activities of the
7 agency regardless of the regs. We are dealing with one at
8 the moment, 50.59, where if you have any ideas as to how to
9 make that quick change, they would be welcome, because we
10 are having a heck of a time. We have this design basis
11 analysis that is the fundamental --
12 DR. GARRICK: Right.
13 COMMISSIONER McGAFFIGAN: -- sort of stylized
14 analysis that underlies that and the whole -- the whole of
15 Part 50, really. And the Commission, sort of naively, in
16 its SRM said, well, you might be able to look at some of the
17 work you did on Reg. Guide 1.174 and try to define --
18 CHAIRMAN JACKSON: Right. Go forth and do good.
19 COMMISSIONER McGAFFIGAN: Define minimal and sort
20 of the same sort of notion you just threw out, and we are
21 not there yet.
22 DR. GARRICK: Yes.
23 COMMISSIONER McGAFFIGAN: Dr. Aposotolakis in the
24 ACRS has thrown something across the transom that may help,
25 but we are struggling with how you build in, even in our
12
1 analysis, a risk-informed analysis to deterministic
2 prescriptive regulations.
3 DR. GARRICK: Yes.
4 COMMISSIONER McGAFFIGAN: And the design basis
5 accident analysis.
6 DR. GARRICK: Well, I have to live my colleagues
7 here, and for me to really get into 5059, --
8 COMMISSIONER McGAFFIGAN: I know.
9 DR. GARRICK: -- I might spend the rest of the
10 day. But I think -- I am a great believer in the top-down
11 approach. I think if we come to grips with some
12 fundamentals and some policy issues, and some methods, and
13 the staff begins to embrace those and get trained in them,
14 that, you know, we will see solutions that we didn't see
15 before.
16 Now, I will comment on a couple of things that you
17 mentioned as I go along here. Fortunately, the questions
18 you have asked has allowed me to cover most of what I have
19 just covered. So I think we are in pretty good shape.
20 So let me return to the exhibit on risk and risk
21 assessment. I am a great believer that in any science if
22 the science is to move forward, you have to have some way of
23 measuring and risk is no different than that, and the more
24 the measurements can be in terms of fundamental principles,
25 first principles, the more broadly it will apply to systems
13
1 that we have to worry about, so I think the encouragement
2 here is to not get locked into a single measurement
3 necessarily that constitutes what we mean by risk, because
4 it usually does not quite do the job.
5 One fundamental that we have seen work very well
6 in the applications arena is something we call the triplet
7 definition of risk. Whether we have been analyzing the risk
8 of importing agricultural animals and the implications that
9 has on disease rate or whether we are analyzing the space
10 shuttle or a chemical refinery or a nuclear power plant, the
11 triplet definition of risk has applied and been a very
12 constructive framework within which to ask the important
13 detailed questions -- what can go wrong, how likely, what
14 are the consequences approach in practice has seemed to work
15 very well.
16 Given that that is kind of what one might assume
17 is a definition of risk, I also like to look upon it as
18 containing the definition of deterministic safety analysis.
19 Even in the old days when we were doing safety analysis of
20 nuclear power plants, long before PRA, we used to ask the
21 doublet question -- what can go wrong and what are the
22 consequences? -- so in the context of the triplet, what we
23 like to say is it's not a question of deterministic versus
24 probabilistic. It is a question of whether or not you want
25 to deal with the question of uncertainty and likelihood of a
14
1 safety analysis and if you do a safety analysis becomes a
2 risk analysis -- so that is an example of a general kind of
3 fundamental notion.
4 CHAIRMAN JACKSON: Go ahead.
5 COMMISSIONER McGAFFIGAN: Before you leave the
6 slide, the triplet definition of risk, if we adopt -- I
7 remember being taught risk is probability times consequences
8 for an individual event. It's the same thing -- what can go
9 wrong --
10 DR. GARRICK: Yes.
11 COMMISSIONER McGAFFIGAN: But how important is it
12 that we develop a common definition across agencies, health
13 agencies, FDA, FAA, EPA, et cetera so that we are not
14 speaking past each other?
15 There is this report I think Gil Olman put out a
16 year or two ago --
17 DR. GARRICK: Yes.
18 COMMISSIONER McGAFFIGAN: -- about risk and I
19 think it talked about some of this stuff, but we are -- are
20 you suggesting we just go ahead or do we try to foster a
21 common language or how do we do what we do in the context of
22 what everything else is doing?
23 DR. GARRICK: I don't know that I would suggest
24 that we force anything. I think that it is a concept that
25 has worked well and generally concepts that work well are
15
1 adopted and spread and become standards.
2 I have never been to one to think that there was
3 so much wisdom as to be able to know what the ultimate
4 definition ought to be, so I would think that if the agency
5 has lots of success with this way of thinking, this kind of
6 definition, that it would be adopted by others.
7 As a matter of fact, the definition I am finding
8 is finding its way into a number of other arenas, including
9 defense and NATO -- I have seen it in NATO documents --
10 chemical and so forth, so I think that there is enough
11 evidence out there that the idea has enough confidence
12 behind it or it wouldn't be suggested, that its acceptance
13 is not taking anybody out on a limb very far, but my
14 preference would be that the language would be standardized.
15 At least we would move in that direction.
16 COMMISSIONER DIAZ: Yes. Besides the definition
17 of risk, of the triplet, in your set of fundamental
18 principles is there anything else you could put on the table
19 that would be more specific what you mean by fundamental set
20 of principles?
21 DR. GARRICK: Yes -- well --
22 COMMISSIONER DIAZ: Measures needed to be
23 interpreted in terms of --
24 DR. GARRICK: Well, yes. One thing that I'd put
25 out on the table in the risk business is that I am very much
16
1 a believer in evidence-based risk assessment. I think that
2 what we want to do is to have our risk assessments be
3 impersonal, not be dependent upon opinions and politics,
4 religion, or anything except the supporting evidence, so I
5 think the evidence has to speak for itself.
6 I think that in order to do that you can very
7 often enhance that process by the tools you select to
8 process that evidence, and they need to be transparent and
9 that not only means transparent with respect to the specific
10 exercises that you go through, but transparent with respect
11 to the logic that you employ.
12 You know, this is the thing that sets risk
13 assessment apart from a lot of the other analyses that have
14 risk principles in them, and that is that usually in the
15 risk field we are trying to calculate something about which
16 we have very little or no information, and so what we have
17 to do is map that requirement, that number or that outcome
18 that we want down to where we have some information, and it
19 is that mapping that needs to be visible and if the logic is
20 visible and the information is clear, then of course you
21 move in the direction of transparency.
22 CHAIRMAN JACKSON: Let me ask you two questions.
23 I mean I think I understand what you are trying to say.
24 One theoretically could say that superficially
25 there seems to be an inconsistency between, say, using PRA
17
1 to complement our traditional deterministic approaches,
2 which is what we talk about sometimes, and secondly, using
3 what you say is treating deterministic approaches or
4 analyses as a subset of risk analysis --
5 DR. GARRICK: Right.
6 CHAIRMAN JACKSON: -- and it seems that the
7 resolution of that apparent inconsistency is in your triplet
8 definition, namely that you are basically arguing that a
9 deterministic analysis or approach answers the first and the
10 third question and that PRA answers or attempts to answer
11 all three.
12 DR. GARRICK: Right.
13 CHAIRMAN JACKSON: So it is in that sense that the
14 deterministic analysis is the subset --
15 DR. GARRICK: Yes.
16 CHAIRMAN JACKSON: -- but it's also in that sense
17 that PRA is the complement that allows you to add in an
18 answer to the third --
19 DR. GARRICK: Right.
20 CHAIRMAN JACKSON: -- to the second question, is
21 that right?
22 DR. GARRICK: Yes. That's right.
23 CHAIRMAN JACKSON: Okay. Then the second question
24 I wanted to ask you is you spoke about evidence-based risk
25 assessment and of course one could raise the question of the
18
1 use of expert panels and expert judgment, and is the point
2 you are making that this mapping needing to be made visible
3 and therefore the transparency of the logic, as you call
4 it --
5 DR. GARRICK: Right.
6 CHAIRMAN JACKSON: -- is that the way that one
7 justifies and makes the best use of expert judgment?
8 DR. GARRICK: That's part of it, and you will
9 notice I did not use the word "data" because data conjures
10 up certain specific things in people's minds, and data is a
11 piece of evidence, but it is not the totality of evidence.
12 The laws of physics are evidence, logic is
13 evidence and expert elicitation outcome is evidence --
14 CHAIRMAN JACKSON: Okay.
15 DR. GARRICK: -- so I think that is what I was
16 referring to.
17 CHAIRMAN JACKSON: Okay.
18 DR. GARRICK: All right. Let's go to risk
19 assessment and defense-in-depth.
20 We have written to you much about those topics.
21 One of the things that we see as an advantage of a
22 risk-informed approach is the opportunity to add clarity to
23 the concept of defense-in-depth, the opportunity to move in
24 the direction of quantifying the contribution to performance
25 of all lines of defense.
19
1 Of course, when we talk about quantification we
2 are really not talking about necessarily a property of the
3 real world so much as we are about the knowledge of the
4 individual or individuals and their ability to express that
5 knowledge, and in order to express knowledge about rare
6 events you have got to have a mechanism and a form to do
7 that that captures the fact that there's lots of things you
8 don't know or the fact that there are uncertainties, so
9 quantification doesn't necessarily mean a number. It means
10 capturing the information in a form that conveys what you do
11 know as well as what you don't know, and some of the lines
12 of defense you are going to know a lot less than others, and
13 if you have a way of communicating that, then you have a
14 real heads-up on the notion of defense-in-depth.
15 On risk-informed, performance-based terms, the
16 committee is very much in agreement with the positions we
17 have seen articulated by the Commission on the fact that a
18 risk analysis is not necessarily decision analysis. Many
19 more things often go into a decision.
20 In risk there is always the opportunity to define
21 your risk parameters in such a way that they embrace issues
22 of cost and issues of schedule. That kind of activity has
23 carried with it a whole new field called performance risk
24 analysis or programmatic risk analysis, but one has to be
25 very careful about using risk in decision-making and making
20
1 sure that it is in its proper context.
2 Performance-based -- there are major differences
3 between materials and reactor licenses in the case of
4 performance-based regulations. I think we spoke to that at
5 the opening, that there's things that have been established
6 by tradition through the reactor field that have to be dealt
7 with in probably an evolutionary manner to move to the
8 risk-based way of thinking.
9 I think one of the primary compromises, if you
10 wish, of the doublet view of safety analysis is that
11 interpretation of design basis. I think if we had not come
12 up with the concept of a design basis accident, I think the
13 coupling between safety analysis and risk analysis would
14 have been much easier to see.
15 Regulatory burden -- I think that most people who
16 are mature about this discipline and practice it look to
17 relief in regulatory burden. They certainly don't look to a
18 relief understanding what the safety is, on the contrary
19 convinced that there will be much more knowledge about the
20 safety, but that eventually there needs to be some
21 efficiencies as a direct result of risk-informed practices
22 and those efficiencies need to take the form of changes in
23 the regulations.
24 So as to my closing comments, I think that we have
25 indicated a number of times that we think the risk view is
21
1 essential to judge the overall safety of a repository. It
2 provides us the perspective we need.
3 I think one of the things that is sometimes
4 underestimated is the experience base in the waste field.
5 While the use of probabilistic methods in the waste field
6 has come relatively late, the amount of activity has been
7 intense and the expenditure of effort, resources in the last
8 10 years, primarily through two projects -- the Waste
9 Isolation Pilot Plan and the Yucca Mountain, proposed Yucca
10 Mountain repository.
11 As a result of those activities we have learned an
12 enormous amount about how to apply these methods to a
13 geologic system, and as we said, one of the things that is
14 very important in evolving and transitioning to a risk way
15 of thinking is to not prescribe yourselves out of the
16 business. We need to retain a certain amount of
17 flexibility.
18 As to the details, even though we have been
19 arguing in my whole discussion here about the importance of
20 fixing some principles and the way we do some of the
21 analyses and the details of some of the methods -- that
22 aspect of it needs to be flexible.
23 I think that's all I want to say about the subject
24 and I am certain available for questions.
25 CHAIRMAN JACKSON: Okay.
22
1 COMMISSIONER McGAFFIGAN: I am not sure you ever
2 answered the Chairman's question about where the low-lying
3 fruit are in reactor space, but maybe your answer was that
4 your colleagues would get mad if you took all afternoon on
5 the subject, so --
6 DR. GARRICK: I think one of the areas is to --
7 okay, I will answer that.
8 I think the design basis accident philosophy
9 approach to regulation is sometimes a barrier to the
10 introduction of a risk-informed approach, and I think that
11 is a specific that you started to look at the regulations in
12 the context of design basis that you would maybe appreciate
13 that this is the one activity, this is the one analysis,
14 effort that has compromised, if you wish, an otherwise
15 doublet approach to safety analysis, and I know why it came
16 about and how it came about and that it was useful but it
17 created partitions that were artificial.
18 We got into class 9 accidents, severe accidents
19 and what have you, and these sort of artificial interfaces
20 that don't really exist in nature. And that we started
21 regulating against a design basis accident as if we did that
22 we would never have a severe accident. And we of course
23 learned that that's just not the case. So that's one major
24 issue that I would love to work with you on.
25 CHAIRMAN JACKSON: Well, would you -- we'll give
23
1 your colleagues plenty of time. Would you care to speak to
2 50.59?
3 DR. GARRICK: Well, I think that what you're
4 trying to do, namely it's like somebody has said if somebody
5 comes in for a change, even though we are not under a
6 risk-based regulatory process right now, such changes cannot
7 be realized anymore without some level of a risk analysis.
8 And I would like to see the 50.59 activities move more
9 aggressively in that direction to where there was increased
10 dependence on that, and I think also there would be great
11 signals sent out to the licensees if with that came a real
12 examination of 50.59 in terms of its deterministic, in terms
13 of its traditional requirements.
14 I think that one of the things that is causing
15 quite a bit of anxiety, and I'm sorry I wasn't to the
16 meeting last week, is that many people are discouraged about
17 risk on a couple of counts. One is this whole idea of
18 keeping a comprehensive risk assessment current is viewed as
19 an extensive burden, and, two, and this involves the NRC,
20 it's not clear to a lot of licensees just what the benefit
21 is, that if they have to go ahead and comply with all of the
22 so-called deterministic requirements, they're not so sure,
23 given the maturity of the industry, that they want to engage
24 themselves in a research-oriented kind of activity just for
25 the sake of building confidence in a risk-based approach to
24
1 regulatory practice.
2 So my view on this is that probably the pilot
3 programs you have are useful. They provide a lot of
4 insights and problems learning about the application of risk
5 to a whole family of issues, everything from hydrogen
6 recombiners and their necessity to the utility of a graded
7 quality assurance program. But I think that the thing that
8 would really advance the cause would be some rather
9 significant backoff, if you wish, or modification if you
10 wish of a regulation that is a heavy burden, on the basis
11 that you're now confident that what was being sought as a
12 result of that regulation is more than offset by the new
13 methods and the new practices.
14 CHAIRMAN JACKSON: Okay. Thank you.
15 Dr. Fairhurst.
16 DR. FAIRHURST: Thank you very much.
17 What I'm going to address is clearly a restatement
18 of material communicated in a letter in April. That was
19 based on a presentation in March from the Office of Nuclear
20 Regulatory Research concerning interim guidance and the
21 support of the final rule on radiological criteria for
22 license determination -- license termination, sorry.
23 I first lay out the several general observations.
24 One, that obviously decommissioning is a subject that's
25 going to be of continuing and probably growing regulatory
25
1 importance. Secondly, that the license termination issue is
2 a complex one, varies very widely from case to case from
3 very simple determinations to really quite complex
4 situations. And the NRC resources required to deal with it
5 are correspondingly quite varied.
6 Then the next observation was really a picking up
7 a little bit on what Dr. Garrick's constant philosophy is
8 that we need to be dealing with a risk-informed,
9 performance-based criterion. This is another case where the
10 changes that are envisaged are along those lines. That's
11 not saying there is some need for -- there is a need for
12 regulatory consistency with respect to the use of the total
13 dose standard basing things on health effects, having some
14 flexibility in the regulatory approach because of this
15 complexity, and also in this particular case recognizing the
16 role of Agreement States. They, too, feel they have a stake
17 in it.
18 An issue that was brought to us and which I know
19 you're very familiar with, but it was raised first by the
20 industry, nuclear energy industry, was this question of dual
21 Federal regulation, and that this is a serious problem and
22 one that is not easy to deal with, but somehow is going to
23 have to be dealt with.
24 The main recommendations in our letter, first we
25 were somewhat overawed by the complexity of the regulatory
26
1 guidance. I think the word we used was it was formidable.
2 I think you responded in kind and said yes indeed you
3 understood it and that maybe it needs to be -- need to take
4 some advantage in this electronic age of finding ways to
5 make it more comprehensible, user-friendly, and a little
6 more menu-driven format.
7 Another issue that we felt we needed to bring out
8 was that the ALARA approach maybe should be considered to be
9 in some cases leading to unnecessary conservatism, and we
10 feel that if you could meet the 25 millirem all-sources or
11 pathways limit, that should be sufficient. I think in your
12 answer to us you mentioned a concern or a feeling that in
13 some cases if it was a simple thing to do, then one could
14 perhaps go lower if it was a question of just wiping things
15 down. But I think we still hold to the notion that that 25
16 millirem should be for most cases sufficient to meet what
17 we'd call ALARA.
18 COMMISSIONER DIAZ: Could you please elaborate on
19 the reason why you believe that it's possible or it's
20 justifiable?
21 DR. FAIRHURST: Well, yes. In the -- first of
22 all, the doses that one receive from 25 millirem from all
23 pathways I think generally would be considered to be of
24 little concern as far as health effects.
25 COMMISSIONER DIAZ: Yes.
27
1 DR. FAIRHURST: There is also I think the feeling
2 that the formula rem standard ground water but if you use
3 the 25 millirem all pathways, it probably will in many cases
4 satisfy the formula rem. I'm not an authority in this, but
5 it's what I've been led to understand, that if you look at
6 the requirements that are being suggested by people that
7 this is not a major deviation from those in most cases.
8 COMMISSIONER DIAZ: But it's the ALARA interface,
9 what I'm concerned, we always put the ALARA interfaces and
10 the additional satisfaction of --
11 DR. FAIRHURST: Well, yes, you know, ALARA, as low
12 as reasonably achievable, and one can then argue, Dr.
13 Garrick wants evidence, wants facts, reasonable is a very
14 subjective word, and the question is what is reasonable.
15 And you can force somebody out of business perhaps
16 financially by pushing them to an enormous amount of effort
17 for very little benefit.
18 COMMISSIONER DIAZ: We have a long history of
19 using ALARA.
20 DR. FAIRHURST: Pardon?
21 COMMISSIONER DIAZ: We have a long history of
22 using ALARA.
23 DR. FAIRHURST: Oh, yes. Yes, I'm just saying
24 that --
25 COMMISSIONER DIAZ: We've managed to keep it
28
1 within bounds.
2 DR. FAIRHURST: Right.
3 CHAIRMAN JACKSON: Well, I mean, isn't also if I
4 go back to what Dr. Garrick was saying, that if you talk
5 about using risk analysis or, you know, today, particularly
6 within the ALARA framework, does that not offer a way not to
7 abandon what has been a cornerstone of how we've done our
8 business, but at the same time address the issue of
9 unnecessary conservatisms from a cost-benefit point of view?
10 DR. GARRICK: Yes, and one thing that's very
11 important, and I'm sure that Charles was going to comment on
12 this, is that when we say in reference to this specific
13 issue that the 25 rem is acceptable, that's not saying that
14 we don't believe in ALARA. ALARA is a rational way to look
15 at things.
16 CHAIRMAN JACKSON: That was my point.
17 DR. GARRICK: If you can meet a standard and
18 spending 10 cents reduce it by 10, of course you would do
19 that.
20 CHAIRMAN JACKSON: Right. That's all.
21 DR. GARRICK: Yes.
22 COMMISSIONER McGAFFIGAN: This is a more generic
23 question, but I will point out you were listened to by the
24 Commission. Our staff requirements memorandum on this
25 particular point uses the word "may." It isn't quite as
29
1 definitive as definitive as you, but in addition if the
2 licensee complies with the 25 millirem dose criterion using
3 the screening methodology, the D and D code which itself is
4 quite conservative, the licensee may have met the intended
5 ALARA requirement. May have met. We didn't, you know --
6 DR. GARRICK: Yes.
7 COMMISSIONER McGAFFIGAN: Therefore additional
8 demonstration of compliance may not be necessary. So we did
9 listen, but we also wanted to take into account by using
10 those mays the circumstances where for 10 cents you get a
11 factor of 10 --
12 DR. GARRICK: Yes. Absolutely. Absolutely.
13 CHAIRMAN JACKSON: Okay.
14 DR. FAIRHURST: And the final point that was made
15 in the letter was that we felt that the D and D code that is
16 being considered should have some flexibility for change if
17 one finds, for example, that the foundations on which it's
18 built change, such as the linear no-threshhold hypothesis.
19 And your response I think was that if that is changed, it
20 will have other ramifications apart from just modifying the
21 D and D code, and we know it will.
22 We also recommended that it would be useful to try
23 to take some test sites, complex test sites, and go through
24 the implementing guidance and see how it works out in
25 reality. There was a suggestion made that there might be
30
1 some level of conservatism by using generic parameters and
2 it might be possible and simple to locally add regional
3 parameters, it might reduce the conservatism.
4 I might add in conclusion that yesterday we heard
5 a presentation from the NRC staff about developing a
6 standard review plan, and it appears that things are moving
7 quite well along where they are about to test it on a
8 complex site and they are considering a number of things to
9 improve flexibility. So I think this is on course.
10 CHAIRMAN JACKSON: Yes. Thank you.
11 COMMISSIONER McGAFFIGAN: Did they give you a copy
12 of our SRM too at some point, because a lot of that was
13 directed so that -- just so you know that your advice is
14 listened to, a lot of the thoughts in the SRM I think and
15 part of all of us was the result of your work and very much
16 appreciated.
17 DR. GARRICK: We are encouraged.
18 DR. FAIRHURST: So we will give you an update
19 later, I think, not just back-patting but we did very much
20 appreciate your response and comments to us on that. It was
21 helpful. It tells us that there is somebody listening and
22 responding. Thank you.
23 CHAIRMAN JACKSON: Of course. Somebody up here
24 even likes you.
25 [Laughter.]
31
1 CHAIRMAN JACKSON: That always helps.
2 Dr. Hornberger.
3 DR. HORNBERGER: Thank you, Chairman Jackson. As
4 always, it is a pleasure to be here.
5 My task today is to report to you on some of the
6 work that the ACNW did in looking at the waste-related
7 research program within NRC. And this was, as you know,
8 ACNW input to an ACRS report. ACRS was asked to review
9 safety-related research and they asked ACNW to look at the
10 waste-related portion.
11 The Office of Research has a fairly modest
12 program, mainly in decommissioning and decontamination, and
13 the ACNW did hear presentations from staff of the Office of
14 Research on that.
15 The NMSS, of course, classifies their work as
16 technical assistance, the work they do with the Center for
17 Nuclear Regulatory Analysis. But we are familiar with that
18 work mostly because we have been keeping track of the work
19 related to Yucca Mountain, and a lot of that work we judge
20 as quite innovative and very important, and so we classify
21 it -- or we decided to include that under research. And,
22 so, of course, we have had regular presentations and
23 interactions with staff of NMSS.
24 We also had a meeting where we had some briefings
25 from the Department of Energy with regard to their waste
32
1 management research. This is not the Yucca Mountain, but
2 this is a program that was done between their Office of
3 Research and the Waste Management Division to sponsor basic
4 research. And we also heard from EPRI, the industry side of
5 the house, on how they conduct their research program.
6 So that is the background, just so you know what
7 we did to come to some of the observations that we had --
8 that we have listed.
9 The observations with respect to NMSS then, as I
10 very quickly summarize out of the report, it is obvious that
11 the Department of Energy has the big job in terms of coming
12 forward with a license application for Yucca Mountain and
13 their research program, obviously, has to show that. So
14 that their research budget is much, much larger than the NRC
15 budget.
16 We took -- one of the reasons we took a look at
17 EPRI was because EPRI has a very -- also a very modest
18 research program, and we were interested in the way they
19 handle it. Of course, from the industry side, they have
20 lots of flexibility, they have almost no constraints, and so
21 they use performance assessment to prioritize the topics
22 that they go after and then they simply go out and find the
23 best person that they can to do the work that they want to,
24 and they contract with that person. And, clearly, the NRC
25 simply can't have that kind of flexibility.
33
1 But one of the observations that we did have was
2 that the research and technical assistance programs within
3 NRC really do have to be focused and flexibility and carry
4 the respect of the scientific community. And, obviously,
5 the NRC has to continue to have national and international
6 stature in the whole waste management area.
7 CHAIRMAN JACKSON: Let me ask you a question. I
8 noted that you made a number of some specific
9 recommendations and one of them was, this goes back to
10 high-level waste, you said -- well, and more broadly,
11 actually. You said that the performance assessment model
12 should be structured to represent repository performance as
13 realistically as possible.
14 I mean is there an implication there that the NRC
15 is not using realistic assumptions or realistic models? Or
16 is this just kind of an overall --
17 DR. HORNBERGER: No, actually, I think that we did
18 have that comment in a previous letter and we continue to
19 believe that the NRC, the staff must continue to strive to
20 be as reasonable as possible -- as realistic as possible,
21 excuse me, and to ferret out any conservatisms that are
22 built in and make sure that they are appropriate
23 conservatisms.
24 CHAIRMAN JACKSON: So it is really a question of
25 following a line with some modulation, --
34
1 DR. HORNBERGER: Yes.
2 CHAIRMAN JACKSON: -- as opposed to that they have
3 been on entirely the wrong track?
4 DR. HORNBERGER: Oh, no. In fact, quite the
5 opposite. We think that they are very definitely on track.
6 CHAIRMAN JACKSON: Okay. I just wanted to be
7 sure.
8 DR. HORNBERGER: I think the first bullet in terms
9 of recommendations, really, really should -- NMSS should
10 continue to focus their technical work.
11 CHAIRMAN JACKSON: Okay.
12 DR. HORNBERGER: They have been doing an excellent
13 job, by the way, in using the TPA, their total performance
14 assessment code, to look at the priorities, to continue to
15 assess the key technical issues and the sub-issues. And
16 they have used it -- I had a chat with Margaret Federline, I
17 guess in April, on this, and she said, yes, they do look at
18 these results and they do have -- they try to maintain as
19 much flexibility as they can to redirect work at the Center
20 as appropriate. So --
21 CHAIRMAN JACKSON: Well, a concern I had had
22 relative to the TPA was the data that the NRC had available
23 to it, because in order to be realistic, you have to have
24 data that tells you something about the site you are trying
25 to model. Do you have any comments or concerns in that
35
1 particular area?
2 DR. HORNBERGER: Yes. I mean, clearly, of course,
3 the DOE, their TSPA suffers from exactly the same problem.
4 So it is not just NRC TPA, but it is the DOE and, of course,
5 EPRI uses their total performance model and they have
6 exactly the same kind of constraints.
7 I think that there are clear areas where the
8 database is sketchy, shall we say, and I think that Ray
9 probably will highlight at least a couple of areas where we
10 really -- we think that probably the database with regard to
11 engineered systems, in particular, definitely needs work.
12 The NRC obviously can't afford to collect all of
13 those data, they have to be very select in terms of what
14 they focus on. And I think that is the focus and
15 flexibility issue that we raised with respect to the
16 high-level waste.
17 CHAIRMAN JACKSON: Is there more opportunity with
18 making use of data that DOE itself generates, but in our --
19 in the models?
20 DR. HORNBERGER: Yes. Oh, absolutely. Charles
21 and I just were up on the seventh floor at lunch and had a
22 demonstration of the three-dimensional geological model for
23 -- that was developed by DOE. And the NRC is verifying this
24 and basically considering what the criteria will be for them
25 to accept it into their own use. And so DOE invested a huge
36
1 amount of money to develop a tool that is I think going to
2 be very useful for the NRC, as one example.
3 CHAIRMAN JACKSON: Thank you.
4 DR. HORNBERGER: So, in terms of the
5 recommendations, as I said, the continued focus of the
6 technical work by using the TPA. We have had clear evidence
7 over the past several meetings that the DOE design continues
8 to evolve and we anticipate that it will evolve as we go
9 into the future with changes. Therefore, the flexibility
10 with the Center has to be maintained in terms of definition
11 of the tasks. As I said, the main flexibility that we
12 observed with EPRI is that they had freedom to engage
13 anyone, any expert in the world without constraints as to
14 prior work with DOE or anything else, and NRC doesn't have
15 that.
16 Nevertheless, we do feel strongly that outside
17 experts, engaged appropriately in a surgically precise
18 manner, again, can enhance both the acceptability, and when
19 you get advice from world experts, really leading experts in
20 the world, I think that it does have -- it reflects
21 credibility onto the program by having these excellent
22 people from the outside concur with you.
23 And there have been a range of letter reports and
24 this last bullet really comes from a letter that we wrote to
25 you on comments on performance assessment capabilities,
37
1 where we, in fact, again, identified, because of this switch
2 -- not a switch, but the enhancement of interest in the
3 engineered part of the system of the repository, which we
4 believe is going to continue to become ever more important
5 as we -- as DOE goes forward, that the NRC staff really does
6 have to make sure that they have the right capabilities,
7 either here or at the Center, or that they have the
8 flexibility to engage help as they need it.
9 The next observation with respect to NMSS, again,
10 it is really a repeat in the sense of the point that I just
11 made. It is imperative that the outside world not view NRC
12 analyses as overly simplistic. And, again, we think the
13 ACNW believes that one way to help out in this is to engage
14 prominent waste engineers and scientists in the resolution
15 of waste management problem.
16 And, of course, we understand that funding has
17 been an issue for years. We discussed, I think, a year ago
18 about the decrease in funding for certain -- curtailing work
19 on certain KTIs and this can throw monkey wrenches,
20 obviously, into programs, and people do have to live with
21 that. We don't have an infinite resource here. But, at any
22 rate, we think that the Center funding has to be such to
23 ensure that this ongoing effort is maintained.
24 Our observations with respect to the Office of
25 Research, really, the first bullet here on the observation
38
1 has to do with priority setting. We heard the presentations
2 on the research. We are impressed by the work that is being
3 done by the Office of Research. But we thought that setting
4 priorities and how priorities were set were a key, and it
5 was unclear to us in our discussions whether the current
6 structure for setting priorities was what we would consider
7 rigorous.
8 We were told that certainly the staff experience
9 and knowledge had gone into setting the priorities, and
10 these people have had many years experience, and there is
11 reason to believe that they are on top of things.
12 Nevertheless, whenever -- especially with such restricted
13 resources, you really want to make sure that you focus on
14 the priorities. So our recommendation to the Office of
15 Research, that we see a need for a structured organization
16 for identifying the priorities and make sure that peer
17 review is involved, and that it focuses on the users,
18 because, after all, it is an applied program, if you like.
19 So that summarizes our input on research.
20 CHAIRMAN JACKSON: Thank you.
21 Dr. Wymer.
22 DR. WYMER: Thank you. My presentation today is
23 on the near-field environment, performance of engineered
24 barriers, particularly as they relate to the Yucca Mountain
25 Repository. And a big part of what I will present is based
39
1 on a working group meeting that we held June 10th and 11th,
2 a two-day meeting where we brought in experts from outside,
3 as well as DOE and NRC and from the Center, and had
4 presentations.
5 We think this topic is particularly important
6 because of the increased attention paid by DOE to engineered
7 barrier system performance. And it is important to the NRC,
8 of course, because they have to keep up with things and have
9 to license that repository, so they have to understand what
10 DOE has done.
11 We also got a lot of input from the working group
12 with respect to what are the really important technical
13 issues, and there was a lot of sort of ad hoc discussion
14 that wasn't even on the agenda that raised some areas that
15 I'll get into which we thought were particularly important
16 and relevant.
17 So, going to the next viewgraph, we have some
18 general observations to start with, then I will give some
19 specific insights that were obtained out of the working
20 group. First, the Yucca Mountain Repository is different
21 from other planned repositories around the world in that it
22 is in an unsaturated and oxidizing environment, which really
23 changes a lot of things with respect to corrosion, with
24 respect to chemistry. Whereas, most of the repository
25 designs are in a saturated environment which is primarily a
40
1 reducing environment, the chemistry is quite, quite
2 different. So that's an important difference and it puts
3 Yucca Mountain kind of apart from all the other repository
4 design considerations.
5 The other thing that George mentioned is that it's
6 like shooting a running deer. The EBS design continually
7 changes as the Department of Energy picks up on new facts,
8 new importances, new emphases arise, and so every time we
9 hear from them there's something new and it's generally in
10 the right direction and we are glad to hear it, but it does
11 make it a moving target so it's kind of hard to keep up with
12 the design.
13 Consequently and concurrently that means that the
14 NRC Staff has to be quick on its feet and has to have
15 flexibility to stay abreast of this evolving situation.
16 The Department of Energy talks about a robust
17 depository and our understanding of what robust means is
18 that it is simply enough that it is not going to collapse
19 under its own complexity and that the defense barriers, that
20 barriers are decoupled so that if one fails, everything
21 doesn't fail, so robustness implies as much simplicity as
22 possible and as much decoupling as possible of one barrier
23 from another so that you don't have in the language of the
24 reactor you don't have common mode failure.
25 We think it is important, and we are not sure we
41
1 see good evidence of this out of the EBS design options, we
2 think it is important that there be a top-down systems
3 engineering approach rather than a bottoms-up. By that we
4 mean that you need to set the overall goals and the overall
5 design features at the beginning and build toward those
6 rather than seeing a lot of details emerge and let those
7 form your design. There is probably quite enough of that
8 going on as there should be and attention should be paid to
9 that in the NRC's review of the situation.
10 Then something that emerged that wasn't really on
11 the working group agenda but there was a lot of discussion
12 that it emerged as a very important issue had to do with the
13 preclosure issues of the repository. That thing may stay
14 open for 100 years. DOE talks about 100, 200 years -- they
15 get a little unrealistic in my view, but nonetheless they
16 are talking a long time into the future keeping that
17 repository open and during that time there are a lot of
18 issues that come up having to do with heat loading and
19 retrievability of waste packages and during that time the
20 repository performance features can be confirmed or denied
21 and the NRC needs to be certain I think that it pays
22 attention -- we think -- that it pays attention to the
23 preclosure aspects of the repository development, which one
24 of our expert panelists said should be an evolving thing.
25 He even advocated continual changes in the design
42
1 of the waste packages and then some features of the
2 repository as information is gained over the 50 to 100 years
3 before closure -- so this is an area where little attention
4 has been paid by either DOE or NRC as far as we have been
5 informed to date and we think that it deserves attention.
6 On the next exhibit here, we get into I guess
7 near-field environmental issues, and by that we mean
8 anything from the concrete liner of the drift on in --
9 anything inside there is what we define as the near-field.
10 One of the things that came out and our first
11 reaction, my first reaction to it and I think maybe the
12 committee's, was that gee, this is kind of obvious, why are
13 you telling us this, is that it is very important how much
14 water comes in and how much contacts the waste. Well, you
15 know, that is what we call a privileged glimpse at the
16 bleeding obvious, but when you think about it and you think
17 about what DOE is planning, it turns out to be worth paying
18 attention to. They are talking about a drip shield. They
19 are talking about potentially backfills and they are talking
20 about the effects on solubilization and transport of fission
21 products and all this relates to water, so anything you can
22 do to control the water is important and that is beginning
23 to get a fair amount of attention, and I will say a little
24 bit more about it.
25 There was some concern expressed about the
43
1 abstraction from the PA models, from the near-field process
2 level models, into a PA model. The concern was, first, is
3 the fullblown model adequately supported by data, a point
4 that we dealt with a little bit earlier, and second, is the
5 abstraction to this more simplified model done well? Does
6 it really incorporate all of the salient points in the
7 process level models? -- so we thought that attention needed
8 to be paid to that. Now there is a great deal of attention
9 going into that but nonetheless it was brought up and we
10 thought it deserved mentioning here.
11 The near-field chemistry is near and dear to my
12 heart and there is a lot of chemistry discussed, even though
13 one of the participants characterized the meeting as a
14 "corrosion meeting" -- he was a corrosion expert and my
15 answer to that was to a hammer everything is a nail -- and
16 he felt it was a corrosion meeting.
17 Actually, there is a lot known about the chemistry
18 of the water entering the repository but there is not much
19 known at all about what happens to that water when it starts
20 hitting things inside the repository, especially at
21 mechanistic level. There's a lot of empirical and anecdotal
22 information but there is not a lot of true basic
23 understanding of the chemical reactions that the in --
24 flowing water will bring about as it contacts in particular
25 the fuel material.
44
1 Those reactions are extraordinarily complex and
2 are poorly, poorly understood on a fundamental level.
3 The next exhibit here deals with corrosion. As I
4 said, one of the participants felt it was a corrosion
5 meeting and it was very heavy on corrosion, and partly that
6 is because there is a lot of expertise on corrosion both
7 within the NRC and its contractors and at DOE.
8 There are good people doing good work on corrosion
9 and there is a lot of interaction between those people but
10 you need to distinguish, we feel, between a good expertise
11 and a basic understanding of corrosion issues and specific
12 understanding about specific corrosion problems relating to
13 specific materials. That gets into the next point on this
14 exhibit, which has to do with the wonder alloy C-22. That
15 is a high nickel based allow which has received a great deal
16 of attention. It is extremely corrosion resistent.
17 I call it a wonder alloy. It is sort of a --
18 without tongue-in-cheek, it's a very good material.
19 However, the information base with respect to corrosion is
20 limited with respect to the amount of time that people have
21 been studying this material -- something less than two
22 decades, which is a whole lot less of course than people
23 have looked at iron and titanium and other kinds of alloys,
24 so there was a lot of stress being put on the use of this
25 alloy and it probably will play a very important part in
45
1 DOE's analysis.
2 CHAIRMAN JACKSON: Commissioner Diaz.
3 COMMISSIONER DIAZ: Just out of curiosity, has
4 anybody been trying to look at single crystal alloys at all
5 because of their tremendous resistance to corrosion and
6 diffusion?
7 DR. WYMER: No. As far as I know, that has not
8 taken place. Of course, that would be a mighty big single
9 crystal but --
10 COMMISSIONER DIAZ: I have seen them big enough in
11 Russia. They do make them big.
12 DR. WYMER: No, that has not -- that wasn't
13 brought up and we're not aware of anything.
14 COMMISSIONER DIAZ: Interesting. There is a
15 program from STIO that gives a nickel alloy, single crystal
16 alloys, as being done now, last four, five years.
17 DR. WYMER: I know that single crystals are
18 sometimes much more resistent to corrosion.
19 COMMISSIONER DIAZ: Much more -- and they're
20 trying to put them in jet engines.
21 DR. WYMER: Even with the microcrystalline
22 materials the corrosion resistance is high for this
23 material. It is based primarily on the existence of an
24 oxide layer because this alloy like all other metals --
25 COMMISSIONER DIAZ: Right, right --
46
1 DR. WYMER: -- most other metals is not --
2 COMMISSIONER DIAZ: And there would be no
3 diffusion and so it is a tremendous advantage.
4 DR. WYMER: Despite the fact that this looks very
5 good, one or two of the corrosion experts raised concerns
6 having to do with localized or crevice corrosion that might
7 occur when you get -- by evaporation concentrations of
8 chloride iodine and other kinds of things that enhance
9 corrosion.
10 One of the speakers brought up a very interesting
11 observation which deserves to be proven or disproven. That
12 is, he said that there is for C-22 a temperature regime
13 during which corrosion can occur. Above that temperature
14 and below a temperature it is practically nonexistent. I
15 mean the corrosion is very low, which suggests that by
16 judicious arrangement of conditions you can avoid that
17 temperature regime for long periods of -- to exist for long
18 periods of time and thereby greatly enhance the lifetime of
19 the material.
20 So that they're knocked down or verified.
21 CHAIRMAN JACKSON: Yes.
22 COMMISSIONER McGAFFIGAN: How quickly can you
23 knock down or verify that? Is it relevant to licensing of
24 Yucca Mountain, or is it a 20-year research project?
25 DR. WYMER: I can't answer that question
47
1 authoritatively, but my feeling is though that you could
2 certainly ferret out a major difference between being in the
3 temperature regime and being out of the temperature regime
4 in a fairly short period of time.
5 COMMISSIONER McGAFFIGAN: Where is the temperature
6 regime where corrosion may occur according to this?
7 DR. WYMER: It's fairly low.
8 DR. HORNBERGER: It's 100 to 120 C.
9 DR. WYMER: Maybe 80 to 120 or something like
10 that. It's fairly low.
11 COMMISSIONER DIAZ: I'm sorry I'm smiling. We
12 were working at 1,400 degrees Kelvin.
13 DR. HORNBERGER: Well, it won't get quite that
14 hot.
15 DR. GARRICK: At a little different time constant.
16 COMMISSIONER McGAFFIGAN: Could I also -- if 80 to
17 120 degrees centigrade is where the risk range is, is it
18 easy to -- I mean, presumably you wouldn't want to be above
19 that, that would be difficult to control, or maybe that is
20 where you end up, if there's a lot of heat in the mountain
21 maybe you end up above 120 and never have to worry about
22 coming below it. But how -- which way were you going to try
23 to control?
24 DR. WYMER: One of the -- I don't know, but one of
25 the considerations is that if these alloys are as good as
48
1 they are claimed to be, then even in the corrosion regime
2 they may well be stable for times long enough that high
3 temperatures due to the decay heat are not important
4 anymore, in which case you might drop down below that. So
5 that's one consideration.
6 We need to know more about this particular point,
7 because it is apparently important.
8 Another point was brought up with respect to
9 corrosion of the outer layer. The C-22 is a thin inner
10 protective layer in the waste package. There's a much
11 thicker outer iron or steel layer which is really the main
12 container for the waste. And that will corrode.
13 One of the experts brought up the fact that well,
14 suppose you get a hole in that container and it rusts and
15 the rust is on the inside rather than the outside, there's a
16 volumetric change as you go from the metal to the oxide, and
17 it'll expand and crush what's inside. And it may in fact
18 bend, break, fracture, and some other ways do harm to the
19 inner container, C-22 or whatever it is, whatever's chosen.
20 And that has not been addressed in detail.
21 Also, the effect, when this happens, when you get
22 iron oxidation, the effect of ferric ion on corrosion is the
23 important factor.
24 Then one of the experts brought up the issue of
25 weld integrity. He says we've got to have a couple miles of
49
1 welds there, and nobody really knows much about corrosion of
2 welds. They know a lot about corrosion of massive
3 materials, but welds are a horse of another color, and they
4 always behave differently from the bulk material.
5 Am I overrunning my time?
6 CHAIRMAN JACKSON: No. Go ahead.
7 DR. WYMER: So the point was brought up that it's
8 important to pay attention to some of these more practical
9 aspects like weld integrity and their impacts on long-term
10 performance of the waste package, waste canisters.
11 Then the whole issue of backfills is an important
12 one. You can control ingressive water with backfills to a
13 certain extent. You can control chemistry in the repository
14 by using certain kinds of backfills having reducing
15 properties or chemical properties to retain elements that
16 might otherwise transport rapidly out of the container.
17 And then finally some of the experts question the
18 use of taking credit for the fuel cladding, the Zircaloy
19 cladding on the fuel as part of what you rely on to prevent
20 release of the fission products, and indeed we said well,
21 we're still thinking about that. We're not sure.
22 CHAIRMAN JACKSON: Doesn't that also put
23 constraints relative to whether damaged fuel could go into a
24 repository?
25 DR. WYMER: Sure, it does. Sure.
50
1 CHAIRMAN JACKSON: Because that issue has come up
2 particularly since the repository is, you know, it's
3 primarily for commercial fuel, and the issue is there, but
4 also for other spent fuel.
5 DR. WYMER: Sure.
6 CHAIRMAN JACKSON: And then the issue of the
7 condition of the fuel, which includes its cladding comes
8 into play.
9 DR. WYMER: That's right.
10 CHAIRMAN JACKSON: And the impact on the overall.
11 DR. WYMER: Yes. Bending or cracking or any of
12 these things is important. Yes.
13 CHAIRMAN JACKSON: Okay.
14 DR. WYMER: Then the final point was -- area that
15 was discussed was the release of fission products and
16 actinides from the fuel itself and the transport of those
17 materials, and one of the invited experts particularly
18 pointed out the fact that when you let the water reach the
19 fuel and the water is saturated with oxygen, as it will be
20 under normal conditions, then you're going to get oxidation
21 of the UO2 to some higher oxide, and also the radionuclides,
22 of which there'll be about 3 or 4 percent in that fuel, can
23 also -- some of those also can oxidize, depending on what
24 they are. Because normally they'll be in an oxidation state
25 governed by the fact that they were born in UO2 and there
51
1 was that much oxygen available and no more.
2 So the formation of these oxidation products could
3 in fact affect the release rates of the fission products and
4 actinides and therefore the source term ultimately for the
5 dose.
6 And finally there was the issue of secondary
7 phases and of colloid formation. The secondary phases is
8 not exactly the same as the formation of oxides. In this
9 case they were talking about specific stable long-term
10 stable secondary phases that would incorporate inefficient
11 products or actinides within their structure. And this
12 could dramatically change their release of fission products,
13 actinides, but not much is known about that, and there is no
14 good thermodynamic data base to use as a basis for
15 calculating what the stable phases might be.
16 And finally colloids and pseudocolloids are I
17 think clearly going to be of importance, and that was
18 discussed at some length. A colloid is something like a
19 plutonium polymer. A pseudocolloid is something like clay
20 or iron which forms a colloid which then absorbs physically
21 or chemically a fission product or an actinide, which then
22 would move the way the colloid moves rather than as the way
23 an ion in that material would move. And we felt that
24 attention needed to be paid to those kinds of things because
25 they could have a dramatic effect, the secondary phases, for
52
1 retarding movement, colloids for enhancing movement.
2 And then we thought that we wanted to know more
3 about -- and I'm sure more is known, but we don't know it --
4 more about the rank ordering of the importance of these
5 various barriers to movement in the repository one with
6 respect to another so that we know what's the 800-pound
7 gorilla and what we don't care about.
8 CHAIRMAN JACKSON: Sure.
9 COMMISSIONER McGAFFIGAN: A fairly fundamental
10 question comes from this presentation and our presentation
11 by the staff a few weeks ago about performance assessment in
12 this area, and that is how much of a grip are we going to
13 have on these engineered-barrier issues by the time we're
14 licensing, and will a conservative licensing process with an
15 array of expert opinion have to ultimately perhaps not guess
16 that the C-22 is going to be quite as good as claimed, and
17 how do we -- how is this all going to come down. The staff
18 seems to have -- and I don't have the exact transcript of
19 the meeting in front of me -- but some real concerns about
20 overemphasis on engineered barriers at the current time in
21 some of the DOE work. So I wonder if that's shared.
22 DR. WYMER: Why don't you, John?
23 COMMISSIONER McGAFFIGAN: Well, either one of you.
24 DR. WYMER: I'll take his lead.
25 DR. GARRICK: Well, it is a difficult problem.
53
1 But I recall the same anxieties when we first started
2 looking at the reactors in terms of the contribution of
3 mitigating systems, that there was great skepticism about
4 our ability to be able to quantify, for example, the worth
5 of a containment system or a high pressure injection system,
6 and much progress was made on that in a relatively short
7 period of time.
8 And I think when we started focusing on that, and
9 we started dealing with the question of what is the real
10 worth of containment, for example, because that was a
11 classic, similar argument, that we don't know how much the
12 containment -- we can design it to certain pressures and we
13 can make it robust. But it wasn't too long before we were
14 able to put some quantification to the whole process and
15 suggest that for some containments, the capacities of those
16 containments were anywhere from 1-1/2 to 4 times their
17 design basis. And it was an extremely important
18 breakthrough to get -- to begin to get those kinds of feels
19 and senses of what the defense mechanisms were.
20 I think the same is true here. I think that right
21 now it is new territory. It is a different problem. It is
22 -- the processes involve extremely long time constants.
23 They are serial for the most part, rather than parallel.
24 They are passive for the most part, rather than active. But
25 I am confident that if we just stop worrying about it and
54
1 start focusing on dealing with the question of how much
2 value are we getting from a drip shield or backfill, or an
3 outer barrier, 100 millimeters of steel versus 50, or 50
4 millimeters of C-22 versus 20, I think we can -- I think we
5 will be surprised.
6 There has been a lot of advancements made in what
7 I would call structural mechanics from a probabilistic
8 perspective and I am more confident than most people.
9 COMMISSIONER McGAFFIGAN: Could I follow-up? You
10 mentioned the word time constants, and one of the issues --
11 I mean if you take -- if, hypothetically, we are working
12 with a 10,000 year period, which is what we worked with, and
13 that may not -- there are longer periods. One can consider
14 the Academy talked about longer periods.
15 But one of the problems with these time constants
16 is you can -- if you really believe the analysis for 10,000
17 years, you sort of -- everything is nice and tightly
18 contained right there at the site, and there is no -- there
19 is no source term going very far. And how robust that
20 judgment is is going to be the heart of the licensing
21 process, if, indeed, there is a lot of emphasis on the
22 engineered barrier.
23 But at some point these things break and we will
24 have to look at what happens once the geologic system is
25 providing the containment.
55
1 DR. GARRICK: Right.
2 COMMISSIONER McGAFFIGAN: And how things go. But
3 to some degree, because of the time constants, you can get
4 into a situation where, if 10,000 years is the licensing
5 period, is the period for analysis and deciding whether to
6 grant a license, the problem gets defined away, and then it
7 just pops up at 60 or a 100 or --
8 DR. GARRICK: The compliance problem gets defined
9 away, but the risk problem does not.
10 COMMISSIONER McGAFFIGAN: Right.
11 DR. GARRICK: Right.
12 DR. WYMER: Well, I would like to throw in my two
13 cents on that. We can identify, and I have, a half a dozen
14 areas of potential concern and things that deal with the
15 adequacy of engineered barriers. But it is very possible, I
16 think likely, that by not particularly sophisticated
17 analyses, quite a few of these things will be laid to rest
18 as being below the horizon, and there will only be a few
19 that will stand out as peaks that we really -- that really
20 deserve attention. And that's why we make the point that
21 this rank ordering is -- early on, is important, because
22 those things which even on a semi-quantitative or almost
23 qualitative basis, you can rule out, reduce the field
24 substantially, or on the basis of the fact that DOE is not
25 even going to rely on those things in the first place.
56
1 So it seems to me that there is a -- we are just
2 before making a major simplification in what we need to be
3 concerned with. And if attention is paid to these, some of
4 these issues that we have raised here, they can -- some of
5 these will just be thrown aside and they won't even turn out
6 to be important.
7 COMMISSIONER McGAFFIGAN: One last question.
8 Whose job is it to bring about that major simplification? I
9 mean you are recommending it. But is that DOE's job to
10 bring it about?
11 DR. WYMER: It's DOE's job to recommend it. It is
12 NRC's to be sure that they are good recommendations.
13 DR. GARRICK: Speaking of recommendations, as you
14 know, this particular work is work in progress, and we
15 intend to send you a letter and to make some
16 recommendations.
17 CHAIRMAN JACKSON: Okay. Dr. Garrick.
18 DR. GARRICK: It's an interesting dichotomy. The
19 essence of reactor safety is the presence of water. The
20 essence of repository safety is the absence of water. You
21 would think we could get it right somehow.
22 CHAIRMAN JACKSON: Well, the presence of water can
23 also be a problem.
24 DR. GARRICK: Well, in some reactors, a special
25 problem.
57
1 CHAIRMAN JACKSON: That's right.
2 DR. GARRICK: And under some temperature
3 conditions.
4 I want to talk a little bit about planning.
5 Planning is something you kind of really hate to do. But
6 when you have done it, --
7 CHAIRMAN JACKSON: Oh, darn.
8 DR. GARRICK: -- you are really glad you did it.
9 That's the case.
10 CHAIRMAN JACKSON: Good. Because it is over or
11 because --
12 [Laughter.]
13 DR. GARRICK: Well, partly because of your
14 leadership, we have moved in the direction of trying to
15 become much more formal in our planning. The ACNW has
16 always attempted to prioritize and plan its activities for
17 the forward year and years. But it was -- this year was the
18 first time we attempted to get a little more structure and a
19 little more formal in the whole process.
20 We tried to lay down some rules that were the
21 basis for our planning activity. We wanted to be darn sure
22 that we didn't get ourselves so tied down to our plan that
23 we were not in a position to offer advice as a result of
24 some major changes and we did not want to get in a position
25 that we couldn't respond quickly to change. So we had that
58
1 as one of our major commitments and rules for the planning
2 process.
3 The one thing that does come from a plan is the
4 ability to kind of look at yourself and measure against some
5 sort of a baseline, and we have been doing that. We have
6 established the plan as input to our operating plan. And,
7 of course, our operating plan has such metrics in it as
8 timeliness of our information, its quality, its efficiency,
9 its effectiveness, et cetera.
10 Also, we, in this year, in a little more formal
11 manner, completed a performance evaluation of ourselves.
12 That was documented in a SECY document on June 1st. The
13 status of our planning activity is that we were extremely
14 pleased that the Commission also read that letter and
15 responded directly to us, and those comments are very
16 helpful and have to do with the fact that perhaps our
17 planning was a little too narrow in scope, maybe it didn't
18 match up with all of the elements of our charter, and we
19 intend to take those comments as source material for the
20 planning activity that we will engage in later this year.
21 We have received Commission requests for new work
22 as a result of exposing the plan. For example, in the
23 low-level waste area, the issue that has already been
24 brought up this morning of criticality at Envirocare and a
25 generic consideration of criticality in low-level waste
59
1 sites is something we are currently addressing.
2 We have, certainly, been addressing the issue of
3 risk. The comment was made to get outside of the box, if
4 you wish, and look at some topical issues like reactor
5 vessel handling and what to do about used reactor vessels.
6 The Trojan reactor vessel has been mentioned in particular.
7 And, of course, clearance levels are another
8 example of things that have been mentioned that we maybe
9 ought to be prepared to deal with. And, of course, we have
10 to be cautious about managing our scope because we have
11 resource limitations just like everybody else. And in
12 regard to that, there was a memorandum to the Chief
13 Financial Officer concerning additional resources for fiscal
14 year 1999 to give us increased confidence that we can,
15 indeed, respond to these requirements.
16 The Committee is very pleased to report that we
17 have issued letters on all of our first tier priority
18 topics. The first tier priority topics included such issues
19 as viability assessment and site characterization,
20 risk-informed, performance-based issues, engineered barrier
21 systems, decommissioning and research.
22 In kind of the spirit of accomplishments, we
23 provided recommendations and advice on a rather large number
24 of issues such as defense-in-depth. We wrote you a letter
25 in October of last year. Multiple barriers in March of this
60
1 year. The use of PRA in the waste field, this was the issue
2 of what lessons can we carry from the vast amount of
3 experience in the reactor field to the waste field, and we
4 wrote a letter on that. And on the subject of the effects
5 low-level ionizing radiation, prompted by Commissioner Diaz,
6 and we wrote a letter on that.
7 One of the things that the Committee has been
8 relatively sensitive to and quite active in is trying to
9 heighten the awareness and the need for attention to the
10 engineered barrier system issue in high-level waste disposal
11 and the growing apparent dependence on engineered systems in
12 -- being in the demonstration of the performance of the
13 repository, and we have been very active in addressing that
14 issue.
15 One of the highlights of the year and, certainly,
16 one of the most technically stimulating activities we have
17 engaged in in the last couple of years was the working group
18 that Ray Wymer was the lead person on, that we had in early
19 June, and we think that working group activity generated
20 some extremely valuable source material for us to address
21 much more intelligently the issues surrounding increased
22 dependence on engineered barrier systems.
23 One of the things that the Commission has reminded
24 us to do from time to time is to be aware of international
25 activities in our work and in our gathering of source
61
1 material as a basis for our advice. We have done a number
2 of things in direct response to that. One of the things we
3 certainly are pleased that happened is that we got a member
4 of the Committee, namely, Dr. Fairhurst, who has a vast
5 amount of international experience and seems to know
6 everybody in this business, and that has been extremely
7 helpful in organizing a number of things, including a trip
8 that we -- and a meeting we expect to have with the German
9 RSK later this year.
10 Future activities, we expect to issue to you a
11 major letter report on engineered barrier systems. We also
12 expect to issue letters on such topics as post-disposal
13 criticality, the NUREG, 10 CFR Part 63, total system
14 sensitivity analysis. In fact, we have completed that
15 letter at this meeting.
16 The interesting issue of importance measures and
17 the whole question of can you really do importance measures
18 for systems typical of repositories. The issue of
19 decommissioning. And, of course, we expect to send you some
20 advice on the viability assessment.
21 CHAIRMAN JACKSON: Commissioner McGaffigan.
22 COMMISSIONER McGAFFIGAN: Could I ask one question
23 on the post-disposal criticality issue? I know you got
24 briefed on this yesterday, and I understand you asked some
25 penetrating questions. If you go back to your risk-informed
62
1 definition of risk, the triplet model of risk, we have got
2 something there that is vanishingly small, although we
3 could, I guess, try to quantify it, and you questioned, and
4 I think appropriately, trying to quantify vanishingly small.
5 The consequences from the Oak Ridge study, even if
6 it happens, are not enormous. And so the question, from a
7 regulator's perspective, and the reason you have been asked
8 the question, obviously, is we -- the Commission is asking
9 is it -- When is enough, enough?
10 DR. GARRICK: Yes.
11 COMMISSIONER McGAFFIGAN: Do you have -- not
12 trying to get the letter report out of your mouth right this
13 moment, but do you have an initial impressions as to when
14 enough is enough in this area?
15 DR. GARRICK: Well, I do. I think this is an
16 ideal example of what we were talking about earlier, of an
17 analysis that should be risk-informed. Even though, to do
18 it quickly, we may be faced with a lot of uncertainties, I
19 suspect we still would learn a great deal about it. We are
20 going to probably encourage that kind of an approach be
21 taken. We are not very sympathetic to an extensive research
22 activity based on what we have heard so far.
23 COMMISSIONER McGAFFIGAN: Okay. Thank you.
24 DR. GARRICK: We have mentioned the issue of
25 international technical meetings. Dr. Fairhurst continues
63
1 to do that and be our ambassador, but we will enhance that a
2 little bit and see what he is up to when we all go to
3 Germany.
4 We expect to hold a stakeholders meeting in Yucca
5 Mountain vicinity to enhance public participation. You
6 recall that that is one of our goals, is to offer advice on
7 how to enhance public participation. And we expect,
8 finally, to conduct increasingly comprehensive
9 self-assessments.
10 COMMISSIONER DIAZ: Excuse me. A few moments ago
11 you mentioned that you have already reviewed or considered,
12 or read about the clearance of materials and the potential
13 development of a rule. Are you prepared to engage in this
14 issue of the clearance of materials? You don't mention in
15 your future activities.
16 DR. GARRICK: We are prepared to engage. I think
17 that's what advice committees are prepared to do.
18 COMMISSIONER DIAZ: That's good.
19 DR. GARRICK: It is not a comfortable issue and a
20 lot of people would just as soon that we not engage, but we
21 will. We will engage.
22 COMMISSIONER DIAZ: All right.
23 DR. GARRICK: I think that completes our
24 discussion. We are sorry we ran over a little bit, I guess.
25 CHAIRMAN JACKSON: Okay. That's all right.
64
1 COMMISSIONER McGAFFIGAN: My only comment is I
2 think next time they are going to have him do risk-informed,
3 performance-based at the end of the agenda rather than the
4 beginning.
5 [Laughter.]
6 CHAIRMAN JACKSON: No, I told them they would have
7 all the time they needed. You had all the time you needed.
8 DR. HORNBERGER: Yes, we did.
9 CHAIRMAN JACKSON: Well, let me just say that the
10 Advisory Committee's views on the matters you addressed
11 today are of tremendous value and importance to the
12 Commission as we are trying to deal with the complexities of
13 a number of technical and policy issues.
14 You talked about risk-informed and
15 performance-based regulation, which you know is an important
16 area.
17 DR. GARRICK: Yes.
18 CHAIRMAN JACKSON: On the issues associated with
19 licensing activities for high-level waste repository,
20 decommissioning, which is becoming increasingly important,
21 and other materials-related areas.
22 I want to commend you for the high quality of
23 today's briefing and of the work you do, and just to tell
24 you that the Commission does appreciate your efforts.
25 And so, unless there is any further discussion,
65
1 this meeting is adjourned.
2 DR. GARRICK: Thank you. Thank you very much.
3 [Whereupon, at 3:16 p.m., the meeting was
4 concluded.]
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
[an error occurred while processing this directive]
[an error occurred while processing this directive]
