118th Advisory Committee on Nuclear Waste (ACNW) Meeting, March 28, 2000
UNITED STATES OF AMERICA
NUCLEAR REGULATORY COMMISSION
ADVISORY COMMITTEE ON NUCLEAR WASTE
***
118TH ADVISORY COMMITTEE ON NUCLEAR WASTE (ACNW)
U.S. Nuclear Regulatory Commission
11545 Rockville Pike
Conference Room 2B3
White Flint Building 2
Rockville, Maryland
Tuesday March 28, 2000
The committee met, pursuant to notice, at 8:32
a.m.
MEMBERS PRESENT:
B. JOHN GARRICK, Chairman, ACNW
GEORGE M. HORNBERGER, Vice Chairman, ACNW
RAYMOND G. WYMER, ACNW Member
PARTICIPANTS:
MILTON LEVENSON, ACNW Consultant
HOWARD J. LARSON, Acting Associate Director,
ACNW/ACRS
JOHN RANDALL, ACNW Staff
JOHN T. LARKINS, Executive Director, ACRS/ACNW
RICHARD K. MAJOR, ACNW Staff
AMY SHOLLENBERGER, Public Citizen, Critical Mass
Energy Project
CHRISTIANA H. LUI, NMSS
JAMES R. FIRTH, NC, Division of Waste Management
MS. DEERING
STEVEN KRAFT, NEI
LINDA A. VEBLEN, NRC, RES/DRAA
DR. BRADY. C O N T E N T S
NUMBER DESCRIPTION PAGE
1 Strategy for Developing NRC
Sufficiency Comments 110
2 Industry Perspectives on Yucca Mountain
Sufficiency 137
3 Yucca Mountain Review Plan Development
Update 162
4 Characterization of Radioactive
Slags 199
5 Charts/Graphs . P R O C E E D I N G S
[8:32 a.m.]
DR. GARRICK: Good morning. The meeting will now
come to order. This is the second day of the 118th meeting
of the Advisory Committee on Nuclear Waste.
My name is John Garrick, Chairman of the ACNW.
Other members of the Committee included George Hornberger,
Ray Wymer, and Consultant, Milt Levenson.
This entire meeting will be open to the public.
Today we're going to first review the NRC Staff's plan for
the development of a strategy to produce site
characterization sufficiency comments on the Department of
Energy's Yucca Mountain site recommendation.
We're going to hear a periodic briefing on the
development of the NRC's Staff Yucca Mountain Review Plan,
and we will review two projects by NRC's Office of Nuclear
Regulatory Research on, first, the radioactive content of
slag that is produced as byproduct of the manufactured
metals; and, second, research on uranium plume attenuation.
Richard Major is the Designated Federal Official
for the initial portion of today's meeting. This meeting is
being conducted in accordance with the provisions of the
Federal Advisory Committee Act.
We have received one request from the Nuclear
Energy Institute, to comment on the Staff's site sufficiency
discussion. Should anyone else wish to address the
Committee, please make your wishes known to one of the
Committee Staff.
It is requested that each speaker use one of the
microphones, identify himself or herself, and speak with
sufficient clarity and volume so that he or she can be
readily heard.
Unless other Committee members have some opening
remarks, I think we will move right into the agenda. The
Committee Member that's going to lead the discussion on the
next two agenda items, namely, the strategy for site
self-sufficiency and the Yucca Mountain review plan is
George Hornberger, so, George, it's your show.
MR. HORNBERGER: Thanks, John. Our first topic is
the strategy for site sufficiency, and, James, are you going
to do this? Is Bill your assistant, or is he going to
introduce you?
MR. FIRTH: I'm going to be running through the
presentation. I basically want to talk to you about our
broad outlines for our strategy for developing sufficiency
comments.
And right now, we're in the development stages, so
we'll be preparing this, and I will get to the schedule a
little bit later.
The purpose of the review is basically to evaluate
and comment on DOE's progress related to the sufficiency of
data analyses and the design for the license application.
What we're going to be doing is considering both
what DOE has at the they released the site recommendations
Considerations Report, as well as their plans for either
augmenting their documentation or collecting data.
Our focus is going to be on the foundations for
DOE's safety case, and their performance estimates. So
we're going to be focusing on the data and the conceptual
models.
So, does DOE have the data and understanding for
putting into a license application? The way we're
structuring the review is to fold this into all of our other
pre-licensing interactions with the Department of Energy.
So this review is going to be integrated with all
of the other work that we're doing along the way.
And one reason why specifically we're doing a
sufficiency review is that the Nuclear Waste Policy Act
requires the Department of Energy, in any recommendation to
the President of a site, to include preliminary comments of
the Commission considering the extent of at-depth site
characterization analyses, and the waste form proposal, and
to what extent they seem sufficient for inclusion in the
license application.
And, again, these are preliminary Commission
comments.
And looking at how this review fits into our
strategy for licensing, since we're doing a very broad
review of DOE's data and conceptual models for developing
our site sufficiency comments, that it's going to provide a
progress report of where DOE stands on data analyses and
plans, their understanding of the interactions between the
geology and the engineered systems, and the status of our
KTI resolution process.
And this is something that I will talk to a little
bit more later, and that we will be trying to come back to
the Committee after we meet with the Department of Energy in
April.
And one thing that I wanted to emphasize is, even
though we're doing a very broad review and we're looking at
the sufficiency of DOE's data analyses and design, this is
not a licensing review. We're not going to be able to go
into the same depth or scope as we would for a licensing
review.
So what we're trying to do is give a picture of
where things stand. Do we feel that there is enough there
in terms of understanding and data for developing a license
application.
And the way that we frame this is, when DOE comes
in with a license application, they are going to need to
build an adequate case to support a regulatory decision on
whether construction can be authorized.
So this means that the data design analyses,
quality assurance, all of these together have to build a
sufficient case for the Commission to grant the construction
authorization.
We realize that there is going to be an
opportunity for DOE to provide additional information; they
can augment what they have submitted in the initial license
application, and if a construction authorization were given,
DOE would have an opportunity to collect data through
performance confirmation or actual conditions as they build
the repository.
And the Staff is going to be using a lot of
information to review the -- make the decision in terms of
data and design and analyses. What's going to focus this is
that DOE is developing a Site Recommendation Considerations
Report that they are going to publicly release.
At this time, they're going to ask for NRC to
develop its preliminary comments on the adequacy of at-depth
analyses and the waste form proposal.
In addition, DOE is going to have a repository
safety strategy, technical basis documents such as their
analysis model reports and process model reports. They will
have a total system performance assessment supporting the
site recommendation.
They are going to have QA audits that are
underway. They're going to have a total system performance
assessment and methods and assumptions report.
We're going to use all of this technical details
that are the building blocks of DOE's Considerations Report,
as well as what is in the Considerations Report itself.
So we're going to be using a lot of different
information in terms of developing our assessment.
What the Considerations Report will do is, it will
provide a description of the proposed repository, including
the preliminary engineering specifications, the description
of the waste form proposal, and packaging. And this is also
going to include an explanation of the interactions and the
relationship between the engineered system and the geology.
There is going to be a discussion of data obtained
during site characterization, as well a discussion of the
analyses related to repository performance.
Our review objectives are basically to provide
preliminary comments on where the data analyses appear to be
sufficient or insufficient, and if any areas that additional
data or analyses may be needed, what is that information?
When would that be needed? Are the conceptual
models supported? Because if the conceptual models are not
supported, that's indicating that either another approach
may be needed for DOE to take, or additional data may be
needed.
And what is the status of DOE's QA efforts?
And since we're trying to develop preliminary
comments on the sufficiency of the site characterization
analyses and the waste form proposal, we need a yardstick by
which to measure that. And the 10 CFR Part 63 sets up a
risk-informed, performance-based framework, and it
identifies the information that needs to be included in the
license application, and it establishes an overall
performance objective.
Then we start implementing the regulations with
the Yucca Mountain Review Plan, which continues the
implementation of the risk-informed performance-based
framework.
And there the amount of information and support
that we would be looking for in particular areas will
reflect the degree of conservatism that DOE is using in
certain areas, their treatment of uncertainty, the
importance to the licensing case, as well as the risk
contribution.
So we're able to apply a graded scale in terms of
areas where we want to focus and make sure that we have
greater assurance in DOE's case that they have laid out.
Again, I want to get into the scope of what we're
going to be reviewing. Again, we're going to be focusing on
the building blocks of DOE's assessment of repository
performance, so we're going to evaluate both the at-depth
site characterization and the engineering design.
We're going to be taking a very broad view of the
data that would apply and the analyses that would apply for
site characterization. So we'll look at at-depth data,
analog data, laboratory data, expert elicitation.
So all of those pieces, we're going to be looking
at in terms of is there enough information to support DOE's
safety case?
Then we'll look at the related analyses that DOE
has assembled, as well as the conceptual models and plans
for refinement.
And although we are going to be looking at DOE's
screening analyses, throughout the repository system of the
engineering, the geology and the interactions, what we've
decided is that we would focus on the interactions between
the engineering and the geology.
And this gets back to one of the requirements that
DOE has under the Nuclear Waste Policy Act, and given the
evolving nature of DOE's design, we felt that by paying a
little bit of additional attention to the interactions, it
will give us a good feeling about whether DOE has
sufficiently understood its current design and how that
might interact with the geology.
So this is an area that we're going to be focusing
a little more on in terms of the analyses of features,
events, and processes.
In terms of the performance assessment, we're
going to be using the performance assessment analyses to
look at the risk-informed performance-based context for
evaluating DOE's data and analyses.
So we're using PA as a way of focusing our
analyses of the building blocks, the data and the analyses
that will be going into DOE's assessment, and we'll be
looking at the relative importance of those building blocks
to DOE's overall assessment and their possible eventual
licensing case.
We will also evaluate DOE's conceptual models that
they use to describe repository performance, and here, we're
going to be looking at does the data support the models that
they're using, and how are they treating uncertainty in the
models and input?
And one reason we're looking at the treatment of
uncertainty in the models and input is, how they treat that
needs to be considered and evaluating whether they have
enough data and analyses.
If they are conservative, and take very
conservative bounds, then we may need as much information,
but as DOE takes additional credit for certain areas and
those are important to DOE's licensing case, we would want
to focus in on those areas to see whether that is supported.
In terms of quality assurance, we're going to
assess DOE's progress towards qualifying data, models, and
codes. We'll evaluate DOE's capability of qualifying those
things that they will rely in the license application.
The way we're reviewing this is, we're going to
have to be looking at DOE's schedule for license
application. So we'll look at h how much DOE has qualified
at the point that they release the Site Recommendation
Considerations Report, as well as their plans for continuing
to increase the amount of qualifications.
DOE has also indicated that they would provide an
assessment of what has been qualified, the effect of the
group of the information that has not been qualified in
terms of how significant that is, as well as what the
impacts that may have on their Site Recommendations
Considerations Report.
And as we review quality assurance, we may notice
through our technical reviews, that we may want to look at
little more closely at some of DOE's data and qualification.
But we would see that this is a limited effort,
and there will be certain thresholds. We're not going to
just go through and try and reconfirm DOE's qualification.
If there is something that arises as a question, then we may
go back and look at that.
I wanted to highlight a couple of things in terms
of what the review is not: And we're not planning on taking
any position on DOE's dose calculations, so we will be
looking at their performance assessment and the building
blocks in terms of the data, the conceptual models, but
we're not planning on taking a position in terms of whether
we agree or disagree with DOE's final dose calculation.
Also, we don't feel that it's NRC's role to be
evaluating DOE's compliance with 10 CFR 963, so we're not
going to be getting into that in terms of our review.
This review is being focused on the adequacy of
the site characterization, data analyses, and the design in
terms of does DOE have enough there to develop a license
application, and so we're not going into looking at all of
the other things that DOE is pulling into their site
recommendation process.
And I want to talk a little bit about our
schedule. Basically, since DOE is developing their
technology documents now, what we want to do is be using all
of our review activities that are underway now to have them
focus and lead to developing our site sufficiency comments.
So, what we are trying to do is develop a strategy which is
underway and what I am briefing you about today. We are
going to then develop guidance that will implement this
strategy and give the staff direction in terms of how we
would use the strategy. And our objective is to have the
guidance and the strategy be developed in parallel to the
extent we can, because we see great benefit in having the
guidance come early rather than right before DOE's ruling --
considerations report.
Then we are dependent on DOE's schedule. We will
then move from our preliminary activities into a review of
the site recommendation considerations report, which the
current schedule has being released in mid-November. Our
goal is to develop the staff comments on sufficiency by the
9th of April and we would transmit that in terms of a paper
to the Commission. DOE's current schedule would call for
NRC comments being provided to DOE by the 25th of May.
And one thing I will emphasize here is that there
is going to be an opportunity for ACNW and the staff to
interact throughout this process, and I will talk a little
bit more later in terms of how we see that happening, but
there is opportunities from now until we basically prepare
our paper and provide that to the Commission to really have
substantive input.
If the committee is going to want to have input in
terms of providing input to the Commission, given that the
comments have to go to DOE at the end of May, you may want
to look at having your comments to the Commission by the end
of April to allow that to be considered before we send
anything to DOE.
And, again, as I said earlier, we are basically
doing the pre-licensing activities now. Our plan is to
provide DOE early review -- feedback on the reviews that we
conduct now. And what this will do is it will establish the
basis for, where do we feel that the information appears to
be sufficient? Where do feel there is additional that is
needed? And this would allow DOE to have some opportunity
to consider that in developing their future plans for
collecting additional information, augmenting their analyses
and so forth.
We are observing their QA audits, and what this
does is it gives us a view in terms of how the information
is developing, and also on their quality assurance efforts.
Also, we are trying to make sure that all of the issues
within the KTIs are resolved at the time that DOE comes in
with their license application.
So what we have underway and that we hope to brief
the committee on in the summer of this year is our KTI issue
resolution process. And this is a process where we are
trying to work with DOE to identify what information is
needed and to resolve the open items that we have before DOE
submits a license application. And we have plans to meet
with DOE in late April to discuss this process.
And the interactions that we are proposing to have
with ACNW is today's briefing on the strategy. And from the
strategy, we are going to be developing the details by the
end of June. So there is opportunity for you to provide us
input in terms of the scope and how we are structuring the
review. Then we will develop the staff guidance. We would
hope to come back to the committee after the guidance is
completed. And given its current schedule of being
completed in the end of September, we would look at coming
back to the committee in October, and this would allow us to
brief you on the guidance before DOE releases the
considerations report. So this will help us so that we are
not focusing on briefing you at the same time we are trying
to pull everything together in terms of our site sufficiency
comments.
Then we are looking at having a similar construct
to what we used with the viability assessment, that as time
goes on we would allow one-on-one interactions with the
committee members on areas within their technical expertise.
So this would allow you to stay well informed in terms of
how our review is progressing, as well as to provide early
input to the staff. And we have this starting, in essence,
in November and ending in April. If you want to meet
earlier to talk about our preliminary interactions and
review efforts, we can do that as well.
Then we would look at briefing you in mid-April,
which, again, if you want to have input to the Commission,
you would want to have your comments submitted to the
Commission at the end of April, so you may want to look at
that timing.
We are also looking at trying to look at how we
are going to involve stakeholders in this review. What we
are planning on doing as part of our KTI issue resolution
process is to hold public interactions in Nevada. And so we
will meet with DOE on various topics and we will try and
have those out in Nevada.
Then we are looking at holding at least one public
meeting that will address our rule and approach to the
sufficiency review and, currently, we are looking at
attaching that to another previously planned public meeting
out in Nevada in the summer of this year.
Then in terms of how we are going to document our
results, we are going to develop preliminary statements on
the sufficiency of DOE data and analyses for license
application. Again, we are going to be looking at the data
that is in hand, as well as DOE's plans. So, we are going
to be considering how DOE is going to go from where they are
at the time they release the considerations report, as well
as to when they will submit the license application.
And while there is going to be -- while we want to
provide a balanced view of where DOE stands in terms of
where things are sufficient, or where things may
additionally be needed, we are going to provide less
documentation on the areas where we feel that there appears
to be enough. So, even though we want to give a balanced
view of where DOE stands, we are going to be spending more
time documenting where we feel additional information is
needed, because we feel that that is our burden, that we
have to give some indication of why we feel additional
information is needed. Where there appears to be sufficient
information, we are going to acknowledge that, but we are
not going to spend the effort documenting why we feel that
there is enough.
We are going to be documenting the details in
terms of where things are enough in the IRSRs, so we are
still going to be working on establishing the basis and
preparing for an eventual review of the license application
by looking at how DOE's case is progressing, as well as
documenting where we feel additional information is needed
or where things appear to be adequate.
And we will comment on any significant open
issues. And to be a significant open issue, basically, the
benchmark is that those would be open items that would
prevent the license application from being docketed
potentially. So, we will comment on those and raise those
at the time we do our sufficiency comments.
And, again, to summarize the things that I have
gone through, this is not going to be a licensing review.
We are not going to have the time or resources to go into
the same level of depth or to cover everything that we would
in a licensing review. We are going to be focusing on DOE's
data analyses. Okay. So what we are going to do is be
focusing on DOE's data, design analyses. We are going to be
evaluating sufficiency in the context of our
performance-based, risk-informed approach to licensing. The
review of models is going to be primarily limited to the
upstream pieces of it. We are not going to be emphasizing
how DOE is going to get to their final dose calculation, but
we want to make sure that they have enough data and
information to get their.
The review is going to be fully integrated into
our licensing strategy and KTI issue resolution process.
And since this is going to be a broad look at DOE's program,
it is going to be a progress report on their progress
towards preparing a license application. And that concludes
my presentation, and we will welcome any questions that you
may have.
DR. HORNBERGER: Thanks very much, James.
It strikes me, from, well, a fairly casual
observation, and I think it did come through to me fairly
strongly, even in your presentation, that the sufficiency
review in some ways could be considered a natural extension
of the issue resolution status process. Is that a fair
generalization?
MR. FIRTH: Yeah. What we are doing is we are
having it take place, basically, within the issue resolution
process.
DR. HORNBERGER: Right.
MR. FIRTH: Because the process is going to
identify what information is needed. What do we feel about
the case that DOE is assembling? And that is going to
naturally lead into our sufficiency comments. It will be an
extension because we will be able to look a little bit more
broadly at the interactions with our sufficiency review.
But, in essence, they are very well integrated, and the
sufficiency review is just one point in time where we are
going to assemble all of the information in one place in
terms of where we feel things are sufficient or
insufficient.
DR. HORNBERGER: So, in one sense, it is a
synthesis, which, of course, the individual IRSRs don't
necessarily see the integration. So I understand that.
Now, the other side of it, the question that I would have
is, is it also, do you also envision then providing, say,
more detail? That is, if one looks at the issue revolution
status reports and you see, well, all right, what needs to
be delivered, and you have statements in there, well, the
data and the conceptual models need to be sufficient. Are
you now going to go in and, say, pick out examples where
your conceptual model for flow and fractured tuff is
insufficient?
MR. FIRTH: Yes, we would want to get into that
level of detail in terms of identify those areas where we
feel something more is needed. And it could be because the
documentation process has not fully assembled the case that
DOE will have at the time of licensing. It some cases it
may be that we feel that there is not enough basis to
support a conceptual model. Also, if there is limited data,
DOE would want to look at alternate conceptual models to
look at the range of performance. So, we will be looking
at, and trying to say in our sufficiency comments, these are
the areas where we feel that something more is needed, and
try and characterize that in terms of whether it is going to
be data that is needed, a new approach, or what-have-you.
DR. HORNBERGER: Do you have any sense, James, on
when -- you outlined a process where you are going to focus
in on some key points, rather than -- you are going to come
to some decision relatively early on that there are some
areas that are relatively sufficient, and you are going to
pay less attention. Do you have a sense as to the timing of
identifying the issues that you really want to hammer on?
MR. FIRTH: Okay. What we are doing now is
looking at DOE's repository safety strategy, the TSPA-SR
methods and assumptions report. We are going to be meeting
with DOE in early June about their TSPA-SR. So, at that
point in time, we will probably really start to get a good
look at the information that DOE has in terms of their
numerical analyses in terms of what appears to be more
important for their case. So we will be using all of those
pieces of information, but it probably won't be until early
June that we will start really getting a full picture.
DR. HORNBERGER: John.
DR. GARRICK: James, I appreciate that a
sufficiency review is not a licensing review. The one thing
I am struggling a little bit with is, in the context of
taking a risk-informed approach, a performance-based
approach, it is very difficult for me to imagine how you can
reach closure on the sufficiency of data, analysis and plans
without doing some of the things you say you are not going
to do, particularly with regard to decomposing the bottom
line results, which would be the dose calculations, into the
contributors, such that one could see more clearly where the
uncertainties are, where they are coming from, and,
therefore, reach a more supportive conclusion relative to
the sufficiency of data, design and analyses.
MR. FIRTH: Yeah. It is a difficult point in
terms of how far you go. The one thing that we do want to
stay away from is taking a position on DOE's dose
calculation itself. So, what we also want to do is make
sure that we can inform our review, so we are going to be
looking at DOE's TSPA, what the results are of that in terms
of the contribution of the different components. So we will
be doing some of the decomposition. But we are not going to
be pulling apart DOE's actual calculation to the extent that
we would do in a licensing review.
So, our hope is that we would be able to give our
preliminary comments at the point of the sufficiency review,
but it is only -- and it is starting the process of pulling
apart DOE's whole safety case. It is not going to be until
licensing that we will pull everything together for the
review to definitively say this is what really is needed and
this is actually adequate. So we are going to, at that
point in time we are going to have to have everything
assembled.
We are going to be trying to move towards that
during our sufficiency review, however, we are not going to,
again, pull apart everything and disassemble DOE's whole
performance assessment. But we will be looking at it. We
will be considering that, the results of that, in terms of
prioritizing those conceptual models in terms of what is
important.
DR. GARRICK: Yeah.
MR. FIRTH: And generally look at how much leeway
is there.
DR. GARRICK: Yeah. Now, the point simply being
that this will be an opportunity to test the genuineness, if
you wish, of the NRC really invoking or adopting a
risk-informed strategy.
The other thing I wanted to mention, just maybe an
extension of what George was just getting to, obviously,
through the technology exchange meetings that have been
taking place, and the issue resolution reports and studies,
there has been a lot of processing, if you wish, already of
information and what has been going on in these areas of
data, design and analysis. And all of these things are not
going to be equally important. So I suspect that the
attendees of these exchange meetings are pretty savvy right
now on what the most important issues are, for example,
something like QA, which seems to have been kind of a
chronic issue and problem for a long time.
Is there feedback being developed and will the
sufficiency review contribute to this, that allows the
licensee to get a running start on the issues somewhat in
terms of their importance, even though it is not a licensing
review and you don't have that application yet or
what-have-you?
MR. FIRTH: Yeah. Basically, as we are moving
through the interactions now, we can help DOE understand
areas we feel are important. And they also see things from
their own performance assessment in terms of what they feel
is important. But one thing that we wrestle with is, as DOE
changes what they want to emphasize in their safety case,
that also changes what is important. As the design gets
modified, that also may change what is important and to what
degree. So, we can give them information, and the
sufficiency review will be a concrete example of things that
we feel are most important, and for DOE to emphasize, to the
extent they feel that they need to, but it is also subject
to the changes that DOE makes in their own program.
DR. GARRICK: Thank you.
DR. HORNBERGER: Raymond.
DR. WYMER: I realize that you are well along on a
long and arduous path of Issue Resolution Status Reports and
discussions of Key Technical Issues, and there has been a
lot of input and lot of meetings held. So things have been
talked about and considered at great length But, usually,
where you get into trouble is about the things that nobody
thought of. Despite the fact that a lot of thinking has
gone on, it has been by a fairly circumscribed group of
people for the most part, who have particular points of
view, particular backgrounds and knowledge.
And now you are considering things in the
aggregate, sort of putting the whole package together. And
while it is fairly easy to comment on what is in front of
you, what is much harder is to come up with what is
important that is not in front of you, as I am sure you
know. And, so, what I am getting to ultimately here is, to
what extent in this review will you bring in outside people,
maybe even people from overseas, who are considering the
same kinds of problems, just to make sure that the whole
thing is really covered, now that you are getting down to
sort of the nitty-gritty of this pre-licensing activity?
MR. FIRTH: Yeah. In terms of developing the
strategy, we have not specifically identified a mechanism
for doing that. That is something we can consider as we
develop the strategy more fully and the guidance.
One thing that the staff is doing is evaluating
DOE's analyses of their -- the features, events and
processes that they wish to exclude from the performance
assessment, and there is your review, what is being screened
out. A question that you have to ask is, is the initial
list complete? And one thing that we did in the near-field
environment is to audit what DOE has done in terms of their
database of features, events and processes, identifying that
there are some things that DOE may want to also consider.
So, we are trying to put into our process, is that
initial list that DOE has complete or is it not?
DR. WYMER: Yeah. A particular area that has been
a concern to me is this business of coupled processes, as
you know, and that is a very complex area and requires a lot
of good people doing a lot of good hard thinking, I think,
in order to be sure that it is adequately covered.
MR. REAMER: In my profession, process is
important, and this going to be a public process. This is
going to be built on interactions with DOE that we hope will
stimulate feedback, not only from the potential applicant
here, but also from the state, and potentially anyone who is
interested in this project and wants to see it done safely.
So that would be one way that I would hope that we
would be enlightened and hear more from others. Hopefully
you'll be able to attend our technical exchange later this
month on issue resolution, and continue to make the
suggestions that you're making.
So in any event, in addition, we know that
parallel to our review -- and our review is focused on
sufficiency for a license application. We're not, under the
statute, asked to take a position on this site
recommendation, one way or another.
We're out of that. We have a potential licensing
proceeding in front of us, and objectivity that we need to
be sure that we bring to that.
So, the process doesn't ask us to take a position
on the site recommendation, and we're not going to do that.
But parallel, DOE will be running a public process as well
on their site recommendation document. And I think it
behooves us to be aware of what is coming out in that in the
way of technical feedback, as well, and make sure that gets
factored into our thinking.
DR. WYMER: Thank you. Milt?
MR. LEVENSON: I have a question that may just be
for clarification. On your slide where you list the
technical basis for DOE's recommendation, one of the bullets
is description of the waste form or packaging proposal.
Is it really intended to be "or," or is it "and."
And the context of my question, of course, is that the waste
form is not singular. There are at least four major
different waste forms. I wondered what part of that
diversity gets into this picture?
MR. FIRTH: I mean, this gets back to a little bit
in terms of what the Nuclear Waste Policy Act puts in place.
So that was done well before the point where we are now in
terms of what would go into a potential Yucca Mountain
repository, so it's focused on that.
We would expect that DOE would include both the
waste form in terms of the different waste forms, as well as
the packaging that they would use. This is the minimum, but
we expect that there will be actually more than that.
MR. LEVENSON: You will be looking for multiple
waste forms?
DR. HORNBERGER: Yes, I mean, that would be what
would be needed in the license application.
James, I have just one other question: In one of
your slides, you had mentioned the treatment of uncertainty,
and that you're going to be looking at that. And you
mentioned the degree of conservatism in the same breath.
And I'm just curious, because one of the things
that has always concerned me is having a treatment of
uncertainty doesn't lead me inexorably to say, well, if
you're uncertain, then you have to be exceedingly
conservative in assumptions and models.
Did you mean that?
MR. FIRTH: Basically what we would consider is
DOE's model and they way they've implemented that,
conservative? If it's very conservative, even if there's a
large number of uncertainties, we would not need to spend as
much attention on that.
If it's very uncertain, then DOE isn't obligated
to take the most conservative route, but we would look at
how they have treated the uncertainty.
To the extent that DOE appears to be taking a
non-conservative approach or to be taking additional credit
for beneficial aspects, we would want to sharpen our pencil
in terms of the review to see is there a basis for what DOE
is doing, or are they being overly optimistic.
So while DOE is not driven to the most
conservative, if they do take a bounding approach, even if
it's uncertain, we won't have to pay as much attention to
it, because it will be easier to say that it is
conservative, so we're not going to have to spend a lot of
time evaluating all of the details in terms of the process.
If that's only going to make it show how
conservative it is, we don't need that to evaluate the
conservatism that DOE is taking in terms of whether -- the
approach that DOE has taken is conservative or not.
DR. HORNBERGER: So, a hypothetical portion of a
question and then a followup: In my best of all possible
worlds, DOE wouldn't use any bounding or conservative,
overly conservative or conservative assumptions; they would
simply take the uncertainty as they understand it into
account, and do their analyses and present the results.
If my definition of the best of all possible
worlds came to be, would the NRC have a problem with that?
MR. FIRTH: What we would do is, we would focus on
areas that are most important to DOE's licensing case. So
it's -- and evaluate, does DOE have enough basis in terms of
is their description of the uncertainty appropriate?
There is the treatment of the uncertainty, but
there is also getting down to the effective annual dose
calculation in terms of would that be significantly changed
by how DOE has treated uncertainty or developed their
approach?
That's the main issue in terms of looking at
compliance with our regulations, so is it resulting in a
significant change that we would need to evaluate the
regulatory compliance?
DR. HORNBERGER: Good. Thank you very much, James
and Bill.
MR. FIRTH: Thank you.
DR. GARRICK: We're going to hear from NEI.
DR. HORNBERGER: Okay, Steve from NEI has some
comments. Is Steve here?
DR. GARRICK: Yes.
MR. KRAFT: Thank you. I'm going to use the low
tech apparatus here. I have some additional copies.
With all the preliminaries out of the way, thanks
very much for the opportunity to speak to you all today. I
am fascinated and interested in and pleased to hear what Mr.
Firth had to say about the NRC's views on sufficiency.
I have some commentary on that. As we moved
through it, I was jotting down in my notes here, that at the
appropriate point I'll make some comments.
The Nuclear Energy Institute, of course, takes
great interest in this project. As you know, we've appeared
twice before this group in recent memory, and have found
that they have been very useful exchanges. We learn a great
deal from watching this body deliberate, and ourselves, and
we greatly appreciate your interest in this topic and the
work that you are doing.
I think that the country has gone to a very
interesting location in this program. This program has
become extraordinarily public in its debates, and that has
only helped. And NRC has had a lot to do that.
We appreciate that. There is an early warning
system in place now in terms of what the issue are, how they
are going to affect the program, how they are going to
affect the nation, and I think that's all very, very
positive, and NRC and this body are to be complimented for
being a part of that and bringing that about.
Why is sufficiency important? Why is such a
decision important?
A decision on Yucca Mountain is urgent.
Competition is reshaping the nuclear industry. Nuclear has
proven that it can compete, and environmental energy supply
stakes are high.
Five years ago, internal to our offices, and even
to some extent, in public, we were making several statements
about why we needed to solve the, quote, nuclear waste
problem, unquote.
Nuclear power plants would start shutting down for
lack of storage space. We couldn't pursue the economic life
of our companies, we could not buy and sell nuclear plants,
we could not seek license extension.
Well, maybe we've solved the problem, because all
those things are happening. The only thing on the list that
hasn't happened is the purchase of a new plant, and I
suspect that will happen overseas in some few years.
So what has happened in the environment? What has
happened in the broader environment?
It's not that we've solved the problem. I think
what has happened is that the good news is that our industry
has done an extraordinarily good job of managing its nuclear
waste, high level and low level.
The bad news is that we have done an
extraordinarily good job of managing our nuclear waste. Our
country is at its best in times of crisis, and we have
proven that over the centuries, time and again.
There is no nuclear waste crisis, but there is a
need to move forward with solving the problem. In our
discussions with the individuals on Wall Street, with the
individuals in the regulatory community, with the
individuals in our member companies who are pursuing what
appears to be a renaissance in the use of nuclear energy in
this country, and ask them about, well, how do you make
these decisions in the face of the nuclear waste "problem"?
The answer is not, oh, forget it; it doesn't
matter; the answer is, there is an expectation and a
confidence that the nation will make the right decisions and
move forward and solve the problem. The problem still
exists, but it's not preventing us from going forward in
ways that we want to go forward.
It will eventually work against the nation and the
use of nuclear energy. That is a very, very important
point.
As scientists and policymakers debate whether or
not we ought to have an allowable dose in the vicinity of
Yucca Mountain, plus or minus some percentage of background,
that argument could delay this program and result in energy
choices that have far-ranging, far more profound health
impacts than the decision being made at Yucca Mountain.
That's an important point that policymakers need
to keep in mind. And when we talk about sufficiency, we
talk about suitability, what we are talking about is how
policymakers will deal with this issue, and what information
and advice they need in order to do so.
That is why NRC's view, and, by extension, the
views of this body, is so very important. NRC is a source
of objective expertise.
And what does NRC think of DOE plans is an
extremely important question. I was pleased to see that Mr.
Firth focused on the issue of their plans.
So, what is science telling us at this point? At
face value, the case for going forward is very compelling.
Potential radiologic consequences are projected to
be so low that they're almost hard to find. This is
documented in the EIS.
Part of the problem with the EIS and part of the
problem with the NRC letter on the EIS, as distinct from the
viability assessment and NRC's review of the viability
assessment, is that the EIS materials were darn near
impenetrable.
The conclusion I just read to you is buried in
Appendix Q on page 435 or something like that.
Decisionmakers, policymakers, can't deal with information in
that way. They can deal with information in ways that bring
to their attention, the important factors and put them into
perspective.
NRC, DOE, and the industry analysis through EPRI,
all confirm that that is, in fact, the case. Even the EPA,
while they didn't talk about radiological consequences, they
did give the EIS a very high rating, and I think that
whatever infirmities different agencies found in that EIS,
are easily correctable as they prepare the final.
Uncertainties will remain. They always will, and
the issue here is not elimination of uncertainty, but
understanding of uncertainty. What do decisionmakers need
to have in their hands in order to react to a decision, make
a decision?
The first thing they need to know is, they need to
know the nature of the decision. Think about what decisions
are. A personal story, and I have used this before, so
forgive me if you've heard it. I'm just a proud parent.
If the decision that my wife and I faced two years
ago to allow our son to participate in Maryland Youth Hockey
was based solely on NHL head injury rates, he'd never lace
up the skates. But that's not how you make the decision,
and that's not how you make the decision in your personal
lives.
And I don't mean to say that making a decision
such as Yucca Mountain is a matter equivalent to whether my
son plays ice hockey, it is an example of how decisions are
made.
Decisions, by their very nature, are holistic. If
the entire decision on Yucca Mountain was the result of a
model, we'd eliminate the policymakers, run the models, and
turn the switch one way or the other.
Decisions, by their very nature, bring in every
bit of evidence that the decisionmaker wants to bring into
it, whether that evidence is presented to the decisionmaker
in some formal way, whether it's a gut feel the
decisionmaker has, whatever it is; that's how decisions are
made.
But in order to do that, the decisionmakers must
have concisely summarized and clearly communicated science.
That is largely the role of the DOE; it is the role of NRC
in its reviews; it is the role of this body; it is the role
of the Board on Radioactive Waste Management, the Nuclear
Waste Technical Review Board, the industry, and anyone else
you can name who has a scientific credential, who might have
something -- the state, didn't want to leave them out. They
have very valid views -- and the counties -- on the merits
of the science.
They also have to have confidence that the
regulatory components are there to move forward, which is
part of what the sufficiency decision is all about.
And they need a sense of perspective. Think about
it; think who the decisionmakers are.
When we wrote the Nuclear Waste Policy Act in
1982, when we wrote it as a nation in Congress, I doubt very
much that Congress had in mind that the President, at the
time this decision went forward, was a geoscientist -- I
doubt it. At the time, he was an actor.
Now, he's a former governor. I don't know what it
is when this decision comes up. Maybe he'll be an oil man.
Maybe he will be a former member of the Senate.
There was never a determination as to who that was
going to be, so the information that comes forward has to
have a sense of perspective for decisionmakers and the
nation to understand, is this risk bigger than a bread box?
Is it smaller than a bread box? What is its significance?
Not whether it's 10-6 versus 10-7, whatever in the Vadose
Zone. You know all the details far better than I.
That kind of information doesn't help
decisionmakers make decisions.
Unfortunately, two bullets were left off of here.
Also what is needed is an understanding of the future
ongoing R&D, because there will be future ongoing R&D.
There will be a confirmatory R&D program following the
license application, following emplacement, if we ever get
there.
And what that will be -- and there also needs to
be an understanding of future technology development. We're
not going to turn off the scientists and the National Labs
and anyone else who has an idea about what to do in the
future about nuclear waste.
I think the decision makers can take great comfort
in knowing that there are all these things in place,
especially a future R&D program to answer the question of,
well, did we make a mistake? And secondly, that there is
the potential for future technological development that
might make it easier to deal with these materials in the
future.
It is important to keep in mind, getting to the
point of what this decision is, is that it is a four step
process going forward. There have been prior steps, but
going forward is a four step process. What is the site
recommendation?
The site recommendation is in fact from the
technical side a relatively limited decision. It is an
important decision. It will be a difficult decision. It is
going to be made at the highest policy levels. It is a
politically-charged decision. But on a technical basis it is
a relatively limited decision. It is not saying we are
going to put nuclear waste in Yucca Mountain forever and a
day and walk away from it. It is merely saying it is okay
to go forward and ask the NRC to review the license
application. That's all this decision is.
There are protections in the process beyond that
point. There are protections in the site recommendation
process, as you know, but there are also protections in the
NRC process beyond that point that will allow anyone who
still thinks the project ought not go forward to challenge
it, be a part of the NRC process, et cetera, et cetera.
Of course the political process is always
available, even though Congress will have acted to allow it
to go forward, and I have just laid out here the four
steps -- the license to construct and the license to
operate. The most important decision will be the license to
close. The earliest that will happen is 2060. It may
happen as late as 2300, so what we are seeing here is a
step-wide process, probably with greater requirements for
additional certainty in the data going forward which will be
developed by the confirmatory R&D program beyond what is
known, now what is known at license application.
What is needed to be known now -- is the site
suitable and is the site sufficient but sufficient to begin
the licensing process. I won't belabor suitability because
it is not what NRC is doing but it is important for the
record to understand. This is the sole language out of 180
some odd pages what suitability is, and from that we have to
discern what is an appropriate suitability recommendation
that DOE has to make.
Remember though what I said about decisions.
Suitability, while a major component and a major input to
the Presidential decision, is not the sole determiner. It
is part of it. It is a major part. It is not the sole
determiner. I can only guess what the President will have
in mind when he sits there with his pen in his hand to
decide to approve or disapprove going forward on the
project, and only the President will be able to determine
what this decision is really all about, subject to review by
Congress and review by the state, et cetera.
It is a conclusion based on science, not merely a
compilation of scientific information. It must be clearly
communicated, as I just mentioned, and the uncertainty --
this is the key point.
It is an understanding of uncertainty, not an
elimination of uncertainty. How important are the different
uncertain matters? I particularly appreciated the questions
posed to Mr. Firth on how getting into the numbers a little
more deeply at sufficiency to understand where the
uncertainties are, but I also completely agree with his
answer, that we will do that to the extent we can, as I
believe he said, but we will not comment on the final
answer, and that is the right thing, because commenting on
the final answer is inappropriate to a stage where there was
no requirement to produce a final answer. If there was, we
would be licensing, not making a site recommendation.
Of course, suitability has to be documented and it
is a comparison against the criteria that DOE currently has
out in draft and will probably publish before the SRCR is
issued.
Well, NRC is in better shape than DOE. They have
got about twice as many words to tell them what sufficiency
is and they are right when they say it is preliminary in
nature and that it is in fact something that they, too, have
to discern as to what this decision is. It is by its very
nature, the fact that it says preliminary comments
concerning the extent to which at depth site
characterization analysis and the wasteform seem to be
sufficient for inclusion in any application to be submitted.
Those are all future words, what happens in the
future, so I think to a large extent NRC has got it right,
and that is a forward-looking decision about DOE's ability
to file a license application, not the licencability of the
project. If it was about licensability of the project, the
law would have said give us a license application, and it
does not say that. It is an interim step. It is, as I said
before, a much less technically challenging decision.
But let's talk for a minute about some of the
things, one or two of the things that I heard Mr. Firth say
that I think are somewhat confusing to me, although I
suspect their heart's in the right place. He indicated that
it is not a licensing review because we don't have the time
and resources. Well, excuse me, if it was a licensing
review and you are not doing it because you haven't time and
resources, that is an unacceptably weak answer.
You are not doing a license review, NRC, because
it is not a licensing review. It is in fact a sufficiency
review and it is not a sufficiency review in the nature of a
Part 50 sufficiency review for a license application. It is
a completely different animal, okay? Frankly, it is wrong
to say that it is not part of site recommendation. Yes, I
agree with what Mr. Reamer said when he says they are not
commenting on the site recommendation, they are not
commenting on suitability, but it is an important part of
the site recommendation decision that the President has to
make, which is in fact is DOE going to, if they follow the
plans they have, going to get from wherever they are at the
point of the site recommendation to filing an adequate
license application. Are those plans in place?
I think that that confusion as to how the
sufficiency determination fits into the greater scheme of
things as part of informing a policy determination needs to
be kept in mind, because it is very different than any other
thing that NRC is used to doing, and they need to be very
careful that they don't err on the wrong side of that line.
They ought to in their review close out questions
if they can that they raised in the VA review, defining the
path forward if there are any new or remaining questions and
determine whether they think they will get there by LA and
of course it is an opportunity for them to explain how the
licensing process will address the uncertainties going
forward. It is separate and distinct from suitability,
which is a point that I just made, and it needs to be
balanced.
Let's talk about balance. When we were
contemplating NRC making a statement on the viability
assessment, which was not something they were required to do
by law but something that they knew as a responsible agency
they had to do, we had several public meetings with NRC
about that. Bill will remember that. The point we made,
which I think they took to heart, was that if you go back
over the decades that you have been working with DOE think
about how many things you have asked DOE to do that they
have done versus the number of things they have said to you
we are not going to do those, and I guarantee you that the
ones they are not going to do or haven't done are in the
vast minority, but as a regulatory agency and as people in
general there is a tendency to focus on the negative rather
than the positive.
We suggested to them that what they do is they
point out all the things that the DOE still has to do to get
it right, but don't be afraid to point out all the things
that DOE has done that they are getting right.
They took that advice to hear and the VA letter
was extremely helpful to policymakers to allowing the
project to then go forward beyond that point.
The EIS letter, however, was a disappointment in
that regard. I heard the EIS presentation to this body and
I was, frankly, shocked to hear the kind of negativity that
was presented, and by the time that letter came out it still
focused on here are the five or six things we don't like,
and left to the reader to figure out what they did like.
If you read that letter between the lines, it is
pretty clear they liked a lot about the EIS. One simple
example -- in transportation what they are critical of was
not radiological impact. Well, that must mean that they
thought the radiological impact analysis was just fine.
They never said that, but that is a conclusion you can draw.
I would just encourage NRC to take a balanced
picture, a balanced view to what they are going to say about
the SR, and of course it is a legally flexible approach, and
I think that Mr. Reamer's profession is very adequate at
doing things that are flexible and making sure that all the
bases are touched in ways that the agency itself is not
hamstrung in future licensing decisions because it said "x"
or "y" at an earlier stage.
In conclusion, it is our view that this is an
extremely important time. Sufficiency is one of the
elements that will lay the groundwork for a decision on
going forward on Yucca Mountain. It was important to inform
that decision. NRC should not shy away from the role they
are playing in informing that site recommendation decision.
We need to keep this decisionmaking process on
track. We need to encourage the agencies to keep their
meeting schedules and keep their interactions going and
things like that, and then, most importantly, we need to
prepare to implement the decision, whatever that decision
is. Remember, the law allows that decision to go both ways,
as it should.
Now it has got one sentence in it about what
happens if the President makes a decision saying not to go
forward, but we need to prepare to implement that decision
however it comes out.
Those are my prepared remarks. Thank you very
much.
MR. HORNBERGER: Thank you, Steve.
Are there questions from the committee? No?
DR. GARRICK: I would just like to make one
comment. One of the criticisms levelled at industry in the
waste field is that industry has not exhibited the same
level of interest in waste, if you wish, as they have in
doing what they have to do to keep their licenses in place
and operate their facilities.
Part of the reason that sometimes I hear given is
that when the Nuclear Waste Policy Act of 1982 came about
the industry relaxed a little bit because that was supposed
to be the solution -- for DOE to take the waste off their
hands and dispose of it.
I am sort of one of those that has been a little
critical at the absence of industry on this issue, and the
absence of visibility of the industry on this issue. Do you
see that changing? Do you see -- for example, we seldom
hear from a utility executive on their concerns and
interests and activities associated with trying to make a
contribution to solving the waste problem. We do see NEI.
I have talked to a few utility executives and have
sensed that they probably made a mistake, at least from some
of them, of relaxing, if you wish, if I can call it that, on
the waste issue in the mid-'80s when they should have
continued to be very visible, very active, and proactive.
What is your thought on that?
MR. KRAFT: Well, Dr. Garrick, in the spirit of
sufficiency as a forward-looking decision, let's look
forward. Any views I would express on the past are purely
my own and I would not want them to be taken as NEI's views
or the industry's views of the past.
I will say this on behalf of my industry, that
when we deal with industry problems that affect everybody
there is a tendency in the industry to look to the central
organizations to deal with it in a more concise, consistent
and to some extent cost-effective fashion, and we see that
in Part 50 as well.
I see the -- the only answer I can give you is
that you have had your personal conversations with the
executives of my industry and I have sat in their meetings
and I can only tell you if you sat in their meetings I think
you might have a somewhat different impression of their
interest in the topic and their interest in helping DOE be
successful, and I do see that there is more being put on the
table in that regard, if I can use that term.
We have a group now in NEI of senior executives
who form the committee who are working with DOE. We have a
group of senior management people who are meeting with DOE
every now and again about how they are -- DOE and TRW --
about how they are managing projects. The most recent of
those meetings occurred about a month ago. We are going to
be taking certain individuals within the agency and exposing
them to practices at the best nuclear plants to help them
understand how they operate.
We work tirelessly in Congress to obtain the
needed appropriations -- it is that time of year so we are
working on that now for this agency as well as for DOE.
I think it is instructive that there are two
groups in the industry that have pursued their own interim
storage projects, one in Utah and one in Wyoming, the Utah
one being several years ahead of the Wyoming one, both of
which are on a path for success, and I think those are
really very quiet activities as many of the industry
activities are in trying to help out DOE.
If what you are hinting at, and let's just put it
out in the open, is that we have focused primarily on the
legislation that might or may not become enacted as opposed
to helping the program, I would say that the legislative
activities by their very nature are simply more visible.
The other activities are not.
I suspect that what happens in the industry in
this waste program is that it tends to be a rather esoteric
exercise not normally within the day to day activities of
the typical utility senior nuclear officer and they would
defer to competent and expert staff in the central
organizations to deal with it.
We certainly hear a lot from them about helping
DOE be successful, so maybe it is just more of a visibility
question, Dr. Garrick, and I think we will have to correct
that.
DR. GARRICK: Okay, thank you.
MR. HORNBERGER: Thanks very much, Steve.
MR. KRAFT: Thank you.
MR. HORNBERGER: Amy Shollenberger has asked for
time to comment and this would be a good time. Amy?
MS. SHOLLENBERGER: Thanks, Dr. Hornberger, for
allowing me to speak. My name is Amy Shollenberger and I am
here representing Public Citizen's Critical Mass Energy
Group. In the spirit of Mr. Kraft's suggestion to start
with the positive, I will say there is one thing that I
agree with him on and that is that there is no crisis that
needs to push Yucca Mountain project forward. I was really
happy to hear him say that.
It seems that a lot of the policy-makers maybe
need to be told that a little more loudly. And I think it
is especially true in light of yesterday's presentation
showing the new planned ISFSIs around the country to store
waste on-site, so I just wanted to start with that comment.
DR. GARRICK: I think there is a very important
distinction. I think what he said, there is no nuclear
waste crisis. I don't think he said there was no crisis
relative to the need for Yucca Mountain.
MS. SHOLLENBERGER: Well, a lot of lawmakers on
the Senate and House floors have claimed that there is a
nuclear waste crisis and that is why we need Yucca Mountain,
and I would just like to point out that NEI is saying that
that is not the case.
On to my other comments, first of all, I think it
is just really interesting that it is the industry actually
here telling you all what the decision-makers need to make
their decisions. And I think it is important to note that
the decision-makers, at least those in Congress, supposedly
represent their constituents and not necessarily only the
industry, although it is very clear that the industry makes
it a lot easier for them to get elected.
I think that I would like to add that what
decision-makers need to make decisions is they need to know
how the decisions affect their constituents and, also, the
taxpayers who are also their constituents. I think the
hockey-helmet thing was a good example, because it is true,
you need to have all the pieces of the puzzle to make the
good decision. You need to know how likely it is that your
child is going to crack his skull open before you decide
whether or not he can play hockey, and you don't only need
to be told that there is padding on the walls, so if he runs
into the wall, it is not going to crack his head open. You
have to have the whole picture.
Also, I would just like to mention on the record
that, in regards to suitability, there is a petition that
was sent to the DOE signed by over 200 groups asking them to
disqualify Yucca Mountain, and that petition was based on
the guidelines in 10 CFR 960, which right now there is a
proposal to change, as you all know. But both 10 CFR 960
and the Nuclear Waste Policy Act had -- well, the Nuclear
Waste Policy Act called for individual disqualifiers for
Yucca Mountain and those disqualifiers were listed in 960,
and the petition was based on those disqualifiers.
The petition was, of course, denied by DOE, but it
is important to know that part of the response was a
proposed rule change to eliminate disqualifiers in 963,
individual disqualifiers.
So I think it is just really important to keep
that sort of in the forefront as you move forward and look
at sufficiency and suitability, that there is a large group
of Americans who believe that the information that is being
put forward is not sufficient to make a decision and that
Yucca Mountain is not a suitable place to store radioactive
waste.
Also, I would just like to thank the NRC for its
EIS comments. I think that, for once, it showed that the
NRC was willing to stand up and say what was right instead
of sugar-coating the comments to make DOE feel good. And,
for one, was really glad to see those comments. I saw you
all struggling with how they were going to be worded, and I
would have liked to see them a little stronger. I heard
some things in the meetings that actually were more strong
than what you put in your letter, but I was really happy
with the letter, and I would just like to thank you for the
work that you did on that. Thanks.
DR. HORNBERGER: Thank you, Amy.
DR. GARRICK: I wanted to just comment on
something that Amy said that I think is very important, and
I will give my spin on it, and I will do it in the context
of Steven Kraft's exhibit that he showed us, what
decision-makers need in order to act.
I think the thing that is needed more than any
other single thing, and that supersedes everything on the
list, is the will of the people to want to solve the
problem. I think the reason nuclear waste lingers on
without a long-term solution is there is not a will out
there to solve it. There is no a sense of urgency.
Steven points out that maybe one of the reasons
for that is the good job that industry is doing in managing
the nuclear waste, and I think, in general, that is so,
particularly in the civilian side of the business. But I do
think that the underlying and overarching problem that
exists with respect to making a decision here is the lack of
a public will to do so, and that we just can't escape that.
If the public really wanted to do something, be it store it,
dispose of it, Yucca Mountain or what-have-you, clearly, it
could be done.
DR. HORNBERGER: Any other comments?
[No response.]
DR. HORNBERGER: Discussion? The last item on the
agenda here is to discuss elements of a possible ACNW report
on this topic. My own view is that that may be premature.
It may be premature to have such a discussion. If there is
no objection to that statement?
DR. GARRICK: Yeah. Unless the committee, some --
any of the members feel that there is an issue that has come
up that would warrant some remarks at this time. Ray.
DR. WYMER: I doubt personally there is anything
that warrants a report, probably some additional discussions
with the staff, but not a report.
DR. GARRICK: Yeah. I think some additional
discussion on this issue, that we got into a little bit of
conservatism versus uncertainty, and the clarification of
what all that means in the context of a risk-informed
approach. I think there clearly needs to be more discussion
about that, because I am still not convinced that the NRC
has their heart in a risk-informed approach, because a lot
of the process and a lot of the regulations continue to be
incompatible with a genuine risk-informed approach. But,
nevertheless, progress is being made. The steps are smaller
than some would like, but at least they are in the right
direction. But I do think some more discussion on that
would be very constructive.
DR. HORNBERGER: I am sure that surprised you,
Ray, that John raised that.
DR. WYMER: Right out of the blue.
DR. HORNBERGER: Other comments?
MR. LEVENSON: I suppose since we are making
expected comments, I need to make mine on conservatism, and
that is we have to recognize that large uncertainties are
not, in any case, automatically a reason for conservatism.
If the consequences of the uncertainty, even at the limits,
are all fully acceptable, you don't need to add more
conservatism. Uncertainty, by itself, is not necessarily a
detriment.
DR. GARRICK: That is exactly correct, and that --
I am impressed that you have moved along so --
[Laughter.]
DR. HORNBERGER: Okay. I think we are caught up
with this topic for now.
DR. GARRICK: Okay. I think we will take a break
now and come back in 15 minutes.
[Recess.]
DR. GARRICK: We would like to come to order now.
The next item on the agenda is the Yucca Mountain Review
Plan. George Hornberger will continue as the member leading
the discussion.
DR. HORNBERGER: Okay. As John said, we have the
YMRP on our agenda for attention this year, and Christiana
Lui is going to give us a briefing. Christiana.
MS. LUI: Thank you, Dr. Hornberger.
I guess I am not going to have my branch chief
here supporting me, I am flying solo today.
DR. GARRICK: He is here.
MS. LUI: Okay. I am Christiana Lui, I work for
Bill Reamer in the High Level Waste Branch in the Division
of Waste Management, and today we are giving you an update
of the Yucca Mountain Review Plan development effort.
We last briefed the committee on November 18th,
1999 and, basically, during that particularly briefing, we
gave you the approach that the staff is using to develop the
YMRP. And, also, during that particular briefing, we laid
out what are the major components of the YMRP. Here I am
just reiterating that framework that we have adopted for
YMRP.
On the next page you will see a chart, a schematic
for the framework of the major components in the YMRP. As
we go through this particular presentation, I am going to
come back to this diagram from time to time. Basically, the
YMRP is divided into introduction, where we lay out the
purpose and scope of the RP, of the review plan and the
review strategy.
And the next chapter is acceptance review. What
is where we will evaluate whether DOE has submitted a
complete license application, and the basis for that
particular chapter will be a comparison to paragraph 63.21,
that is where we lay out the content of license application.
And the next chapter is the general -- is the
review plan for general information, which is basically laid
out in 63.21(b), and there are five components to it,
general description, schedule for construction, receipt and
emplacement of waste, physical protection plan, material
control and accounting, and a brief description of the site
characterization work DOE has conducted.
And the real main focus of the review plan will be
the chapter on safety analysis report review. And in that
particular chapter, we have divided the review plan into
three major sections, preclosure safety evaluation,
postclosure safety evaluation, and evaluation of the
administrative and programmatic requirements.
Before we go any further, I would just like to
tell you where we are, the current status, and we will come
back to a schedule at the end of this presentation. The
staff is currently completing the Revision 0 postclosure
sections, and we intend to have this particular postclosure
portion of the review plan accompany the Draft Final Rule to
the Commission by the middle of April this year. So
approximately two weeks from now, it is going to accompany
the rule to the Commission.
And we are working on the preclosure sections and
all the other chapters of the review plan, and Draft
Revision 0 of the preclosure sections is coming to us from
the Center in the middle of April, and we will provide a
quick review, and the Revision 0 of the preclosure sections,
we believe, will be done by the end of May. And all the
other sections of the review plan, such as QA, such as
material control and accounting, the general information
portion, we are looking at completing the Draft Revision 0
by the end of April and have a Revision 0 by the middle of
June this year.
We intend to basically make the YMRP Revision 0,
all the sections, publicly available after management
approval. We are contemplating about transmitting the
Revision 0 to DOE as information copy, and we want to put
the Revision 0 on the web site. I will come back at the end
of the presentation and talk about the public comment and
the other revisions that we are planning to give you a more
complete picture. Right now I just wanted to let you know
where are in the process.
Just to reiterate, the principles that we have
adopted for development of the Yucca Mountain Review Plan,
most importantly that NRC is responsible for defending the
license decision and DOE is responsible to ensure the
adequacy of its license application and safety case. They
are the ones with all the resources to carry out all the
site characterization and experimental work, and they, in
their license application, need to provide a sufficient
safety case.
And 10 CFR Part 63 is a risk-informed,
performance-based rule, and we fully intend to make the
Yucca Mountain Review Plan a risk-informed and
performance-based review plan. And what we are doing right
now is using the total system approach and an integrated
approach to formulate the review plan.
I will talk about the postclosure part, but it
will be later, in great detail and give you a sense of how
we are carrying out this particular process. And at the
same time, we are incorporated all the experiences and
knowledge that we have accumulated during the prelicensing
consultation period and using a risk insight to help us
formulate this review plan.
The purpose of the review plan is to provide
guidance to the NRC staff, our methods for conducting and
documenting the license review. This is not the equivalent
of regulation. In other words, it does not have the force
of law. In the review plan we present at least one approach
for compliance demonstration, or, basically, how the staff
is going to conduct the license application review. And
other approaches are definitely acceptable if DOE can
demonstrate the appropriateness of the alternative
approaches.
And for each section of the review plan, for each
of the topics that we are evaluating, there will be five
subsections. Areas of review that will provide the scope,
basically, what is going to be reviewed in that particular
section of the review plan. We lay out the review methods
that tells how the staff is going to conduct the review.
And acceptance criteria, what the staff will find acceptable
and the acceptance criteria are based on the regulatory
requirements in the rule. We will present the general
conclusions and findings in the evaluation findings portion
to echo the areas of review that we have identified at the
beginning of each of the review sections. And we, of
course, will provide all the references that we have cited
in that particular section.
Now, I'm going to turn the attention to the
preclosure, which will be on the left-hand side of these
charts.
As I have mentioned before the Draft Revision 0 of
the preclosure sections will be coming from the center in
middle of April, and we fully intend to provide a quick
review and have a version out by the end of May.
Basically in the preclosure sections, we establish
a set of criteria and review methods based on whether the
preclosure performance objectives can be met. The
preclosure performance objectives are identified in Section
63.111 of the rule, and 63.12 lays out the technical
criteria for an acceptable preclosure safety analysis.
More importantly, we want to emphasize that DOE
has the flexibility in selecting design details and methods
for compliance demonstration. In the rule, we did not
prescribe any of the design criteria in the review plan, and
we will not be doing that either.
I note that that was one of the major concerns
when we first developed Part 63 that people were asking, if
you're going to have a risk-informed performance-based rule,
are you going to be dropping all the prescriptive detail in
the review plan?
We don't intend to do that; we want to provide DOE
with the flexibility in defending and constructing its own
safety case.
However, where appropriate, we fully intend to
rely on existing guidance documents, and we are in the
process of working with the Spent Fuel Project Office and
Fuel Cycle Facility folks to help us identify what are the
major components that we need to pay attention to in the
preclosure portion.
And the evaluation will include the adequacy of
site characterization, repository design, construction,
operation, monitoring, and closure.
And also in the preclosure performance objectives,
we have identified that DOE needs to provide a plan for
retrievability, and they also need to -- and the design also
needs to accommodate the implementation of a performance
confirmation program and in the preclosure part, we will be
looking for those pieces that DOE is required to address.
Now, I'm going to talk about the postclosure
portion of the review plan. And this is where we have the
most detail that I can discuss with the Committee today.
Like in the preclosure case, in the rule, we have
postclosure performance objectives established for
postclosure safety.
We are developing the acceptance criteria and
review methods, based on whether these post-closure
performance objectives can be met. And the evaluation will
include the adequacy of ODOE's work such as site
characterization, field testing, laboratory testing, and
natural analog investigation.
Multiple barrier analysis, that is also another
performance objective that we have laid out.
Demonstration of repository resilience to human
intrusion events and also -- performance confirmation
programs.
I now want to turn your attention to the
performance assessment portion of the review plan.
Basically let me go back to the diagram again. I'm looking
at these particular blocks where we have all these detailed
components supporting the evaluation of performance
assessment.
We have divided up the review of performance
assessment into four major pieces: System demonstration,
system description and demonstration of multiple barriers;
scenario analysis; model extraction, and lastly,
demonstration of the overall performance objectives.
There are many different ways we can sequence the
review. But the logic that we have come up with sequence
our review in these particular order is that DOE has
already, by the time of license application, DOE has already
completed all the iterations and the required analyses.
Therefore, right up front, we want them to tell us
what they are relying on, i.e., the barriers, in meeting the
postclosure performance objectives on individual protection
standards.
And that will help the staff to focus our review
in the subsequent portion of the performance assessment.
And in the scenario analysis portion, we want DOE to tell us
what they have included or excluded from the consideration,
and the probability of the scenarios.
Like Dr. Garrick's paper, in Dr. Garrick's paper,
the risk triples, has scenarios, probability, and
consequences. Here in this particular part, we want DOE to
identify what are the scenarios that they are considering
and what are the associated probabilities of those
scenarios?
And once we have a good handle on what DOE is
considering in their compliance demonstration calculation,
we will be going into detail, looking at the model
abstractions portion. That's where DOE would conduct its
consequence analysis.
And at the end, we want DOE to put together the
scenario, probability, and the associated consequences to
give us the risk estimate.
That's going to be evaluated in the last portion
of the performance assessment review.
Now let me go into more detail for each of those
four subsections. Multiple barriers: We have formulated
acceptance criteria and review methods that focus on whether
DOE has identified all the barriers that they are taking
credit for in the compliance demonstration.
Also, DOE is required to describe and quantify the
capabilities of the barriers, using the information coming
from the total system performance assessment. They can use
intermediate outputs, or they can use sensitivity analysis
results. Basically they need to quantitatively describe how
the barrier is going to contribute to the performance of the
repository over the compliance period.
And DOE needs to include technical basis to
support the assertion of all the barriers' capabilities.
During the public comment period, we did receive a
fair amount of comments on the clarity of the multiple
barriers requirement. And in the final rule, the staff is
doing -- is clarifying the requirements on the final rule in
this particular area.
Therefore, we will develop additional criteria and
review methods that are consistent with what is going to be
included in the final rule, understanding that the
Commission will have to make a decision on what options they
want to go with, and based on that particular decision, the
staff will carry the work in this area further.
The next piece is scenario analysis. In this
particular portion, we are focusing on a methodology for
inclusion or exclusion of features, events, and processes in
the compliance demonstration, the informational scenarios.
There are five steps that DOE needs to carry out:
The first part is DOE needs to identify a comprehensive list
of facts that are applicable to the Yucca Mountain site.
Understanding that the subsequent steps here --
DOE does not require to carry out any of these subsequent
steps, once they have identified the initial list of facts,
but there are thousands of facts that are applicable to the
Yucca Mountain site.
By doing the grouping and applying the screening
criteria, that will basically streamline the performance
assessment process and make the analysis more transparent at
the end.
However, if DOE likes not to carry out any of the
subsequent steps, they can certainly incorporate all the
facts applicable to Yucca Mountain into their performance
assessment.
And our understanding is that DOE will be
characterizing or grouping the FEPs together to basically
form FEP groups. And DOE can perform the screening of these
categories, based on two separate criteria:
One, it has been laid out in the rule that if the
probability is below 10-4, then DOE can screen that
particular event out. Or DOE can perform a consequence
analysis. It does not have to be a very detailed PA,
however, if they can perform a bounding analysis to
demonstrate that the exclusion of a particular FEP is not
going to impact the timing or the magnitude of the dose, and
based on that particular rationale, they can exclude the FEP
from consideration.
Once DOE has a reduced set of FEPs, then we want
them to start putting all these FEPs into scenarios. And
again, they can screen, they can perform the screening based
on the scenario classes, using the two criteria I have just
mentioned before.
And also in this part, we want to examine DOE's
assertion or their technical support on the probability of
disruptive events.
The first thing is that DOE needs to provide a
very defendable definition for what is being included,
whether it's in a single event or a particular event group,
to make sure that the characterization or the slicing up is
technically defendable.
And we will be looking at the data, models and
uncertainty in the probability estimates, based on how DOE
has formatted its FEB Division, how DOE has come up with
this definition.
So we're giving DOE tremendous amount of
flexibility in coming up with a defendable FEP case going
into the performance assessment.
The next piece I want to talk about is model
abstraction. I'm sure that you have seen this diagram many,
many times, and I probably don't need to spend a whole lot
of time on this now.
But I just want to give you an idea that the model
abstractions portion is being divided up based on the lowest
tier of this flow-down diagram. This is basically where
staff has utilized the experience and knowledge form
reviewing DOE's analysis, doing our own analysis, and
finding out what are potentially important to the repository
performance.
And we want DOE to address all these different
pieces. However, the level of the detail of these 14 topics
is going to be different, because we want to incorporate the
risk insight in terms of how much we go into basically
specify what DOE needs to do.
And that level of detail will be commensurate with
their impact on the performance.
Again, the model abstractions portion is based on
the integrated subissues, the lowest tier of the flowdown
diagram. We use five general technical criteria that focus
on data and model justification, data uncertainty, model
uncertainty, model support, and integration.
In data model justification, basically we're
looking at whether DOE has conducted sufficient site
characterization, analog investigation, field and laboratory
testing, to basically define the models and the associated
parameters that go into the performance assessment.
And in data uncertainty, we are looking at whether
the parameter ranges in the performance assessment have
captured the uncertainty existent in the database.
And in model uncertainty portion, we are looking
at whether DOE has considered alternative conceptual models
that can be explained, based on the existing information.
And in model support, we are looking at the
evidence that the DOE has to support the models that they
eventually used in the PA.
And in integration, we are basically looking at
whether DOE has properly handled the interface between the
various components of the TSPA. If you look at the 14
pieces here, none of them is a stand-alone piece that does
not have any relationship with the prior piece.
For example, if you look at the quantity and
chemistry of water contacting waste package and waste forms,
and the model of radionuclide release coming out from the
waste packages, those two are definitely related because the
solubility and the release rate will depend on how much
water actually gets into the failed waste package.
Therefore, the integration piece is looking at the
interfaces between the 14 pieces.
And the approach that we have taken in formulating
the model abstractions portion is to extract the review
methods acceptance criteria from the issue resolution status
report to strengthen the five generic technical criteria
that we have chosen.
If you flip to the last page in your handout
package, you will see a chart that will require a decoder.
And you can find the decoding information on the flowdown
diagram.
The top line, you see ENG-1, 2, 3, 4. Here we
have labeled ENG-1, ENG-2, ENG-3, and ENG-4. That's a
shorthand for the 14 ISIs or integrated subissues that we're
looking at.
And on the left-hand side, we see all the KTI
abbreviations, and the three pages preceding to the last
page, I have given you what these KTI subissues stand for.
So by using these particular charts, staff basically is
integrating the information from the various IRSRs, going to
the particular KTI subissues, and integrate the information
from the KTI IRSRs, based on the total system approach.
I know that is probably a lot of information to
digest, and that's why I provided you with the detailed info
there to give you an idea and a sense of the commitment that
we have in terms of looking at the issues from the total
system standpoint, incorporating the risk insight and
integrated amount of various technical discipline.
Moving on to the last piece under performance
assessment -- by the way, before I start here, I just want
to mention that information in the cross reference -- in the
cross -- flow chart, and the subissue definition has all
been attached at the end of the TSPA Issue Resolution Status
Report Revision 2, which was released the end of January, so
that has been in the public domain for about two months now.
Okay, moving on to the last piece, overall
performance objective, here we're looking at the compliance
to the individual protection standard. This is where we
will come together and make a determination of whether
consistent assumptions, data, and models have been used in
DOE's compliance demonstration calculation.
And this is also where the probability, scenarios,
and consequences all come together to form the risk
estimates over the compliance period.
In this part, we are also looking at the human
intrusion analysis. Basically we are looking at whether DOE
has used consistent approach compared to its PA for doing
the human intrusion calculation.
The only exceptions on the deviation from what DOE
has used in the PA calculation is where it will be
appropriate to modify because of the intrusion scenario.
And again, this is also an area where we have
received public comments during the Part 63 comment period,
and the staff is working on clarifying the human intrusion
analysis requirement. Once the Commission has made its
final decision on what issue will be put into a final rule,
we will further develop this piece to make sure that we have
a consistent review approach compared to the final rule.
Now I want to talk about the last piece of the
safety analysis report evaluation. The difference in this
piece is that we do not have performance objectives for
administrative and programmatic requirements as in the
preclosure and postclosure case. In this particular section
we are looking at mainly the procedural matters and there
are numerous existing acceptable programs in the agency, and
we are planning on using those existing programs modified to
the extent necessary, so that they will be suitable for the
high level waste repository at Yucca Mountain.
Also, I want to mention that a lot of the
information contained in this section is going to directly
impact the preclosure safety evaluation, therefore there is
going to be a lot of looking back and forth between the
preclosure part and the administrative requirements to make
sure that during operation the preclosure performance
objectives will be met.
In the evaluation in this particular section we
include QA, training, recordkeeping, normal operation,
emergency planning, and physical security. Those are the
big topics in this particular section.
As I mentioned previously, the draft Revision Zero
for this particular section will be coming to us from the
Center by the end of April and we are hoping to have the
Revision Zero done by the middle of June.
Scheduled activities -- I have mentioned that we
are in the process of establishing coordination with other
NMSS divisions and program offices for review of the
preclosure safety and the administrative and programmatic
procedures. We are consulting with NRR in terms of the
emergency planning procedures and working with SFPO in the
fuel cycle, the Part 72 and Part 70 folks, on the preclosure
safety analysis portion.
We have assigned technical leads to integrate the
multidisciplinary teams and build consensus for each of our
review sections.
We will continue to work on the level of detail,
integration and incorporation of risk insights.
In the future revision of the Yucca Mountain
Review Plans will be modified as necessary so that you will
be consistently implementing the final Part 63 because now
we are on this schedule that we don't really know when the
Commission is going to be making its final decision on Part
63. Therefore, we will be keeping an eye on that particular
progress and make sure our future revisions capture the
final position in Part 63.
I would like to bring your attention to a third
bullet -- continue working on level of detail integration
and incorporation of risk insights. As I have stated at the
beginning of this presentation, we fully intend to make the
Yucca Mountain Review Plan risk-informed and
performance-based. However, we also need to give sufficient
guidance to the Staff and indirectly to DOE on what is
acceptable.
We don't want to be prescriptive but we also don't
want to get into a situation where it is bringing another
rock situation.
Compounding with the fact that the design is still
evolving, the knowledge base is still evolving, and there is
an inherent uncertainty for a 10,000 year repository, I
believe this is where this particular committee can provide
the most recommendations and guidance to the Staff on
helping us achieve the goal of having a risk-informed and
performance-based review plan, without being overly
prescriptive providing sufficient guidance so that it will
be clear to everybody what is necessary to demonstrate
compliance and at the same time there is an appropriate
level of flexibility for DOE.
I just want to conclude this particular
presentation by giving you the schedule. Looking ahead, the
Revision Zero of the Yucca Mountain Review Plan -- here I am
talking about the postclosure sections only -- will be going
to the Commission with the draft final Part 63 by the middle
of April and we are planning on holding meetings with DOE,
these are public meetings.
Right now on the book there will be a PA technical
exchange in early June and we are also planning on an
Appendix Y meeting on YMRP and the license application in
the middle of June.
In Revision 1 of the Yucca Mountain Review Plan,
that's where we have all the sections together. We are
planning on having that particular version out by the end of
FY 2000 and we will formally invite public comments on that
particular version, so anything we release prior to the
Revision 1 is going to be for information only, and then
after we release Revision 1 we will be holding public
meetings and also meeting with DOE to explain the approach
that we have taken in the Review Plan and also to solicit
comments.
In Revision 2 of the Yucca Mountain Review Plan we
will address the comments that we have received and
incorporate any information, whether it is going to be on
design or site characterization, into consideration.
MR. HORNBERGER: What is the date for that one,
Christiana?
MS. LUI: Revision 2 is September 30th, 2001 --
sorry, that is a typo.
MR. HORNBERGER: Otherwise you would do it very
quickly.
[Laughter.]
MS. LUI: That's okay. That is an obvious mistake
so that you won't notice anything else.
[Laughter.]
MS. LUI: That's it. That is the end of my
prepared presentation. I will be happy to entertain any
questions you may have.
MR. HORNBERGER: Thank you, Christiana.
MS. LUI: You're welcome.
MR. HORNBERGER: The first question that I have is
that you were probably in the audience earlier this morning
and you heard James give a presentation on the site
suitability report, and there is obviously or it is obvious
to me that there's an awful lot of overlap.
MS. LUI: Yes.
MR. HORNBERGER: I understand the different
objectives, but there has to be an awful lot of overlap.
Can you just give us some insight on how you are
coordinating?
MS. LUI: Okay. I believe a while ago that we
have sent a letter up to -- I have to remember who was the
recipient of that particular letter. I think it was a
letter back to the Commission saying that we intend to use
the Yucca Mountain Review Plan to formulate our sufficiency
comments. If you remember that -- of course, I lost that
particular page -- but the five technical acceptance
criteria that we are using for model abstraction evaluation
includes data and model justification, data uncertainty, and
model uncertainty and model support and integration.
That is going to be the emphasis of how we are
looking at the -- or to formulate our sufficiency comments.
we are not going to carry everything to the last piece.
That is where we are going to be looking at the overall
performance objective but we want to look at the report and
make a judgment whether we can proceed with the review that
we have in mind when the license application comes in.x
MR. HORNBERGER: Okay, so the teams are clearly
working together then?
MS. LUI: Yes.
MR. HORNBERGER: I have one other question, sort
of a clarification for myself.
On one of your slides you mentioned that you were
going to be looking at postclosure performance confirmation.
Do you have ideas on what postclosure performance
confirmation would look like? Are there any requirements or
is there any guidance or could you just give us your
preliminary thoughts on what would be required?
MS. LUI: Okay. Remember that DOE's responsible
to define its safety case. Therefore, let me just look at
this chart here, we have these two pieces in the postclosure
review. One part is that there is going to be we require
DOE to establish a program and schedule for closing any of
the open issues, and that is clearly linked to the
performance confirmation program that they are going to be
implementing.
In the performance confirmation we certainly
expect them to focus on those areas that they will need to
further clarify or have more information in order to support
a defendable safety case such as the heater test is still
ongoing and the results are now going to be available by
license application time. Therefore, that would be one area
we are going to be carrying into the license application and
when the information becomes available to us, well, to DOE,
we certainly will expect DOE to utilize that information to
update its performance assessment.
MR. HORNBERGER: Okay. Perhaps it is a
misunderstanding on my part then. When I think of
postclosure, I think of physically after the repository has
been closed. That is not what you are talking about when
you talk about postclosure performance confirmation?
MS. LUI: No, that is not what we are talking
about at all. No, no, that is not.
Even though in Part 63 we do require DOE to
establish permanent oversight --
MR. HORNBERGER: Right.
MS. LUI: -- but the performance confirmation
program we are talking about here is not after closure. It
is more to strengthen its safety case for postclosure.
MR. HORNBERGER: Okay, got you. Questions from
the committee? John.
DR. GARRICK: We have talked quite a bit at this
meeting about the positive aspects of the public review
process associated with draft Part 63 and I assume that that
has spun off as far as the Review Plan is concerned, and
also you note in your presentation that there will be public
comments on the revisions to the Review Plan.
Are there any other activities between Part 63 and
the Rev. 1 for example where there has been deliberate
effort to get public involvement and participation in the
creation of the Review Plan?
MS. LUI: We did not have -- I mean up until this
point we have not conducted any public meetings outside from
the public meetings that we had with DOE, so in terms of the
framework of the Review Plan pretty much we looked at how
our Part 63 is structured --
DR. GARRICK: Yes.
MS. LUI: -- and also our conversation with you,
from time to time, in terms of helping us focus on what we
need to look at, so the structure portion, no, we have not
gone out explicitly to ask for public comments, but that
does not mean that during the Revision 1 public comment
period that they will not be commenting on the structure of
the Review Plan.
DR. GARRICK: Yes.
MS. LUI: But between Revision Zero and Revision
1, because Revision Zero is going to be for information
only, we do not anticipate that we will be actively seeking
public comments.
DR. GARRICK: Okay.
MS. LUI: However, if management decides that we
can go ahead and put the Revision Zero on the website,
whatever informational comments that we receive on Revision
Zero we will take that into consideration when we address
the comments together when we address the comments on
Revision 1.
I think Bill has something else to add.
MR. REAMER: Bill Reamer, NRC Staff. I would just
add one more point, kind of to build on something that
Christiana mentioned during her presentation. That is, our
plan to use the Review Plan in the sufficiency comments as
well as issue resolution and both of those processes,
sufficiency comments and issue resolution, are public
processes that involve technical exchanges, meetings with
DOE that are public, so we hope that they will also produce
feedback on the way in which we are following up and
implementing this draft review plan and could lead to
improvements or changes in the review plan as well.
DR. GARRICK: Okay, thank you. A little more on
the technical side -- when you talked about the scenario
analysis with respect to postclosure safety evaluation, you
indicated that you have a screening criteria or a cutoff for
scenarios of 10 to the minus 4, I believe it was, that you
mentioned, the number?
MS. LUI: The probability, but the bottom line is
that the frequency is less than 10 to the minus 8 per year.
DR. GARRICK: Yes.
MS. LUI: So over the compliance period it would
be 10,000 years, so I converted that to 10 to the minus 4
probability.
DR. GARRICK: And of course these I assume are
mean values that you are talking about?
MS. LUI: No, we are looking at -- DOE needs to
look at the full range.
DR. GARRICK: But I know you are looking at a
range, but the specific number is a central tendency
parameter?
MS. LUI: I don't want to commit ourselves to just
looking at mean value at this point because there is a large
uncertainty associated with any of the geologic processes
and we really need to look at the supporting evidence DOE
has before we will all be able to say whether it is a
legitimate exclusion or DOE is basically dicing up its cases
to the point that everything is going to be below 10 to the
minus 8 per year.
DR. GARRICK: Yes, okay. I think that is a good
answer, because among other things if you have an Epsilon
amount on the proper side of the screening number, then one
of the things you really want to look at of course is the
variation in uncertainty between the scenarios.
MS. LUI: Right.
DR. GARRICK: The other thing I wanted to just get
your early comment on is one of the elements that you are
going to be looking at when you start looking at uncertainty
of course is modeling uncertainty, and I think that was on
one of your exhibits.
Can you give us a very abbreviated glimpse of how
you are going to do that?
MS. LUI: Okay. I think doing a couple of public
meetings we have with DOE -- I mean we have had with DOE, we
definitely are not advocating assigning probability to the
alternative conceptual models. What we want DOE to do is
that for the alternative conceptual models that would
actually lead to pretty significant differences in the
consequence estimation, we want them to incorporate the
results of those alternative conceptual models and we want
to look at all the results rather than have them assign
probability to alternative conceptual models and combine
everything together, because to us that is not transparent
and we will not -- what we really want to do is to
understand what are the bases and if the current information
cannot lead us to basically discriminate one from the other,
then we definitely want to see the results of all credible
alternative conceptual models represented individually.
DR. GARRICK: And you think this will give you
some insight?
MS. LUI: Right.
DR. GARRICK: Yes.
MS. LUI: Right.
DR. GARRICK: One of the things that could be a
major issue here is this issue of quality assurance.
Quality assurance I notice was on your list of
administrative and programmatic rather than technical, and
of course many of us are much more inclined to think of
quality assurance as an important issue from a technical
standpoint more than an issue from a documentation and
creating a record standpoint.
I guess the question that I have is that being an
issue that is well-known and that has existed for quite some
time, I assume that there has really been a great deal of
exchange already on just what the NRC is looking for in
terms of an acceptable quality assurance activity.
Is that correct?
MS. LUI: Okay -- I think, Dr. Garrick, you are
probably leading to graded QA?
DR. GARRICK: Yes.
MS. LUI: Okay. Well, we had a meeting with DOE I
think two weeks ago talking about a graded QA effort.
Basically what we will be looking for is that there will be
a set of minimum requirements DOE will have to meet no
matter what. However, DOE can use the graded QA approach in
terms of identifying what are the important structures,
systems and components for the safety case, and the level of
rigor for those that are important to the safety case we
will certainly be paying a whole lot more attention to
compared to the ones that have been identified to have
minimal impact on any of the bottom lines.
Even though the QA is in the administrative and
programmatic requirements area, it indicates that a lot of
the pieces in that particular section are the underpinnings
for the pre-closure and post-closure. So the level of rigor
that we're going to be looking at is going to be dependent
on how these supporting pieces feed into that preclosure
safety case and postclosure safety case.
CHAIRMAN GARRICK: Okay; thank you.
Raymond?
DR. WYMER: Yes; as usual, your presentation was
very well-organized and presented.
MS. LUI: Thank you.
DR. WYMER: I particularly appreciate -- it will
be useful to me -- the presentation on the KTIs and how they
tie together. Having said all that, let me get to a
question.
MS. LUI: Okay.
DR. WYMER: Could you say just a bit more -- this
isn't a question, exactly, but could you say just a bit more
about the acceptance criteria? I think that there will be a
great many people who will be interested in those, and
particularly, the Department of Energy will be interested in
those. And I wonder what you can say about the degree of
detail that will be presented a little bit more.
MS. LUI: Okay; let me pick an area where you will
be interested, so it will have to be one of a couple
processes. Let me pick ENG3, the quality and chemistry or
water compacting and packaging waste forms. Basically, if
you look at the chart here, if you just count the number of
dark boxes, ENG3 is probably one of the most complex
integrated sub-issues that we have. So what we do is we go
into each of these IRSRs and look at the acceptance criteria
and review methods that have been documented in those IRSRs
and organize them based on the five generic technical
criteria in terms of model and data justification, data
uncertainty, model uncertainty, model support and
integration with other portions of the integrated
sub-issues.
DR. WYMER: So from that, I would assume that
there will be quite a bit of detail of --
MS. LUI: There's a lot of detail.
DR. WYMER: -- the acceptance criteria.
MS. LUI: And because also, according to our own
sensitivity studies, these particular ISIs happen to be
probably one of the very top ISIs, meaning that from both
the uncertainty perspective and also how you could influence
the dose -- I mean, how you could influence the risk
computation, this is definitely on top of the list, so we
would have a lot of detail associated with these particular
integrated sub-issues.
DR. WYMER: Thank you.
MS. LUI: Okay?
MR. LEVENSON: I guess one of the advantages and
disadvantages of being last is most everything has been
said, both in challenging you and how good the presentation
was, but I do have two left-over questions.
MS. LUI: Sure.
MR. LEVENSON: One, under administrative and
programmatic requirements, you list physical security.
Exactly what is the scope of that? Physical security for
the property? Safeguards? What's intended there?
MS. LUI: Unfortunately, you asked a question that
I really don't have a whole lot of background in, but I will
try my best to answer your question. The physical security
here is looking at during the operational period, the OE's
program in terms of maintaining access to the site, and I
believe that we do have existing programs in the agency.
MR. LEVENSON: Well, let me give you the context
of my question.
MS. LUI: Okay.
MR. LEVENSON: There is a possibility that Yucca
Mountain may be used to dispose of excess weapons-grade
plutonium, and that could potentially, if you're thinking of
safeguards, make a significant difference compared to
defense-vitrified logs coming from Savannah River. So, I
just wondered how -- what this issue is and how it might
expand for what might be going to Yucca Mountain.
MS. LUI: Okay; the level of detail that we
currently have, I understand, in the review plan is that
based -- because earlier on, we do ask DOE to identify what
kind of waste DOE is going to be receiving at the Yucca
Mountain site, and based on the content of what's coming in,
the DOE needs to develop a program that's consistent with
the level of security necessary in order to have confidence
that there is going to have a sufficient amount of safety
and safeguard oversight.
MR. LEVENSON: Okay; so, in essence, it's DOE's
responsibility to define the physical security program --
MS. LUI: Yes.
MR. LEVENSON: -- based on materials, and they
would like to expand it if, indeed, they add weapons
plutonium.
MS. LUI: Right; but however, there are certain
minimal requirements DOE will still have to meet based on
the existing agency's programs.
MR. LEVENSON: Yes; okay.
The second question I have has to do somewhat with
the question John asked, and that is how much, how far you
get into checking what's really in the models, and the
context of that question is two and a half years ago, when I
was doing a study for the National Academy in connection
with the research reactor fuel disposal, I discovered that
at that time, the Yucca Mountain PA did not have either
conservation of mass or conservation of energy in any of
their models. And as a result, they were overestimating
consequences by huge amounts, and I presume by now they've
put that back in. But will your review be in enough depth
to catch things like that?
MS. LUI: I think so, because in our review plan,
we have stated in our review method where appropriate, staff
is going to use our TPA code to do our other calculation,
because if there are -- for example, during the VA review
process, we have looked at some of the graphs and charts,
and trying to understand how has DOE come to those
quantitative numbers, and in those cases, we use our TPA
code to come by all of the calculations. So if there is
something that is really obviously inconsistent about one
module to another, basically wanting to intermediate -- once
that will be open to intermediate to another, then, staff is
going to go into more detail in looking at how DOE has
conducted its analysis.
MR. LEVENSON: So you're confident that your TPA
code does conserve mass and energy?
[Laughter.]
MS. LUI: I think so. I would say that Tim has
worked really hard, and also, the other PA staff was working
really hard to make sure that our code is going to be
suitable to the license application review.
CHAIRMAN GARRICK: The next thing, Milt will be
asking about the continuity equation.
[Laughter.]
DR. HORNBERGER: Questions from staff? John?
Others? Carol?
MS. HANLON: Dr. Hornberger, committee,
Christiana, I just wanted to thank you for your
presentations. I think it will be very helpful. And I
wanted to thank you for mentioning the upcoming interaction
in June, where we will be working with Christiana to look at
the Yucca Mountain review plan in the context of the license
application to make sure that it's consistent.
And also, I wanted to mention that we have
forwarded eight letters culminating on the IRSR. Each of
the IRSRs that we have, we forwarded them to Bill Reamer
last week, and that was at the invitation of the NRC to
comment on those IRSRs. We regret that they're late in the
process. We realize that they are late. We worked very
hard to make sure that our comments were integrated, and it
provided an integrated approach to you. So, some of the
areas where we have commented were for questions or for
simple clarification. Other places where we may have
commented were where we thought perhaps the techniques
discussed were more prescriptive than you may have meant,
and in some other areas, there were some other areas where
we felt that the performance-based aspects may have been
vague.
So we provided you with those. We hope they will
be helpful in financing the Yucca Mountain review plan and
being part of the acceptance criteria. So, thanks for
looking at those as you move forward, and again, we're sorry
they're late.
MS. LUI: That's okay. I just wanted to emphasize
that for revision zero, what we are trying to do is to
basically synthesize and integrate what all of the work that
has been done has done up to this point, and the two
revisions that were -- and on revision one, we do intend to
look at the level of detail, whether they are appropriate
for the particular topic we are looking at using the
risk-informed and performance based approach, and that's
also where we're going to be interacting with the ACNW to
also get their recommendation.
DR. HORNBERGER: Great; thank you very much,
Christiana, and thanks to your silent assistant over here
for his comments.
[Pause.]
DR. HORNBERGER: Back to you, John.
CHAIRMAN GARRICK: All right; I think what we
would like to do, as many of you know, we're in the middle
of writing several letters, and there's some word processing
that I'd like to do on one of those letters. I think what
I'm going to do now is essentially recess the meeting until
our appointed time of 1:00, where we will talk about
radionuclide content of slag and prepare ourselves better
for the -- take advantage of this time to prepare ourselves
better for the 3:15 session on the continued preparation of
ACNW reports.
So with that, I think we'll adjourn.
[Whereupon, at 11:14 a.m., the meeting was
recessed, to reconvene at 1:07 p.m., this same day.]. A F T E R N O O N S E S S I O N
[1:07 p.m.]
CHAIRMAN GARRICK: Let's see if we can come to
order here.
Leading our discussion this afternoon on
radionuclide content of slag and uranium plume attenuation
is committee member Dr. Ray Wymer.
Ray?
DR. WYMER: The next presentation starting this
afternoon off is the status report on the characterization
of radioactive slag, and we're going to hear about leach
rater of uranium and thorium from slag, and I was a little
surprised that it was limited to those until I looked at the
list of the site decommissioning management plan items.
It's something like two-thirds of those are, in fact,
uranium and thorium slag, which explains why you're focusing
on those. I would expect that in the future, as the
clearance rule comes along, there will be other types of
slags that will rise to the top in importance, but for right
now, these are the two front-runners.
MS. VEBLEN: That's right.
DR. WYMER: This is Linda Veblen. Please give us
your report.
MS. VEBLEN: Well, thank you, and I'm happy to
have the opportunity to speak to you. I did bring some
samples here of slags that are not radioactive, but we have
one -- just I thought you can pass them around and look at
them -- this is slag from Maryland that's about 100 years
old. It's from Ashland, and this slag is from Sudbury,
which we will see in one of the next slides just to give
you --
DR. WYMER: From where?
MS. VEBLEN: Sudbury, Ontario, Canada.
Okay; so, I'd like to acknowledge my colleagues.
Dori Farthing is a Ph.D. student at Johns Hopkins
University. Dr. O'Donnell has been looking at this, and
Professor Veblen at Hopkins has provided a lot of input and
also the facility and people from the department.
Well, as Dr. Wymer mentioned, there are a lot of
mountains of waste slag being produced in the United States,
and these contain toxic metals and, in some cases,
radioactive waste or radioactive elements. And so, one of
the concerns is how stable is this waste? And we know from
areas around the world where there are large slag piles that
some of these toxic elements are leaching into the
groundwater and contaminating groundwater and land areas.
So the NRC has several decommissioning sites that
contain slags. There are about 17 of them, and they contain
uranium and thorium in quantities that exceed the old
regulatory limits, so -- and the owners of these sites would
like to have the site material license released from the
NRC, so the NRC, then, needs to try to understand or to
determine what the long-term stability of these slags are
and how uranium and thorium will then leach from the slags.
Well, so what? And how difficult is that to
determine uranium and thorium leaching rates? Well, there
are three different standard leach tests that are currently
being used. One is ANSI, which -- 16.1, which is use for
leachability of solidified low-level waste, and in this
case, a cylindrical plug or an intact piece of slag or
low-level waste is placed in deionized water and leached for
a period of time, and then, the leachate is decanted off and
determined how much has come out of the slag, and generally,
in these cases, we see that not much has come out: 10-12 to
10-10.5 for a thorium leach rate.
And these studies were done by the center. They
were also studying the slags and trying to determine which
of these methods might be the best for characterizing the
uranium and thorium loss from the slag.
DR. HORNBERGER: What are the units on that rate,
Linda?
MS. VEBLEN: I believe it's grams per meter
squared per day. The EPA method, the TCLP, is the second
one. In this case, the sample is crushed to a fine grain
size and leached in acetic acid at a very low pH. The third
type of test is also using crushed material and with a
slightly than higher closer to neutral pH. And as we see,
there is a range of leach rates, anywhere from 10-8 up to
10-12. That's four orders of magnitude, which makes it kind
of difficult when you're running performance calculations to
determine which one to use.
So, why are we doing this? What's the bottom
line? Well, we didn't know what was in a slag. What is a
slag in the first place? Three years ago when we -- four
years ago, we were starting this, people said, well, it's
glass, or it's this and that. Well, is it? We don't know.
What's in a slag? What are the phases? Where is the
uranium and thorium? And even the licensees didn't know
where these were.
So where is the uranium and thorium? Is it evenly
distributed? These are some of the questions that we're
trying to find out. What's the leach rate for a slag, and
how does it change with time? Or does it change with time?
And what standard tests should we use, if any? And, as I
mentioned before, there are 17 of these SDMP sites. Are
they all the same? Is it the same slag? The same phases
that are in them? And if they're not, can we use the same
licensing criteria, then, for all of the different slag
sites?
And so, ultimately, we really wanted to identify
or understand how a slag weathers. With that in mind, we
come to the objectives of this study, and that's to identify
the solid phases in the slag, both radioactive and
nonradioactive and get an idea of the weathering mechanisms
in the slag, try to determine that, and also estimate what I
call in situ leach rates or instantaneous leach rates from
the slag for input into the RESRAD code for performance
calculations. And these are not actually leach rates, but
what we would be determining is the mass loss over distance
for a period of time, and I'll show you later how we might
be able to calculate a leach rate from that.
I'd like to run through some of the sites, what
they look like. This is an example of one of the SDMP
sites. They were smelting tin slags that had been brought
over from Malaysia and were extracting niobium and tantalum
from these slags. That's done by crushing up the tin slags,
adding fluxes and heating it to 1,500 to 1,700 degrees
centigrade. That forms essentially a lava. It's poured
off; allowed to cool, and while it's still in its molten
state, of course, the niobium, the tantalum and the heavy
metals drop to the bottom.
They then go in and break up the slag with
sledgehammers; take off the metal separate, which is also
known as the blooms, and then throw the waste slag away.
This is the site, one of the sites we're studying, and you
can see the slag pile right along the edge of this major
river.
These are pictures from 1967. This is an example
of a slag pile being built, and these are what the loughy
slags look like. This is the slag that's been -- the pour
that's finally been cooled and broken up into about 10
kilogram-sized blocks. There are railroad tracks right down
here, and the river is just beyond that.
And then, this was a picture taken three years ago
by the center. We were up there investigating the site, and
you can see that we're parked on top of the slag pile
roughly about where that is, so it's changed a lot. There
is a lot of vegetation that's grown up on it.
What do slags look like? Well, I just passed
around some samples. These are two of the samples from the
SDMP site. Notice that they're blocky, glassy. There's an
upper weathered portion in both of these. These slags
resemble the salts, quickly-cooled igneous rock.
We have another site that has what I'm calling
reprocessed slag, where you start with that blocky slag.
They crush it up in a bow mill; leach it, then, with
hydrofluoric or hydrochloric acid to extract the last
remnants of niobium and tantalum and then throw the waste
back into a settling pond, and it settles out along with the
leachate. And so, what we end up with here actually are
sedimentary rocks, but they're made up of slag components.
The grains are glass or slag phases. And you see, it looks
just like a very nice crossbedded, layered sedimentary rock.
It even has a nice weathering line that's about a centimeter
to two centimeters thick on the surface.
DR. HORNBERGER: Is it all glass or amorphous, or
do you have some crystalline stuff?
MS. VEBLEN: There is a lot of crystalline stuff
in both of them.
Okay; so, we set out then to determine, identify
the phases, and these were the research methods that we
used. We're looking for elemental variation and the
distribution in the slag. We did that using a light
microscope; x-rayed the fraction to determine the
crystalline phases that are in that slag. We used an
electron microprobe, both SEM and wavelength dispersal of
the EDS analyses to get quantitative chemical analyses on
about a micron spot size, so we can do very detailed
chemical analyses as we go across these samples.
And finally, to understand really the weathering
mechanism, we've gone to using the TEM, which can analyze
the slag on an angstrom level, and we can look at here
elemental variation on a very small scale. We can also look
and determine the species of an element, for instance,
uranium, whether it's in a +4 or +6 state, which I think
would be a powerful tool to use in this weathering study.
We haven't gotten to that point yet, but hopefully, we will.
So, these are the different tools, the Hans scale,
the light microscope, and with the light microscope, we're
looking at very thin slices of the rock, and they're about
20 millimeters by 40 millimeters and 30 microns thick. We
use that same thin section on the electron microprobe, but
then, finally, when we go to the TEM scale, we have a copper
grid that's about 3 millimeters in diameter and glue that
onto the thin section and then bore a hole in the center of
the copper grid with an iron beam and thin it down to
electron transparency; so, several angstrom-thin layers.
This is just a pretty slide that's an example of
one of the SDMP slags. These are aluminum, chromium,
titanium spinels, almost similar to a ruby, which is mostly
an aluminum spinel basically. We see perovskites, which are
opaques, in here. The white is glass, and we also have
magnesium iron spinels, and I'll show you electron
microprobe back-scattered image of this a little bit later
and show you what's happening with the weathering here of
these.
This is an example of one of the other slags;
slightly different chemistry. The white is glass. These
dark, opaque areas are perovskite, and we have a gehlenite,
which is a calcium aluminosilicate, and basically, just
using two techniques of a light microscope, plain polarized
light and cross-polarized, we can identify the glass phases
and the crystalline phases, which is very helpful, because
you can't do that with an electron microprobe. You can do
that with a TEM, but you're on a much finer scale.
So this is the same area that we're observing.
Here's a scale bar of 100 microns, and this is what the thin
section looks like; the slag looks like under a plain
polarized light microscope. There's lots of glass, this
white area. We see beautiful uhedral crystals of this brown
uhedral crystal, which turns out to be a clonopirixine, but
it's a rare clonopirixine that contains zirconium titanium
and is similar to pirixines that are seen in meteorites.
The long, black dendritic crystals are the
perovskites, just an image in cross-polarized light, again,
to give you an idea of what's glass versus what are the
crystalline phases. Then, we go to a back-scattered
electron image. This was taken on the microprobe, electron
microprobe, and the electron microprobe basically shoots an
electron beam down on the sample, and the electrons are
back-scattered, and the heavier elements the electrons much
more easily, so we end up with a much higher intensity.
So what this tells you is that when you look at a
back-scattered image, the very bright phases, then, are
those that have the heaviest elements in them. So we know
right away that this long dendrite contains heavy elements,
and it turns out that it contains uranium and thorium. This
is one of the perovskites. We can also tell from this that
this phase that I've labeled -- well, I didn't label it;
sorry, it's gehlenite -- doesn't have many uranium and
thorium. And the glass, which is the medium gray area in
here, does contain uranium and thorium.
So from here, we can go down, then, and actually
obtain quantitative analyses, elemental analyses, on little
points. In fact, that's a point that we analyzed right
there, that little tiny dot, and there's a little tiny dot.
MR. LEVENSON: Are all of these on one particular
slag?
MS. VEBLEN: This slide are all on one slag. This
is all the same image at the same scale. But we've done
this; we have multiple slag samples, so we've looked at
quite a few.
The bottom right is a high-resolution TEM image,
and we're essentially looking at angstrom scale here. These
little tiny white blebs in here represent silica tetrahedra
that are strung together in a chain, and this is a
clonopirixine. This, actually, is the old TEM that's at
Hopkins. The new TEM that went in has essentially -- can
resolve about an atom, so we could be looking at uranium
atoms on that scale. But the real beauty of the TEM for
this particular -- one of the beauties for this particular
study is that we really didn't know what this clonopirixine
was, what kind of phase it was.
We knew its elemental chemistry, but we didn't
know any structural information about it. So, you go to the
TEM; you can do electron defraction there on very small
crystals, and this is actually an electron defraction
pattern from that very crystal right there, and that's what
showed us that it was in fact a clonopirixine but just one
that has very rare chemistry, and I think, Dori, that this
actually has some thorium in it, very small amounts. Dori
is the graduate student working on this.
This is another thing that we have done is looked
at a back-scattered electron images and x-ray mapping. You
can go in with a certain area and set up wavelength
dispersive and energy dispersive analyses to determine
silica, calcium, all these different phases. And then, you
just basically plot up that particular element on a map, so
let's go over to this bottom right image, which I've labeled
thorium, and what this shows, then, the bright areas show
that those phases contain thorium. The relative intensity
corresponds to the relative amount of thorium that's in a
particular phase, so right here, from this map, we can see,
yes, thorium is in the perovskites; thorium is in the glass;
it's not in this, you know, there's not much in some of
these other phases.
One thing I would like to point out in this
particular slide is if you go up to the back-scattered
image, we notice that this light gray area is glass. Coming
off there are some crystals, and there are also some holes,
and there are some very bright spots. And what's happening
here is the glass is devitrifying. As it devitrifies, the
volume changes. It decreases, so it's opening up a little
bit of porosity. Now, the uranium and thorium is in the
glass, but what looks like is happening here is that as the
glass devitrifies or dissolves, whatever it's doing, the
uranium and thorium and in this case cerium tend to go into
a silicate phase, so they're held in that phase, it seems.
So that's a nice thing.
All right; these are the phases, then, that we've
identified, and I don't expect you to -- you know, we're not
going to go through this. It's just -- I wanted to show you
the number of phases that are in these things. In some of
the slags, there are 20 individual phases. So it really is
a mess when you try to characterize an x-ray defraction or
something else.
But one of the things that we noticed right away
was that the phases are kind of interesting. We have
calzirtite, zirconolite, perovskite, spinel, and this is the
chemical formula for them over here; barium aluminate, which
is like a barium, well, okay, barium aluminate and glass.
And these phases are all found in SYNROC. Now, SYNROC is a
synthetic rock, hence its name, but it's a ceramic waste
form that's being considered for high-level waste, perhaps
the Hanford waste tanks and also for excess plutonium.
So we thought that actually, the study of these
slags might be a nice analog to the study of SYNROC. There
is a lot of data on SYNROC that's been done in the
laboratory, but SYNROC has only been in existence for the
last 30 years. So this might be a way that people could
look at long-term leaching of SYNROC. Vice-versa, it's a
way that we can use -- determine leach rates for the slags,
because they are so similar to SYNROC.
DR. HORNBERGER: Just a clarification.
MS. VEBLEN: Yes.
DR. HORNBERGER: On the previous slide, Ce,
cerium?
MS. VEBLEN: Ce is cerium. There is a lot of
cerium.
DR. HORNBERGER: A lot of cerium?
MS. VEBLEN: One thing I didn't -- I didn't show
you the analyses we did, but we analyzed for 29 elements,
and these are -- the slags are all very heavy in light rare
earths. I mean, in some of these, there's 18 weight percent
cerium -- not in the bulk slag but in the slag phase.
DR. WYMER: The slags must come from a very
peculiar and specific kind of refining.
MS. VEBLEN: Well, actually, a lot of them come
from carbonotites, which are, you know, fairly unusual rocks
that are very high in rare earths. The slags were
originally smelted for tin. They extracted the tin from
them. But there was so much rare earth and everything else,
that that went into the waste, and they figured, well, they
can certainly turn around and use that. So that's why we
ended up with these.
DR. WYMER: Okay.
MS. VEBLEN: Okay; so, then, our ultimate goal
here is to determine a leach rate, and we do that by
measuring the elemental variation in weathered SDMP slags
with a microprobe and the TEM. Also, we have looked at the
elemental variation in slags, these SDMP slags that were
leached by one of our research contractors at Pacific
Northwest National Lab, and I'll show you some examples of
those, so we looked at solid phases there and also looked
for alteration.
And finally, we can get an idea of what long-term
leach rates might be by analogy to archaeological slags, and
we've collected tons, it seems, of archaeological slags. We
also can get an idea of leach rates based on what leach
rates might be for SYNROC. And finally, we can look at
leach rates or study leach rates of slags by looking at
natural minerals such as hibonite, and Dori will be doing
some of that for her Ph.D.
This is an example of a weathered SDMP slag, and
I'd just like to show you this back-scattered image, and
what I'd like you to notice is that, of course, we have a
hole there but some large crystals. These are those red
spinels that I showed you in one of the first thin sections,
and these long areas that used to be dendrites, which are
now -- these are actually holes -- this was the area where
the magnesium iron spinel was, and in this outer edge of the
thin section, it's been leached out preferentially, I would
say.
We can look at the interior, and you see these
light or gray areas. These are the spinels that are intact
that have not yet been leached. But up in the upper slide,
which is at the weathered edge, they're weathered out. So,
there's evidence of weathering along the grain boundary.
This slide just shows the transition from where
these -- well, here's a nice one where the magnesium iron
spinels are still intact, and here's the same grain, the
same crystal, but it's finally been leached out here. So,
what happens with this? Well, if there is preferential
leaching of a particular phase, the fluids move along the
grain boundaries, and within the grains, it can open up
fluid access to the interior of the slag.
Some examples of microprobe data that we have:
this is variation, elemental variation in glass, and I've
kind of flipped this slide around; I'm sorry, but the right
side of each graph is the outer edge of the slag, so that's
the weathered edge. The left edge of the chart is 40
millimeters, and what we're looking at is the variation of
aluminum, titanium, zirconium, all of these elements:
silicon, calcium, thorium and uranium as you go from a
weathered edge into an unweathered area of the rock, the
slag.
And what we notice is that in the outer 10
millimeters, silica is depleted compared to what is seen in
the interior, the same as calcium, thorium is depleted, and
uranium is slightly depleted. And from that, I've measured
-- and do I have it here? I didn't plot it up. We have
measured the difference in mass between uranium and thorium
over that distance. You can use that, then, and this -- I'm
sorry; this graph is wrong. It should be concentration over
distance. We get a plot. We can put a point there. We
know that what the concentration for a certain distance is
at a particular period.
If we go to the next slag that's 100 years old or
50 years old, we can plot its concentration loss or its
elemental loss over a certain distance and likewise do that
for numerous slags. We may find a line, but this is
hypothetical data. We haven't gotten to this point yet, but
we're moving in that direction.
Another way of looking at the leach rate is to
determine kind of a bulk leach rate, and these are measured
modal abundance of one of the particular slags. There are
about 40 weight percent gehlenite, 36 weight percent glass.
Notice the calcium and thorium bearing phases; perovskite,
calzirtite and pyrochlore are all total less than about 18
weight percent. They do contain most of the uranium and
thorium. So, if we were to weather out one particular phase
or leach out one particular phase, this much of 25 percent
of the uranium and thorium would be released of that phase
were gone. If you weathered out all of the glass, 25
percent of it would be gone. If you weathered out all of
the perovskite, that would be gone.
So, this is data that might be used also in RESRAD
to give us an idea of how things might change with time.
Just real quickly, some examples of the slags that
were leached at PNL. These are slags that have hibonite in
them. These are these big brown crystals, and they're not
brown, they're gray, dark gray. There's lots of glass;
again, zirconolite, perovskite and rutile in here. And what
we're looking at, this was a crushed. The slag was crushed.
You can see the size of the particular grain here is maybe
500 microns across, and it sat in water for a period of
time, and what we're doing is looking at the -- seeing if
there is any depletion of elements along this outer edge, so
we do traverse this across here to see if there has been
leaching.
Also, PNL did some leaching experiments at pH,
leach where the pH ranged from 0.05 up to 12, and these are
SEM images of the surface of these slags, what happened to
them after they'd been leached. The top -- yes, SEM image
is glass that's been leached at a pH of 2, and this shows
classic hydration and corrosion of the glass, where you get
these cracks. There's dissolution going on. We actually
see precipitation of some secondary phases on the surface of
the glass.
We move down to the bottom image, and this is --
was leached at a pH of 6, and again, it doesn't seem like
there's a lot of damage to the glass here, but there are a
lot of secondary phases that have grown on the surface. And
finally, at a pH of 12, again, the glass looks to be very
corroded, and we're left with these rutile perovskite
needles sticking up in the air.
Just some more images. This is what the slag
looks like unleached. There's pH of 5. That's been leached
in deionized water on the glasses, dissolving a little bit,
leaving zirconolites.
And so, finally, we can compare this data -- glass
that has been unleached and glass that has been leached.
Now, this is a little confusing. The two graphs on the left
are weight percent oxide versus distance across a thin
section, and they had two graphs on the right. I plotted
weight percent oxide versus pH, okay? So, on the first one,
really, the unleached, I'm just showing this to show that
there is not a whole lot of variation between -- of any one
particular element: silica, aluminum, calcium, cerium.
Thorium varies all over the place. Uranium is fairly
constant.
But you get to the leached glass, I mean, that is
that -- at the middle pHs, it's fairly constant, but at a pH
of 4 or below, calcium decreases; silica actually increases.
Aluminum decreases as well. But the key thing here is that
at about a pH of 4 and below, thorium decreases rapidly, and
uranium -- we lose uranium as well.
MR. LEVENSON: Is there any explanation for the
spiked behavior of the zirconium?
MS. VEBLEN: Zirconium? I think to some extent,
it depends where you are, how far away from a crystal, a
crystalline, the crystal you are in the slag, that what
happens if a gehlenite was growing in the slag, it doesn't
take the zirconium; it doesn't take uranium or thorium. So
it tends to exclude those as it crystallizes, and you get
this huge chemical profile building up in front.
So if I were a couple of microns like right next
to the grain boundary, there might be a lot of zirconium in
that particular glass, whereas, if I was further out in the
glass away from the crystal, it may be slightly less, so
it's just --
MR. LEVENSON: And it doesn't happen to any of
these other elements?
MS. VEBLEN: Well --
MR. LEVENSON: Zirconium seems to be unusual in
its behavior there.
MS. VEBLEN: What does?
MR. LEVENSON: Zirconium.
MS. VEBLEN: I'm sorry; zirconium.
MR. LEVENSON: Yes.
MS. VEBLEN: Yes; I don't know; I can't really
say. I mean, this is my guess about what's going on with
it, but I think it's essentially exclusion of these elements
from crystals as the crystal is growing.
DR. WYMER: It seems to be pH dependent.
MS. VEBLEN: It's definitely pH dependent.
DR. WYMER: Which would argue against it being an
exclusion mechanism, don't you think?
MS. VEBLEN: Yes; okay.
So, moving on from there, then, onto
archaeological slags, we've decided to look at these because
they've been smelting tin for 2,000 to 3,000 years, ever
since the onset of the bronze age. So it gives us a good
time scale of interest. For SDMP sites, we're interested in
a 1,000 year period of time, and how did we identify
analogous slags? We looked at bulk chemistry. We do have
tin slags at one of the sites, and so, tin slags were an
obvious thing to look for. We also looked for similarity in
crystalline phase, such as spinel zirolovines, and we looked
for similarity in glass chemistry, because we know from the
literature that glass is not glass is not glass; that cerium
glasses corrode or weather more readily than calcium and
much more readily than titanium glasses, so that if we look
at a particular chemistry, if we find a glass that has a
chemistry similar to these slags, we could use that as an
idea.
Then, okay, so, that's how we've selected sites.
We're in the process of identifying the phases in these
archaeological slags. They weren't all that easy to find in
the first place, and we're identifying alteration; and then,
finally, we go on to quantification of the alteration.
These are two of the slag sites that we studied.
The first one is a tin slag in Cornwall that's roughly 1,000
years old. This is the site of an archaeological dig that
was being studied by Bradford University in the UK, and they
happened to come across slags, and they went and dated it
with some carbon dating, so they have a pretty good idea of
what the date is. Hard to find here, because this area is
highly vegetated. It's very damp, and any slag piles that
are there were there 1,000 years ago most likely are in
somebody's back yard right now.
The second area we went was to Pribram in the
Czech Republic, and it's a site of lead and silver mining
and smelting. They have huge piles there. They've been
doing this for on the order of 500 years, and the piles are
very well dated. There's a student at Charles University
who is studying the slags over there, the mineralogy, so we
thought that would be a good jumping off point. It turns
out those slags aren't quite as similar as we thought they
would be, so we're not looking at those in much detail.
We are looking at Malaysian tin slags. They're at
least 50 years old; have the same phases as the slags we're
looking at. We're also looking at numerous Cornish tin
slags, and they vary in age, as I said, from 50 to 1,000
years. This is just an example of the type of alteration
that we look for. This is a copper slag from Cyprus that's
approximately 1,000 years old, and what you see are these
beautiful uhedral crystals of olivine and black in between,
and the black is where the glass would have been. However,
in this sample, it's no longer glass, but it's
oxyhydroxides; it's devitrified glass, and that's due to
oxidation and devitrification.
We also ran an XRD analysis of this particular
slag, and there is no glass, typical glass hunk that you
would see in an XRD spectrum.
Other types of evidence of alteration, we see
evidence of oxidation and alteration around fractures. This
is a nice wide fracture going down through this particular
slag. Glass away from the fracture is clear, but as you get
closer to the fracture, we see this bright oxidation zone,
and then, finally, perhaps, some alteration and leaching
occurring in this area.
Another thing we look for is secondary minerals.
I showed you some growing on the surface of the other slags.
This is an example of secondary minerals growing in a
vesicle in a basalt. We also look for fluid pathways that
are provided by cooling cracks or fractures, and we do see a
lot of alteration, in some cases, along these. This is
actually -- the lower example is one of the SDMP slags, and
we see alteration along this fracture, indicating that there
is fluid movement or some type of movement through the
fractures.
And the archaeological slags aren't perfect. We
certainly have numerous uncertainties. We don't know what
the original composition of the slag was, and we don't know
the original uranium and thorium content. We have found
slags that are still radioactive, and we have found a lot
that aren't, so we're in the process of doing that.
As I mentioned, one of the problems with tin
slags, if you try to look for really ancient ones, we
haven't found them. We've talked to archeometallurgists,
and for the most part, they find only very small soil size
fragments. The question I have is, well, is that because
the slag has all weathered, and we're left with these
resistant phases, or is it because the Cornish and the Turks
were so smart that they knew there was more that they could
get out of it, and they ground it up and reused it?
And the archaeologists think one thing, and I
don't know. So, that's where we are on that.
Okay; just an example of SYNROCK. This is taken
from Smith, et al. and a scientific basis for nuclear waste
management. It's an example of SYNROCK, which is, as I
mentioned before, a synthetic ceramic waste form that's
being considered for high-level waste in other countries and
for the Hanford tank wastes and weapons plutonium here.
Notice the phases P is perovskite; we've seen that in our
SDMP slags. R is rutile, which is a titanium oxide that
actually can be a weathering product of perovskite, and Z is
zirconolite. We've seen that also. H is a hollandite,
which is a barium aluminate in this case, and M stands for
metallic leads, pieces of metal also known as pearls.
These investigators leached this particular SYNROC
for several days at 90 degrees centigrade and found a leach
rate of 4 x 10-3 grams per meter squared per day, which they
found, then, to be the same leach rate or very similar to
the leach rate of flow-through perovskite, so that told them
that what was happening here is the perovskite is weathering
preferentially.
We know that perovskite is not stable at near
surface conditions. All of these phases are phases that
have formed at high temperature, and they tend to weather
when they're in oxidized conditions and also at low
temperature. Perovskite is one mineral that goes from a
calcium-titanium oxide to the calcium weathers out, and it
forms -- releases calcium and forms anatase or rutile, which
is a titanium oxide.
One of the other things I didn't mention is that
these slags were produced under extremely reducing
conditions, and we notice that as soon as we cut them up,
they begin to oxidize. So, that may be good, or that may be
bad. We may find that initial oxidation will release
uranium, for instance, or certain elements, but since there
are enough reducing materials, then, they have precipitated
out or remained behind in stable phases.
All right; so, conclusions: the characterization:
we've used many different techniques. I would also mention
that we've done, on one and several of the slags, BET
analyses to determine surface area and permeability of these
slags, but I haven't done them on all of them so that was
another thing I'd like to do. And we found that the slag in
some of these cases, whether it's by dissolution of the
glass, and there's also preferred dissolution of spinels and
perovskites along grain boundaries.
Chlorine is found in these slides and glass.
Perovskite storianite, which is a thorium oxide, thorium
dioxide, and pyrachlora, which is a calzertite; it's a
thorium, well, anyway, you've got it in there, in your list
of minerals.
[Laughter.]
MS. VEBLEN: And as I mentioned before, glass and
perovskite are not stable under near surface conditions, so
we'd expect those to be some of the first to weather out,
and I've said here that although thorium is present in
several stable phases, such as thorianite, it is also
present in glass and these unstable phases, and as that
dissolves, there is a question as to what happens to the
thorium. We don't know yet. But we do know that the
dissolution of the glass and other unstable phases provides
fluid pathways for water or different fluids to get deep
within a slag.
Okay; uranium, then, is found in calzertite,
pyrochlore, perovskite and glass, and it does appear to be
leaching from the glass at a rate of -- and I've got a
question mark. We don't really know what that rate is just
yet. We haven't finished tackling it yet. We have done
characterization of the solid phases of the PNL leach tests,
and one of the things that PNL found was that they saw
calcium and aluminum coming out at rates that they --
calcium, specifically, they said it's not a calcium
carbonate; what is it?
Well, it turns out it's the hibonite, which is a
calcium aluminate. The calcium is leaching from that. So,
by studying the solid phases, we can help to identify
controlling phases for solubility calculations, calculations
that might be done, an EQ36 type of calculation, where we
look at solubility-controlled leaching or
solubility-controlled uranium and thorium release. So, we
know, for instance, we could calculate the solubility of
these different ones: perovskite, calzortite, pyrochlore,
we put them into a code like RESRAD or a code like EQ36.
So, we're back to so what? What's the bottom
line? Well, what is a slag? Well, I think we've taken a
look at these tin slags, and we have an idea of what they
are now, and we've looked at the slags; we see that they're,
you know, a certain number of phases that are in there.
They're not just glass, and they're not just crystalline,
but it's a mixture of both.
What's in the slag? Multiple phases. Where is
the uranium and thorium? Well, it's not evenly distributed.
It's in discrete phases: perovskite, calzertite, glass,
pyrochlore, thorianite. So, the release of that will be
determined by the rate at which those minerals degrade.
What's the leach rate for a slag, and does it change with
time? Well, we're beginning to determine that for
archaeological slags. My guess is sure, it changes with
time, but I don't really have an answer for that.
What standard tests should we use, if any? And I
think that this research will help us discuss this with NMSS
and the center at PNL, all groups that have been involved in
the study, and hopefully, we'll be able to provide NMSS with
some useful information on that.
There are 17 SDMP sites. Are they all the same?
No, they're not all the same, but we have managed to
characterize the slags by their bulk chemical analysis.
Now, this is very similar to what a petrologist does, when
they go collect a rock, crush it, get a bulk chemical
analysis, and because they've studied so much about a
particular rock like a basalt and know what phases are in
it, and based on the chemistry from many, many previous
studies can then go in and characterize an unknown rock
based on its chemical analysis, and I think we're beginning
to see that with these slags.
For instance, I took the bulk chemical analyses
that we did; XRF data, plotted them up on a ternary diagram
with the calcium aluminum, and a thalmycie within certain
areas. So, from the numerous slags, we can say well, it
looks like we've got two or three different types, and if
someone else were to come in with a chemical analysis, we
could put it on a plat and say oh, well, it's probably a tin
slag, or it's probably this kind of slag or whatever, and
these are the phases that might be in it. So that's kind of
where we're heading with that.
And how does the slag weather? You've seen a
grain boundary diffusion, glass to solution and preferential
leaching of certain phases. Future work? There is still a
lot to be done. I don't think we've answered any of the
questions, all of them to my satisfaction. Let's see: we
still have to calculate some estimated leach rates, and
that, I will be doing. We'll be applying those leach rates
to RESRAD, and I've been running the RESRAD codes for some
of these sites, and we'll continue with the microprobe
analyses on the elemental variation of the slags, both
weathered and the archaeological slags, and yes, then, Dori
will be continuing this study on archaeological slags for
her Ph.D. dissertation at Johns Hopkins University, which,
yes, she'll be doing on her own.
And then, finally, we can determine long-term
alteration mechanisms, and this is a very interesting area
of research, I think, applicable not only to the slags but
to any soils or other solids that contain uranium or thorium
or elements of concern. So, I think that's all I have to
say. I'd be glad to entertain questions.
DR. WYMER: Well, thank you very much for that
presentation. There's some nice research that you're
carrying on or you and other people.
MS. VEBLEN: Thank you.
DR. WYMER: I have two observations. I don't
expect you to be able to respond to them necessarily, but I
want to say them anyway. One is while I think it is very
unlikely that SYNROC will be used to fix the Hanford tank
waste site, they won't vitrify it, and I think some of your
results lead to results in some of these phases would be
directly applicable to leaching of some of the vitrified
waste from the Hanford tanks, because I recognize a number
of the phases as being similar to those that are found in
the vitrified waste. So that's one thing.
The second point is that this slag is in amount,
while it may be in an absolute amount like the large, it's
trivial compared, of course, to the flash from coal fire and
steam plants which also have uranium and thorium in their
decay products, and, of course, there is no mandate to do
anything about --
MS. VEBLEN: Right.
DR. WYMER: -- those, and it's sort of like
sticking a Band-Aid on when what you need is a tourniquet,
so that was my second observation.
Let me ask if there are any comments.
CHAIRMAN GARRICK: The last time you were here,
and you were talking about future research, you did mention
microbial action and a great deal of interest in knowing
what the impact would be on degradation and stability. Have
you anything to report on that?
MS. VEBLEN: No, other than we see evidence of it,
but we haven't identified what it might be.
CHAIRMAN GARRICK: Do you expect -- has there been
any analysis that would indicate what the expectations might
be?
MS. VEBLEN: For the slags?
CHAIRMAN GARRICK: Yes.
MS. VEBLEN: I just could go from literature, you
know, what we might be seeing, but I really don't -- can't
say. I would love to get further into that, but we really
are at the point where we have to get moving further on the
TEM stuff, and that's where we could start analyzing that a
little more.
CHAIRMAN GARRICK: Okay.
MS. VEBLEN: It's taken a long time. As I
mentioned, the phases were fairly unusual, and we analyzed
for a lot of elements, so the analysis, the microprobe
analysis, has been very difficult. I thought oh, when I
first, you know, started working on this problem, I thought
oh, they're like basalts; it'll be easy, you know. We'll
have some beans, this, that, no; they're really different,
and it's been quite a learning experience for me, certainly.
CHAIRMAN GARRICK: Yes; you had mentioned the last
time trying to get some indication of whether microbial
action would precipitate out some of the --
MS. VEBLEN: Yes, it might.
CHAIRMAN GARRICK: -- toxic substances.
MS. VEBLEN: There is a nice volume of the
material research, not material research, the mineralogical
society of America just did a short course on uranium
minerals, and they had several papers in there on uranium
and thorium microbes that tend to -- in some cases, they
actually help to precipitate out the uranium, thorium, and
other cases. They mobilize it. So it really depends on
which bug is present.
CHAIRMAN GARRICK: Thank you.
DR. HORNBERGER: You say you did see some evidence
of --
MS. VEBLEN: Yes.
DR. HORNBERGER: -- microbially mediated -- these
are what? Weathering etches?
MS. VEBLEN: Well, what we're seeing are very
small precipitates.
DR. HORNBERGER: Precipitates?
MS. VEBLEN: On the surface; we had someone at
Hopkins about a year ago that was a microbial geochemist,
and she took a look at them and said yes, it looks like, you
know, it could be microbial activity, but I haven't honestly
done any more. I'm sorry; I'm just --
DR. HORNBERGER: No, I was just curious. Just
curious. So, you don't know what it is, then, going after?
I mean, typically, I mean, what little I've read of Jill
Banfield's work and stuff --
MS. VEBLEN: Right.
DR. HORNBERGER: -- it appears that the microbes
are going after something in particular phases that are of
interest to them.
MS. VEBLEN: Right; they certainly like elements
that have multivalence states, because they use that.
DR. HORNBERGER: They use that as an energy.
MS. VEBLEN: Exactly, so, you know, certainly,
they could be going after the uranium.
DR. HORNBERGER: Right.
MS. VEBLEN: And it's in the glass where we, you
know, we find these precipitating on the glass. The glass
would be certainly much easier to extract the uranium from
than, for instance, a calzertite, most likely.
DR. HORNBERGER: I assume that all of the leach
tests that are done are done abiotically; is that right?
MS. VEBLEN: I would think so. I would hope that
they were autoclaved or something.
DR. HORNBERGER: Right.
MS. VEBLEN: But I really don't know.
Brett, do you know with the Setter which tests --
I saw him in here. Put him on the spot.
John, do you know?
Nobody knows.
Do you know?
DR. WYMER: Milt?
MR. LEVENSON: One question out of ignorance base.
Are these slags principally from tantalum and tin mining and
so forth, are they typical of what exists at the sites that
have licenses, NRC licenses?
MS. VEBLEN: Well, I think of the 17 sites that
have slags, a good half of them have niobium and tantium
slags there.
MR. LEVENSON: But that's not the basis of the
license, right? The license was for something else at the
same site.
MS. VEBLEN: Well, no, the license was because
their ore material that they used for the smelting contained
uranium and thorium, and it contained uranium and thorium in
levels that exceeded what was allowed at the time. So, 30
years ago, they had to obtain a nuclear materials license.
DR. WYMER: Any other questions? John?
DR. RANDALL: Yes; I know that you've got samples
from some SDMP sites, and you weren't able to get some from
other sites.
MS. VEBLEN: Right.
DR. RANDALL: How does that, not being able to get
some of those other samples, limit your research results?
MS. VEBLEN: Well, I think it would be useful for
NMSS if we had samples of most of the sites, but I know that
there is one site in particular that we can't get a sample
from, but they've described it, and it sounds very much like
the sedimentary reprocessed slag that we found. So, you
know, my guess is that we could perhaps apply that, but
without knowing a chemical composition, you really can't say
a whole lot. You have to have some information on this.
DR. HORNBERGER: It strikes me that you're going
to have a tough time with your archaeological reconstruction
of leach rates, primarily because of the problems that you
already pointed out. You know, if you don't know what you
started with --
MS. VEBLEN: Yes.
DR. HORNBERGER: -- it's hard to say the path that
you've gone on.
Now, I suppose what -- I assume that what you're
going to do is make some assumptions of what you started
with by looking at more recent samples and then
reconstructing that way.
MS. VEBLEN: Yes; there are two ways that we're
thinking of going. We actually have started trying to smelt
some of these things ourselves in the lab, not with a lot of
success, because I've had it up to about 1,550 degrees
centigrade, and it's not melting it yet, so we -- you know,
that's a little bit of an experimentation. But we did
collect ore from the sites in Cornwall, so we can go and
smelt those and see what, you know, if you start with an
original composition like this, what ends up in the slag.
The other thing we can do is get an idea of the
isotopic composition of the original ores that were being
used and make an assumption of how much of that original
uranium and thorium in the ore went into the slag, and that
would be, you know, basically just based on partitioning,
melt versus solid.
DR. HORNBERGER: It also strikes me that you'd
have to make well, I think for example, Michael Velbel's
work on weathering, sort of a whole history in terms of what
leaches out when, and it's, as you know, a very complicated
geochemical problem.
MS. VEBLEN: Yes, that is.
DR. WYMER: If that's all, thank you very much for
an interesting presentation.
MS. VEBLEN: Thank you.
[Pause.]
DR. WYMER: Our next presentation is on historical
case analysis of uranium plume attenuation of uranium plumes
from ore bodies and from contaminated sites, and considering
the wide diversity in the geology and hydrology and in the
types of sources of uranium, the results that you're going
to hear are a little bit remarkable, I think. This
presentation will be given by Dr. Patrick Brady from Sandia.
[Pause.]
DR. WYMER: For those of you I've forgotten, Dr.
Patrick Brady will make this presentation, I think.
DR. BRADY: Yes.
[Laughter.]
DR. BRADY: I'm part of a large group that's
working at Sandia on the prediction of metal sorption in
soils. This is a project that's been funded through the
Nuclear Regulatory Commission for several years. There's a
whole host of people on it, some whose names are written;
some people just recently, because I realized I left them
off.
The important names to remember here or the
important name to remember is Carols Colon. He's my postdoc
who's done a lot of the difficult work here going through
the data. Now, the overall objective of our work is to
follow a semi-classical approach to understanding how plumes
move in the subsurface.
DR. WYMER: Why don't we put the mobile mike on
him?
DR. BRADY: I'd rather stand.
[Pause.]
DR. BRADY: We follow a semi-classical approach,
where we presume that if we understand what happens between
radionuclides and mineral surfaces, we might be able to
predict sorption better in the field. Now, sorption is
critically important because for a lot of the radionuclides
we care about, it's the primary sink. So, in theory, if we
can understand sorption, we can understand a lot of other
things, like how big plumes get and what's the relative risk
they might present.
Now, in the process of looking at the mechanistic
controls on sorption, we also gain clues as to what types of
characterization are needed and what types of remediation
are possible and what kind are not. Now, tomorrow, I'm
going to talk about the mechanistic work we've done with
spectroscopy, with molecular modeling and with performance
assessment code SEDSS to try to take a crack at what are the
possible variations in the parameters that go in one end of
a reaction transport calculation.
What I'm going to do for about the next 30 minutes
here is to focus instead on what nature tells us the answers
have to be in the case of uranium, and the tool I'm going to
use is a historical case analysis. It's one we've found
very useful for doing a couple of things: one, identifying
mechanisms that control transport in the subsurface, and
two, I think you'll find that this looks to be a singularly
compelling way to communicate risk posed by plumes to
stakeholders without presuming a great deal of technical
knowledge.
So, that being said, I've got to point out that
the historical case analysis approach is not original.
We've taken it from Dave Rice at Lawrence Livermore. This
is a top-down approach that looks at plumes and worries
about the mechanisms later, and what the Livermore group did
-- some of you may be familiar with it -- they were funded
by the State of California and a number of other agencies to
look at the benzene plumes that emanated from leaking
underground fuel tanks where no remediation had been done.
Now, several underground plumes were examined, and
the only question that was really asked was how big do the
plumes get, followed by a secondary question, which was are
these plumes stable or not? Now, what they found was
somewhat surprising. They found that the plumes moved --
and I should point out this was after the fuel tank was
removed, and there was no fluid product left. It was just a
dissolved plume. The dissolved plumes tended to move out to
about 200 feet maximum. They would become static, and then,
they would collapse.
Now, as scientists, we look at this and say that
makes perfect sense, because fuel hydrocarbon components are
quite biogradeable. Indigenous microorganisms are very
effective at breaking them down, ultimately, to CO2. What
was striking, though, was how much these plume lengths
tended to cluster, despite the wide variation in hydrologic,
geochemical and microbiological parameters that were
inherent in the data set.
Now, biodegradation was ultimately ascribed to the
plume stasis. I'm going to, in the case of uranium, I'm
going to lump all four of these together under the umbrella
of natural attenuation. These are the processes which tend
to decrease the bioavailable concentrations of a particular
contaminant in the subsurface. Biodegradation seems to do
it for a lot of the fuel hydrocarbons. When I refer to
natural attenuation for uranium, though, I'm referring
primarily to sorption, dilution and formation of mineral
phases, which I forgot to leave in.
All right; the other thing that was striking about
the Livermore study was the impact that it had on
regulators. Almost immediately, the State of California
ceased all active treatment at dissolve phase leaking
underground fuel tanks. That's a $2 billion market. It
just basically vanished in the space of about 6 months. A
majority of the states in the U.S. have followed suit. The
last time I counted, nine months ago, I think it was around
37 states that had monitored natural attenuation as a de
factor presumptive remedy for fuel hydrocarbon plumes.
EPA has subsequently issued monitored natural
attenuation guidelines for contaminants other than fuel
hydrocarbons, well, fuel hydrocarbons, coordinated solvents,
metals and radionuclides as well, and a lot of it all came
from this historical case analysis approach.
Keeping in mind the mechanistic differences
between attenuation mechanisms, biodegradation versus
sorption plus mineral growth, we proceeded on the hypothesis
that we could apply the same approach to the inorganic
contaminants, and this is a hypothetical graph of what we
thought plumes would look like. This was possibly a year
ago. We said, well, lead is a priority pollutant; lead has
very high KDs and is a high solid sorption coefficients that
typically goes very short distances in the environment. We
guessed that if you looked at all of the plumes that one
could find good data for and plot up the number of sites
that were a given length, lead would plot here.
Things which sorbed less effectively, such as
uranium, would spread out. There would be sort of a
chromatographic separation, and we'd see slow movers here;
fast movers out there. One of the upshots of this talk is
going to be that this conceptual model needs some
modification to make it actually explain the data, but this
is what we thought we'd see, and this encapsulates the focus
of our study: uranium, strontium and cesium. The uranium
results, I have today.
What we did was we tried to --
DR. HORNBERGER: Actually, for lead, I mean, that
conceptual model obviously doesn't work as an example,
because as long as you have water movement, you don't have
stability in the sense of being frozen in time.
DR. BRADY: So you're saying that ultimately, this
thing will just start moving along. We don't see many
examples of that, and you'll see for uranium here that we've
got cases where there have been, well, I'm going to come
back to this.
DR. HORNBERGER: Unless you're talking about
precipitating out stable phases, if you're talking -- I
mean, you're talking as if this were just sorption.
DR. BRADY: Yes; let me just -- for lead, lead
tends to form hydroxycarbonates, but as you say, that would
keep on going. Reversible sorption, you would still have an
advancing plume. Irreversible sorption is one of the
largest factors that affects lead. In other words, lead
sticks to the surface; becomes overcoated and stays there.
In effect, it's an insoluble phase that no longer sees the
groundwater that's therefore entered.
Now, I'm not going to talk a whole lot later on
unless people keep asking questions, but irreversible uptake
actually applies to a lot of these things. Uranium is one
of the -- irreversible uptake effects these three more than
uranium, but that does give the otherwise seemingly
incorrect assumption of anchoring plumes. That's what
happens when you have irreversible sorption.
All right; so, what we did was we looked for every
single uranium plume that we could find data on. The data
we looked for was groundwater concentrations, and we
typically found these at the UNTRA sites and at natural
analogue sites, uranium ore bodies, the Oakland natural
reactor and what have you.
I'm going to hit this stuff towards the end, so
I'll come back.
Now, let me give you a little bit of chemical
background on uranium. This is important to look at here,
because it's almost illegible on the handouts that I've
given you because of the printer size. The upshot here is
that oxidized uranium, the most mobile forms of uranium,
tend to sorb right around pH 5 to 8, 5 to 7. Above that pH,
uranium forms carbonate complexes, and it becomes anionic.
Since most of the mineral surfaces are anionic to begin
with, there is an electrostatic repulsion. Hence, there is
manual retardation that occurs that way.
Down here, below pH 5, preuraneal, which is
positively charged; it sees positively charged mineral
surfaces at the low pHs, and there's a repulsion as well.
In other words, the only place where uranium tends to drop
out of oxidizing solutions is right around in here. Under
reducing conditions, uranium forms lots of insoluble phases.
Okay; this came out of the most-recently published
EPA guidelines for Kds for inorganic contaminates, from
King, Krupka, et al. at PNL. This shows the pH. PH is
where Kds are measured from 3 to 10. There's a minimum and
a maximum. Essentially, these folks look at every measured
Kd they could find for uranium, and what they see kind of
follows out, drops out of the speciation diagram I showed in
the previous slide.
Basically, there is maximum sorption about pH 6
and 7. And it drops off at low pH and at high pH. I'm
going to come back to this, but I should emphasize: uranium
is one of the more mobile of the inorganic contaminants.
The anionic contaminates protechnitate and iodide are much
more mobile, but of the cations, uranium tends to move a lot
further and a lot further than things like lead, cadmium or
cesium or strontium.
I just want to briefly point out what the phases
are that uranium shows up as in subsurface. Urananite, the
reduced form, the reduced form of uranium typically goes in
urananite. Pitchblende shows up in some of the ore phases.
Schoepite is a hydrated uraneal oxyhydroxide. It's
theorized that this might limit transport of some uranium at
some of the ore bodies. Let's see; other important ones
here: uranophane, uranium silicate and soddyite are
probably two of the most important of the other solid
phases.
All right; our objective at this point was, again,
to see how big the plumes got, and there were a number of
problems that we had to deal with that added uncertainties
to what we measured. First of all, there's little long-term
monitoring. We'd like to have had -- if we could have had a
time series monitoring such that we could look at a plume as
the source as it emanated from the source; dilution occurred
at the edges; the thing spread out, became static and then
stayed there, collapsed, this whole story would be a lot
clearer. There is not enough data to do that at any single
site that we are aware of.
At every single site, you end up with spotty
monitoring oil locations. The DOE well, it seems like
there's almost a three-strike rule. They analyze three
times, and then, they either yank the well or lose the
location. So we never have the perfect site to say how the
typical plume goes. So when we tried to determine what the
life cycle of a uranium plume is, we're limited.
Another one of our big problems is rivers. A lot
of the data set comes from the UMTER sites. Most of the
UMTER sites are very close to rivers. Sometimes, this
truncates our plumes. Our ultimate objective is to be able
to give some idea as to how far dissolved uranium is going
to move from a point source. Well, if a river truncates
your plumes, you really don't get a whole lot of useful
information. But it turns out that there are only a couple
of sites where this is a problem, and our primary friend
here was the fact that in the west, where most of the UMTER
sites are, not all of the streams are gaining. Some are
losing; in other words, the plumes don't always, by default,
go right into the rivers that are adjacent to them. Quite
often, they go parallel; sometimes they go away.
So, some of our data, it's an annoyance rather
than an obstacle. This is something we knew going in. The
geologies, the hydrologic parameters, permeabilities,
hydraulic gradients are all going to be vastly different for
all of the sites. We're using both the milltailing sites,
DOE plumes as well as the natural analogue sites. These
things are geologically quite different. Their ages vary by
several orders of magnitude.
Now, keeping that in mind, we wanted to see if
there were some general features that described all of them.
And the way we measured them was we looked at the 10 to 20
part per billion contour, and where we found a plume, we
assumed this was the plume, and the source was somewhere in
here. The maximum axial difference was the plume length if
the border is defined by the 10 to 20 part per billion
contour.
We tried to err on the side of greater plume
length. Now, the last thing -- this assumption, we assumed
that the plumes were at steady state. This assumption is a
tough one. We only had one site where we had 15 years of
sufficient monitoring data that indicated that the 10 to 20
part per billion contour was not moving. Now, so, when I go
forward, keep in mind this has an asterisk on it, and if
anyone can think of a better way that we can independently
verify that this is true, I'd like to know it.
Let me show you three or four of our sites. This
is typically how it was done. This is a view of the city.
We take the UMTRA report, their contours. That right there
is the -- it's the 10 to the 20 part per billion contour.
The big point to get from this slide is that we're just
taking the maximum value.
I mentioned river truncation being a problem.
Riverton, Wyoming was one of those places that gives us an
anomalously small plume. Note, though, that there is a fair
bit of spreading away from the river, so, you know, it's not
a completely gaining stream.
This is one of the better sites. This is Slick
Rock, Colorado. It was a two-fer. We had two plumes there.
Note that the plumes spread parallel to the river.
All right; in each of these cases, we take that
measurement and then consider them all as a group. I
forgot. I've got to show at least one of the natural
analogue sites. This is from the Alligator Rivers project.
We've -- there is -- this has been funded by NRC for several
years, so there is a great deal of data. There is basically
an ore bodies being weathered; there is a plume that extends
out this way. The maximum axial plume length is on a -- I
don't have it shown here, because it's better seen in plaid.
All right; these are all of the data plotted up in
the histogram fashion. The red ones -- this is the number
of sites. The red ones are the natural analogues. There
are a couple of natural reactors here. We have Pacos de
Caldas, Cigar Lake, Condara. At Pocas de Caldas, those are
all down on this side. Now, over here are all of the UMTRA
sites, and these are the ones that are neither UMTRA sites
or natural analogue sites. These are typically DOE sites:
Weldon Springs, Lawrence Livermore; many others. Fernald
and Hanford are incorrectly put up here. They should be
down here.
I put on a leach from Konigstein, Germany. This
is where sulfuric acid was used to leach out the uranium
inside the aquifer, not on top of the milltailings. Now,
there are bound to be sites that there is good data for that
we have missed. When I spoke to the EPA this morning, they
had a couple of sites that they didn't provide data for. We
expect to get some more data from Savannah River sometime in
the near future. We don't expect this picture to change.
There are a couple of important features about this. First
of all, we don't see that bell-shaped curve like Rice, et
al. at Livermore saw for benzene and the fuel tanks.
Now, part of that is due to the fact that there is
some skewing, really small sites, well, the increment of
measurement is almost half a kilometer here, so if you had a
site that was -- things -- since the increment of
measurement is about half a kilometer, then, if it was a
plume that was 10 meters long, it would get buried in here.
The upshot is we can't see incredibly short plumes.
Let's see; what else is there that's --
DR. WYMER: In each of these cases, the source
stays put.
DR. BRADY: Yes.
DR. WYMER: And in the case of the petroleum
tanks, you took the source away.
DR. BRADY: Yes; but the source has been taken out
of most of the UMTRA sites, too. So, they have shipped away
the milltailings, and we've got fresh recharge going
through, and it's --
DR. WYMER: Well, it's time-dependent.
DR. BRADY: Well, the -- I think part of it is the
semantics of how one defines the source. We would like to
have plumes that came out of single spots and moved, but the
plumes for the UMTRA sites, these are sometimes -- sorry,
the sources, the milltailings piles are sometimes hundreds
of meters across. Now, there are a couple of points about
that. One, it means that these maximum plume lengths often
include the imprint of a factory, and so, again, if we're
searching to find how far is uranium going to move from a
single point source release, it would be a lot less than
this.
All right; going back to, I think, the third or
fourth slide, all of these sites differ greatly in their
hydrologic parameters and the time and extent of source
loading. The natural reactors in Gabon were over a billion
years old. Most of the UMTRA sites, a couple or two or
three decades. The DOE sites were typically 10 to 15 years
old. The hydrologic conductivities, we haven't looked at
the measurements. My guess is they're all over the board.
The fluid chemistry where we can find data, we might be able
to put together a clear picture.
The point here is that although a lot of the input
parameters that go into a classical transport model would
vary by several orders of magnitude, it looks like the plume
lengths seem to cluster. Now, this is -- we think this is
more than fortuitous. I think it suggests that basically,
the uranium chemistry is the more important control. The --
so, at this point, we're basically looking to find any more
data we can to add to this. But in the meantime, it
suggests that plumes for uranium tend to go out to about two
kilometers top and then stop.
The small amounts of data we do have that look at
the temporal movement of the plumes suggests that these
things reach data states in about 5 to 10 years. That comes
from the UMTRA sites. And this suggests that if we are to
consider long-term transport of the uranium, I would argue
for all of the other inorganics as well, we've got to change
the way we model or rather change the way we think of the
inorganic plumes.
This is a classic, the classical approach. A
plume starts at the source. Groundwater flow in that
direction moves it off; dispersion spreads it in a couple of
different directions. From what we can see, for uranium, it
looks more like an ore body case. And sure, the data set
included ore bodies, but if it had just included UMTRA
sites, we would have gotten about the same results.
So, we think that in fact, what these contaminant
plumes are, they're more like ore bodies. There's a
concentrated source; there's a halo that seems to be stable
over time. Again, that's the weak point. We'd like to know
what happens over 10 to 100 to 1,000 years. The gap between
the UMTRA sites and the natural analogues makes this
somewhat difficult to bridge.
All right; lastly, what we're doing right now is
trying to again expand the uranium plume database. We
honestly don't think we're going to find any 10-kilometer
long uranium plumes, and we don't think that's an accident.
Although uranium moves faster and further than a lot of the
other radionuclides, there are substantial chemical
processes that cause its retardation, and I mean retardation
in the biggest sense: the formation of ore minerals,
irreversible reversible sorption, what have you.
We are about neck-deep into doing the same thing
for the strontium and cesium plumes. Now, the time factor
becomes less of an obstacle here, because all of our plumes
occurred in the last 40 to 50 years. I was promised a view
graph for this talk by Dr. Colon, but I never got it. I can
tell you what we've seen so far. It's probably better that
I tell you what we've looked at so far first. We looked at
strontium data from Chalk River and the Canadian program.
We looked at strontium coming out of various low-level waste
facilities.
There is a strontium plume at Brookhaven that
we've got data for, and there's about 10 other ones. The
data is not nearly as good looking as it is for the uranium.
The cesium, we're getting a lot of those analyses from the
Hanford tank farm leaks. And I will quote the folks who do
the monitoring at Hanford and who spotted the, I guess, it
was the last spring or the spring before with I think it was
the cesium that got so much press.
We described what we were doing, and we asked
them, well, how long do you think the plumes get? They said
the strontium, it probably goes 40 meters, the cesium maybe
20. That begs a couple of questions; I've pointed to here
the exceptions. There are exceptions to small plumes for
cesium, and cesium is transported as a colloid quite
frequently. Strontium is not. A lot of the attention has
been paid to the colloids. I think if we look at the great
mass of the cesium data, once we get that done, in 3 months'
time, we're going to see something like this, and then,
we're going to see a bunch of outliers showing colloidal
transfers.
But otherwise, I think we're going to see a much
more compressed plume trajectories for cesium and strontium,
and that's going to be a direct outgrowth of the fact that
both soared much more strongly than uranium; both are taken
up irreversibly much more readily than uranium.
All right; lastly, the references -- the two
Livermore reports, the idea on which this was based, are
listed there in your packet. This is our Webpage. Since
some of the things are illegible on the stuff I handed out,
that will all get posted on our Webpage as soon as I get
back.
In conclusion, if we can confirm that this is all,
in fact, the way uranium plumes work, we think we'll have a
useful tool for considering the potential transport and
potential remediation of uranium. We'd say uranium, the
maximum movement, oh, it's on the order of about 2
kilometers. It's very easy to explain to somebody who lives
4 kilometers away and is worried about the uranium plume
about the level of risk they're exposed to. It's also very
easy to tell somebody he's inside of two kilometers, too,
without a whole lot of extra modeling.
The other aspect of this approach that we think
will be useful is that we believe it drives these
discussions towards the technical realm. If this were the
Hanford plume, and I lived here, I think the argument would
be couched in terms of not DOE's polluting my water, but it
would be more one of what makes my site different than all
of the others? In other words, if we can provide a broad
picture of the natural life cycle of plumes, this might
couch what is and is not a risk somewhat more simply for
stakeholders.
And that's all I have to say.
DR. WYMER: Thank you very much. It's an
encouraging presentation.
DR. BRADY: Yes.
DR. WYMER: I presume with respect to cesium,
since it doesn't form colloids, you're talking about
pseudocolloid transfer.
DR. BRADY: Yes; getting stuck on the sites, yes.
DR. WYMER: Yes.
DR. BRADY: Yes; I'm sorry, but when I mean that,
yes, it's going on the silicate lattice with its --
DR. WYMER: Okay.
DR. BRADY: I suspect that's what happened at
Hanford.
DR. WYMER: That's reasonable.
Are there any questions? John?
CHAIRMAN GARRICK: Well, I was just curious. Has
your work had any impact on the more recent performance
assessment modeling, particularly with respect to waste
package degradation rates, the corrosion model?
DR. BRADY: No, because we've -- this was -- our
deadline was the end of February; correct me if I'm wrong,
Ed, but this has all been done in the last 5 weeks, in
getting the NUREG report done. If the question you're
asking is what does this mean for Yucca Mountain, we haven't
had enough time to think about it. I can sketch what one
would do. You'd go back and compare what were the absolute
masses of uranium and planned for one, observed in the
other, and make some assessment of whether the same process
has prevailed for this suite is likely to occur at Yucca
Mountain, but I haven't done that, because that wasn't part
of our charge.
CHAIRMAN GARRICK: Yes; well, I was just curious,
because when we had our working session on engineered
barriers, we got a considerable amount of information on the
importance of secondary phases with respect to the
solubility of uranium and some of the fission products, and
it sounds like at least with respect to uranium and what it
does in the reducing environment, even though the mountain
is an oxidizing environment, the mechanisms at the
mechanistic level, it's not clear that that couldn't be a
substantially reducing environment, and some of the data
that you have could be kind of interesting in terms of
addressing some of the uncertainties of the effects of these
secondary phases.
DR. BRADY: Yes; I should point out that most of
the -- well, you're right. Typically, uranium is -- it is
more retarded and is rather less mobile in reducing
conditions --
CHAIRMAN GARRICK: Yes.
DR. BRADY: -- and more mobile in oxidizing
conditions.
CHAIRMAN GARRICK: Right.
DR. BRADY: And you're also right; if you look at
our knowledge of the thermodynamics of the various uranium
phases, it's just not where it needs to be. Most if not all
of those sites that I showed are in oxidizing environments.
So one could develop a story from there. I think Cigar Lake
is fairly reducing.
CHAIRMAN GARRICK: Maybe an opportunity.
DR. BRADY: Yes.
DR. WYMER: George?
DR. HORNBERGER: Just, first of all, a comment. I
think that your cartoon where you compared plumes and ore
bodies is a bit misleading.
DR. BRADY: Oh.
DR. HORNBERGER: Because if you think about the
way you defined the plumes, it was with a fixed
concentration, and so, on your top schematic, it's
impossible that that would continue to grow. You don't --
you wouldn't -- you'd be violating conservation of mass. It
might grow for a short while, but then, it has to shrink to
nil.
DR. BRADY: Right.
DR. HORNBERGER: Just because of dispersion.
That's number one.
DR. BRADY: Right.
DR. HORNBERGER: The second thing is that at these
UMTRA sites, as you pointed out early on, you have dilution,
and if you take dilution into account, then surely, your
upper cartoon doesn't hold.
DR. BRADY: Yes; yes; I'll apologize right now for
that being anatomically incorrect.
[Laughter.]
DR. BRADY: Yes; I could have put in all of the
isopacks, but I don't know if I'm addressing your question
there but --
DR. HORNBERGER: It wasn't a question; it was a
comment.
DR. BRADY: Okay; yes, you're right. But the big
point that I wanted to make out of this slide there is that
these plumes tend to get out and stop fairly quickly. When
I think of a plume, I think of something that is -- its
potential for movement is almost unlimited.
Now, these things don't seem to be all that
mobile, national the's the upshot here, and if you use a
straight Kd model like the world uses right now, you will in
fact predict that the remaining concentrations, albeit
lowered, can leave off, and it will be a chromatographic
front. You don't see that.
DR. WYMER: Which makes your analogy with an ore
body pretty sound.
DR. BRADY: Yes, again, keeping in mind that
that's wrong there.
DR. HORNBERGER: And the other, well, partly
comment, partly question, because of that, what you just
said, it strikes me that you have to distinguish here
between dilution on one hand and some form of however you
want to characterize of what you called irreversible
sorption, because you simply need processes that you either
form an insoluble phase, or you sequester a soluble phase
behind an armoring that prevents it from being dissolved.
And you have to distinguish, then, between those two
mechanisms, because I still think that even if you took a Kd
model with dilution, and you defined your plume by a fixed
concentration that you would not predict it going off
forever and ever.
Your prediction would be that as long as you had
the source there, it would be relatively stable, and when
you took the source away, it would just all go away.
DR. BRADY: Yes; well, I guess presumably, you'd
fix the concentration by, say, the presence of schoepite or
uranophane or something like that, okay? In response to the
other part that it would go away, maybe we haven't waited
long enough for the UMTRA sites to go away, because -- and I
kind of alluded to it, you know. We'd really like to know
what's going to happen in 250 years. All accounts right now
say it's not going.
Now, as for determining mechanisms, this came up
with the VA this morning. One could not use a graph like
the one I showed to make site decisions. One could use it
to say this is what we typically expect. And as a property
owner, what I would say is prove it to me. Prove to me that
those mechanisms that you've seen there apply here, whether
it's irreversible sorption measures or a leach test or XRD
at a spot schoepite formation or what have you. So I
absolutely agree with you.
DR. WYMER: Anyone else?
[No response.]
DR. WYMER: Well, thank you very much. I think
that's -- although there's a lot of science yet to be done,
it's an encouraging sort of gross result.
CHAIRMAN GARRICK: We've got a question over here.
MR. LESLIE: Oh, Dr. Wymer, I wanted to actually
address Dr. Garrick's comment. DOE is looking at the Nopaul
I site in terms of using it as a qualitative information for
their license application. They are planning to drill that
site within the next 6 weeks looking for a plume from Pina
Blanca.
DR. BRADY: Very good.
MR. LESLIE: Brett Leslie from the NRC staff.
DR. WYMER: Actually, some of the stuff I've seen
sort of looks like maybe there is some plume information
already available out there. I don't know whether he has
that or not.
DR. BRADY: I can't remember if we had the Nopaul
stuff. If it's not on that graph, we don't have it, but I
know that we looked. Let's see; if I can find the graph --
DR. CAMPBELL: I'll provide you the information.
DR. BRADY: Okay; thanks.
DR. WYMER: Anybody else I missed? Did you want
to say anything?
DR. CAMPBELL: I think going back to Linda's
presentation, the ability to characterize this stuff going
from this macro scale, very large macro scale approach, down
to the microscopic approach, where you can actually identify
particular mineral phases which are taking up uranium or
thorium or whatever you're interested in is potentially a
very powerful tool for establishing a mechanism for the
phenomena that Pat's data to this point seems to be
indicating. I would say that Pat's got to have more
information about ore bodies and other stuff to kind of fill
in the details about uranium and for example, how far do
uranium deposits move with time? Now, I don't have a handle
on that, but they may actually move further than a couple of
kilometers, or maybe all of the uranium is coming from a
halo of within a few kilometers of the ore pocket. That's a
question you might address.
DR. WYMER: I want to support one other thing you
said. I certainly, too, believe that there is a great lack
of good thermodynamic information; that you simply don't
have data that we need to have in order to do the kind of
analysis that we would like to do.
DR. BRADY: I agree.
DR. WYMER: Well, if that's all the questions,
thank you very much.
That's the end of this.
CHAIRMAN GARRICK: All right; the committee has a
great deal of letter work and report work to do, so I think
we're going to take advantage of that time, since there are
no comments or questions, and we will move into a report
writing phase, but before that, we'll declare a break.
[Whereupon, at 2:44 p.m., the meeting was
recessed, to reconvene at 8:30 a.m., Wednesday, March 29,
2000.]
Page Last Reviewed/Updated Monday, October 02, 2017