Advisory Committee on Nuclear Waste Audit Review of Chemistry Issues for the Yucca Mountain Site Considerations Report
Official Transcript of Proceedings
NUCLEAR REGULATORY COMMISSION
Title: Advisory Committee on Nuclear Waste
Audit Review of Chemistry Issues for the
Yucca Mountain Site Recommendation
Docket Number: (not applicable)
Location: Rockville, Maryland
Date: Wednesday, February 21, 2001
Work Order No.: NRC-078 Pages 1-252
NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers
1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005
(202) 234-4433. UNITED STATES OF AMERICA
NUCLEAR REGULATORY COMMITTEE
+ + + + +
ACNW AUDIT REVIEW OF CHEMISTRY ISSUES
FOR THE YUCCA MOUNTAIN SITE RECOMMENDATION
+ + + + +
FEBRUARY 21, 2001
+ + + + +
+ + + + +
The ACNW Audit Review Committee met at the
Nuclear Regulatory Commission, Two White Flint North,
Room T2B1, 11545 Rockville Pike, at 8:30 a.m., Dr.
Raymond G. Wymer, Chairman, presiding.
DR. RAYMOND G. WYMER, Chairman
DR. JAMES CLARKE, Member
DR. PAUL SHEWMON, Member
DR. MARTIN STEINDLER, Member
. ACRS STAFF PRESENT:
DR. ANDREW C. CAMPBELL
DR. TAE AHN
DR. JOHN BRADBURY
DR. RICHARD CODELL
DR. GUSTAVO CRAGNOLINO, CNWRA
DR. BILL DAM
DR. CARL DIBELLA, NWTRB
DR. BRET LESLIE
DR. TIM MCCARTIN
AGENDA ITEM PAGE
Opening Remarks by Chairman Wymer. . . . . . . . . 4
Overview of Waste Package Chemistry Issues
in TSPA. . . . . . . . . . . . . . . . . . .14
Alloy 22 Corrosion . . . . . . . . . . . . . . . .33
Chemical Environment on Waste Package. . . . . . .52
Ti-alloy Corrosion . . . . . . . . . . . . . . . .72
Discussion of Issue Resolution and
Key Concerns . . . . . . . . . . . . . . . 110
Overview of Near-Field Chemistry Issues and. . . 120
TSPA-SR Source-Term Model
In-package Chemistry . . . . . . . . . . . . . . 155
In-Drift Chemical Environment. . . . . . . . . . 166
Discussion of Issue Resolution and
Key Concerns . . . . . . . . . . . . . . . 167
Radionuclide Transport in Near and Far-Field
Environment. . . . . . . . . . . . . . . . 180
Discussion of Issue Resolution and
Key Concerns . . . . . . . . . . . . . . . 210
Discussion of Defense-in-Depth and Multiple
Barriers Issues. . . . . . . . . . . . . . 236
General Discussion and Comments. . . . . . . . . 251
CHAIRMAN WYMER: Let's start. I want to
kick the meeting off by reading some prepared comments
that I have, and that will be the last formal thing
that we will do, I think, and we will get into the
So I will go ahead and read this stuff and
it will take about the amount of time that I have
allotted for it. And part of it is background
material that I think everybody knows, but it is sort
of for the record.
The Yucca Mountain repository site
characterization activities are specified in NC
geological repository regulations. NRC repository
licensing requirements are contained in the proposed
Part 63 of the Code of Federal Register.
And the process that the NRC carries out
is as follows. The NRC strategic planning assumptions
call for early identification and resolution of issues
related to potential licensing of the repository.
Considerable pre-licensing work is carried out by DOE
and NRC both separately and jointly, to identify,
clarify, and resolve issues associated with site
characterization and performance.
To facilitate this process, NRC has
identified what are called Key Technical Issues, and
publishes Issue Resolution Status Reports on the Key
Technical Issues based on an issue resolution process.
The process is carried out through -- and
I call them formal pre-licensing consultations with
DOE. These consultations are required by law and are
carried out in open meetings.
During the consultations, DOE orally
presents information on technical issues to NRC staff
and contractor personnel. The information presented
is supported by DOE technical documents, though not
necessarily at the time of the presentations.
Questions on the presentations are
permitted by the public, as well as by NRC
representatives. At the conclusion of the
presentations, NRC staff and contractor personnel
caucus to discuss the DOE presentations.
The purpose of the caucus is to determine
what, if any, additional information NRC believes is
required from the DOE for NRC to provisionally close
The deliberations of the caucus are
presented by NRC staff to DOE at the time of the
meeting, and DOE responds at that time, either
agreeing to provide additional information, or taking
exception to NRC's requests. This exchange between
NRC staff and DOE is iterative over time. That is,
they do it many times on the same topic.
The DOE Yucca Mountain repository Yucca
Mountain repository safety strategy relies on
engineered and natural barriers, and natural
attenuation -- for example, radioactive decay -- to
contain and isolate the radioactive wastes from the
DOE has identified four waste system
attributes as being most important for predicting
engineered and natural barrier performance. The first
is limited water contacting the waste package. The
second is long waste package lifetime.
The third is slow rate of release of
radionuclides from the waste forms, and the fourth is
concentration of reduction of radionuclides during
transport through engineered and natural barriers.
In this working group, we will address all
four of these attributes to the extent that they are
chemical in nature. The first attribute, limited
water contacting the waste package, is related
chemically to corrosion of the titanium trip shield
which covers the waste packages.
It is also related to climate, to the
design of the repository and to fuel emplacement,
which affect repository temperature and temperature
profiles for hundreds of years, and to a certain
extent the paths followed by water in the repository.
The second attribute, long waste package
lifetime, is related chemically to corrosion of Alloy
22, the outer waste container material. In DOE's
present repository safety conceptualization,this is
the single most important factor in determining
This attribute is also related to
container fabrication, to damage that could be caused
by material falling from the walls of the drifts
containing the waste packages, or to mishandling of
Drifts could be damaged by earthquakes or
by volcanism, as well as by less extreme events, such
as thermal cycles or water damage.
The third attribute, slow rate of release
of radionuclides from the waste forms, is chemical in
nature through the solubility of the waste forms in
the water contacting them, to colloid formation, to
secondary phase formation, to temperature, to redox
reactions, and to the rate of water contact with the
waste forms, which in turn are all related to
corrosion of the drip shield, corrosion of the waste
packages, and corrosion of the cladding, in the case
of spent fuel.
The fourth attribute, concentration
reduction of radionuclides during transport through
engineered and natural barriers, is related to the
chemical species of the radionuclides released from
the waste form and to the chemical nature of the media
through which they move.
In this working group, we will concern
ourselves only with those media within the drifts;
that is, with corrosion products, with organic
material, if any, with the rock beneath the waste
packages, with the inverts and with the waste
If in the future backfill is considered
for the drifts, then these media will also be
important in radionuclide transport.
The purpose of this meeting is to discuss
selected chemical issues in the near-field, and to
reach a consensus among members of the working group
that will lead to a written evaluation of the NRC
staff process and activities in formally resolving
selected parts of the Key Technical Issues related to
the chemistry in the repository near-field.
In addition, the adequacy of the
abstractions of the models used to address the
technical issues into the Total System Performance
Analysis will be addressed, as will the extent to
which the working group believes the NRC requirement
of Defense-in-Depth will be met by DOE.
If deemed appropriate, the working group
will also comment on the degree to which conservative
assumptions challenge the credibility of the analyses
of coupled, thermal, hydrological, and chemical
phenomena in the near-field.
A final point to be addressed is how well
NRC has been able to prepare itself for
contingencies; that is, to prepare for the unexpected,
or to changes or changes in emphasis in the DOE
These goals will be reached in part by
exploring the issues identified in the four attributes
discussed above, by critically examining the
information requested and obtained from the DOE, and
developed by the NRC staff and the Center for Nuclear
Waste Regulatory Analysis, and by providing written
comments based on what we learn.
Now, those are sort of my formal comments.
Let me add to that fact that what we are doing is what
we in the Advisory Committee on Nuclear Waste call a
We are looking at specific key technical
issues. There are too many technical issues for us as
a committee to address all of them. So in order to
assess the process that the staff goes through in
evaluating DOE's proposal, we are taking four.
Each of the four members is taking a
vertical slice of a key technical issue, and our key
technical issue relates to chemistry in the near-
And by a vertical slice, we mean that we
are looking broadly at how things are done in the
process, and in detail, and in this particular case
the chemistry that has been studied, and the chemical
processes that are explored.
So we are looking at the chemistry issues
in the near-field in depth so far as we can, and in
this particular case, the chemistry issues with this
group of consultants with staffer, Andy Campbell.
So that is what we mean by a vertical
slice. So we will not look as a committee, as an ACNW
committee at all of the KTIs, but only a selected few,
and from these, we will try and gain some feeling for
how well the process works, and we will comment on
The members here of this group are Dr.
Paul Shewmon, Dr. Martin Steindler, and Dr. James
Clarke; and Dr. Andrew Campbell is the staff member
who is every bit as important and involved in this as
-- and maybe or probably more so as the rest of us.
So, with that --
DR. CAMPBELL: And more importantly, I am
the designated Federal Official for this meeting,
since we are conducting this as an open meeting. So
now that that is over with --
CHAIRMAN WYMER: Let me say that we could
have held this as a closed meeting since there is only
one member of the ACNW present. We decided not to
hold it as a closed meeting in keeping with the NRC's
policy of openness, and permitting the public and
interim people to come and see what we are doing, and
hear how we do it, and how we go about it.
So we think we are in complete compliance
with all of the Federal Advisory Committee -- what is
DR. CAMPBELL: Federal Advisory Committee
CHAIRMAN WYMER: And I think we are in
agreement with all of the FACA requirements. However,
I want to retain the flavor of a small group of people
discussing openly their opinions.
I hope that nobody in the committee here
is reluctant to someplace along the line take an
extreme position with the expectation that the other
members of the group will beat him back to a more
rational position, because I think that is probably
the way you get at the issues best in this sort of an
So what some of us might say -- although
we are not going to be crazy, but what some of us
might say will not necessarily be what appears in the
final report, but is merely a device, a mechanism, to
more fully explore the issues.
I will permit comments and questions from
the audience. However, in the interest of getting on
with it and the time being so limited with all the
topics that we have to cover, I would ask that those
be held until the end of the day.
We will allow time this afternoon for
comments and questions. However, we would like to
feel that we can draw on the expertise of those
present in the room and at the center when questions
come up that we don't have the answers, and we have
not been able to dig out the answers to, and we think
that somebody else might know the answer, because we
have not been emersed in this for the last 7 or 8
years like some of the people have, and we are not
necessarily as familiar with the details as we would
like to be.
So with that, let me ask Andy if he has
any comments that he wants to make from the staff's
point of view.
DR. CAMPBELL: Just some housekeeping
items. In terms of the meeting, I have not received
-- and I don't think Ray has received -- any requests
by anybody to speak. But if somebody does desire to
speak, contact me or let Ray know.
But let me know and then we will arrange
for some sort of time for you to be able to speak.
But what we prefer to do is to do that at the end of
the day if anybody wants to make any points.
CHAIRMAN WYMER: Right.
DR. CAMPBELL: But certainly we will be
willing to accommodate somebody's schedule, for
DR. CAMPBELL: Yes. If somebody has to
leave early, and it is something that they really feel
should be said, and that they feel strongly about, or
that it is a factual matter that we have gotten wrong,
and they want to put into the record, fine. We will
certainly allow for that. We are not going to be
rigid, but we are going to be firm.
DR. CAMPBELL: Okay. What I have done
-- and I guess I am next on the agenda here -- is the
intent -- of course, the one thing I am missing is a
As I put together some view graphs that
are basically -- and in fact they are just excerpts
out of TSPA or some of the AMRs and PMRs that I
thought illustrated some of the key issues that we are
going to talk about, in terms of waste packages, and
I am sure that Paul is going to have additional things
to say or comments.
But this would be a way of getting
started, and so I am going to hand these handouts out,
but I am going to attempt to do this via Powerpoint,
and we will see how successful or not it is if we do
it that way.
So what I am going to do is that I am just
going to go ahead and go over this. The main point of
this slide is just to make sure that we all know what
the layout of the proposed layout of the drifts are,
in terms of the types of packages, and the kind of
loading that they are doing.
The first package you see there is a PWR
package that contains -- I think the PWRs contain 44
CHAIRMAN WYMER: Lots. No, that is the
DR. CAMPBELL: It is a lot. It is a lot.
But there is a difference between the PWR and BWR
bundles, and it is not 44. The next package are the
co-disposal packages, which consist of stainless steel
flasks that contain high level waste flasks, along
with defense nuclear waste or spent fuel.
These packages are line loaded. I believe
Naval reactor fuel also goes into the repository,
where they put the shipping cask inside the Alloy 22
stainless steel disposal cast.
And then the whole business is covered
over with a titanium drip shield. They are going to
line load these things so that hey are end to end
basically, with a very short distance between waste
DR. STEINDLER: What are they resting on?
DR. CAMPBELL: They rest on a pallet like
device which consists of Alloy 22, which is at the
contact with the waste package, I believe, and
CHAIRMAN WYMER: Is it stainless or
DR. CAMPBELL: It may be carbon steel, but
it is a steel cradle if you will.
CHAIRMAN WYMER: Yes, it's carbon.
DR. CAMPBELL: But my understanding is
that there is going to be at the contacts an Alloy 22
DR. DIBELLA: It says stainless steel here
in the picture.
DR. CAMPBELL: That's what I thought.
DR. DIBELLA: And underneath there is
stainless steel beams.
CHAIRMAN WYMER: Well, let's go ahead.
DR. CRAGNOLINO: The stainless steel tube
supports it underneath?
DR. CAMPBELL: Yes. Underneath that is
the invert, which I will show a picture of in just a
minute, but it is basically -- originally the liners
of the drift -- the drift support was going to be a
concrete liner, and that was in the VA design.
They redesigned the repository after the
viability assessment, and basically what you are
looking at are steel drift supports and rock bolts,
and that sort of thing.
CHAIRMAN WYMER: And grouting.
DR. CAMPBELL: And the kinds of things
that you see in there right now.
CHAIRMAN WYMER: The rock bolts is grouted
DR. CAMPBELL: And they are grouted in,
right. The invert itself originally in the VA design
was a concrete pad, and in this design it basically is
a steel framework that is filled with crushed top rock
from the Yucca Mountain environment.
And the design after VA originally
envisioned putting backfill over the dip shields, but
DOE backed off from that after doing some further
calculations and deciding that the insulating effect
of that backfill would cause the cladding temperatures
to exceed 350 degrees centigrade, which they have set
as an upper limit to preclude creek rupture of the
cladding inside the containers, and inside the
This is a cross-sectional view of how they
envision this will look through time. You have some
rubble, and eventually of course the drift supports
aren't going to prevent everything from caving in, and
so there will be some rockfall around and on top of
the drip shields.
The potential for water dripping in comes
from -- well, the current models look or focus
primarily on the crown drip as a focal point for
Some of the processes that would occur is
in the dripping water you may have colloids, and there
may be chemical properties of this water that are
different from the natural water, either the pour
waters or the percolating water from the surface,
because of interactions with the steel and the
support, or in the rockbolts and the cement, and the
grout and the rockbolts.
That is a possible source of chemistry
changing. The gas content in the drip, they evaluate
not only the water content, but also the oxygen, the
CO2, and the nitrogen.
And the CO2 content, and the oxygen
content, and the water content, are all related to one
another as a function of temperature. The drip shield
is titanium as I pointed out. The emplacement pallet
as we discussed is Alloy 22 and stainless steel.
The invert here underneath the package is
crushed tough. So it is local rock. And that's
pretty much it.
DR. STEINDLER: We expect water to be in
liquid form, but not right away.
CHAIRMAN WYMER: That's right.
DR. CAMPBELL: That's right.
DR. STEINDLER: That looks like a picture
that is what, 5,000 years old?
DR. CAMPBELL: Or much longer frankly, in
terms of water dripping on the waste package itself.
I have a slide that I pulled out earlier on that.
This just shows the details, some of the
details of the design. A key point here is that DOE
has focused on the welding. They have a double-welded
lid, a double-lid closure, and the package itself is
annealed prior to or during the manufacturing process.
So the goal of that is to relieve the
residual stresses created when they put the whole
thing together. But because they are welding the lid
on, there will be stresses associated with these lid
And they have added this double-lid
because of an attempt to have a defense-in-depth type
DR. SHEWMON: Annealed in air, or in
DR. CAMPBELL: I don't know how they are
going to anneal it, and maybe --
DR. SHEWMON: There are some advantages to
letting it oxidize. I just wondered.
DR. CAMPBELL: They are talking about
laser pining as one.
DR. SHEWMON: That is for the weld for
stress relief. I was thinking about the vessel
DR. CAMPBELL: Oh, the whole vessel? I
don't know the details of how they are going to
DR. SHEWMON: A different question while
I have got you interrupted. One place for galvanic
problems is between the titanium and the steel. I
don't see anything about how they are going to
DR. CAMPBELL: Well, it looks like it is
resting on a steel beam, but I don't have the design.
DR. DIBELLA: There is a C-22 shoe on the
bottom of the titanium drip shield.
DR. CAMPBELL: Could you identify
DR. DIBELLA: Carl DiBella, and I am with
DR. CAMPBELL: So there is a C-22 shoe
here, and then this has C-22 here. The point is that
in the DOE models, this is the locus of stress
corrosion cracking in their model, is at the lid,
although what they do is that they treat it a little
differently, in terms of allowing water to come into
the package even through a stress growth cracking.
This basically shows the key flow paths
in this scenario. Go ahead, Martin.
DR. STEINDLER: Let me go back to
chemistry. The sense of the double-weld is only a
delay find rather than an attempt to change chemistry;
is that right? Is that the way that you see it?
DR. CAMPBELL: I think so.
DR. STEINDLER: All I am really asking is
the difference between a single cover and a double
cover is only in the time that it takes to penetrate
the whole thing.
CHAIRMAN WYMER: I think that's right.
DR. STEINDLER: Rather than a shift in
chemistries. So they don't have an interior plate
with a different material?
CHAIRMAN WYMER: No, I think it is just to
make sure that it stays closed.
DR. CRAGNOLINO: Could I make a
clarification? Dr. Gustavo Cragnolino from the
Center. They use to cover in order to have some sort
of remedial action after the water, and the internal
cover after being wet is going to be submitted to a
process of shut pining, doing a reduced compression of
stresses, and removing the potential for cracking the
And while the second cover is going to be
underneath that shield, and I think that is what they
tried to make a distinction in terms of the
construction. And answering to the question of the
initial weld, the weight package, the outer container
of Alloy 22, and --
DR. SHEWMON: Do you think the corrosion
resistance would be better if they did it in air?
CHAIRMAN WYMER: Let's not engage the
audience in a discussion.
DR. CRAGNOLINO: It is only for
CHAIRMAN WYMER: That's fine, and I
appreciate that, and don't hesitate to do that, but I
don't want to get into a discussion.
DR. CRAGNOLINO: Okay.
DR. CAMPBELL: Okay. The key locations,
in terms of the modeling, are the drips, gas and
seepage drips, and so that essentially becomes what
they call location one coming into the drip.
Another key location is the top of the
drip shield, where drips are falling on to the drip
shield. Initially for the first few hundred to few
thousand years, this is hot. It is above boiling for
the few hundred years.
So any water that would come back into the
repository and drip on here would evaporate. And in
fact I will show later that the temperature of this
system is above ambient for many tens of thousands of
And for a long time it is tens of degrees
above ambient. So there is always going to be a
thermal radiant from the fuel rods out.
CHAIRMAN WYMER: You are going to have to
kind of hurry, Andy, because you are running out of
DR. CAMPBELL: Okay. And the next
location is the top of the waste package, and then
location four is the waste forms themselves, and that
includes the cladding. Location five is flow through
the invert into the unsaturated zone underneath.
So those are the basic key modeling points
that they are going to follow, and that we are going
to be talking about in terms of water chemistry.
This is their concept for general
corrosion of the waste package, and general corrosion
that causes failure in catches in the drip shield.
There is humid air general corrosion after the
humidity gets above a certain point.
I think that is around -- anything above
50 percent. Once you get failure of the drip shield,
you can get drips directly falling on the waste
package, and it is generally thought that those drips
would occur at the top of the waste package, or at
least the most damaging ones.
And that's where they tend to model the
formation of these general corrosion patches, and then
you have stress corrosion cracking at the welds around
Again, this is all from TSPA, and this is
their calculated temperature for various infiltration
rates. You can see initially at closure the
temperature goes up to about -- this is at the surface
noise package, to a little less than 180 degrees
And then decays away with time, and so
that by 10,000 years, you are looking at temperatures,
depending upon your infiltration rate, in the
neighborhood of 50 degrees centigrade; and by about a
hundred-thousand years, you have essentially decayed
So even for long after the main thermal
pulse is gone, you still have temperatures that are 10
or 20 degrees above the ambient temperature.
DR. SHEWMON: What are the units on the
various lines? Millimeters per year bin?
DR. CAMPBELL: Well, those are different
-- and I don't have it here, but those are different.
They divide the repository up into different bin
infiltration, and each one has its own flux rate,
depending upon the rock properties above it.
DR. SHEWMON: So is that millimeters of
rock or millimeters of water?
DR. CAMPBELL: No, this is millimeters of
DR. SHEWMON: Okay.
DR. CAMPBELL: So they are modeling a
range of percolation rates.
DR. SHEWMON: I understand. You have
answered the question.
DR. CAMPBELL: This is the cladding
temperature and that is 350 there. So they are trying
to keep the cladding temperature below 350 degrees.
This is what happens to relative humidity
in the repository during the thermal pulse. It goes
way down, and then comes back up. So even at a
thousand years, you are above 80 percent relative
So clearly fairly early on, even while the
packages are still warm, they are accumulating a film
of water of them, but the humidity does not reach a
hundred percent until close to a hundred-thousand
And this shows the percolation flux.
During the thermal pulse, you are heating the rock up,
and the idea of the current design is that the areas
-- the distance between drips is about 80 or 85
meters, or something like that.
And the boiling front only reaches a few
meters into the drip wall. But you are still going to
be moving a fair bit of water around by heating up
that amount of rock, because the rock is about 10
percent by volume of water.
So there is a potential for a percolation
flux during the thermal pulse, and one potential
scenario is a reflux scenario, where some warm water
can come back down through a fracture and come in on
top of the drip shield, and possibly even get on to
the way it is packaged.
DR. STEINDLER: Can you go back to that
relative humidity flux, or is that going to screw
DR. CAMPBELL: Which flux?
DR. STEINDLER: The relative humidity.
That's the one. I guess I am raising the point that
reaching the relative humidity of a hundred percent at
a hundred-thousand years, or whenever, is not
particularly germane to the onset of chemistry.
DR. CAMPBELL: No.
DR. STEINDLER: Because you have got a
significant film, and relative humidity is quite a bit
DR. CAMPBELL: Essentially about 50
DR. STEINDLER: Well, 50 may be a little
low, but certainly a thousand to 5,000 years, you have
got a significant amount of mineral movement in that
thin film that goes on, and there is lots of
experimental evidence that glass, if it were opened,
begins to react pretty thoroughly at those
And so we are talking a potential for
glass reactions, if we are through it, at times that
are less than the compliance time. That was my only
CHAIRMAN WYMER: Yes, and which is germane
DR. STEINDLER: Yes, I thought so.
DR. SHEWMON: When they got started on
this a long time ago, they wanted to design it as a
matter of policy for very hot fuel, but they haven't
been putting it in, and they won't be putting it in
until it is going to be a lot cooler.
Did they take that into account at all, or
is this 300 degrees C limit gotten with very hot fuel?
DR. CAMPBELL: What they are going to do
is blend. They are going to try and fix --
DR. SHEWMON: So do you think that is a
reasonably realistic number?
DR. CAMPBELL: Yes.
DR. SHEWMON: Okay.
DR. CAMPBELL: And that' based upon --
DR. SHEWMON: Well, with the other limit,
it looked pretty cold. You might change in other
directions, which would be more damaging, but go
DR. CAMPBELL: What they plan to do is
create a particular mix, in terms of putting waste
packages -- because they are putting them end-to-end,
they are going to determine the heat output for each
And then they are going to try and blend
it in such a way --
DR. SHEWMON: You have answered the
question. Thank you.
DR. CAMPBELL: -- to get a constant or
relatively even distribution. Oh, let me back up.
CHAIRMAN WYMER: You are out of time,
DR. CAMPBELL: Okay. Well, let me make a
couple of points here. What you see here, these
increases, these are the increases in percolation flux
due to the climate change model that they built into
And basically what they are doing is that
they are modeling an increase around 600 years, in
terms of percolation flux, and then another one at
about 2,000 years.
So they are modeling into this -- and this
is pretty much based upon the Molenkavich cycles,
which are the perturbations in the orbit of the earth,
which are generally thought to control on a large
scale going from glacial to inter-glacial periods.
That in about 2,000 years they are
modeling an increase to a more glacial type of
climate, and that will cause an increase in
percolation flux. And you notice that that is really
going to be the driver, in terms of the water.
This is a short duration event relatively
speaking, and unless they have got this completely
wrong, it is not that different than after about 2,000
years, the high end of the percolation flux.
Now, of course, you can have the lower end
of it, and that is much lower in terms of percolation.
That's unreadable, but that is just a chemistry of
basic major ion chemistry, and some minorized species
for these various periods that they are modeling.
DR. STEINDLER: I think that table we
ultimately need to recover again, because I think
there is an important issue, and that is the little
line that says additional constituents from complex
thermal hydraulic chemical model. It is not used in
the normal calculations.
CHAIRMAN WYMER: That was a point that I
was going to make later.
DR. STEINDLER: That is the one that
contains the fluoride.
CHAIRMAN WYMER: I will make a comment on
that now. As far as I know, the Department of Energy
is using the simple -- or what they call the
simplified model because it seems to agree better
with the experimental information that they have.
Whereas, the complex model doesn't, and to
me that is a poor reason to use it. What they should
do is understand why the complex model doesn't agree
better than the simple model.
And obviously there is something bad or
something wrong with the complex mode.
DR. STEINDLER: Or the experiments are not
CHAIRMAN WYMER: There is always that,
DR. STEINDLER: A simulated J-13 doesn't
do much for me.
CHAIRMAN WYMER: That's true. Okay.
Andy, finish up.
DR. CAMPBELL: This is just their
conception model, and the way they model corrosion is
they develop these for both the drip shield and the
waste package, 300 square centimeter patches, at least
for general corrosion.
So the concept is that you end up with
patches that can allow evaporative water into the
system, and again the same thing for the waste
Again, this is the general corrosion
model, and that is one of a number of figures of
general corrosion rates that are in the TSPA model.
This is a CDF Alloy 22, with various variabilities.
And you are looking at corrosion rates
that are between 10 to the minus 5, and 10 to the
minus 4 millimeters per year on that scale, and that
is for general corrosion.
And this is unreadable to everybody, but
you have a copy on the last page, and these are just
percent of packages breached. It is for the drip
shield and the number of patch breaches per failed
drip shield, and percent of waste packages, patch
breaches, per failed waste package, and which is some
measure of how many of these packages are failed. I
don't have anything else --
DR. STEINDLER: But that is not a
chemistry issue is it?
DR. CAMPBELL: No, this is a TSPA. This
is an output from TSPA.
DR. STEINDLER: But what I am saying is
that package failure per se does not attack
immediately the question of what chemistry is inside
the waste form. It is still on the outside.
CHAIRMAN WYMER: That's right.
DR. SHEWMON: Some people could say the
corrosion of the metal is a chemical question.
CHAIRMAN WYMER: I would say that. Okay.
I think we are up to Paul here aren't we?
DR. SHEWMON: Okay. Well, a little
philosophical comment at the start there; that if you
pile enough conservative assumptions end-to-end, you
can get an unreasonable answer.
Let me start with a story, a story about
a King of India 1,600 years ago. He was a dutiful
son, and he wanted to honor his father. They had a
lot of good metallurgists around. So he put them all
to work making a monument for his father.
And this turned out to be a 20 meter long,
6 ton, bar of wrought iron. And they erected it
outside of what is now Delhi, and has been there
standing in the weather, monsoons every year, and a
certain number of holy cows going by doing what holy
cows do, a lot of dust.
And the column apparently hasn't rusted,
and it has a patina on the outside of what a
metallurgist might call coherent oxide, and it hasn't
corroded like anything the rate that the models used
in these documents that we are looking at would assume
corrosion has occurred.
And so after 1,600 years the corrosion is
quite slow. If we go back to under repository
conditions, which were above water, and not too humid,
a corrosion rate of Alloy 22 should be much slower
than that of the Delhi column.
Yet the corrosion rate should be
unmeasurably and unbearably slow, and yet the
conservative DOE assumptions are that it goes at a
rate which is appreciably faster than that of the
wrought iron, at a rate characteristic of corrosion
dissolution, driven by an applied voltage in deaerated
salt water, strongly acidified.
I think that this is a mistake and that
the engineering column is a good engineering analog,
to use a phrase which is the DOE's, and that I will
come back to later.
Now, to do something to find something to
measure, the research people must find ways to compare
alloys in time in which they can get results for their
quarterly reports, and so they do this by devising
tests under aggressive conditions which give
dissolution, or cracking, or failure.
This stress corrosion test require a
stress tending to pull the crack open. That is, there
is an active stress in it, and in an aggressive
And the aggressive environment they used
in this sharply cracked, highly stressed, sample is to
make it quite acid, dew point 7, and they put it in a
boiling magnesium chloride and water mixture, about
120 degrees C, and they can crack 316 in this, but
they can't crack the 3-C-22. So they know that the C-
22 is more resistant.
However, neither the stainless nor the C-
22 shows cracking in hot acidified sodium chloride or
lithium chloride solutions. And you know from our
last meeting that if you raise the pressure, lower the
Ph, add lead ions, and try like heck, you can indeed
get stress corrosion cracking in the stainless C-22
I personally have difficulty seeing the
relation of this, if any, to the performance of the
waste containers in air in Yucca Mountain, but let's
Crevice corrosion. Crevices often
generate a more corrosive environment than flat
surfaces. Differential aeration under water will give
an anode or dissolution in a lower oxygen region of
the alloy leading to localized attack.
Chloride ion accentuates this in the
nickel-chrome-iron-moly alloys of concern here.
The simulation test used here is to put two teflon
washers in a deaerated chloride solution, heat the
solution, apply a voltage until dissolution or
corrosion occurs in the crevice.
The voltage is then reduced until
passivation occurs. The minimum chloride concentration
and voltage required for crevice corrosion to be
initiated at a given temperature, like 95 degrees C,
is higher for Alloy 22 than 316.
That is, it is more corrosion, or it is
more resistant to crevice corrosion. Alloy 22
exhibits passive behavior over a wide range of
voltages, chloride ion concentration, and Ph in
There is no evidence of localized
corrosion was detected. These are in experiments done
by Cragnolino. If the average corrosion current was
measured with a steady applied voltage, and
circulation to carry away the ions, the authors say
the corrosion rate corresponds to a lifetime of 60 to
That is to remove the two millimeters
thick wall of Alloy 22, and that is where these
corrosion rates are obtained that we see, and led to
the last perc that Andy showed. But that is in a cell
with a voltage, and with extreme conditions, always
under water, and always with a voltage applied.
This is not a best estimate, but a minimum
estimate. The rate is increased by the reduced
oxygen, the flowing salt water to remove ions, and the
Thus, a more realistic life, which
actually will allow dry-out periodically, and didn't
have all these things on it, would probably lead to a
life that was appreciably longer.
Conclusions. I believe the DOE and NRC
staffs have no sound scientific basis for their
predictions of the rates of the corrosion of the waste
package in the Yucca Mountain repository, and as a
result have grossly overestimated the corrosion rates
of the waste package.
The DOE claims, quote -- and this was in
a letter that Andy sent us recently -- there is no
information or analogs that exist on the performance
of engineered materials for the necessary time frames
of performance for the waste package.
I disagree and would suggest that
conditions of the metallic iron nickel meteorites
buried for a hundred-thousand to a hundred-million
years in dry soil, or in Kansas or Iowa, which is not
And exhibit negligible corrosion provides
a very good analog and strongly indicates that the
rates of corrosions the DOE and the NRC staff have
assumed for the waste package are too high by many
orders of magnitude. Let's say a thousand for a round
The waste package should remain tight for
at least a million years in Yucca Mountain, and this
is without a drip shield. Now, the use of such a
meteorite analog would be more accurate, yet
conservative, since these iron nickel alloys would
corrode much faster than Alloy 22 in the aggressive
test DOE is using for guidance now.
Many of the Iron Nickel fragments are from
inches to feet in size, and have been recovered from
sites all over the world, and I have listed some
Some of these are wet and some are dry.
I can see no use of such information in the reports
put out by either of the groups we are reviewing. So
I think what they have done by taking these cell
measurements is put themselves into a world where they
feel that they have to dissolve it somehow, and then
they take that as the minimum rate for what they think
they will find in nature. End of report.
CHAIRMAN WYMER: Okay. Let's talk about
that. You think that is still true in the case of
trace catalytic materials?
DR. SHEWMON: Well, there is lots of
traces in wherever these things are buried. I mean,
I don't know what meteor crater in Arizona is like,
but the way the rainfall is, is probably the same or
about as Yucca Mountain.
I wouldn't be surprised if what it is
sitting in is about the same as Yucca Mountain, though
I don't know the chemistry.
CHAIRMAN WYMER: So we would really have
to know those things in order to make more than a
qualitative comment about it?
DR. SHEWMON: You would, but there have
been meteorites dug up that have been there for -- my
Britannica, which was my reference on this -- a
hundred-thousand years to a hundred-million years.
And some of these are in dry places, like
Northern Australia or Southern United States, but some
of them are pretty wet places.
CHAIRMAN WYMER: Well, they say -- DOE
says that they take these very conservative positions
and they assume these corrosion rates, which are much
higher than they should be, and they still come out
So do you think that there is perhaps a
loss of credibility or believability in some sense
because they are taking such an extremely conservative
DR. SHEWMON: Yes.
CHAIRMAN WYMER: And is this a negative or
DR. SHEWMON: Well, I don't know. I think
it is just wrong. It is whatever you want to use is
a more polite way of saying that it is conservative.
But that was the point of my opening
statement, which is that by striving always for
conservative answers, in a chain-of-events, one can
pile these on top of each other and find very
unreasonable and unrealistic estimates.
CHAIRMAN WYMER: And that is one of the
comments that -- for example, the advisory board has
said that they haven't really come to grips with
quantitatively evaluating the conservatism that they
have in there in their system.
They have not really added it all together
in a way that is understandable and credible.
DR. SHEWMON: In the reactor business, you
try to get them to say best estimates, and nobody says
best estimates here.
CHAIRMAN WYMER: And that is sort of a
philosophical point of view as you said at the
beginning, and something that has bothered me is that
they keep coming out with the conclusion -- and
probably right -- that the Alloy 22 will last a long,
long time, and that corrosion will not be a big
problem within 10,000 or maybe a hundred-thousand
But by taking these very conservative
positions, they sacrifice something in believability,
and I have not been able to decide whether I like that
or not. I would like them to do a very realistic
evaluation, and the best evaluation that they can, as
it would be more believable.
And if they came up with a hundred-
thousand years for the lifetime, then fine. That's
sort of a million years. Great. But to come out with
11,000 years, which is only a thousand years over the
10,000 year limit, is somehow -- it sort of rubs me
the wrong way.
DR. SHEWMON: And I don't know what the
committee is going to do with this, or what they can
do with it, since DOE is the one that is supposed to
design it, and NRC comments, and you comment to NRC.
DR. STEINDLER: But therein exactly lies
the problem. I don't recall reading -- and I don't
claim to have read everything that NRC has written,
but I don't recall NRC coming up with or the staff
coming up with the same kind of general comments,
saying to DOE that you guys have lost your mind.
You are way over-conservative on the
lifetime of the metallic barriers. Now, maybe that is
not their function. The other end of this thing is --
CHAIRMAN WYMER: And I suspect it's not.
DR. STEINDLER: -- that if the
conservatism is adequate to meet whatever reasonable
assurance ground rules that are used, then you don't
But I have another question, and that is
whether is it feasible on the basis of experience to
devise a catalytic corrosion process that would be
accelerated greatly over what you have just indicated,
and would those catalytic components have any chance
of being in the drip water, which is moderately ill-
defined as far as I can tell?
DR. SHEWMON: I think the catalytic
components, we wouldn't use those words, but chloride
would serve the function of accelerating the process.
I guess you could call it a catalyst. It doesn't
change its nature.
Fluoride is even worse. You will get to
that perhaps in the titanium part, where it comes up
DR. STEINDLER: But let's focus in on
Alloy 22. Well, besides the chloride, which we can
argue about, depending on the concentration --
CHAIRMAN WYMER: Lead is the other one.
DR. STEINDLER: That's exactly the point.
DR. SHEWMON: Lead is the other one that
comes up, but where they -- and I don't know what
experiments with people like Gustavo are going to do
with lead in place of the normal things.
But what they had for accelerator tests
were such gosh awful pressures and temperatures, and
so on, that --
CHAIRMAN WYMER: The question that you ask
there, Paul, I think is what is the trade-off for
extreme conditions for a short term test, and much
milder conditions for a lot longer period of time, and
that's what you have to try to get at.
DR. STEINDLER: It depends on how smart
you are about the mechanism.
CHAIRMAN WYMER: Yes, and it does come
down to mechanisms in my book always, and that is
something that the Nuclear Waste Technical Advisory
Board point they came up and that I agreed with, was
that as best you can, you should determine what the
mechanisms are because its only when you understand
the mechanisms that you can extrapolate with
confidence for the future.
And having said that, I will also say that
I know that mechanisms in things like corrosions are
extremely difficult to determine, the true fundamental
DR. STEINDLER: But the point that I am
trying to make is -- and not very well, I guess --
that while it may seem on the surface that one can
throw rocks at DOE and hence the NRC staff for not
objecting to this extreme conservatism, I wonder
whether the uncertainty in relation to things like
led, for example, isn't sufficiently large so that the
conservatism used by DOE, and apparently accepted by
NRC, is okay in terms of reasonable assurance.
DR. SHEWMON: But if you do that, what
they have done is sort of put their head in a sack and
-- or put a hand in the sack and pulled out a number,
and said, gee, it is conservative, and so that must be
Maybe there is things that we don't know
that I guess I sort of gave up the boogie man thesis
some time back, and don't like to see it used here.
DR. STEINDLER: Well, corrosion is not my
strong suit, and so I can't argue with you too
successfully. I guess all I am trying to do is to
defend, if that is a necessary term, defend the NRC
staff from the charge that you guys have let the
ultraconservatism of DOE slide past without objecting
DR. SHEWMON: It seems to me that the
meteorite thing --
CHAIRMAN WYMER: I don't know that they
DR. SHEWMON: The meteorite thing has been
in a variety of environments. I asked Andy about it,
and he said, well, people have found smaller particles
of this stuff in luvial mixtures and they are very
Now, relatively to what, I guess, but at
least they are still there, and they haven't gone
DR. STEINDLER: Well, the trouble with
those things is that you don't know what their
conditions are that they have been subjected to over
DR. SHEWMON: But they dig these things up
out of the ground, and in a wide variety of places.
DR. CAMPBELL: Has there ever been -- and
are you aware of studies of corrosion rates of iron
nickel meteorite fragments. I mean, that seems to me
what you are saying here, is that that is a natural
analog study that probably ought to be done.
DR. SHEWMON: No. But people have tramped
around in the iron nickel alloy systems for a
generation or two, and I am sure if there are more
corrosion resistant than Alloy C-22 or even 316,
stainless steel, we would have learned it.
It is not a novel system, and so it is
more corrosion resistance than carbon steel, and
probably more than wrought iron, but certainly not in
chrome bearing high nickel alloys. That I am sure of.
CHAIRMAN WYMER: Well, DOE is in the
unfortunate position of having to back up as best they
can technically anything they say. You can sort of
make these -- and pardon the expression -- handwaving
arguments about, gee, this stuff is really corrosion
And it has been there a long, long time,
and it is similar material. But that doesn't cut it
as far as providing something in a document that they
can support scientifically and credibly to the
DR. SHEWMON: They had booked for
engineering analogs that have been in business for
this long, and I think the meteorites are a superb
one. It is conservative, and it would corrode faster,
and it has been there for a hundred-million years in
CHAIRMAN WYMER: But what they don't know
though is what it would do under the controlled
conditions that they try to run these experiments at
relating to what they expect the repository conditions
DR. SHEWMON: Well, what they run the
experiments at is not what they would find in a
repository. It is not in an electrohooded cell and
circulating in acidic chloride solutions, and that is
what bothers me.
CHAIRMAN WYMER: Well, my impression of
that was that they were trying to find the potential
at which corrosion would start, and then were stating,
okay, those potentials are never reached in the
repository. So that is sort of how the argument went
as I understood it.
DR. CAMPBELL: Well, let me play devil's
advocate a little bit here. The pillar of Delhi,
which I don't know, but if it is in the area of India
that I am thinking of, is probably subject to a
monsoonal type environment.
DR. SHEWMON: That's right.
DR. CAMPBELL: Relative humidity may be
significantly lower than most of the time frame for
the repository in that environment. So it goes
through these wetting and drying cycles, but during
the wetting cycle, you are looking at a short duration
event and then it dries out very rapidly.
So it is going to develop some sort of
patina on it that becomes a barrier to further
corrosion. We also don't know whether it was treated
with anything that would help that.
DR. STEINDLER: Wait until you find out
that somebody goes along and paints the fool thing
every three years.
DR. SHEWMON: It has a fair amount of
silicate inclusions that get hammered out in these
things, which are thought to give wrought iron better
than modern steel.
CHAIRMAN WYMER: You have some factual
DR. CRAGNOLINO: Yes, I would like to
provide some updated information on the New Delhi
pillar. The New Delhi pillar was saved to sustain the
condition -- and this is an important consideration,
to know what that is, because essentially for as you
said hundreds of years, and thousands of years, it was
exposed to a relatively dry type of environment in New
Very low relative humidity, and I mean
that it was perfect condition with oxidation in the
air. However, there are now two peculiar concerns,
the stability of the corrosion problem on the New
And with the process of oxidation in
India, and in particular in the areas surrounding New
Delhi, the air has become polluted, polluted with
This is one problem. And this can be
discussed in more detail, because the Indian people
are very concerned over this, with the air impurities
and people are concerned about it.
The other problem that was called to the
attention of -- and unfortunately he is not here to
verify this in more detail, but the Indian people were
going through a time and took advantage of the
situation that the British came there.
And the British were concerned with this
problem with -- and decided to build a concrete
support, and in order to do the work better, they
built a concrete support there.
Now, the particular problem is with the
interface, and --
CHAIRMAN WYMER: Interfaces are always a
DR. CRAGNOLINO: Yes. And this is a
problem that they have, and there are people in India
who are trying to grow away from this, and produce --
and through the internet -- and I am making this story
very long, but this is a fact -- got involved and he
is providing technical support to these people.
CHAIRMAN WYMER: Well, I think we can
conclude from all of this that the use of these
analogs is probably only appropriate if they are
tested under the relative conditions.
DR. CRAGNOLINO: In environment type
conditions, and where they have very well defined
conditions, and this is the moral that I get.
DR. SHEWMON: And do you take as these
conditions underground someplace, or in the
CHAIRMAN WYMER: Now, don't get me wrong,
Paul. I think that these natural analogs are very
suggestive of what will really happen. I don't think
there is anything wrong with the general philosophy of
what you are saying, except that I don't think that
DOE can use them, and NRC can't use them either,
unless they are more sharply scientifically defined.
Not that they may not be valid, but they
will not be accepted I think is the point.
DR. STEINDLER: Well, I confess to a
significant amount of confusion. Where are we?
CHAIRMAN WYMER: We are to my section of
DR. STEINDLER: I know. We are seven
minutes past that time.
CHAIRMAN WYMER: Well, I am going to catch
us up, because I think that my discussion on the in-
drift chemical environment, which I have prepared
handouts for the group here that I will read through,
are only designed to provide a factual -- a DOE
DR. STEINDLER: No, what I was driving at
was in the context of what you eventually want to put
into a conclusion from this little exercise, do we all
believe that in a sense chastising, mild or otherwise,
both the Department of Energy, which is not I think
your function, but certainly the staff, for allowing
this extreme conservatism in corrosion rates to stand
unchallenged is the question.
We don't have to decide it today, but I
think that is the focus of the question.
CHAIRMAN WYMER: My view -- and we will
discuss this and come up with a consensus, but my view
is that extreme conservatism diminishes the
believability of the analysis.
It doesn't necessarily impact whether or
not the repository is going to hold the waste, because
there are a lot of factors involved there, and it
doesn't even necessarily negate the conclusions about
the corrosion of the material.
But it diminishes your confidence in DOE's
analysis, I think, and insofar as the NRC goes along
with that, it diminishes my confidence in that. It is
sort of similar to the arguments about whether or not
the errors are acceptable in the analysis.
And whether or not the experiments are
rungs that will get better results so that you can get
some of the conservatism out of it. Now thee is a
push right now to get better and better results so
that there is less and less error in those results.
And while this may not affect the validity
of the use of the repository, it does reduce the
scientific credibility. It is philosophical as much
as anything else, and I think we ought to at least
comment on that, that there is a point there.
What it means is that they are not doing
as tight a job as they should do, or as good of a job
as they should do. We will wrangle about that.
DR. STEINDLER: We will argue about that
DR. CAMPBELL: Let me add a couple of
things here as a seaway into your thing, into your
segment, Ray, is that the NRC staff has to evaluate
what DOE presents to them.
So inevitably, and because their goal is
to be a regulator, they have got to focus on -- okay.
DOE has given us this series of concepts, models,
data, and so on, and we have to evaluate that in the
context of what we know.
We can't go back to DOE and say, hey, you
guys are nuts in terms of this conservatism that you
built into the model. It is not NRC's position really
to tell DOE to go back and redesign this, and get more
So they have to pretty much take it as it
is given, and evaluate it in that context.
DR. STEINDLER: But you are saying that
the evaluation can only be on one side. In other
words, is the value too low is the only question they
can ask. You can't ask the question is the value too
high, which is what the issue is.
DR. CAMPBELL: They can if it gets into an
issue of -- or in my opinion at least, and this is my
opinion, if it gets into the area of challenging the
whole concept of defense-in-depth, and that because of
the conservatisms built in that you really don't have
a handle on how other systems will --
DR. STEINDLER: You are moving me out of
DR. CAMPBELL: Okay. Well, that is the
question really, is can the NRC staff say that this is
just too high.
CHAIRMAN WYMER: Well, I think it is an
observation that we would make rather than a damning
comment that we would make.
DR. STEINDLER: Well, let's not overlook
the fact that there is uncertainty.
DR. CAMPBELL: Right.
DR. STEINDLER: We have not, I think, said
to the staff or to anybody else that this evaluation
or the acceptance of the DOE position is wrong.
Perhaps what I would call for is an enhanced
commentary about the uncertainty on the corrosion.
CHAIRMAN WYMER: That might be a very
DR. STEINDLER: I am still looking for
some good answers to catalysis.
CHAIRMAN WYMER: It might be a very
appropriate way to answer that.
DR. STEINDLER: And the uncertainly in
their data has got to somehow temper the staff's
approach to whatever DOE hands them. So I can argue
on both sides actually.
I can argue that if the staff is given
this extremely low corrosion rate to look at, and
let's assume they hire somebody like Paul, who looks
at the thing and advises them that this is an absurdly
low high corrosion rate in relation to what the real
world appears to be.
They have to add the uncertainties in the
whole thing and say, look, we need reasonable
assurance. So our window is a lot larger.
CHAIRMAN WYMER: Right.
DR. CAMPBELL: Of course, along those
lines, the key issue is going to be the environment
inside and on top of the drip shield and the
container, and the chemistry of this water coming in.
And has DOE characterized the chemistry of this water
and the chemistry on the surfaces of the drip shield
and the waste package in a way that truly bounds the
conditions that it will see.
I mean, you have cited the Delhi pillar,
but as Gustavo has pointed out, conditions change, and
the environment changes, and now instead of having
this long lifetime, we are now probably looking at a
relatively short lifetime if those conditions
So one of the areas of uncertainty is how
well, or how good a job have they done in terms of
characterizing this environment right there, and that
to me is the key to corrosion.
DR. SHEWMON: They always approximate it
by an electrolytic cell, where they have got flowing
solutions, and water all the time, and that just isn't
the situation here.
The humidity may be 80 percent, but that
is not flowing salt solution with an applied voltage.
So go ahead.
CHAIRMAN WYMER: Well, pursuing just a
little bit more my philosophical uneasiness. I feel
the same uneasiness about the use of bounding
conditions which are probably perfectly valid, and
they do bound the conditions probably that could
But if you use those instead of
information that you could use to reduce the
uncertainty, and so that that whole approach is not
satisfying. It may be adequate, but it is not
But it doesn't mean that the conditions
aren't bounded, because I think that they probably
DR. STEINDLER: We are going to get into
an argument about this.
CHAIRMAN WYMER: Well, are you really?
Okay. Let me -- I won't have a whole lot to say here
because Andy in his opening comments, where he made a
nice discussion of what the situation is, pretty well
covered what I was going to say about the in-drift
I will go down through a list of the TSPA
model and what the extractions and processes relate to
as they are relevant here. Down in the middle of the
page there, the model ingrates these things more or
The water and cement reactions, gas and
water, evaporation and condensation of water,
precipitation and dissolution of salts, microbial
action, which I think is a red herring, corrosion and
degradation of EBS components.
Water in the invert, and water in the
colloids, and these are things that the TSP model
integrates. And the modeling period is divided into
three regimes which are meant to simplify the model,
and make it possible to do the calculations in a
And also trying to catch the periods
during which major changes occur. And the first
period that is looked at is 50 to a thousand years;
and the second one is a thousand to 2,000 years; and
the final one is 2,000 years to a hundred-thousand
years, or more.
So they do try to capture in an overall
way the time periods for which they examine what
pertinent processes there are that are taking place,
and what the temperature and humidity, and so on,
conditions are that are relevant in those time
Now, one of the major criticisms that I
have about the invert chemical environment is that
they use simulated J-13 water, and I don't for the
life of me know why they didn't go out to the well and
full a couple of 55-gallon drums with water and use
that instead of simulated material that does not
necessarily have everything in it that was in the
And this gets to Marty's catalysis issue,
these trace elements that are not necessarily included
as simulated water. I think the simple fact of the
matter is that we don't know what actual J-13 water
would do, although there is a strong -- I have a
strong feeling that it would not be a whole lot
different from the simulated J-13.
But in fact I don't think we really know,
and it seems to me that if you can experiment with the
real stuff that you ought to.
DR. STEINDLER: You are defending the
wrong groups here. In defense of the folks who do
experiments, I would say in the last five years that
they have used crushed tough calibrated J-13 water.
Now, you can argue that in the two --
well, what they do is that they basically take J-13,
and let it sit on crushed tough for two weeks, and
filter it off.
And in the two weeks that they use to
calibrate this stuff, you can argue that you may not
be getting a full compliment at concentration. On the
other hand, also recall, please, that J-13 is a
simulation of what they expect in the repository, and
whether that is a good simulation is another story.
And actual pore water, to get a pore
water, is a real chore. I mean, the notion of a 55-
gallon drum of pore water is a little bit difficult.
Thirst water you may able to get away with, but pore
water is tough to come by.
I am not nearly as unhappy about the use
of simulated J-13 for a lot of experiments. It's when
the concentration of the traces that they are looking
at, which unfortunately happen to make some difference
in the downstream answer of what this whole thing is
about, is significantly lower than the normal trace
composition of things that they have ignored. That's
when I begin to at least wonder about it.
DR. CAMPBELL: One of the things that --
and I thought I had sent them on to you, Ray, is one
of the things that they have done with this simulated
water is that they have these wonder ICP mass
lectromers and other things that can do enormous
amounts of data collection on every element that you
can lay your hands on.
And apparently there are databases perhaps
unpublished by DOE of the trace constituent in the
waters that they used in the experiments. So even if
you don't exactly have the water -- and it is a guess
anyhow that it is from a J-13 well.
I mean, that is a guess that that would be
something akin to or close to the actual water that
would be essentially dripping on to the drip shield
and waste package.
You have at least measurements of trace
species in these waters that could be used to at least
understand the impact at those concentration levels.
The real issue in my mind is have they
characterized this environment well enough in their
thought processes to have a good analog to what is
going to be actually accumulating on the surface of
the drip shield to the waste package.
And in my mind it is an evaporative
environment for very long periods of time, and that
you will tend to have fairly concentrated solutions on
CHAIRMAN WYMER: Well, I have a comment
about that that I picked out of a report, a most
recent report that I could get my hands on. It says
that water evaporating into drips can lead to
temporary accumulation of up to a few kilograms of
soluble salt per meter of drip.
Now, that sounds like a lot to me,
depending on the proximity of the repository, and to
the repository center, and the infiltration rate.
Edge locations had less salt accumulation because of
less heat available.
And it goes on to say that salts would be
deposited in the backfill, which they don't have
anymore, but the report said this, and in the invert,
and that seems to me to be a lot of soluble salt per
DR. STEINDLER: But if you will look at
the question of where is the soluble salt when in fact
a cladding of the fuel is breached, which is when the
rubber hits the rope.
My sense of following this down to the
time interval is that the large accumulation of
evaporates is gone.
CHAIRMAN WYMER: They expect it to go down
DR. STEINDLER: Right, and it is gone, and
it has accumulated at least outside the waste form
when the waste form begins to corrode.
CHAIRMAN WYMER: I think that's right.
DR. STEINDLER: That simplifies the
corrosion picture a little bit, because you don't have
to begin to guess at what the concentration of sodium
nitrate is, for example.
CHAIRMAN WYMER: But it is suggestive that
there might be some of these trace elements quite well
DR. STEINDLER: Well, the problem comes
under your domain, because it is going to pile up
somewhere else. It may pile up somewhere else, and
since I am looking at the in-waste form chemical
dissolution issues, I don't have to deal with that I
don't think in any significant fashion. But
downstream in the unsaturated zone I may have to.
CHAIRMAN WYMER: Well, here is another
comment that was made, and that is the redissolution
of precipitates is difficult to model accurately
because thermal chemical models lack data support for
extreme concentration of temperature conditions
because of the distribution of the flow in the EBS
depends on change in backfill properties, which we can
take out, and the nature of the seepage from the host
rock, which we can't take out.
So they are really saying in this
particular report that they really can't model very
well or accurately they say because they lack data.
DR. STEINDLER: This is the kinetics, you
CHAIRMAN WYMER: This is for the buildup
of concentration of the salts.
DR. STEINDLER: So it is kinetic issue and
not a thermonomic equilibrium issue?
DR. SHEWMON: It is easier to precipitate
than it is to put it in solutions; is that what you
CHAIRMAN WYMER: Well, the redissolution
of precipitates is difficult to model accurately,
because thermo-chemical models lack data support for
extreme concentrations. It just means that they get
a precipitate, and then they put stuff on it that
would change the nature of the precipitate, and they
don't have the thermo-chemical data to see what those
changes would be, what the nature would be after those
DR. STEINDLER: And these are precipitates
and not evaporates?
CHAIRMAN WYMER: These are precipitates
and presumably it would be on the drip shield, or if
that fails on the equation, the package; as well as in
the invert and underneath the package.
DR. SHEWMON: If it dripped on to the
shield and then the water went off as a vapor, is that
an evaporate or a precipitate? It seems to me it
could be both.
DR. STEINDLER: It is an evaporate.
Precipitates are formed when you get uranium, and --
well, when it is dissolved out of the fuel that now
reacts with a whole bunch of other material, and you
uranium and minerals.
That is a precipitation process and that
becomes important because occasionally you precipitate
things that you really don't want downstream.
Plutonium, for example.
CHAIRMAN WYMER: Well, I do have a table
here where they have made 19 separate analyses of J-13
well water and then averaged them all to give you the
-- and the analyses are pretty well --
DR. SHEWMON: Are these stimulants or the
CHAIRMAN WYMER: This is J-13 well water,
and they analyzed for aluminum, boron, calcium,
chlorine, fluorine, iron, bicarbonate potassium,
lithium, magnesium, manganese, sodium, nitrate ion,
phosphate ion, silicon sulfate and strontium.
DR. STEINDLER: And no fluoride.
CHAIRMAN WYMER: Yes, I mentioned
fluoride. Fluoride is 4.4 milligrams per liter.
DR. CAMPBELL: That is an average, right?
CHAIRMAN WYMER: That is an average of 19
separate analyses, which range anywhere from 2 to 2.7,
depending on the analysis, but not bad. So that is
about all I really wanted to say about that, because
you have already seen quite a bit about it, and what
Andy has done.
They have not really done the trace
element of the analyses, and I do have some
information about the lead content.
DR. STEINDLER: What is your view on the
role of the cement that is holding the rock bolts in?
CHAIRMAN WYMER: That holds the rock bolts
DR. STEINDLER: Yes. Do you think it is
CHAIRMAN WYMER: I think it could be,
depending on the location. One of the things that is
important here that is not dealt with very well in the
models because of the difficulty dealing with it is
the microstructure of the thing.
Suppose you have a couple of rock bolts
grouted in directly above the waste package, and water
drips out of those and reacts with the cement that is
holding the rock bolts in?
They don't really catch that very well in
the model. They catch it with respect to whether or
not it ultimately winds up beneath the waste package
and might lead to the plugging of fractures.
But they don't deal at all with the
chemical environment that it might produce on the drip
shield or on the waste package in these very awkward
conditions, and the fact that it has a petition of the
cement mixed with.
DR. STEINDLER: So the design by the
Department of Energy should be that there are no rock
bolts directly over the waste package?
CHAIRMAN WYMER: Just let the rocks fall.
The heck with it.
DR. STEINDLER: No, just let the steel
handle it, and just put the rock bolts on the side.
CHAIRMAN WYMER: Yeah. So that is sort of
a minor point, but --
DR. CAMPBELL: Let's redesign the
CHAIRMAN WYMER: Yes, but which is not our
DR. STEINDLER: Well, I guess what I am
trying to find out is whether or not you think that is
a long enough issue so that it could begin to
influence the corrosion rate of the fuel and glass,
and all the other junk that is important?
CHAIRMAN WYMER: Well, I always come up
against the fact that 10,000 years is a long time, and
I certainly don't have any feeling for what these
chemical effects that are not observable in the short
term might be in 10,000 years.
My seat of the pants feeling is that I
hope that it doesn't amount to much.
DR. STEINDLER: That isn't quite my point.
CHAIRMAN WYMER: What is your point?
DR. STEINDLER: My point is that if the
corrosion of that grout, or whatever it is that they
use, is done and over with, and all the rock bolts
have fallen out in a sense, before the next bunch of
water arrives at my waste form, I can argue that in
terms of corrosion rate of the waste form, it doesn't
make any difference.
CHAIRMAN WYMER: Yes.
DR. STEINDLER: Is that sustainable as far
as you are concerned?
CHAIRMAN WYMER: It is arguable, but I
don't know if it is sustainable.
DR. STEINDLER: Anything is arguable. I
have been there.
CHAIRMAN WYMER: Again, that is an
argument that lacks factual --
DR. CAMPBELL: One of the things that I
came across, and I can't recall exactly where it was,
and maybe it was in the IRSR, but maybe it was in one
of the DOE documents, was that they are going to look
at evaporative processes and the effects on chemistry,
and they are looking at those.
The scenarios right now don't necessarily
take into account evaporative processes when they are
calculating the solubles or the mobilization from the
And in fact they have a pretty wet
environment that occurs, but that is your topic of
CHAIRMAN WYMER: Well, the environment on
the waste package is a separate discussion.
DR. CAMPBELL: But I think what we have
identified here is an issue that the chemistry on the
waste package is still highly uncertain, and probably
needs to be defined better in terms of the scenarios,
or waste package and drip shield corrosion, and then
CHAIRMAN WYMER: Because of uncertainties
in the chemistry of the water that hits the waste
DR. CAMPBELL: Because of that, and
because of the uncertainties in the scenarios in which
you have water on the drip shield and waste package,
and whether it be a film of water most of the time,
with occasional drips, and how those two different
scenarios can play out with time.
And you have basically -- and let's ignore
that the load humidity, high temperature regime that
is relatively short in duration for the time being,
although you need to look at that in terms of a couple
of processes --thermo-hydraulic, chemical, and a
couple of other processes.
But for the longer term, the main concerns
are what are the long term chemistry for this moisture
film on these two barriers, and then what is the
impact of water dripping on to those barriers.
DR. STEINDLER: Do you get the impression
that the NRC staff is ignoring that issue?
DR. CAMPBELL: No, not at all.
DR. STEINDLER: So they are as unhappy or
as concerned about that as we might be or that we seem
DR. CAMPBELL: It is one of their issue
CHAIRMAN WYMER: I think there are
practically no issues that we could think about or
talk about that haven't been considered and discussed.
DR. STEINDLER: Well, that's my view, and
I just wondered what they are coming up with.
CHAIRMAN WYMER: There isn't anything that
hasn't already been thought about and discussed at
some length by the staff. I think that is certainly
true. And I have some other comments along that line,
but I will save those until a little bit later.
But I think the issues have been
thoroughly thought of, and whether or not the
experimental information is adequate to do the issues
that have been obtained is a question, and that is a
matter of sources more than anything else.
Well, let's push on here, and there are
other things that will come up as we go along. Let's
talk about taking a break.
DR. CAMPBELL: That sounds like a good
DR. CRAGNOLINO: May I provide some
information about natural analogs for --
DR. CAMPBELL: Make it short.
CHAIRMAN WYMER: Let's defer it if you
don't mind. I want to hear it. Anything you want to
say is relevant. Keep it in mind and we will get to
it. Let's take a break and come back at a quarter
(Whereupon, a recess was taken at 10:03
a.m. and the meeting was resumed at 10:17 a.m.)
CHAIRMAN WYMER: All right. Let's start.
I guess you are up again, Paul, on Titanium alloy
DR. SHEWMON: Okay. I was surprised to
find out that this thing is a 15 millimeter drip
shield, which is a respectable piece of titanium. It
is Grade 7, which has 2/10s percent palladium added to
it to help avoid hydrogen going into solutions in the
The alloys proposed for Yucca Mountain
-- well, okay, because this catalyzes the hydrogen and
reduces the hydrogen pickup, and Gustavo says it helps
or works, and so I will take his word for it.
Titanium is quite corrosion resistant in
air, water, and sea water. They build ships out of it
and have not had any trouble with it. Passivation
under water occurs in hours to days, and titanium is
active, and so contact with iron will give an
electrolysis of water.
And it would seem to me that in air that
titanium would last forever, but there is no
engineering analogs, and so they make conservative
With an applied voltage, as you might get
from the galvanic corrosion with iron, you can
-- the titanium can be made to dissolve in chloride
solutions and dissolve faster in fluoride, plus
There is little tendency to crevice or
localize corrosion, and so they are interested in
general corrosion. There is reasonable talk about
hydrogen embrittlement, and the outline that I got
from Andy suggested that I talk about this
So let me talk about hydrogen induced
cracking. I find in the NRC notes that this can occur
only if you have all three of the following. You have
to have some potential which will tend to make the
water break up in the contact with titanium, and
galvanic voltage is enough for this.
You must be above 80 degrees C, or else
the hydrogen won't diffuse into the titanium, and you
have to be at either acidic or basic, less than 3 or
greater than 12 Ph.
Hydrogen induced cracking occurs in
engineering applications like aircraft. I don't think
it will happen here, but let me tell you why. Where
it occurs is in high strength alloys with sharp
notches and high stresses.
They then get tearing and the tearing can
be accelerated by a generation of some titanium oxide
and hydrogen that is free to go into the metal.
And so you end up with an accelerated
crack growth under applied stresses. But the stresses
and the notches are minimal in the drip shield. The
roof collapse could cause this, but many of these
stresses would be compressive.
Thus, it is difficult to see how hydrogen
induced cracking could be a concern. Let me emphasize
-- well, okay. I find this hard to give credence to
because the phenomenon never gives spontaneous
cracking or indeed fragmentation of the drip shield.
What it means is that it is not as hard to
drive a crack through it when you have applied stress
and you are tearing something apart. I don't see how
there can be the substantial stresses and strains in
this geometry that are required, and so it seems to me
that hydrogen-induced cracking is really something
that would be absent.
But then I have got it summarized here.
Cracks in Titanium are absent, crack opening stresses
are absent; water, required for hydrogen charging, is
rare and transient.
But the AMR says they know all this, but
assume that it fragments anyway just to be
conservative. So how can you argue with conservatism.
CHAIRMAN WYMER: And that gets back to
this same philosophical question that we have been
raising periodically here, which is to what extent
does that give you a feeling of disquiet.
It is contrary to the scientific method,
but it probably is a safe and conservative way to go,
which really doesn't challenge the viability of the
DR. SHEWMON: And they also go back to the
general corrosion, which again they get out of a
galvanic cell with circulating fluids, and aggressive
media, and that then is taken as a bound on what could
be the general corrosion rate.
CHAIRMAN WYMER: My reading on everything
that you have been saying so far, Paul, is that you
think that they are very conservative, and that
actually there will be no significant corrosion
problems in a 10,000 year time period. Is that a fair
assessment of your position?
DR. SHEWMON: Yes. At least 10,000.
DR. STEINDLER: You don't think there is
an electrolytic problem at the foot of this thing?
DR. SHEWMON: There is certainly the
potential for galvanic cell there, but even if you
broke it up there, the shield still functions.
CHAIRMAN WYMER: Now, suppose you had a
couple of rock bolts that were cemented and grouted
in, and the grout slowly over time -- and we have got
a lot of time here -- dissolved and ran off the drip
shield like the picture shows there, and combined at
the foot of the drip shield, where it rests on the
And how you have got these cement
ingredients there. Does that not change the picture
considerably, and does that not make it desirable to
have some sort of an experimental analysis of that
DR. SHEWMON: Well, if we are talking
about hydrogen embrittlement, the question and the
criteria that I have got at the top would be does it
change the Ph to be low or high.
CHAIRMAN WYMER: Well, is hydrogen
embrittlement the only thing we should concern
DR. SHEWMON: Well, that is the voltage
driven problem that you have down at the bottom, yeah.
Up at the top, you have got general corrosion and it
is not under water. And I don't see how it could stay
CHAIRMAN WYMER: Well, it wouldn't be
emersed, but it would presumably have a film of water.
DR. SHEWMON: Yes, and is that an
effective electrolytic media that will carry away ions
CHAIRMAN WYMER: Yes.
DR. SHEWMON: Okay.
DR. STEINDLER: Especially in high
amenities where you have got more than a couple of
monolayers, and we have made minerals on glass that
way very fast.
DR. SHEWMON: Well, then maybe you could
get some general corrosion up there. I don't know
what the applied voltage would come from, but the
transport media is there.
This thing is not allowed to dry out. We
have got enough source of water in the surrounding
soil so that it is not dry like the surface. It is
always wet or saturated.
DR. CAMPBELL: In the absence of air
circulation through the repository, the natural
condition is to approach a hundred percent, or to go
to a hundred percent humidity.
And once they close it up -- as long as it
is open and they are circulating air through the
system, they are drying it out and they are keeping it
dry. And that is more analogous to these analogs,
where you have a dry environment that occasionally
gets some moisture on it, but by in large is dry.
Or, for example, archeological artifacts
that are in caves in Nevada that are dated at almost
10,000 years, because of those dry environments, they
tend to be preserved.
But once you close the repository up,
there is enough water and moisture in the rock, and
percolation flux, that the air trapped in there will
go to about a hundred percent humidity, except during
this thermal pulse, when you are driving that moisture
DR. SHEWMON: But you still need something
to drive this, and it is corrosion resistance, and
more corrosion resistant than meteorites.
Now, if you have got a voltage applied,
yes, but if you haven't, then you get back to whether
that is realistic, and do you corrode through that
fast. And I guess it could be, but I have trouble
CHAIRMAN WYMER: What happens if the --
you probably can't answer this, but let's talk about
it. But what happens if, let's say, the basis did
corrode away to the point where the drip shield sat on
top of the Alloy 22? What about that interface? Are
there any galvanic problems there?
DR. SHEWMON: I suppose if you have got
monolayers of water there, but then you have to get to
these other criteria, and by that time is the
temperature above 80 C? Is there something that would
make the Ph high or low?
I am not sure that hydrogen charging would
be your problem. There maybe some galvanic corrosion
and dissolution, and the titanium has to get carried
DR. CAMPBELL: And by high, you mean above
10 or 11?
DR. SHEWMON: Well, 12 is what it says in
the NRC report I got.
DR. CAMPBELL: So, 12, and cement pore
water type of pHs to get into that range. I mean, one
of the scenarios that really doesn't show is that with
time the supports are going to corrode and lose their
DR. SHEWMON: Which supports?
DR. CAMPBELL: The drip supports, and so
you could have not only rock fall, but you could have
over longer periods of time --
CHAIRMAN WYMER: A collapse of the
DR. CAMPBELL: Yes, a collapse of the
support on top of, or a rock fall falling out and
flying on top of the drip shield. Eventually you are
going to have bangs and dents, and material on top.
DR. SHEWMON: That might influence the
general corrosion. I don't think it will give
hydrogen cracking, because even if you have the
hydrogen there, if you have not got a stress and
strain to drive it, you won't break it up.
CHAIRMAN WYMER: Is hydrogen a pretty key
issue with respect to the titanium?
DR. CAMPBELL: Not with general corrosion.
It is with the hydrogen induced cracking.
CHAIRMAN WYMER: Sure, but we are talking
more broadly than that. That is one of the issues.
DR. SHEWMON: Well, I am trying to
differentiate the two, and say that the hydrogen
cracking, which they assume will occur, I don't see
how it can. The general corrosion could well be
driven by the galvanic or accelerator.
DR. CAMPBELL: Because of the environment
there, where would the hydrogen come from?
DR. SHEWMON: Water.
DR. CAMPBELL: But you have an oxidizing
DR. SHEWMON: The titanium is active
enough to take the oxygen from the water.
DR. CAMPBELL: So the titanium itself is
going to act as the catalyst to generate it?
DR. SHEWMON: It is going to act as a
getter, but if it is going to get past the surface of
the titanium, it has to be hot or warm.
CHAIRMAN WYMER: And alkyl generally chews
away at these oxide protected coasts doesn't it?
DR. SHEWMON: It can, yes.
CHAIRMAN WYMER: At the risk of
randomizing our discussion here, let me ask Andy if he
will show that view graph about the temperatures of a
function of time again.
Well, okay. We did in previous meetings
talk about this temperature regime, where corrosion
can take place in a regime, and if it gets hotter than
that, it dries out. And if it gets colder than that,
then it is kinetically too slow to make any
So there is a regime of temperature and it
looks to me like for the first -- well, sort of like
for the first 80 or so years you are in that regime.
And then you get into it again after a couple of
hundred years, and you stay in it for a few hundred
DR. SHEWMON: What causes the spike, and
where are we measuring this temperature?
CHAIRMAN WYMER: The circulating arrow.
DR. CAMPBELL: If the spike comes from
closing up the drips, or closing up the repository,
when you cut off the ventilation and close it up, then
you will get that spike in temperature.
In fact, the temperatures prior to that
are probably not very realistic the way that they
CHAIRMAN WYMER: There are hundreds of
years, and maybe thousands of years, where the
temperature is in the corrosive temperature range, if
that regime is a true regime, and people seem to think
that it is.
DR. STEINDLER: Above 80 degrees, is that
what you are saying?
DR. CAMPBELL: Yes. My recollection is
that that was something that was brought up in our EBS
working group 2-1/2 years ago, and that that was, I
believe, crevice corrosion that they were concerned
DR. SHEWMON: In titanium?
DR. CAMPBELL: No, no, no.
CHAIRMAN WYMER: That's why I said I am
randomizing the discussion, Paul. But I neglected to
bring it up when you were talking about that. So I
think in fact that there is a temperature regime is
important, and the fact that you are in it for pretty
long periods of time potentially here is important.
DR. CAMPBELL: I may have a view graph of
CHAIRMAN WYMER: It's not as though you
are out of the regime for most of the time.
DR. CAMPBELL: The temperature on the drip
DR. STEINDLER: On the C-22 or the drip
DR. CAMPBELL: On the drip shield.
CHAIRMAN WYMER: Well, the C-22 though is
what we really care about. That is where the
temperature regime was discussed as being relevant.
So we are not nailing it down too tightly here, but
the drip shield is not going to be a whole lot
different from the Alloy 22.
DR. CAMPBELL: The peak in temperature
there, the solid line, is the alloy or the waste
package, and the dotted line just below it is the drip
wall temperature. So the drip shield is probably not
going to be that different than the waste package
CHAIRMAN WYMER: That is very similar to
the graph that you just showed and it tracks it pretty
DR. CAMPBELL: Yes. This is just one
slice of that, one of the bins.
CHAIRMAN WYMER: The first 80 or 90 year,
you are in that regime, and then you get into it again
after about a thousand years.
DR. CAMPBELL: Now, 5 meters above the
crown of the drip, these are the temperatures, and so
you get a very strong radiant from the drip wall to a
few meters in.
CHAIRMAN WYMER: Well, the only point I
wanted to make in bringing this up was we are in the
corrosion regime for it for quite a while. That was
the only point, and so we can proceed to talk about
DR. CAMPBELL: Paul, what are the key
issues in your mind in terms of more general corrosion
effects on titanium and the uncertainties of that.
DR. SHEWMON: No. I think they are
probably more credible than the C-22, because it is an
active material, and you have got water, and it is in
the discussions of how protective the oxide layer is
over these long periods of time. And there is no
analog, and I don't know, but it may indeed be true.
CHAIRMAN WYMER: Were you able to
determine from what you read whether or not it is the
position of DOE that the titanium drip shield will
last 10,000 years or longer?
DR. CAMPBELL: It's longer.
DR. SHEWMON: Yes. He gave something
there, and it started up in the 20,000 year period as
I recall. I don't know whether that --
DR. CAMPBELL: The last line.
DR. SHEWMON: And that is all general
corrosion; is that right?
DR. STEINDLER: What do you mean by less?
DR. SHEWMON: Maintain some --
DR. STEINDLER: Well, all I need is a
small hole for liquid to get into my waste package and
begin that process. I don't have to collapse the
DR. SHEWMON: Well, I don't think that
once you get past the titanium shield that you are
going to go through the C-22 as fast as you do the
DR. STEINDLER: Okay. Well, at least the
point is that is where you start counting, in terms of
CHAIRMAN WYMER: Now, what would lead to
a hole, something like a rock bolt dropping and
DR. STEINDLER: No, no, no.
CHAIRMAN WYMER: And wouldn't that be an
enhanced corrosion area to lead to a hole?
DR. STEINDLER: No. You get uniform
CHAIRMAN WYMER: I don't know. That's the
DR. SHEWMON: Done with statistics.
DR. STEINDLER: Done with statistics?
Okay. Well, that takes care of me.
DR. SHEWMON: I don't know what they do to
get their randoms. It is too large a spectrum and
conditions, I guess.
DR. STEINDLER: I am trying to see how old
my fuel is before somebody finally says, okay, you
have got water dripping on your oxide.
DR. CAMPBELL: In the last graph on the
view graphs that I handed out, those are the kinds of
time frames for the top one, and that is from TSPA.
DR. STEINDLER: And I didn't understand it
there either. Fraction corrosion failure.
DR. CAMPBELL: Well, I wrote that just to
try and summarize what these slides are showing, but
these are the various percentiles for failure on a
DR. STEINDLER: Well, I can argue that we
ought to be looking at 10 or 20 percent breaches since
it is statistical. And at 10 or 20 percent, I now
find that I am dripping on my waste package.
And pretty soon the waste package is going
to have 10 or 20 percent penetration, and again
statistical since you guys in the corrosion business
seem to be entirely statistical.
I am still trying to find out -- and in
effect I don't care about the time particularly,
because we are well past the compliance time, but I am
interested in the temperature.
If the only thing I have to deal with --
and I am focusing in on the waste package, but if the
only thing I have to deal with is reasonable
solutions, ground water, et cetera, dropping on 25
degrees centigrade and irradiated at 1O2, that is one
If I have to worry about the thing being
150 degrees initially, I get a somewhat different
CHAIRMAN WYMER: And you have to be
DR. STEINDLER: Especially in the gap
release and the release of material in the grain
DR. SHEWMON: You can go out to 10,000
years here,and you are down to 40 degrees centigrade.
DR. STEINDLER: So you think I am safe
that length of time?
DR. SHEWMON: Well, I don't know about
that, but I think you are quite below temperature by
the time that the liquid comes in contact with it.
CHAIRMAN WYMER: Well, what kind of
activation energies for the corrosion process are we
talking about? How steep is the curve for the
DR. SHEWMON: Well, anything that is
active at these temperatures has to be some process
which has a low activation energy, because everything
with a high activation energy doesn't work anymore.
So what the source of hydration steps they
come in contact with, or are going on here, I don't
know. But I think the activation energy isn't a
useful way to get at it, because there has to be
different processes with a spectrum, and the high
activation energies won't go, period. You're out.
CHAIRMAN WYMER: It depends on where it
DR. SHEWMON: I am very familiar with that
sort of thing. We could ask, but I don't think they
would find it too useful. They do find hotter
solutions go faster.
CHAIRMAN WYMER: Well, if you do have
different mechanisms at different temperatures, then
of course all bets are off.
DR. SHEWMON: Well, you do have different
DR. STEINDLER: Do I get dissolved
titanium dripping on my outer end waste package?
DR. SHEWMON: Yes. Conservation of matter
is our policy.
CHAIRMAN WYMER: Not if you look at the
DR. SHEWMON: Oh, okay.
CHAIRMAN WYMER: Which is another issue
for another day.
DR. STEINDLER: What sort of
concentrations would you expect? Well, are we looking
at the solubility of Ti02, too? Is that the limit?
DR. SHEWMON: Well, is there titanium
hydroxide? And whatever it is, we are carrying it
away in this demolecular lader, and it builds up
someplace. And 15 millimeters is a lot of titanium.
That is one of my complaints with the electrolytic
It always gets the products away, and
washes it away so that you never get into the buildup
of this barrier that happens in the real world.
CHAIRMAN WYMER: One of the complexities
of this whole thing that makes it so hard to grapple
with, and I am sure the staff and everybody else has
had the same problem, is this time dependent factor.
If things happen early, and if something
really goes badly wrong, which is not expected, but if
it does, the first few hundred to a thousand years or
so, then you don't necessarily have an oxidizing
environment, because you have a hell of a lot of iron
in this repository.
And until it is oxidized, you have a
reducing environment, and titanium, of course, as it
dissolves first, it is Titanium-3, which is a powerful
reducing agent. It is a strong reducing agent.
And then you have, of course, you have
ferris ions. So if something does happen early before
all the oxygen depleting materials are used up, then
it is a reducing environment.
And that is not what has been considered
in any of these considerations. Now, it is unlikely
that anything will happen in these early stages while
there still is iron around in a reducing environment.
But if it were to happen, then this is a totally new
DR. SHEWMON: I don't think it is unlikely
at all. I think it is highly likely. Unless you
expect the world to corrode nice and uniform across
this whole thing, and the cladding and all the rest of
the stuff immediately disappears as the water attacks
the actual waste form, which sounds to me to be even
more of a ridiculous conservative approach, I think
you have a real chance of at least a portion of the
corrosive attack on the fuel in the glass to be in a
CHAIRMAN WYMER: Yes, there is a
DR. SHEWMON: And I will raise that issue
CHAIRMAN WYMER: But certainly not uniform
corrosion is an issue here. We are all familiar with
the fact -- and to be simplistic -- that when we drive
our cars through the salt in the winter that the whole
car doesn't corrode. It corrodes around the running
board and under the fenders, and of the joints.
So non-uniform corrosion is well known,
but that is under conditions where you have non-
uniform conditions of the surface, and the metal, and
we have some of that here.
DR. STEINDLER: Well, you do have some of
that, that's right.
DR. SHEWMON: Let me bring up something
different, and I guess it has to do with the
permeability of the earth over this mine, which some
of you may know more about than I do.
But I remember going out to Arizona a long
time ago in a different incarnation almost, and
somebody was studying air coming out of vents in the
ground, and they didn't know where it went in, but
they knew it slowly came out here.
And I guess the thing that I carried away
from that is that the air, or the earth above this is
permeable. There are passage ways through it. Radon
does come up in our basement out of the ground or
And is that over these times fast enough
to counteract this reducing environment that you talk
about, or is there anything done on that?
CHAIRMAN WYMER: Well, the oxidizing
environment is assumed entirely to be due to oxygen in
the air and in the water that comes into the drip. It
is not really considered to be necessarily anyplace
DR. SHEWMON: But we are talking about
after this is closed up. The air can still come into
the drip then?
DR. STEINDLER: Yes, debris.
DR. CAMPBELL: They have done a fair
number of air permeability studies.
DR. SHEWMON: Okay. Good. So we are
talking about a reducing of air environment here.
CHAIRMAN WYMER: No, we are not, and the
temperature changes by themselves are by the pumping
action, and let alone the fact that the thing is
permeable, and the water brings oxygen in with it, or
some, and no nearly as much as the air.
And in addition, once you get into the
transport mode, then you are not necessarily in an
oxidizing condition anymore -- and I will digress from
our topic for a minute here.
But as you go through the invert and
through the material beneath the waste package, and
down into the earth, you can there maintain a reducing
environment I think quite a ways.
DR. STEINDLER: No, I don't think so.
CHAIRMAN WYMER: I do.
DR. STEINDLER: You are going to get
breathing of permeable rock, independent of whether it
is above or below the drip. It is still unsaturated
or in the unsaturated zone.
DR. CLARKE: Probably 300 meters to that.
DR. CAMPBELL: The general consensus is
that this is a thoroughly oxidized environment because
of this permeability.
CHAIRMAN WYMER: But there is a
recognition that there can be local reducing regions.
DR. CAMPBELL: I would say the greatest
chance of that is inside your waste package, where you
have particularly small pin holes, cracks, and
initially small patches, and a large mass of material
that could act as a reducing agent inside the waste
DR. STEINDLER: Well, I will make some
comments about what happens if you are fishing UO2,
and you dump out two oxygens into the system per
uranium, and now let's do a little arithmetic.
CHAIRMAN WYMER: And if you make fishing
products which have an oxygen demand.
DR. STEINDLER: Well, that's what I am
saying. If you then add up all the oxygen demands
according to just their free energy formation.
CHAIRMAN WYMER: It is reducing.
DR. STEINDLER: That's right. Half the
oxygen immediately goes to a whole raft of fairly high
yield fishing products, whose oxidizer is more stable
CHAIRMAN WYMER: That's right. And that
is in fact true.
DR. STEINDLER: And then you can work your
CHAIRMAN WYMER: And that is in fact true.
You don't have enough oxygen to meet the demand of the
DR. STEINDLER: And you also have a five
component metallic alloy, which I think they call
Epsilon Phase, but I am not sure that is quite right.
DR. SHEWMON: What do you mean? You don't
like the use of Epsilon for that, or Epsilon means
something else to you?
DR. STEINDLER: I thought that Epsilon
meant something else, but it depends on whose Epsilon
it is or whatever. So, yes, I think there is a
DR. CAMPBELL: So I think the bottom line
here is that within the drip itself there is always
going to be a tendency, even with reducing agents
available, and materials available inside the drip,
there is always going to be a strong drive towards an
The waste package, until it is essentially
open to the air or the drip, it is going to be --
there could be a significant amount of reducing
CHAIRMAN WYMER: And we are concerned
about the local conditions, and that's where the
chemical corrosion is taking place. It is locally.
Well, what happens globally is not such much the
point. It is what happens specifically locally.
And if you have a global oxide
environment, but a local reducing environment, then
you are going to have a different corrosion regime.
DR. CAMPBELL: What about the effect of
fluorides on titanium? Is there enough fluoride in
the water to -- and especially in concentrated water
to be an issue here?
DR. SHEWMON: Didn't I say that fluorides
are worse here someplace?
CHAIRMAN WYMER: They are almost always
DR. SHEWMON: I don't know what kind of a
scenario -- what do you have to do to get very
concentrated fluoride solutions?
DR. CAMPBELL: Well, the water itself has
fluorides in it.
DR. SHEWMON: Yes, the 10 to the minus 5
levels, and 10 to the minus 6.
CHAIRMAN WYMER: Let's minimize the side
discussions and hear from Gustavo.
DR. CRAGNOLINO: Yes, completing what was
already mentioned, and the issue that you want to
address on spent fuel, but not for waste package, and
neither for the drip shield. This was our analysis
and we don't pay attention to the issue --
CHAIRMAN WYMER: And for a very good
reason, because if the waste package has already
failed, why pay any attention to it.
DR. CRAGNOLINO: This is the issue.
CHAIRMAN WYMER: So that's right. Okay.
Well, this might be a good -- any other observations
or sage remarks here?
DR. CAMPBELL: Sage remarks?
CHAIRMAN WYMER: That's kind of a spice
that you put on things. Maybe this is a good time to
hear from you, Gustavo.
DR. CRAGNOLINO: Well, this is only a
brief remark regarding the comment that Paul Shewmon
made about the possibility of having a good natural
analog for Alloy 22.
And the issue that we confronted on one
side was the fact that the stability, the long term
stability is not based by any means on long term
It is based on direct finds, because a
passive film is not an established structure that
remains there. It is a completely dynamic type of
structure, and it is strictly related or correlated
with the behavior of the environment.
CHAIRMAN WYMER: The problem that I have
with -- well, I will defer that.
DR. SHEWMON: I would like to ask one
question. These meteorites have been taken out of
places like Iowa and Kansas, too. You would say that
that is wet, and it has been wet for millions and
millions of years, and you are saying that it hasn't
got oxygen, and that's why it has survived?
DR. CRAGNOLINO: (Off mike.)
DR. SHEWMON: But why do the meteorites
stay there then?
DR. CRAGNOLINO: Well, I think that is
because as Mr. Wymer stated, because of a particular
condition in the climate, in the weather, and not only
DR. SHEWMON: Well, over a hundred-million
years, you get a fair number of cycles.
DR. CRAGNOLINO: Right. But I think we
can discuss this with more information.
DR. AHN: I would like to add Gustavo has
stated, and more housekeeping information for you. In
the waste form performance studies, actually they
analyzed Penna Blanca (phonetic) uranium deposits and
compared with the laboratory testing over spent
And in the lab testing, they identified a
sequence of passive fuel information the beginning,
and they eventually ended up with their own acidity.
They observed the exact sequence in the
Penna Blanca type over a million years.
DR. SHEWMON: What site was this?
DR. AHN: Penna Blanca. That gives us
very good insight and perhaps we need to reduce the
uncertainty of what the establishment is saying, and
on the other hand, we also look at patterns and
verification or validation.
DR. SHEWMON: What happened? Was this a
meteorite site or what happened at Penna Blanca?
DR. CAMPBELL: It is a uranium body that
has been studied as a natural analog for Yucca
DR. AHN: Perhaps we could get better
insight from the analysis -- regarding the stability
of -- in C-22, another view that we considered.
CHAIRMAN WYMER: One of the things that I
have a question, or a problem, or misgivings about is
the relationship of polarization studies, which do
tell you a lot about under what conditions and whether
or not something is going to corrode on the one hand,
and what they mean with respect to the actual
understanding of the mechanism of corrosion on the
We seem to have somehow substituted
polarization studies for mechanism studies, or we have
used polarization studies instead of going after and
understanding the mechanisms. Am I off-base on that?
DR. SHEWMON: No, that's right, and that
has no build-up of ions, and none of that sort of
stuff that traditionally stops or slows our actions
CHAIRMAN WYMER: Elaborate on that a
DR. SHEWMON: Well, if you put out a very
high voltage to it, you can get what they get
CHAIRMAN WYMER: Right.
DR. SHEWMON: Which means that it slows
down. But with these very slow tests that they do,
they do vary the oxidize potential, and that they have
moving solutions carry the ions away. And I guess
there is not a preferential solution and we could get
DR. STEINDLER: Well, let me just make a
comment. You are looking at either gas solid or
liquid solid reactions. The solid tends to be an
unstable alloy of some sort.
CHAIRMAN WYMER: It certainly is.
DR. STEINDLER: And whose composition is
fairly well defined, but whose chemical activities of
the components are not chemically or very well
So to ask can we get at the mechanism of
this heterogeneous reaction in an unstable system, et
cetera, et cetera, my comment is that I bet you can,
but not if you want to put a repository together in 10
CHAIRMAN WYMER: I agree with that.
DR. STEINDLER: So that, and that general
system is also true in waste form corrosion.
CHAIRMAN WYMER: I know how hard it is to
get true mechanisms.
DR. STEINDLER: I am trying to get you
away from science, Ray. We have got a mountain to
CHAIRMAN WYMER: Well, I am not opposed to
polarization studies. I think that they do give you
a lot of insight into the stability of a system,
provided that they are done under the right
conditions, and with the right temperatures, and --
DR. STEINDLER: I don't mean to cast
dispersions on the need for studies of that kind, but
mechanism studies are very difficult to do.
CHAIRMAN WYMER: And I would point out
that the NWTRB also says that you need to know
something more about mechanisms. Now, it doesn't mean
that you have to fully understand the mechanisms, but
a little more insight would certainly be helpful.
DR. CAMPBELL: Let me add something here
about soil processes, Paul, that may have an impact on
the longevity or not of a meteorite fragment. And one
of the things that occurs in soils is that you get a
tendency towards a reducing environment, particularly
if the soils tend to be saturated, and you have a fair
amount of organic matter there.
As you go down into the soil profile, you
can get a fairly oxygen depleted environment. In
fact, you can get reducing conditions that can lead to
even like methane forming.
So the longevity of these things in a
wetter soil environment can very well be affected by
the removal of oxygen by natural processes, by
bacterial processes in the soils.
And there is a fair bit of difference
between that environment and Yucca Mountain, where you
have a large void space, with interconnected fractures
that are permeatable, and you get barometric pumping,
and you can get oxygen flowing into and through that
And albeit at a slower rate than you would
in open air, but you still have a fair bit of
permeability there that you may very well have a
saturated environment, where these things in Kansas
and Nebraska are found.
DR. SHEWMON: So we get back to meteor
crater, which is probably as porous as Yucca Mountain,
and that is only a hundred-thousand years old, and so
that fits in with your model.
DR. CAMPBELL: Well, it is a dryer soil
environment, and maybe Gustavo -- it looks like he has
a point that he wants to make on this.
DR. CRAGNOLINO: I think I would make a
point the following way. Let's assume that this type
of meteorite is in the right environment, but you
don't know if there are meteorites in other types of
environment. I am going to make the point that to
sustain in some way this point of view that there are
artifacts that have been under relatively reducing
conditions, probably oxidizing at one point in time,
but later on reduced, that were able to absorb
selectively in the oxidizing side layer chloride.
This is the type of oxide hydroxide for --
and they have like a -- and if you keep this in a dry
place, this artifact looks splendid, and covered by
some sort of -- and as soon as you get certain layers
of humidity, they almost explode because they are full
of fluoride, and the oxide cannot preserve it.
We cannot negate the possibility that
artifacts, like the type that you mentioned, like
meteorites, will not be able to sustain conditions in
certain types of environment while in another one, and
this is what corrosion is all about.
DR. SHEWMON: And they will be born with
very dense oxide on the surface, because they came in
under very high temperature conditions, and I don't
know whether that has anything to do with the
CHAIRMAN WYMER: Of course, the only ones
we have found are the ones that are in living
conditions where they can survive.
DR. SHEWMON: That's true.
DR. STEINDLER: Apparently both the staff,
as well as DOE, use a statistical approach for the
corrosion of the surface. How good is that?
CHAIRMAN WYMER: And by that, explain what
you mean by statistical.
DR. STEINDLER: Well, they divide the surf
ace into patches, and the patches don't all corrode at
once, and that is the drip shield, and I can go down
another layer, and there are patches in the waste
package, and they don't all corrode at once.
CHAIRMAN WYMER: And then of course you
say that when there are enough patches that are big
enough that they can release enough stuff that it
matters, then you have got a problem.
DR. STEINDLER: Well, that's what I am
driving at, exactly. Does that make sense?
DR. SHEWMON: It makes more sense for the
drip shield than it does for the package to me,
because the drip shield is going to have different
things dropping on it, and you will have a very
And whether it has to do with the odd
steel bowl, or rocks, or whether there is some paste
that came out of the cement that dripped down on it,
as long as you have got this integral shield over the
top, it seems to me that it is a lot harder to see if
the -- the metal is quite homogeneous.
Gustavo says they don't see crevice or
localized pitting corrosion problems. So with regard
to the build-in, the inherent inhomagey (phonetic)
beyond the metal would be rather low.
But up on this roof there is all manner of stuff.
DR. STEINDLER: All right. So the
statistics on the top are fine, and I am trying to
chase this down to see whether or not the model that
I sense -- and, boy, if you ask me to explain it in
detail, I am in trouble.
But the model that DOE and the staff seem
to be accepting is that you will get penetration of
the drip shield in places. You will get penetration
of the outer barrier, and the stainless steel
underneath it in places.
And you will begin to attack the circular
cladding in places, and now things really get unglued
as far as I can tell. As soon as you get down below
that, all of a sudden the whole system is infinitely
quickly mixed. And evolution out of that now is --
CHAIRMAN WYMER: Not only that, but the
stuff that hits the new material has in it the
ingredients of everything it corroded in getting down
DR. STEINDLER: Yes, but I am just trying
to get up above that, and you made the comment about
DR. SHEWMON: I don't where they get their
randomizing factor, and what they take it for. But I
guess I just --
DR. STEINDLER: You think it is a sensible
DR. SHEWMON: On the top it is, but
underneath it, it is hard to see.
CHAIRMAN WYMER: Well, again, to digress
rather wildly, if you wanted to challenge anything,
you would challenge the 10,000 year period, because
this stuff doesn't really start to happen for a
DR. STEINDLER: Well, that's challenging
in the wrong direction.
CHAIRMAN WYMER: I know that.
DR. STEINDLER: If you were an intervenor,
that's not where you would --
CHAIRMAN WYMER: I realize that I said
DR. STEINDLER: But my question to what
your earlier comments were as to what you think our
function is, is to address the question of does that
make sense, and it sounds that up to a point it makes
CHAIRMAN WYMER: Up to a point.
DR. STEINDLER: It gets a little iffy I
think further into the fuel you go.
CHAIRMAN WYMER: I think we can make a lot
of observations. I think we have to be extremely
careful about the conclusions that we draw with
respect to what it means in repositories.
DR. STEINDLER: I don't draw any
conclusions. That's your role.
CHAIRMAN WYMER: Well, that's all of our
roles, and the conclusions will not be nearly as
radical as our observations I would think.
DR. CAMPBELL: I think Tae Ahn may have a
DR. AHN: Yes. I would like to provide
you with additional information. In our evaluation of
the early program, we have chosen a risk informed
approach, which means that in environmental conditions
that are concerned, for instance, we have randomly
chosen the barometer conditions.
We don't accept a hundred percent of a
highly acidic containing environment. In other words,
there is a distribution of the chemistry, and so I
would like you to consider that factor.
Also, in terms of regarding the
statistical analysis, again we have distributions, and
it is a risk informed approach, and it is not just the
single permissive value of years.
DR. SHEWMON: So this means that the Ph
and fluoride concentrations are different for these
little squares when the rate of corrosion in this
square is calculated? It doesn't have to be physical
in the sense that I was thinking of.
CHAIRMAN WYMER: Well, the whole concept
of risk informed is that it gets back to the business
of conservatism and credibility, and believability.
How risk informed are you if you really
don't understand the processes that make up the risk.
Just how informed are you, and in a sense you are risk
informed. But not as risk informed as you would like
DR. STEINDLER: No, I understand. That's
not a problem.
CHAIRMAN WYMER: We do have 15 minutes
left, and so let's break from what I said earlier, and
field any questions from the group.
AUDIENCE: Just a point of clarification.
There seems to be some concern about when the drip
shield fails and what it means. As far as the
corrosion of the waste package is concerned, we are
assuming the same environment on the waste package
with or without the drip shield.
The basis for that is that there is going
to be a lot dust and stuff like that on the panel
environment before the drip shield is raised, and they
may contain hygroscopic material.
And so when the humidity goes up, you are
likely to find as much acrose film on it that produces
humidity or whatever. So we are assuming the same
environment, and so the corrosion starts as soon as
the humidity threshold gets in.
CHAIRMAN WYMER: Well, that assumption
cannot be strictly true, of course, but it may be true
as to an approximation and that's okay. It can't be
true because in fact the composition of the water has
been changed by the process of corroding the drip
AUDIENCE: That's true, but what I am
saying is that it doesn't have to -- the water doesn't
have to come through the drip shield, because there is
an open environment between the drip shield and the
So when the humidity gets up to 50 percent
CHAIRMAN WYMER: Only the water is
AUDIENCE: Right. But then there is --
DR. SHEWMON: It came in by the gas phase
and not the --
CHAIRMAN WYMER: Most of the dust in our
observations collects on the tops of things and not
AUDIENCE: Well, the drip shield doesn't
replace until the water closure, and the waste package
has been sitting there for quite some time, and that
is an assumption in our model anyway. So I just
wanted to clarify that.
CHAIRMAN WYMER: So you are saying there
may be 300 years worth of dust?
AUDIENCE: Yes, exactly.
CHAIRMAN WYMER: That's a good point.
AUDIENCE: So all I was getting at was
that for the waste package to start corroding, it
doesn't have to wait for the drip shield to corrode.
CHAIRMAN WYMER: My original feeling about
airborne dust was that it didn't amount to much, but
the more I thought about it, the more I thought that,
gee, it does.
AUDIENCE: Well, there is going to be
ventilation going on, and I don't think the
ventilation have got filters in it.
DR. CRAGNOLINO: You may consider in the
future electronic components.
CHAIRMAN WYMER: Right. That's why they
DR. STEINDLER: And that raises the
question that I would have for Paul. Vapor phase
corrosion is one thing and liquid corrosion is
another. Would you equate the two, which is what I
think they seem to be doing, in terms of rates?
DR. SHEWMON: Well, no, if vapor stays
vapor, that you have got this magical monolayer or
whatever that has all the properties of a flowing
electrolyte, or even a stationary electrolyte.
DR. STEINDLER: I see. Okay.
DR. SHEWMON: You still have the problem
of waste buildup that isn't treated very well with
these cell approximations, but you still can bring
DR. STEINDLER: If the mechanism were like
glass, then you would be in trouble, because you can't
pile up enough silicate in glass to slow the reaction
DR. CAMPBELL: One of the things that
certainly I have noticed over the years in various
tours through Yucca Mountain is that you pass by these
placards and other things that when they are first put
through the DSF were nice and clean, and over time
those things have been heavily coated with dust.
And that is a process that is going to
occur when they are drilling these drips and --
DR. SHEWMON: What we need is a monsoon
every so often that will wash it all off.
DR. CAMPBELL: Wash it all out, right.
But over the operation period of the repository, you
definitely are going to have a significant build up of
stuff on the surfaces.
CHAIRMAN WYMER: Yes, I certainly after
reflection arrived at that position, too.
DR. STEINDLER: But, folks, that is a
different kind of material than something that has
been formed by evaporation of a soluble salt.
CHAIRMAN WYMER: Absolutely. It is a
solitious material for the most part.
DR. STEINDLER: So you kind of have to ask
the question what is this dust really going to
contribute on my magic two monolayer thick film on the
waste package or whatever.
CHAIRMAN WYMER: And to what extent will
it be washed off before anything happens. These are
all subtleties that have not been dealt with and are
almost impossible to deal with, and probably are not
important, although we don't know.
DR. STEINDLER: I suppose --
DR. SHEWMON: It is not the J-13 water
that comes in.
CHAIRMAN WYMER: It is not J-13 water for
DR. SHEWMON: It is pure water.
DR. CAMPBELL: And the layer of water on
this surface is not going to be J-13 water either.
CHAIRMAN WYMER: That's right.
DR. CAMPBELL: It is going to be some sort
of evaporative water.
DR. STEINDLER: It will be in equilibrium
with the atmosphere, and so it will have carbonate in
CHAIRMAN WYMER: That is about the one
thing that it can have, yes. The Phs spike up pretty
good temporarily, but they do not, however, ever spike
down in any of the models that we have seen.
DR. SHEWMON: That's interesting.
CHAIRMAN WYMER: And that is an
DR. SHEWMON: Most of these cell
approximations are in acids.
CHAIRMAN WYMER: Well, the one thing about
nitrous acid, of course, is that it is much more
active as a dissolving re-agent. It is more active
than nitrate acid, and it doesn't have the driving
force, but it has the kinetics that are in general
DR. SHEWMON: A minute ago we were saying
that the CO2 in the air would tend to drive the Ph up,
and then we have the nitric acid --
CHAIRMAN WYMER: And the cement.
DR. SHEWMON: And then how did we get it
CHAIRMAN WYMER: Radiolocist of nitrogen
in the air and actual oxygen, or peroxide radicals to
form nitric acid. Wasn't that your statement?
DR. SHEWMON: That is the only thing that
could spike it, yes.
DR. AHN: On the surface of the waste
package, we can include all tests on severe
environment. However, as I mentioned here, in the
risk assessment, those in the distribution, the actual
impact on the performance could be a small fraction
rather than failure, and we need to review the basis
for doing that, and --
CHAIRMAN WYMER: About the only
fundamental objections that I can make as to what has
been done is that it doesn't satisfy me
scientifically. But I think the bounding conditions
and the other assumptions that are made are
reasonable, and they cover --
DR. AHN: And that is the kinds of things
that we are reviewing.
CHAIRMAN WYMER: And it just doesn't
satisfy me that you really don't understand the
mineral, but still it is probably adequate for NRC's
purposes. It is a strange position to be put in for
a scientific area.
DR. LESLIE: Since Andy opened it up, this
is Bret Leslie from the NRC staff, and I guess I made
some notes as Ray started off the meeting this morning
on what this working group is trying to get at, which
is to come up with some further consensus on whether
the NRC process to resolve the issues is appropriate.
And I guess one of the things that comes
to my mind is that this has been a great scientific
discussion, but where has the evaluation of the
agreements that the NRC staff done?
CHAIRMAN WYMER: That will come, I hope,
DR. LESLIE: Okay. Because it looks like
there are several different discussions as you go
along and I am not hearing anything that is saying how
is this resolution process good or bad, and I am just
wondering when that is going to happen.
DR. STEINDLER: But you may have heard
some comments about the staff didn't seem to raise a
particular point, and that in itself I think is
CHAIRMAN WYMER: And that is what we are
digging at now.
DR. STEINDLER: If the staff accepts DOE
without any particular comment as you heard in the
conservatism issue, then that represents a question
that needs to be raised; why did they do that and
should they have done that is an issue that the
committee ultimately -- the ACNW ultimately will have
to decide, either to put in a message to the
Commissioners or not.
CHAIRMAN WYMER: I wanted to detail
chemical discussions in order to get everybody sort of
in the same ball park, and then we are going to back
off and say what does it mean, and is the process
getting NRC to where it needs to be to make the site
suitability, or contribute to that recommendation, and
to the license application.
But first I really wanted to dig into all
these chemistry issues and just see if we brought up
a snake to use an old southern expression.
It is very unlikely that we are going to
get any pythons, but we might get a few small snakes.
That's the way that the process is working here, Bret.
Tomorrow we need to actually address how
is the process working, and is it working, and how
independent of DOE's positions is the process, and how
much, if at all, are you being swept along by the DOE
tide, and there is a massive effort under way, and a
lot of money being spent, and are we being submerged,
or are we keeping our heads above water here.
DR. STEINDLER: I assume tomorrow morning
you are going to start at six o'clock?
CHAIRMAN WYMER: No earlier than that.
Absolutely. I think we ought to break for lunch. We
are due back at one o'clock.
(Whereupon, a luncheon recess was taken at
. A-F-T-E-R-N-O-O-N S-E-S-S-I-O-N
CHAIRMAN WYMER: All right. The first
topic here after lunch is the overview of the Near-
Field Chemistry issues and TSPA-SR Source-Term Model,
by Andy Campbell.
DR. CAMPBELL: Okay. And I am going to
basically do what I did earlier this morning, is we
will come back to this view graph from the DOE and the
FDA, which shows the key areas of concern, in terms of
Basically what I asked Marty to do was to
look at the chemistry inside the waste package, and
then I believe we were also going to talk a little bit
about how that mobilization, potential mobilization of
radionuclide extend and exit through the invert.
So that is basically the portion of the
system that we are looking at now at this point. In
terms of the flow diagram that we are looking at, the
in-waste package chemistry and corrosion, and
cladding, the degradation of the spent fuel, and the
transport of -- the potential transport of
radionuclides basically through the invert.
DOE, you will see, doesn't really have a
release model, per se. What they basically assume is
whatever water gets into the waste package, an equal
amount of water gets out of the waste package.
So they don't have a particular mechanism
or model for the contaminated water escaping from the
waste package. I am going to have to move this up and
One of the degradation mechanisms that
they are looking at is the corrosion of the cladding,
and the interaction of just that fuel with water, the
way the deal with that is not entirely obvious here.
But the fact that the waste package ports
are filled with glue, the assumption is made that the
entire waste package void space is filled with water,
and that is about 4-1/2 cubic meters of water.
It is an operating assumption that they
use in order to do the calculations. So even if --
and the input of water into the top of the waste
package is somewhere based upon their infiltration
models between about 1-1/2 liters per year to up to
150 liters per year.
And that is based upon different
percolation rates, and how much water is diverted and
so on. The assumption is that if water is dripping on
top of the waste package that it goes into the waste
I can't find an easy explanation, and in
the NRC's TPA model there is some diversion factor
that I talked about earlier for water to essentially
roll off the side of the waste package, as opposed to
going in, but it doesn't appear to be a DOE model.
So they have anywhere between 1-1/2 and
150 liters, and in TSPA that is abstracted into three
in-fluxes of water; 1-1/2, 15, and 150 liters per
And so then the water that comes out of
the waste package is an equivalent volume to the
incoming water. But, of course, that is now water
that is equilibrated with spent fuel, and the
materials inside the waste package, and that is all
done with this EQ36 reaction path code.
DR. STEINDLER: Well, I think that it is
important that their code, I believe, assumes
instantaneous mixing of that 4-1/2 cubic meters with
DR. CAMPBELL: This is a classic stirred
bath model. There is no nooks and crannies where you
get different chemistry than you do in the entire
bath. It is basically 4,500 liters of water that
starts out life with a composition similar to J-13.
And a bunch of materials that are going to
be inside the waste package, including certain
fractions of spent fuel available for interaction with
CHAIRMAN WYMER: Which is certainly a bad
assumption, because in order to have gotten through
the steel container, and in order to have gotten
inside rather I should say, it will have to have
dissolved some stuff to get in there, and that will --
the ingredients or whatever that is dissolved will be
present in the water.
DR. STEINDLER: But it only dissolves on
CHAIRMAN WYMER: How much difference this
will make you don't know, and I think that is the
point, that you don't know.
DR. STEINDLER: Well, I guess the thing
that concerned was that you have this large amount of
inventory, static inventory, which is diluted by in
the lowest case 1-1/2 liters in a year, and that has
undergone a small amount of reaction, relatively small
reaction, with the spent fuel, which is instantly
diluted by this 4-1/2 cubic meters.
And out of that soup now comes at some
time in the future, secondary mineral formation,
colloids and so forth, and so on, and it can make a
hell of a difference if that 4-1/2 cubic meters
DR. SHEWMON: Does it run out the bottom,
or does it have to diffuse out the top?
DR. CAMPBELL: Their model does not
account for it. It just magically goes from inside
the waste package to the top of the material -- at the
bottom, or underneath the waste package, and it is
DR. SHEWMON: Well, you know, both of
these assumptions are wrong, but how many orders of
magnitude would it change things? Did they do
anything to try and do that?
DR. CAMPBELL: At this point, they are
committed to looking at evaporative processes, but it
is not clear at all to me that they are going to look
at evaporative processes that minimize the amount of
water in the waste package.
They are just assuming that if we drill
holes in the top of it that we are going to drill
holes in the bottom of it, and that whatever gets in,
Now, I will give you an idea. The NRC
also has a bath model, but it is a spill-over model,
and the location of the whole in the side that spills
out is a sample perimeter. So it randomly samples
between the bottom and the top of the waste package.
So a certain fraction of waste packages on
average are about half-filled, just because of the way
that they do the sampling. And then it assumes that
there is a hole in the side, or up here, or down here,
that allows water out.
And then only the fuel, if I understand it
correctly, in the NRC model, only the fuel that would
be emersed in water could react with that water, or
some fraction of it.
DR. SHEWMON: For example, this gives the
zercoroy (phonetic) zero life around the fuel?
DR. CAMPBELL: No, in both the -- I
believe in the DOE model and in the NRC model, there
is some credit given to the zercoroy for cladding.
The way that is implemented in TSPA -- and I think in
TPA -- is that a fraction of the cladding of the fuel
is available to interact with the water, but not all
of it. Is that correct?
DR. AHN: Yes. Credit was given to
cladding by DOE and not by NRC.
DR. CAMPBELL: Okay. So in the TPA code
there is no cladding added.
DR. CODELL: It is in there.
DR. AHN: Yes, it is in there.
DR. CAMPBELL: That's what I thought.
DR. AHN: But not in this case.
DR. CAMPBELL: The NRC has a series of
alternative models that they have explored in their
own code which evaluate things like if you take credit
for cladding, and how will that affect your results.
And maybe you might address that at some point.
DR. STEINDLER: Now, cladding credited by
DOE is a relatively recent change, right?
CHAIRMAN WYMER: Yes, that's my
DR. STEINDLER: And that is the picture
that I have.
DR. CAMPBELL: But the way that they
present it is that they have some fraction of t
he fuel is available to interact with
water, and that is how they implement the cladding
credit. They do calculations on the side to determine
how much cladding has failed, and how much has not
DR. STEINDLER: And that fraction is a
function of time?
DR. CAMPBELL: Yes. And so not all the
fuel within the rods are available to interact with
the water. But what is available is assumed to reach
equilibrium with the entire 4-1/2 cubic meters of
water inside the waste package.
CHAIRMAN WYMER: Or it reached a steady
state at any rate, and presumably the water is
continually changing with time as well.
DR. CAMPBELL: The volume of input water
relative to the volume of the stirred bath --
CHAIRMAN WYMER: Is very small.
DR. CAMPBELL: -- is relatively small. So
the impacts on the chemistry of the input water is
relatively small. So from a purely calculational
view, you can see why this became an attractive model
to work with.
The concern that I have -- and this is my
own concern -- is that the water that gets into this
system and that can interact with this fuel, is not J-
It is some water that has undergone -- it
may have started out life somewhere in the ball park
of J-13, but it has gone through an evaporative
process, because even until you are several tens of
thousands of years down the road, the fuel is the
hottest thing in the repository.
CHAIRMAN WYMER: Well, it has got a lot of
iron in it, too.
DR. CAMPBELL: So there is an evaporative
process that is not accounted for, and so the
chemistry of this water is going to be more
concentrated than something like J-13, which is a
CHAIRMAN WYMER: The chances are of
reducing the water as well, since it will have gotten
in there by corroding the steel container.
DR. CAMPBELL: Well, presumably whatever
caused the corrosion to the container has left a hole
in it, and you can get water into that hole from the
But again you have got this large volume,
4-1/2 cubic meters of essentially buffer volume of
water in the system.
DR. STEINDLER: But the turnover in the
lowest flux case is 3,000 years. Your pictures came
out better than mine. I couldn't even read the print.
CHAIRMAN WYMER: What is your point,
DR. STEINDLER: Well, at a liter-and-a-
half per year influx rate, with a 4,500 liter
inventory, your turnover is something in the
neighborhood of 3,000 years. It gets to be 300,000
years for the highest flux.
It isn't very clear to me what that
assumption does for them. You know, that you have got
something other than essentially an empty container.
But it does confuse the chemistry.
CHAIRMAN WYMER: It certainly confuses the
chemistry. I think it does allow them to calculate
DR. STEINDLER: Well -- okay. How much
faith have you got in that EQ36 code?
CHAIRMAN WYMER: Well, you know, garbage
in and garbage out. Good data in and good data out.
It is the same old story.
DR. CAMPBELL: I will say that all of the
thermodynamic modeling codes have limitations. In
terms of applications, EQ36 is probably as good as
any. There maybe some that are better, and some that
are worse, but the key issue is the database that you
CHAIRMAN WYMER: That's exactly right.
DR. CAMPBELL: The mechanism and the
processes incorporated into those codes are all not
that different from one equal thermo code to another.
How you make up for limited data, the
biggest problem that I see with all these codes is
that they tend not to deal with co-precipitates. They
tend not to deal with salt solutions and things like
that, which are the real world.
CHAIRMAN WYMER: They tend not to put
everything in the water that is in the water.
DR. CAMPBELL: This is just showing what
I have already talked about briefly, in terms of what
the TSPA code is calculating, and there is a pCO2, the
partial pressure of carbon dioxide, and partial
pressure of oxygen, and Eh, the redox state of the
system, the ionics strength. And then the key species
are fluoride, chloride, and carbonate.
CHAIRMAN WYMER: They just have two time
regimes; one less than a thousand years and one
greater than a thousand years?
DR. CAMPBELL: Basically, because remember
that the key temperatures spike when you get a
significant temperature increases and are in that less
than a thousand year period.
CHAIRMAN WYMER: I was thinking that in
this other thing we had a while ago that they had
three temperature regimes.
DR. STEINDLER: Three time regimes.
CHAIRMAN WYMER: I'm sorry, yes, time
DR. CAMPBELL: Time regimes for the waste
packages. This is the in-package.
CHAIRMAN WYMER: What does that say that?
Does that say at temperature, or what does that say?
I can read the thousand years, but --
DR. CAMPBELL: At failure.
CHAIRMAN WYMER: At failure? Okay. That
is blurred to me.
DR. STEINDLER: But that fluoride is only
true for glass. I don't think they do much
calculations for it, for fluoride and UO2.
DR. CAMPBELL: Well, remember that they
are also looking at high level glass degradation in
the co-disposal containers.
DR. CRAGNOLINO: The fluoride is not
incorporated in order to deal with the solution of the
radiated uranium dioxide. It is used as a surrogate
for cladding. They have a model for the dissolution
of cladding, on the basis of cladding, and this is the
reason that it is there.
But it is not incorporated in the
barometric equation for the dissolution of the
DR. STEINDLER: Well, I sure missed that.
CHAIRMAN WYMER: So did I.
DR. CAMPBELL: And based upon the model,
these are the calculated in-package Phs, and I am
going to have to magnify again to see them. For the
commercial spent nuclear fuel -- and by the way, if
anybody is missing and needs extra copies, I can get
more made in case we need them.
Again, the interesting thing about this,
and that I found interesting, is the uncertainty based
upon the TSPA calculation, the Ph is larger in the
beginning than after the longer time frames.
That was just an observation. But these
are the -- well, somewhere between 4 and 7 of the
first thousand years.
DR. STEINDLER: Is there a message there?
DR. CAMPBELL: And between about 6 and a
little above 7 --
CHAIRMAN WYMER: What sends it down to
DR. CAMPBELL: Particular combinations of
corrosion, water flux, and other conditions.
CHAIRMAN WYMER: From the chromium?
DR. CRAGNOLINO: Yes, and what is in the
in-package calculations would between -- but all the
things that is inside the waste package, materials
that are together, are run and they come out with
CHAIRMAN WYMER: Well, the reason that
raises my interest is because you are getting down now
to Ph ranges where you can with iron reduce plutonium,
and to reduce the plutonium is a very significant
thing, and it is a danger as far as transport is
DR. CAMPBELL: Ray, the time here is a
CHAIRMAN WYMER: I see that.
DR. CAMPBELL: And you do have higher
temperatures in this regime.
CHAIRMAN WYMER: And something has to fail
in a thousand years for any of this to have any
meaning, of course.
DR. CAMPBELL: But the waste packages
ostensibly are -- oh, I'm sorry. I am incorrect, Ray.
This is time sense waste package failure. This is
1,000 years plus, and this was the initial amount of
water coming into the package and reacting with the
iron and stuff, and dropping the Ph down.
Then as more and more water and the
reaction regresses with time, the sense of failure,
you get a steady stay of environment if you will.
CHAIRMAN WYMER: But then you get into
some questions like how much or what is the oxygen
partial pressure over that period of time, and is
there enough iron in there to have for the first
thousand years to have consumed all the oxygen coming
in, and that would make a difference, too, of course
to the whole chemistry of everything.
DR. CAMPBELL: I think that is an
assumption on their part that the water is in
equilibrium with the atmosphere and the drip.
CHAIRMAN WYMER: Which may be a bad
DR. SHEWMON: And the drip is in
equilibrium with the atmosphere and the air above?
CHAIRMAN WYMER: Yes, that is an
DR. STEINDLER: Well, that one is not too
bad. I mean, there have been enough experiments done
in similar kinds of formations that showed the thing
breaths fairly --
CHAIRMAN WYMER: Except that this has
enough iron in it that it would consume oxygen for
maybe a thousand years and still be some more left.
DR. STEINDLER: Well, that's what we mean
by consuming oxygen.
DR. CAMPBELL: This is for the co-disposal
packages, where you have high level waste glass. And
again this is time sense failure of the waste package.
So this is sometime after 11,000 years, in terms of
But the long term Ph that the system goes
to is around 9, between 8-1/2 and 9. So you do have
the higher Ph in the co-disposal package.
CHAIRMAN WYMER: The packages of high
level waste from the very few processing plants and
spent fuel are co-mingled. So that what you
ultimately get in the aggregate is an average of these
Phs based on the weight of the amounts and the
relative corrosion rates.
And 10 percent of the waste approximately
is glass logs, and the other 90 percent is spent fuel.
DR. CAMPBELL: Well, the co-disposal
packages are interspersed with commercial spent fuel
packages. The majority of packages are commercial
spent fuel packages.
CHAIRMAN WYMER: Sure, 90 percent of them.
DR. CAMPBELL: But this is the in-package
Ph. This is the package with the Ph inside a co-
CHAIRMAN WYMER: There is no mingling of
anything, no real mechanism for that.
DR. CAMPBELL: No, not in their model, and
when you think about it, probably not in the real
world, except in the invert itself. But we will get
The way that they model the invert is
basically diffusion through --
CHAIRMAN WYMER: Straight down.
DR. CAMPBELL: Yes, straight down
CHAIRMAN WYMER: Well, I would think there
would be a little lateral fusion.
DR. CAMPBELL: This is the commercial
spent fuel degradation model showing the degradation
rate that they use as a function of Ph in temperature.
So at the higher Ph is the degradation rate, and it is
lower than the lower Phs; and of course the
degradation rate is higher at higher temperatures.
The cladding degradation model looks at
the unzipping function, and the cladding creep, local
corrosion, and actual physical failure of the cladding
due to some seismic event or series of seismic events
over time that cause material to fall on to or into an
open waste package.
The calculation includes the seepage into
and the temperature of the system.
DR. STEINDLER: Your prior one was the
degradation of the spent fuel form itself.
CHAIRMAN WYMER: Are you going to go
through all these view graphs, Andy? You are going to
have to hurry if you are.
DR. CAMPBELL: Okay. Let me hurry up.
Then the fraction of perforated cladding is shown on
the following slide. So as a base, they are assuming
a certain fraction of the cladding is perforated.
DR. SHEWMON: Now, is time zero from the
failure of the waste package? So is this a hundred-
thousand years after the 20,000 years?
DR. CAMPBELL: Paul, I don't know the
answer to that, and whether this is real repository
time, or post-waste package failure time for this
DR. AHN: After this cladding from the
reactor, there is an estimate of the initial phase,
and it runs from one percent to 10 to the minus 2, and
10 to the minus 3 percent.
Current DOE -- well, a couple of months
ago, we used 8 percent failure initially for a waste
package failure due to the -- during the interim
storage period because of high temperatures.
Then they sophisticated a model a couple
of weeks ago, and they were talking about 1.5 percent
initial failure now.
DR. CAMPBELL: This blue line on this is
8 percent by the way. And .1 would be 10 percent. So
this would be a thousand, 10,000 and a hundred-
thousand years after closure.
The next picture is just the variability
of the cladding unzipping rate. So they are looking
at a range of unzipping rates. The next figure is
DR. STEINDLER: Does that one make any
CHAIRMAN WYMER: That's always a good
DR. STEINDLER: It seems to me that the
unzipping rate should be a function of temperature.
You are basically forming a high volume, and the only
way you can get unzipping is really if you form a high
DR. CAMPBELL: If you start corroding the
DR. STEINDLER: But that rate is a strong
function of temperature. By the time you get to a
hundred-thousand or 10,000 years out, that temperature
is down fairly far. I wonder if that reaction still
Because there are two kinds of reactions
that take place. This isn't a simple oxidation to
U308, for example, which was a cladding standard
approach that the --
CHAIRMAN WYMER: It expands, and therefore
it breaks it up.
DR. STEINDLER: Well, I don't think that
is what you have got here.
DR. CAMPBELL: Well, that matrix
temperature is taken into account in this, and that's
why I put it back to that, and that is one of the
DR. AHN: There is another reaction, and
that is hydroxide formation, even at the lower
temperatures, can increase the volume, and I think
that is what they are probably talking about.
DR. STEINDLER: You think that is what
they are doing here?
DR. AHN: Yes.
DR. STEINDLER: Okay.
DR. CAMPBELL: This I just showed because
I was amazed at the huge range of glass degradation
rates that come out of this small uncertainty here,
and it doesn't decrease with time.
DR. SHEWMON: Now, is that a dissolution
rate, or what is this per year unit on a glass
degradation rate? Is it fraction dissolved per year?
DR. STEINDLER: Well, the initial process
is dissolution, but from there you quickly get the
DR. CAMPBELL: Right.
DR. STEINDLER: But I think this is just
the dissolution process that starts the formation of
the other products.
DR. CAMPBELL: So you have about four
orders of magnitude.
CHAIRMAN WYMER: And that is what it looks
like on there, is one per year, and what is that
DR. AHN: It is a fraction per year.
CHAIRMAN WYMER: That is an F, huh?
DR. CAMPBELL: Fraction per year.
DR. CRAGNOLINO: It is one over a year.
CHAIRMAN WYMER: And you are going to get
various silicates precipitated there.
DR. CAMPBELL: Yes, and they incorporate
that in the model. I mean, their model does include
all of that. The solubility model, and the main
radionucleides that they look at are in terms of an
actual solubility calculations are neptunium, uranium,
and americium, as a function of Ph, PCO2, and again
temperature in the in-package chemistry go into this.
CHAIRMAN WYMER: Does colloid formation go
DR. CAMPBELL: Colloid formation comes
after this, but yes. Let's see. What I am trying to
do is just give you an overview of these, and how they
are handling various aspects of --
CHAIRMAN WYMER: Well, they seem to
discuss colloid in terms of what we normally call
pseudo-colloids, and I haven't really seen colloids,
per se, addressed.
DR. CAMPBELL: The main issue is as you
say the pseudo-colloids.
CHAIRMAN WYMER: Who says?
DR. CAMPBELL: Plutonium to degradation
CHAIRMAN WYMER: Why is that assumed? We
all know that plutonium forms nice colloids.
DR. CAMPBELL: There is a very large
amount of glass --
DR. CODELL: I recall in one of the AMRs
that the quantity of plutonium colloids is much
CHAIRMAN WYMER: Then that would be the
explanation, the relative amounts, yeah.
DR. CAMPBELL: There is just a huge amount
of colloids produced through degradation processes
relative to view the natural system, or the true
CHAIRMAN WYMER: Well, do people know what
true plutonium colloids do with respect to forming
pseudo-colloids? To me that seems kind of like a key
question, because I think the first thing to form
would be the true plutonium colloid. So that's the
DR. CAMPBELL: Well, a lot of this is from
the glass degradation process, a lot of it.
DR. STEINDLER: What does it give to the
CHAIRMAN WYMER: Well, there is a lot of
solutious material in there.
DR. CLARKE: But how does it reversibly
attach to another colloid starting out life as a
CHAIRMAN WYMER: Well, there is a lot of
DR. CAMPBELL: Now, this is one of the
interesting aspects of the model, is this diffusion
through cracks. If you -- and I haven't done it
because you just end up with an infinite number of
But if you look at the DOE and TSPA
results, there is a clear change around 40,000 years,
and really before that period of time, between when
the waste packages begin failing due to essentially
stress corrosion cracking, to about 40,000 years, you
have what they call a diffusion dominated system,
where you have essentially small amounts of moisture
diffusing into the waste package.
Then again the assumption is that that
picks up radionucilides and diffuses out. What I
haven't been -- and I am still trying to track down,
is whether or not they are assuming that this waste
package with this diffusion dominated period is also
filled with 4-1/2 cubic meters of water. And I don't
know if anybody has an answer to that.
DR. CODELL: Well, we had a technical
exchange with DOE a month or so ago, I guess, where we
batted several of these things back and forth, and we
did some analyses on diffusion.
And we argued that DOE's model was way too
conservative, and apparently they don't have or did
not have it filled with water. The waste package
isn't filled with water, but there is water film
And that essentially on the inside of the
lid where you can get diffusion, the concentration of
whatever is diffusing is at the solubility limit. And
then it can diffuse through these stress corrosion
cracks to the outside, whereupon it is carried away by
Now, for this to happen -- and if you
don't mind my going on -- the waste package must be
tilted down so that the end cap is exposed up. That
is, one of the supports must fail, and this seems like
a low probability situation to me.
But it has to fail, because there is a
lift around the welds which would prevent liquid water
from the ceiling of the drip, to drip underneath that.
And that is one of the mechanisms. You must have
fresh water to carry this stuff away.
CHAIRMAN WYMER: Of course, the support
time will be gone.
DR. CODELL: Yes, but it seems like at the
very least half of them would fail, and then another
half would fail. But it seemed like a low probability
And then the other thing that really
bothered me about it was that they allowed the
diffusion to occur anywhere along the weld, wherever
the crack might occur.
Whereas, it seemed like the only place you
could really get diffusion would be at the bottom,
because the path for diffusion from the fuel would be
very tortious and very long, except maybe at the
bottom where you might have some crud or sediment
buildup, and you have a more direct category.
CHAIRMAN WYMER: Well, diffusion is one
thing and capillarities is another.
DR. CODELL: Well, this is diffusion.
CHAIRMAN WYMER: Well, presumably you are
getting some water moving all around through cracks
and through edges by capillary action.
DR. CODELL: Well, they are talking only
about diffusion. There are other phenomena here and
that might be, but that isn't in their model.
DR. CAMPBELL: It isn't part of their
model, and the other thing --
CHAIRMAN WYMER: It doesn't mean that it
DR. CAMPBELL: No, and it may be that that
process would dominate diffusion, but it is not in the
current model. The interesting thing about the way
they set up this diffusion model is the boundary
condition is always zero concentration.
DR. SHEWMON: At the external surface you
DR. CODELL: Yes.
DR. CAMPBELL: Right. So there is always
a driving force, a maximum driving force, because in
the real world you might have a diffusion radiant like
that, but eventually that would level itself out
because of the fact that diffusion would take place.
And the other interesting aspect is --
DR. STEINDLER: It is a conservative
DR. CAMPBELL: It is a very conservative
assumption. They don't take credit for degradation of
that radiant. It is always the steepest that it can
And for all intents and purposes, since
they are assuming that this film has some solubility
limits and concentrations are similar to what you get
in the big bath, as opposed to just the humid moist
environment inside the waste package.
The model also assumes through the invert
a boundary condition of zero concentration. So there
is always a driving force, that once the material gets
into the invert that it is always going to be
diffusing towards the unsaturated side.
Now, the other model that they use --
well, I have completely used other Marty's time here.
CHAIRMAN WYMER: Now you are 10 minutes
DR. STEINDLER: Great.
DR. CAMPBELL: And the other model is the
Advective model, where they use patches on top of the
waste package. There are a certain number of general
corrosion patches that are formed on top of the waste
package that allows water in.
And as we already saw, the water fills up
the waste package, and they assume that water comes
out somehow or other, and an equal amount comes in and
For those conditions, you have -- well,
this is kind of a cartoon of that, but advective flow
through the invert. But this really doesn't become a
dominant process until after 40,000 years, when there
is a sufficient general corrosion rate occurring to
allow enough open area on top of the waste package to
allow a significant amount of water in.
CHAIRMAN WYMER: But that is assuming a
11,000 year failure.
DR. CAMPBELL: Yes. Right. But as they
grow those patches, they grow with time. In fact, an
interesting outcome of their -- and it came up in the
context of the TSPASR presentation a few weeks ago
back in January, is that they do something called
neutralization analyses to try and get a handle on the
importance of different engineered systems.
And to do that they assume that a certain
number of patches occur on all the waste packages very
early on, but they don't grow with
grow with time. So the degradation model, which
assumes that those patches only grow with time, in
fact in some long time frame, overtakes the
neutralization analysis, in terms of dose, because the
patches are still growing with time.
DR. SHEWMON: And this is all premised on
a change in the ice glacial cycle, so that there is
always water flowing through this place.
DR. CAMPBELL: The general consensus -- if
I understand it correctly, the general consensus among
people who study climate is --
DR. SHEWMON: The answer is yes; just yes
DR. CAMPBELL: -- is that in the next
2,000 years we are going to go into a glacial climate
that is going to be around for many tens of thousands
of years, 150,000 years or more.
So we are in an unusually dry period for
CHAIRMAN WYMER: Aren't you glad you are
going to be dead, Andy?
DR. CAMPBELL: Okay. Uranium solubility.
These are just outputs of the TSPA model. This is
time and package failure, and this is for commercial
spent nuclear fuel. Again, this is being driven by
that change in Ph that we saw for the spent fuel.
And this is for co-disposal. So this is
the glass fuel. So the uranium solubility in the
higher Phs is high. The colloid model assumes that
you are generating colloids from the degradation of
the waste forms, and that radionuclides are both
irreversibly and reversibly attached to particles or
a certain fraction of the colloid particles, say
plutonium, for example, is always attached to it.
And with a certain fraction of the colloid
particles that plutonium can really exchange with the
aqueous environment. And then presumably if it is in
the aqueous phase, it can then also attach itself to
a mineral surface.
CHAIRMAN WYMER: If it is ionic, which it
DR. CAMPBELL: Right. But in general
then, the irreversibly attached or irreversible
colloids move on average much more quickly than the
reversible colloids, because you have some additional
This just simply shows how they divvy up
the -- how they do the colloids calculation. They do
take in to account some measure of colloid stability.
They have the colloids from high level waste glass,
and from iron oxy, hydrochloride hydroxide, corrosion
products, and from the natural ground waters.
And I think this again is hard to read,
but what I wanted to show here was the role of
colloids, and even on the hard copy it is difficult to
But anyhow it shows the plutonium as the
fraction of plutonium for total release and then the
reversible colloids. So at that point the whole idea
here was to kind of give you a flavor for how the
model is set up and some of the key areas of the
model. And with that, Marty, I will turn it over to
DR. STEINDLER: I don't have anything left
to say. That's fine. I did not look at the corrosion
of the cladding, or the stainless steel can in which
they poured glass, figuring that is a corrosion
problem that I don't know anything about.
So we are going to ignore for the moment
corrosion issues. I first tried to look at the source
term, and that is what I have got for uranium.
You have got a radiated UO2, and we have
a fair chunk of boron sulcate glass, and a literally
dog's breakfast's worth of DOE spent fuel, largely
metallic, but not entirely, and it contains things
like carbide and non-uranium containing material.
DR. SHEWMON: Are we in class or are we in
carbides, or both?
DR. STEINDLER: Both. Glass is strictly
the defense high level waste --
DR. SHEWMON: I understand.
DR. STEINDLER: -- generated by carbide
fuels, thorium fuels, et cetera, et cetera. There is
a lot more obviously than commercial spent fuel than
anything else, which is essentially UO2.
Water with unknown composition gets
through the cladding or the outside container, and
begins to react.
The first issue is in terms of release, is
how much in the way of fission products and what kind
have located in the cladding gap, and that is the gap
between the spent fuel pellets and the cladding.
I wouldn't say that you can get any number
that you want for that, but you can get quite a range,
and I think that is not very well defined. For the
most part, some iodine and -- a fair amount of iodine
and some technetium is brought out by that process.
Let me make a couple of other points. As
I mentioned, if you fish in UO2, you liberate two
oxygens, and half of those, one of those oxygens, is
taken up by fission products whose oxides are
essentially more stable than UO2.
And that generally takes place even in hot
water reactor fuel, and certainly takes place in fast
fuel that has a much higher internal temperature.
The other half of that oxygen gets
distributed between other fission products and
decreasing free energy, or more likely becomes
interstitial UO2, and it becomes interstitial oxygen
dissolved in UO2.
The point that I am making is that the
system tends towards being a reduced system, and in
addition there is this epsom phase that we talked
about before -- five component alloy, which is
metallic, and contains some, but not necessarily all,
of that terrible isotope called technetium.
I have not seen too much discussion on
that particular issue in any of the documents that I
CHAIRMAN WYMER: Can I comment at this
DR. STEINDLER: Well, I was just going to
make the other concluding issue, and that is in the
long run, in terms of the entire inventory of
available fission products, that may not make a great
deal of difference.
And I haven't looked at it from that
standpoint, but it could be the fact that nobody seems
to care is because it doesn't make any difference to
the downstream dose, which is really what people are
CHAIRMAN WYMER: I have talked to some
people in France who do the reprocessing work, and
they point out that there is always metallic
technetium left in the dissolver when they dissolve
that water in reactor fuel, and sometimes you can get
as much as a third of all of the technetium that is
present as undissolved material. And which is a
difficulty in concentrated nitric acid with a
DR. STEINDLER: And with a catalyst is the
CHAIRMAN WYMER: It is a very refractory
material. So that is an ameliorating factor I think
that hasn't even been considered, and it might reduce
the technetium downstream.
DR. STEINDLER: Well, it gets us into the
same discussion we had this morning, namely the
assumptions that DOE is making are conservative, and
as a consequence there isn't much point, I guess, to
arguing about issues which would reduce the technetium
content downstream or the rate. But it is a chemistry
CHAIRMAN WYMER: It is a chemistry issue.
DR. SHEWMON: Is the iodine that is
present after 10 or 20,000 years radioactive yet?
DR. STEINDLER: Yes. There is iodine-129
which has a 15 million year half-life, which is the
key -- well, the only iodine that --
CHAIRMAN WYMER: It is the only one of any
DR. SHEWMON: And the technetium is 99.
DR. STEINDLER: Yes, and it has a 200,000
year, give or take, half-life. I realize that iodine
has been well observed in the clad gap, but there is
enough iodine to be tied up, and there is enough
silver to be tieing up essentially all the iodine if
they had a chance to get together.
And ultimately everything absolves, and so
the question downstream into the unsaturated zone and
beyond is what are the odds that iodide will react
with silver that is migrating downstream. I have not
seen much discussion on that one.
DR. SHEWMON: It all dissolves because it
is infinite dilution finally.
DR. STEINDLER: Essentially. The thing
that puzzles me is that we have been told repeatedly
that the EH of that system is positive by a
Yet, iodide is the only specie that
anybody discusses, and that doesn't make a heck of a
lot of sense. I don't understand why that has been
maintained, again except for the fact that iodide
moves downstream faster than anything else probably.
But as you pointed out early, Ray, it
doesn't sound like good science, and you wonder what
else is wrong.
CHAIRMAN WYMER: Certainly the
observations have been that iodine whistles on through
DR. STEINDLER: Yes. But there is also a
pile of iodate minerals that exist that are reasonably
water stable, and so the opportunity for maintaining
a decent stability with low solubility of an iodine
oxygen compound strikes me as existing.
And I don't know whether that is an issue
either, except that it doesn't seem to hang science
CHAIRMAN WYMER: One of the problems with
iodine is that it does not form many highly insoluble
DR. STEINDLER: Not too many.
CHAIRMAN WYMER: Copper iodide is one of
the winners, and having said that, you have run the
course, unless you get into these more complex
minerals that have iodine tied up with them, which
formations doesn't seem entirely likely. So iodine is
always a problem.
DR. STEINDLER: Well, there are a couple
of iodates that are fairly insoluble. Whether or not
-- and iodates with fission product positive ions, and
so whether or not they exist --
CHAIRMAN WYMER: And I agree with you on
the anomaly of assuming iodide in --
DR. STEINDLER: Well, let's be fairly
clear that the thing that dissolves out of this whole
mess that people are interested in, or at least
transports, is technetium, iodine, neptunium, and
plutonium, as the first-line important nucleides.
CHAIRMAN WYMER: And one of the principal
liberating factors is the formation of the
tricarbonate, and you get it out of the way to release
DR. STEINDLER: Yes. And there is some
Carbon-14, and much further down, you begin to
generate and transport downstream things like radium.
Okay. We have discussed ad nauseam the
whole question of what kind of water do we have. We
won't have J-13 water. The models don't, I think, do
a good enough job that I can see -- whatever that
means -- in addressing trace elements, and their
behavior with very low concentrations of the things
that we are interested in.
So the solution process that we are
talking about here forms materials of concentrations
that are really far down in the mud. Solubility
limited concentrations are really quite small.
Somewhere I have got a list of them, but it is
probably for this discussion not particularly
important what the actual magnitudes are.
It is that the abstraction that DOE has
gotten into, and which apparently works well enough
for them and the staff so that he staff has not
objected too strongly, is that rates are fundamentally
Ph driven, aside from temperature, if oxygen and CO2
are controlled, when they are controlled by
That's not totally true for glass, where
silica is an important influence in the rate. But
essentially these are Ph driven dissolutions. They
seem to work reasonably well.
Glass dissolutions have a strange set of
kinetics as you know. But for the purpose of a
repository type material, glass is a fairly modest
contributor to the total isotope pushed downstream.
Some people don't seem to get too badly
bent out of shape about the fair uncertainties in the
case of glass.
CHAIRMAN WYMER: Well, the saving grace,
of course, with the glass is that the plutonium has
been taken out.
DR. STEINDLER: Yes, but you do have a lot
of neptunium in places, and also a lot of technetium.
CHAIRMAN WYMER: That's right.
DR. STEINDLER: You have got a lot of
technetium everywhere, except for cement in the river.
CHAIRMAN WYMER: Well, there is very
little burnt up stuff, and so a lot of these things
are not there.
DR. STEINDLER: It's not a particular
issue. Okay. What else is there of real importance?
Oh. The fission products that move downstream that we
are not interested in are believed to arrive in
solution by simply congruent dissolution of UO2.
I think that is probably not a bad
assumption. Besides, it doesn't make any difference,
because we are not watching them. I mean, they are
not contributors to the dose. They are elemental
contributors, but they are not contributors to the
Colloids are a different story, and Andy
has kind of outlined what the colloid situation is.
There are two kinds of colloids; those in which there
is a reversible absorption, and colloids which are
nominally called irreversible, but it is not
absorption. It is co-precipitation.
CHAIRMAN WYMER: Those are pseudo.
DR. STEINDLER: Well, whether they are
colloids or pseudo colloids reminds me of how many
angels can dance on the head of a pin.
CHAIRMAN WYMER: Well, if you are going to
talk about reversible and irreversible, then it has
got to be pseudo colloids.
DR. CLARKE: Reversible or irreversible?
DR. STEINDLER: There are two kinds of
DR. CLARKE: I think that's right. There
is a different term in different documents for the
DR. STEINDLER: Yes. Glass is really the
only source of minerals to which you get co-
precipitation, which becomes irreversible. The others
are all obtained from fuel.
There is a bucket of secondary products,
and I simply want to reiterate my puzzlement that in
the DOE models, commercial spent nuclear fuel
dissolves to form copper minerals.
CHAIRMAN WYMER: Do what?
DR. STEINDLER: To form copper minerals.
CHAIRMAN WYMER: That's a novel trick.
DR. STEINDLER: Well, I thought that was
kind of an interesting trick, and so I read it again,
and it is there. What I haven't found where the
And if you are old enough, you recognize
that plutonium at one time was hidden under the code
word copper. But you have to be even older than Ray
in order to --
CHAIRMAN WYMER: Nobody is older than me.
DR. SHEWMON: Hardly a man is now alive
that remembers that famous day and year.
DR. STEINDLER: You're right. And then
they had to distinguish between copper and honest to
god copper when they wanted to talk about real copper.
And in the case of fuel, they do form lots
of silicates. The oxides and hydrous oxides,
depending on what Ph range you are in, of plutonium,
and an oxy carbonate for plutonium, or Neptunium-5, is
an important actor in this thing.
In the case of solid products, and in the
case of things like glass, obviously include borates,
because you have got boron sulfate glass, and nothing
is particularly surprising.
So as this soup dissolves, I hand to Jim,
moving into the unsaturated cell, a pretty dilute
aqueous solution, which is basically a carbonate base.
It has got a Ph, depending on where and when you are
looking at it.
And it varies -- what did we say -- from
4 to 8 about. It has colloids in it that are
important to the folks downstream. It will have
technetium, claimed to be entirely as Technetium-7,
rapidly moving with the waterfront.
And the same thing with iodine. A large
fraction of the neptunium is Neptunium-5, which in the
absence of a large amount of carbonate, will also move
the waterfront. And that is basically what I hand
CHAIRMAN WYMER: And all these things are
modified by whatever secondary phases are formed on
the surface of the fuel that will attenuate, absorb,
or otherwise diminish what comes out the bottom.
DR. STEINDLER: Well, I don't think there
is much claim for excessive absorption on those
CHAIRMAN WYMER: There is not much
claimed, but the question is how much is there.
DR. STEINDLER: That remains to be seen.
CHAIRMAN WYMER: I am not sure it matters,
of course, because if they assume it all comes out,
and it still looks okay, then what is the problem.
DR. STEINDLER: Well, what is the role of
the colloids? The role of the colloids is that they
move a lot faster than stuff that is absorbed and
desorbed, especially with reasonably high distribution
So the concentration of colloids, and the
concentration of actinides on those colloids get to be
a big issue, largely lousy data, and that is my
judgment, and not DOE's obviously.
I think the staff -- and to go back to the
issues at hand, but I think the staff is aware that
the data aren't very good. I have not delved hard
enough into how loudly the staff is complaining that
the data are not very good.
But it could make a significant difference
to the downstream answer. The redux conditions I have
already commented on. I am puzzled by what is
elected, but I can understand if you want to be
conservative, the election of a continuously oxidizing
system can be justified reasonably well.
Whether you would find the technetium
oxide or technetium sulfide that you could form would
remain stable long enough to make any difference in
the technetium downstream. I don't think there is
enough answers on the ability to form technetium and
its rate of oxidation in a system that is as dilute as
CHAIRMAN WYMER: The sulfate is very
DR. STEINDLER: Right. We know that, but
I have not seen any data on oxidation rates. There is
some discussion in a bunch of these documents on the
importance of the surface alpha radiation in modifying
both the Ph, as well as the ionic content, which was
a comment back there.
It isn't the gamma radiation, which at
times is down to the point where it is fairly weak.
It is strictly the alpha flux at the surface. The
folks at the lab have looked at that, and I have not
read their paper, and so I don't know whether that
data is any good. I have to assume that it at least
passed the referees.
I am a little bit disturbed frankly on a
personal basis that trace elements in the water are
not being considered adequately, and that may be
unfair. I will have to look some more. But fluoride,
it seems to me, complexes tremendously with plutonium.
Every good analytical chemist understands
that. I don't see that recognition in the documents
that I have looked at.
CHAIRMAN WYMER: And in an sufficient
amount, it also precipitates it.
DR. STEINDLER: Yes, in those
concentrations. But, I mean, at low concentrations
you can get the Plutonium-4 monofluoride in solution
that becomes inert fairly quickly. So if somebody
assumes this stuff absorbed, maybe that is the wrong
DR. SHEWMON: Inert means it won't absorb?
DR. STEINDLER: Right. I have looked at
very few of the specific things that we were -- that
I guess that I was supposed to have looked at, mainly
what is the staff process and issue resolution.
But my contention is that the staff still
thinks they are looking at science, and that they are
asking questions which you would ask if you were a
referee of a journal article; show me more evidence of
a particular point.
CHAIRMAN WYMER: That's what I always say,
is where is the data. Show me the data.
DR. STEINDLER: Fine. But what I don't
see is -- and it seems to be rather broad, and the
amount of information requested is substantial.
What I don't see is a follow-on sentence
at the bottom of that saying the reason that we need
this answer is because it makes a difference here,
here, and here, and that influences your downstream
dose. I don't see that connection too readily.
CHAIRMAN WYMER: Let me add a little
footnote to your fluoride discussion. There is in
fact, but it amounts to a lot of getters for fluoride,
in the rare earth. So it isn't always plutonium. It
may be only a tiny fraction of it does, because
obviously the insolubility of it varies in fluorides.
DR. STEINDLER: Right. Although I think
the oxides are more stable than the fluorides in that
CHAIRMAN WYMER: Depending on the
DR. STEINDLER: Yes, depending on the
solution or in this system. That in a very truncated
fashion is my view of the world, a very narrow slice
of a narrow slice. What have I left out, Andy? I'm
sure that I have left out lots.
DR. CAMPBELL: You mean that I am supposed
to play --
DR. STEINDLER: No, but aren't you part of
my issue resolution problem?
CHAIRMAN WYMER: Well, I don't see a whole
lot of sense in me going on at any great length about
the in-drift chemical environment which we have been
discussing directly and indirectly since this morning
-- and we all know that --
DR. STEINDLER: Well, let me just make one
comment. Do I sense -- if I address the question,
does it look like the staff is holding DOE's feet to
the fire adequately so that at least in the narrow
area of chemistry of the fuel dissolution process, the
in-waste form chemistry, that the answers are likely
to be correct and good enough for what is to be done,
but they won't pass a journal article referee?
I think that my tentative answer is, yes,
I think the staff has got a fair handle on what the
system looks like, and what it ought to look like, and
what DOE is doing in order to describe it.
CHAIRMAN WYMER: Well, we now really are
talking about the discussion of issue resolution key
concerns here, which --
DR. STEINDLER: Have I jumped in the wrong
DR. CAMPBELL: No, it is the right place.
CHAIRMAN WYMER: But I think that's right.
DR. STEINDLER: And that is my very rough
CHAIRMAN WYMER: That's where I would call
it, too. I think it is a statement that we have
discussed informally earlier, that the issue
resolution process as it is structured doesn't really
have much opportunity for input other than what DOE
brings us as their answers to the issue resolution,
and then the response that the NRC has, and who says
I need more information, more data, and where in the
world did you ever get that conclusion from.
But it is very encouraging to me that the
NRC staff has gone outside that box, and said, for
example, have you guys considered secondary phase --
and this is NRC and the center -- and have you
considered secondary phase formation, and don't you
think it is important.
And DOE says, no, we haven't, and it is
not important, and then they start considering it.
That goes outside the box a little bit, and that is
really not within the formal issue resolution
structure, because it wasn't an issue. It didn't come
DR. CLARKE: It would help me, Ray, if I
understood better what the objective of the issue
resolution process is. If the objective is to resolve
issues that are on the table, that's one thing.
If the objective is more than that, then
that is something else. So, you know, from what I
have seen, I think the issues that are on the table,
however they got on the table -- and I am new to this
process, do get resolved, or aren't in the process of
CHAIRMAN WYMER: They do, yes.
DR. CLARKE: There is a good back and
forth, and there is a spirited scientific exchange at
these meetings, and I think all of that is very
If the process is supposed to do more than
that, and if it is supposed to from time to time
revisit other issues, or if it is supposed to identify
new things, then that's something else.
CHAIRMAN WYMER: Well, I would guess that
it has changed over time with respect to what it did.
I think initially there were -- that there was
probably a flood gate of issues, and the flood gate
was opened up, and out flowed the issues.
And DOE sat there and said, oh, my god,
and it focused down after a while to where there was
agreement by back and forth discussions between DOE
and the NRC. And this is my perception, and if
anybody in the room wants to say it is wrong, please
DR. AHN: I would like to comment on the
issues of the original process with a couple of
examples. One is regarding the secondary minerals.
We discussed this subject with DOE substantially.
However, I don't think we need to be
prescriptive to DOE. DOE has the flexibility to use
their own methods to apply for a license. Therefore,
as long as there current thought is conservative, or
in other words, they don't give credit to secondary
minerals, and not underestimate the performance
objectives of the proposed 63, therefore, we do not
ask in more descriptive ways for this particular
Regarding the radionuclide effect, even
though it will decay away after continual failure
substantially, still there is the possibility on the
surface of cladding from the residual gamma ray, and
that may end up with a nitrogen cessation and lowering
Ph and so on.
In the patch exchange, we raised those
issues and DOE agreed to analyze that. Andy brought
up today the Division 3 IRSR, and that IRSR included
a background of all DOE's AMRs and PMRs, and the
dissolution processes, and I included it, because that
division was prepared after the issue of the
There are numerous subject concerns which
we judge in the agreement for DOE to conduct what we
asked them to do.
CHAIRMAN WYMER: I didn't raise the point
of secondary phase formation so much because I believe
that DOE must have secondary phases, but to point out
that in the NRC there is the ability and desire to
think out of the box a little bit.
That they aren't constrained by this
fairly -- what turned out to be a fairly formal issue
resolution process at this point, and I am sure that
has evolved to that over time with a lot of back and
forths and agreements.
But now it is quite a formalized process,
with very sharply defined key technical issues and
subissues. But to me it was encouraging that
something that was not actually an issue that was
written down that somebody recognized was introduced,
and it suggests to me that the staff and the centers
are thinking creatively about this thing, and they are
willing to throw something else in the hopper if they
see it and think it is significant, and not to be
DR. STEINDLER: Ultimately, if my limited
experience is any indication, both the staff, the NRC
staff, and DOE, will stand in front of a Safety and
Licensing Board Panel and defend themselves against
It is at that point where you find out
-- and I assume we will learn that before that point,
but it is at that point that you find out whether or
not both the staff and the NRC, and the DOE, have left
Because nothing could be more embarrassing
it seems to me than to come to a licensing hearing,
and prepared with 10,000 pages of documents apiece,
and have somebody from the intervenors stand up and
say, guys, you missed an important issue, and here it
is, and you are in trouble.
CHAIRMAN WYMER: And because of that sort
of thing, it seems to me that it would be -- that it
is worthwhile for the NRC and the Center to
periodically stand back. I know that they are all
running like crazy just trying to keep up with things,
and they are overworked and understaffed as usual.
But every once in a while some time should
be taken to stand back and say, okay, we are emersed
in this process, but now that we have explored all
these issues, and we have exposed our mind to
continuing an accumulation of facts, are there any new
things, and to just take a minute, and sit back, and
reflect on whether or not they really have covered the
things that they should cover.
DR. STEINDLER: Well, the Commissioners
are certainly going to ask that of the advisory
CHAIRMAN WYMER: Yes, and that is our
DR. STEINDLER: And they have a right to
get a decent answer out of the advisory committee.
CHAIRMAN WYMER: But we are not in as good
a position to do it as the staff is, because we are
not steeped in the lore of the business.
DR. STEINDLER: I know, because you are
CHAIRMAN WYMER: Yes, but the NRC is
supposed to be independent.
DR. STEINDLER: No, I am talking about the
advice that you give to the Commissioners. The
Commissioners are going to say, you know, has the
staff done -- and they probably care a little bit
less, I assume, about DOE, but has the staff done a
comprehensive job in looking at all of the necessary
aspects of it so that they don't get blindsided when
the intervenors stand up.
CHAIRMAN WYMER: And it seems to be this
issue that we discussed earlier, and I will come back
to it again as being important, that from the point of
view of credibility, taking or wrapping too much up in
bounding assumptions, or wrapping too much up in
conservatism, leaves a point of attack open for
They say that the science is not credible.
Now, maybe this doesn't make any difference, but it is
an argument that can be made. This is not a
scientific method, and it doesn't take a whole lot to
poison people's minds, and to turn their minds, even
though it is down a blind alley, and they want to run
down the blind alley.
DR. STEINDLER: I will have you know that
the Atomic Safety Licensing Board Panels are not
easily poisoned. I've been there.
DR. CAMPBELL: Ray, Tae Ahn has a point.
DR. AHN: Please don't misunderstand the
prescriptive or what I mention to you. The fact is
that in our TPSA code, we used secondary minerals in
the distribution model, and we presented a background,
and our base case model of spent fuel dissolution
included secondary minerals. However, DOE did not.
We did not discuss that issue because DOE
chose a more conservative approach. And I would like
to inform you of that.
CHAIRMAN WYMER: And I think that is very
encouraging personally that the NRC staff has included
things in their code that are not in the original
code, because that demonstrates independence.
And one of the real questions we have been
asked is just how independent are these codes. Are
they really taking different looks at the same thing,
or are they taking the same look at the same thing.
And the more dependence that you can demonstrate, the
more comfortable I can be.
DR. AHN: And also there is another ACNW
comment a year ago, and because DOE chose a very
conservative spent fuel dissolution model, they ended
up with giving credit to cladding. That introduced
another system uncertainties.
On the other hand, we chose the realistic
spent fuel dissolution model, and we took the
protection of secondary minerals, and we do not need
to credit cladding without introducing other
CHAIRMAN WYMER: If you can get a good
result both ways as support.
DR. CAMPBELL: Well, let me chime in here
about a problem that has been nagging me for a while,
and in which I know at least one or two people on the
staff are bothered by it.
And it is in the context of DOE's
neutralization analyses, and when they "neutralize"
the waste package, which I briefly mentioned before,
they get fairly high doses.
And when the NRC in their model does
something equivalent to that, they get doses that are
more than on an order of magnitude lower. And at this
point in time, I do not see why in one case do you get
doses up in the range of a rem when you "neutralize
the waste package," even though it is understood that
that is kind of an artificial process by DOE.
And when something similar in NRC's TPA
code is done, and not even accounting for secondary
phases, but just in terms of the release models and
everything, and they neutralize the waste package, and
they get doses in the range of 30 mill-rem, somewhere
in that ball park.
CHAIRMAN WYMER: That doesn't give you a
warm and fuzzy feeling does it?
DR. CAMPBELL: But the question is why.
What is different about the approach that DOE is doing
with its model and what NRC is doing. And it is not
clear to me -- and I think part of the answer might be
this way they handle diffusion, setting boundary
values that are always zero.
But that may not be the answer, and I
think that --
CHAIRMAN WYMER: That was the mechanics of
the neutralization? The way they do their sensitive
DR. CAMPBELL: It may be, but the question
is has DOE and NRC going through a licensing process
from the pre-licensing process, at some point this
will come up as an issue, with what are the
differences between the models and why should there be
this kind of large difference?
Is it some simple conservatism built into
the DOE model that isn't built into the NRC model, or
is there something more fundamental going on.
CHAIRMAN WYMER: Do you want to address
DR. CAMPBELL: In order to establish the
credibility of that, there needs to be a better
understanding of why those differences occur, because
you get to the question of which is right.
DR. CODELL: Richard Codell. Well, a lot
of individualization analyses would probably answer
DR. CAMPBELL: Well, I know that this has
bothered Tim for a while.
CHAIRMAN WYMER: And Tim doesn't know the
DR. CAMPBELL: I don't know if he does or
doesn't, but I don't know the answer.
CHAIRMAN WYMER: Well, if it bothers him,
he probably doesn't.
DR. CAMPBELL: It certainly is an area of
concern, where you get these huge differences between
the models which ostensibly represent the same basic
system in slightly different ways, or maybe more than
slightly different ways.
And when you do something similar with one
model, and with the other model you get dramatically
different results -- well, if there is an answer, I
would like to hear it. Up to date, I have not heard
a real good explanation for that.
And at first we were, frankly, a little
shocked when we saw these utilization analyses come
out. You know, why is that, and DOE has changed its
model, and the design has evolved.
But fundamentally you are getting the same
sort of dose versus time --
DR. SHEWMON: DOE gets the high value or
the low value?
DR. CAMPBELL: The high, the high value.
CHAIRMAN WYMER: Now, John Kessler, and
the contractors from EPRI have just very recently
issued their total system performance assessment, and
they pretty much agree with the DOE results and have
come out with the conclusion.
I don't know about this particular issue,
but they came out with the conclusion that everything
looks okay, but they are buying into the DOE's
arguments that the waste repository is fine, but that
is a total independent analysis.
DR. CAMPBELL: You are talking about
CHAIRMAN WYMER: Yes, it just came out.
DR. CAMPBELL: But anyhow, with that
scenario, and my scientific curiosity was tweaked a
little bit by what aspects of how they are modeling,
or differences between these two approaches to
modeling in the system are driving those kinds of
Because at an early time frame, you are
looking basically at the difference between something
in the ball park of compliance and something that is
really out of compliance.
And it is only because the waste packages
last that there are other things going on, but because
the waste packages last for long time frames, past
10,000 years, that this really isn't an issue.
CHAIRMAN WYMER: Well, Jim, why don't you
launch into your presentation.
DR. CLARKE: Could we take a break, as I
have to set up my projector.
CHAIRMAN WYMER: That sounds good to me.
(Whereupon, the meeting was recessed at
2:27 p.m., and was again resumed at 2:40 p.m.)
CHAIRMAN WYMER: Okay. My name is Jim
Clark. I am new to Yucca Mountain and new to this
process. I recently joined the faculty at Vanderbilt
University after 25 years in the private sector.
And my objective today is to provide an
overview of the radionuclide transport, and I will
call it issues and understandings as I know it. My
understanding is increasing daily, and I am still at
the connect-the-dots stage, and some of the dots
appearing to be moving.
And so if I mis-speak, you know, please
jump in. I know that John is here, and Bill, and
anyone, please jump in and correct me. But basically
I would like to just quickly overview the transport
And my focus will really be on the
transport processes, and not so much the actual
modeling. But more of the processes and the issues.
And then look at the key technical issue for
radionuclide transport, the sub-issues, and the status
of that situation.
And if we start out with -- and this is
going to be hard to see, as this is from a paper in
published literature. And is sort of a view from
20,000 feet of Yucca Mountain, and from the transport
side, we have the repository right in here, and we
have about 300 meters below the surface, and we have
an unsaturated zone again about 300 meters.
And then we have a compliance point about
20 kilometers down gradient in alluvium, and here
under the repository, and we have volcanic units,
which are welded and non-welded just to give a very
simple explanation, in the unsaturated zone.
We have a transition point between
volcanic units and alluvium, the location of which is
still uncertain, but there is work being done by Nye
County that is attempting to reduce the uncertainty
associated with that.
So, the repository, unsaturated zone,
saturated zone, and alluvium, and a volcanic saturated
zone, and alluvium. Andy spoke about the
classifications that are being used for colloidal
material, and we have had some discussion about that.
As I understand it, the irreversibly bound
colloids are called true colloids, and the
radionuclide is permanently over the time scale of
interest, which is often long, are attached to and are
really incorporated into the colloid.
So that the definition, Ray, I think
really reflects the state of the radionuclide.
radionuclide, and not so much the colloid; and a
reversible bound colloid would be also what is called
a pseudo colloid. Here the radionuclide can partition
between the colloid, whether it is natural or waste
So part of the time it is present on the
colloid, and part of the time it could be in a mobile
aqueous phase, or it could be transported as a
CHAIRMAN WYMER: And I would argue that
there is another colloid, which is a real colloid, as
opposed to a true colloid.
DR. CLARKE: I am not going to argue with
you. The transport assumptions maybe we should review
quickly. If you are an irreversibly bound colloid,
you are transported as a dissolved solute with respect
to advection and dispersion in the zones of water that
However, there are a couple of rules. You
are not permitted to diffuse into the rock matrix in
zones where flow is fractured, controlled, and matrix
diffusion is being considered.
And you can be attenuated through
filtration processes which are being modeled through
a retardation approach.
If you are a reversibly bound colloid,
then you are transported as an IDC when you are bound,
and as a dissolved solute when you are not.
DR. STEINDLER: Do you think as a colloid
moves from an area of EH and PH ionic strength
stability to one, where the principal is unstable, and
then back, that that process will regenerate a
DR. CLARKE: I can't answer that. I am
not sure how to answer that. I think stability issues
are being considered from the standpoint of the amount
And I think for the remainder of this
presentation I am just going to be showing a few
CHAIRMAN WYMER: Let me ask you a
question, Jim. In anything that you have run across
did you see any discussion of what happens if during
the transport of a colloid, however defined, is
chemically altered by a reduction and what this does
to the process, and whether that has even been taken
For example, I read something that said
humic substances in J-13 well water could affect the
oxidation by reducing --
DR. STEINDLER: Isn't that an assumption
on the part of DOE, that there are no colloids in the
incoming J-13 type water?
CHAIRMAN WYMER: Oh, this would be in the
humic acid materials that are present after --
DR. STEINDLER: I know, but they have
defined them out of the system is what I am saying.
CHAIRMAN WYMER: Well, yes, out of the
incoming system, but out of the emulgent system where
you get into transport processes.
DR. STEINDLER: But there are no source of
organics that they are willing to admit to.
CHAIRMAN WYMER: But it could affect
oxidation, but there is no further discussion that I
DR. CLARKE: My understanding at this
point, Ray, is that if you look at the reversibly
bound colloids, they are being handled through a
partitioning approach, Kd, and Kd has been developed
for americium, and that is the one that is being used
for those colloids that would be expected to be
Now, as far as the chemistry beyond that,
I really haven't come across anything, but that
doesn't mean that it doesn't exist.
CHAIRMAN WYMER: Well, I looked at the big
write-up on colloids, and they mention the possibility
of there being organic acids down in the stuff beneath
But they don't say, okay, suppose we
reduce the patched species, and we will chemically
reduce it. What then? Certainly the whole picture
changes, and with colloids that is potentially
DR. CLARKE: They are maybe being looked
at as a process, and to the extent that is being
incorporated into the model --
CHAIRMAN WYMER: But you haven't seen it?
DR. CLARKE: No, but that doesn't mean it
CHAIRMAN WYMER: Well, that's true. There
is so much literature on it.
DR. CLARKE: And again one of my concerns
is that there does appear to be a fair degree of
fragmentation among the issues, and some of the issues
are obviously interrelated and is some critical
The process -- it does appear that the
objectives of the process do appear to be driving the
reports and the format of the reports, so that you can
in looking at an issue find those things that correct
CHAIRMAN WYMER: Okay.
DR. CLARKE: And there are process model
points that are more comprehensive, and there are
analytical model reports that are more focused. But
I haven't seen anything that goes to both points.
In any event, just to very simplistically
talk about the subsurface of the model, the
unsaturated zone below the repository consists of
welded tops and non-welded tops, and the welded tops
would be treated as fractured systems, with the flow
through the fractures.
And the possibility of a matrix diffusion
into the rocks and matrix. The non-welded tops as I
understand it are being treated more as a forest
matrix, where there is flow through the rock matrix
itself, with a distinction between areas which are
zeolitic and which you would expect to have very high
sorption and capacities in vitric areas. And that is
just a very simplistic review.
I am going to skip over to the saturated
zone, and again this is in the book. As I understand
it, the saturated zone is being treated as below
fracture control, or correction, flow and control, or
at least everything that I have seen has indicated
DR. SHEWMON: Is the saturated zone below
the water table?
DR. CLARKE: The saturated zone in the
volcanic units, yes. The saturated zone in the
volcanic units is being treated as fracture flow
control, and the saturated zone alluvium has been
treated as such, and so we have flow in the fractures,
and various things that can happen.
We have vection in the fracture defusing
into the so-called immobile water in the rock matrix,
and it would be an attenuation process for
radionuclides, and we can have sorption on the
In principle, we can have sorption on the
surfaces of the fractures, and we could have
sorptions in the rock matrix. And I think depending
on which model you are looking at, sorption in the
shield is included or not on that kind of a scale.
And when you are in the alluvium, then
this is being handled with an effective porosity, and
these fracture flow models are really dual-porosity
models, and that is the current approach.
And you have flow through with the whole
matrix, with the potential sorption on the surface.
You also have advection as well.
So that the major attenuation processes,
at least two of the major attenuation processes would
be matrix diffusion and sorption. And, for example,
a fracture flow control domain, and if you had no
matrix diffusion, you would have a flow moving in the
fracture with some advection and dispersion.
If you have matrix diffusion, then you
have attenuation of the radionuclides, and diffuse
into the matrix, and the flow direction being this
direction, and with the sorption and matrix diffusion,
then you have a flow direction like this. And you can
get significant attenuation through these processes.
It is hard to see the flow paths, but I
think it is considered to be pretty much coming out of
the repository and going to the southeast, and then
coming back and going to the southwest, and that is
about the predominant flow path based on not only
hydraulic data, but in chemistry data as well.
I will say that in one of the meetings
that I attended there was some concern about that, and
there is some concern on the part of some that
anything coming out of the repository could go deeper
and into the saturated zone.
The other side of that story is that as
you go into the saturated zone with depths, the
vertical gradients are up. So that would support a
plu coming out of repository and kind of riding the
top of the water table.
DR. SHEWMON: And the gradient for what is
DR. CLARKE: The vertical gradient.
DR. SHEWMON: For what?
DR. CLARKE: For what at different depths.
DR. SHEWMON: A change of something for
what, for something?
DR. CLARKE: A change in elevation.
CHAIRMAN WYMER: Is it a gravity motivated
process; is that what you are saying?
DR. CLARKE: No, I am saying that the
force, if you will, would be upward.
DR. SHEWMON: Something is forcing the
water upward through this medium, or are you talking
about the transport or diffusion of an ion?
DR. CLARKE: I am talking about the
pressure levels of the water.
DR. SHEWMON: It's either that I don't
understand that, or it is so obvious that it is
trivial. Go ahead.
CHAIRMAN WYMER: I have a little trouble
with it, too.
DR. MCCARTIN: It is a gravity induced
CHAIRMAN WYMER: Okay. That's what I
DR. SHEWMON: Well, it is a pressure
grade, because of the gravitational field; and if you
go down in water, the pressure always gets higher.
DR. MCCARTIN: This is higher than that.
There is a connection between the upper and lower
rock, such that you are maintaining a higher pressure
for the lower output.
DR. SHEWMON: So one tends to permeate
DR. MCCARTIN: Yes.
DR. CLARKE: If you put welds at different
depths and measure water levels, you will find that as
you go down the water levels go up.
DR. CLARKE: Okay. There is a flow model
which drives the transport model, and what is called
the particle tracking model. And again just an
observation, and I am not sure what we can do about it
in the short term, but there is a fair amount of data
existing and data being generated through this work
that would enable the calibration of the flow model.
The radionuclides are not in the system,
and so we can't in the traditional sense calibrate a
transport model. We can, however, look at the
different pieces and the different processes, and use
laboratory and field tests to get the best definition
of those processes, and that is the approach being
So the particle tracking method includes
radionuclide transport processes of advection and
dispersion, matrix diffusion in fractured volcanic
units, and sorption.
Simulated flow paths occur in the upper
few hundred meters of the saturated zone. And the
they cross the 20 kilometer fence approximately 5
kilometers went of the town of Amargosa Valley, which
I think is a little bit south of Highway 95.
Now, again, the point at which the
volcanic units transition into the alluvium is still
an area of certainty, and that is important because of
the attenuation that you would see in these systems.
And I thought that this might be
interesting. Again, these overheads are taken out of
various reports. The total system performance
assessment-viability assessment, TSPA viability
assessment, this is the information that was taken.
The matrix diffusion was modeled through
what is called an effective porosity, where you have
a fracture porosity, and a rock porosity, and you work
within that range. But you treat the system with what
is called a single continuum.
Dispersion was handled through a dilution
factor, and the flow paths were one dimensional
streamtubes; and if you go over to the current model,
matrix diffusion is now being handled in what is
called a dual porosity approach, an analytical
solution, and dispersion being handled a different way
And the flow paths from the 3-D process
CHAIRMAN WYMER: What kind of difference
do these differences make?
DR. CLARKE: Well, the affected porosity
model is compromised at best, and it would be
difficult to handle a mixture of compounds with this,
and factors for each radionuclide.
CHAIRMAN WYMER: I guess I was asking for
the difference in results of the models. I mean, does
it change the numbers that come out?
DR. CLARKE: I really can't answer that.
DR. STEINDLER: The answer is yes, it
DR. CLARKE: And again I would expect it
CHAIRMAN WYMER: A lot, a little,
DR. MCCARTIN: You mean between the two
different types of models?
CHAIRMAN WYMER: Yes.
DR. CLARKE: I think his correction of
specific prior assessment; is that right?
CHAIRMAN WYMER: Well, you prepared the
two, and I wanted to know if it made much difference
which one you used, and what the answer was that you
DR. MCCARTIN: It probably depends on the
retardation coefficient that is being used. I mean,
when something is really retarded, you change the
I mean, there would still be some
difference for the same retardation values, but if
they also used a different model and different
retardation values, you would probably be swamped by
the retardation changes in the retardation.
CHAIRMAN WYMER: I can understand that.
So perhaps the matrix diffusion might change the ratio
of the materials that had different Kds.
DR. MCCARTIN: I am not sure what you
CHAIRMAN WYMER: Well, I wondered if a
semi-analytical solution changed the ratio of those
materials that had a high Kd, and those that had a low
Kd from the affected porosity model.
DR. MCCARTIN: Right. Yeah. Well, if we
ran both models with the same Kd, there would be some
difference between the two results.
CHAIRMAN WYMER: Maybe because of the
change in the way they handle the ratio.
DR. MCCARTIN: Well, is the representation
CHAIRMAN WYMER: I think I am beating a
gnat to death.
DR. CLARKE: I think it is fair to say
that this is a much better representation of the
system, with dual porosity.
DR. CAMPBELL: There are a lot of
questions about effective porosity models.
DR. CLARKE: As I understand it, the
models are different, and DOE is running its model,
and you folks are running your model, and there are
differences. But you are both taking a dual porosity
approach to a matrix diffusion.
DR. MCCARTIN: Right.
DR. CLARKE: You are taking a kinetic
DR. MCCARTIN: And we don't take much
credit for it. I mean, it is all driven more by the
assumptions of what is the fractured spacing, and what
is the retardation in the matrix.
I mean, those are the things that tend to
-- and I guess I am not aware of how much we have
looked at the difference in any perimeters between the
two of us. We will get to that, but the assumptions
used in the model vary.
DR. CAMPBELL: Correct me if I am wrong,
Tim, but if you use an effective porosity model, and
essentially you have some distribution of porosity,
and you say, well, my effective porosity is blah,
Now, if you use some sort of dual
continuum model, where the fractures say transit most
of the radionuclides, and a particular sweep of those
fractures is really good at transmitting
radionuclides. And an effective porosity model
wouldn't indicate that at all.
It would just say, you know,
radionuclides are being transmitted at some effective
retardation path, and you wouldn't be able to ferret
out a particular set of features that might transmit
it much more quickly.
CHAIRMAN WYMER: And presumably if you did
your effective porosity calculations properly, you
would get the same answer.
DR. MCCARTIN: Well, if you got your
effective porosity based on flux, and most of the
fluxes are fractures, you might be skewed to that end.
I would have to work it out, but --
CHAIRMAN WYMER: I think I have a better
grasp of it now.
DR. CLARKE: I'm sure that you can see
this, Ray, but the effective porosity assumes that you
have got porous medium at that porosity. And this is
a much better representation. These models have
evolved over the years as well, and has diffusion in
the matrix and sorption.
CHAIRMAN WYMER: Okay.
DR. CLARKE: Okay. At this point. Let me
just stop and share a couple of observations. Again,
just based on where I am in this, all of these
attenuation processes really delay the transport.
They really are not irreversible. They delay the
And from what I have seen, I think the
work that has been done to demonstrate whether or not
these processes are ones that would be expected to
occur in this system has accomplished that. I think
there has been a great deal of good work on both
The unsaturated flow meeting in
Albuquerque focused to a good extent on matrix
diffusion issues, and I think the data would support
the efficacy of that process and the system, and
similarly for sorption clearly.
If there are going to be issues and
controversies down the line -- and again I think I am
just stating the obvious here. It is probably more
not through these processes, and in fact attenuated
radionuclides, and should we be looking at them.
And it is going to be more of a capacity
issue, and what is the ability of the system to
effectively attenuate the radionuclides, and to what
extent can they do that.
The data are necessarily based on
laboratory and field studies, and the laboratory
studies do use site specific materials from what I
have seen. I wouldn't say that they are overly
conservatively designed. From what I have seen, they
look pretty good.
And the field tracer studies again used
surrogates to get information, but again I think the
results demonstrate the process. The question is
going to be scaling up, and how much of the system can
we attribute to this.
That strikes me that that is going to be
a function of how well this system is characterized,
which is never enough usually. And so there are going
to be some judgments about how much of this do we take
credit for and in which region.
CHAIRMAN WYMER: I presume, Jim, that
there is a whole tremendous -- say you take a tube
down under the repository, and there is a lot of
sorptive capacity just within a tube straight down.
You are never going to challenge the
capacity of the medium to take up all the stuff that
it sees. That's true, isn't it?
DR. CLARKE: Well, that would be right, I
guess, at this stage. It really is a function of what
goes into the system. I think that's why this
interface is so critical.
And how much is going to be released and
when is it going to be released, and what is the
capacity of the system to attenuate it. If you look
at the work that has been done, from what I can tell,
it's not as if they don't need these natural barriers.
That does not appear to be the case from what I have
CHAIRMAN WYMER: What doesn't appear to be
the case? Are you for or against it?
DR. CLARKE: Oh, no, no. They do need to
take credit for these, and so the issue becomes how
much. I mean, to me, again.
DR. SHEWMON: Why do you assume that there
is enough active relevant surface?
CHAIRMAN WYMER: I didn't assume it, and
that's my question.
DR. SHEWMON: Oh, that's your question.
I thought that was a statement.
CHAIRMAN WYMER: I sort of tended to
believe that since you have 300 meters of stuff down
through there that there is enough capacity. But I
DR. SHEWMON: Well, it depends on what is
there. I mean, if it were all lined with tungsten,
nothing would happen.
CHAIRMAN WYMER: And presumably in the
area where it is going through fractures
predominantly, that washes out, and then it is only
what is left that you have as a medium that has the
DR. CAMPBELL: One of the ongoing projects
that DOE has is this -- what they call their busted
view test, where they are using analogs, and trying to
get a handle on the sorptive capacities and diffusive
capacities of a formation underneath a repository
called Calico Hills, which is a fairly -- well,
portions of it are a fairly friable ash unit, where
flow and transport occur through a porous medium, as
opposed to fractures.
But not all of the area of the repository
is over areas of the Calico Hills will occur. There
are some fraction of the repositories over an area
where it is a harder material, and it is more vetric,
and it has more glass in it.
And there would tend to be more flow
through essentially a fracture network. But a lot of
the units are still fractured rock, and you are
looking at flow through fractures.
CHAIRMAN WYMER: So the sorptive layer is
really a fraction only of the total depth of this
DR. CAMPBELL: Yes. And they build this
into their model, and I think NRC does as well through
having several sub-areas, or a half-a-dozen sub-areas
of the repository, some of which to through a Calico
Hills vitric, and some of it goes through the Calico
Hills that can be more sorptive.
One of the issues is the temperature
effects of the repository on the zeolytes, which are
the reactive phase in that area, and the ability of
those zeolytes to absorb the radionuclides.
CHAIRMAN WYMER: And it strikes me that
there is a lot of competition for those sites, because
you have an awful lot of steel, and you have an awful
lot of uranium relative to the things that you really
want to absorb, and I don't know how much these
competitions have been looked into, or whether the
capacity of the reactive tube is challenged.
DR. CLARKE: I haven't seen much on
DR. MCCARTIN: Yes. It is really a dilute
amount, but for our modeling, generally when you look
at the unsaturated zone versus the saturated zone from
a matrix diffusion standpoint, the velocities in the
fractures in the saturated zone are relatively slow
compared to the unsaturated zone just based on the
So you have got 300 meters at most of
saturated or unsaturated zone fractures, versus 15
kilometers of fractures potentially, and maybe more,
of fractures in the saturated zone, where velocities
And so for our model, as Andy knows, we
have the ability to assimilate matrix diffusion in the
unsaturated zone. We don't do it. Computationally,
it is very taxing, and based on the travel times, it
isn't going to have that big of an effect.
CHAIRMAN WYMER: There are just much
saturates before you get to the bottom that who cares.
DR. MCCARTIN: But part of the benefit is
totally tied to how much retardation there is. And
the biggest thing in the unsaturated zone that I know
when we were looking at could we support matrix
diffusion in the unsaturated zone was that there were
two things that we were aware of.
One was Chlorine 36, and the fact that
Chlorine 36 got down there, and matrix diffusion was
really a strong effect, maybe you shouldn't have seen
that. And then Bill Murphy at the Center did a lot of
work looking at fracture water versus matrix water,
and he saw that there were just two completely
That they are just completely different
chemistries, and once again if matrix diffusion was a
strong influence, you shouldn't see this huge
disparity between the fracture of water and the matrix
And I don't know if John -- well, I know
that goes back 4 or 5 years, and I don't know if
anything more has been learned from that. But with
that information for the user at least, there was,
well, how much do you really want to take credit for
it when you have got 15 kilometers of fractures and
matrix diffusion in the sat zone, with lower
velocities, which makes it an even stronger effect.
DR. SHEWMON: Well, plutonium hit the fan
so to speak a while back because if it migrated
further out of a test site than others. They talked
about it being colloidal, and the colloidal then flows
only in the fractures, and it doesn't get caught up in
the matrix, and it doesn't absorb, is that correct?
DR. MCCARTIN: Well, it doesn't have to
flow just in the fractures. There should be some
screening both in the matrix and --
MR. BRADBURY: Actually, Jim said that
there are filtration processes that DOE takes credit
DR. CLARKE: Which is being handed through
retardation. As I see it, there are four systems in
the unsaturated zone. There is the fractured system,
the welded tuff, and then there is the more porous
And I would agree that in the fractured
system that you have got higher velocities, and you
have the chlorine-36 data and you have all kinds of
reasons not to get real excited about matrix
You do have the porous rock, however, and
you would expect some attenuation there. When you get
into the saturates, you have a long stretch, and we
don't know how long yet.
But you have got a long stretch of
fracture flow control systems, where you have much
slower velocities, and you have got much higher matrix
diffusion potential and dispersion potential.
DR. SHEWMON: You are getting too general
for me. I asked you specifically about the plutonium
and the colloids, and why it was that it being a
colloid all of a sudden explained the results.
DR. CLARKE: I'm sorry, Paul. I thought
we had already answered your question, but the
approach does permit removal or attenuation of
colloids through a filtration process. Colloids are
getting hung up as they are transported through the
DR. SHEWMON: Okay. And that is in the
saturated or the unsaturated?
DR. CLARKE: That would be in both of
DR. SHEWMON: And is the filtration
different from the matrix diffusion that you are
DR. CLARKE: Yes. And the filtration
process really applies just to the colloids. The
matrix diffusion applies to dissolved material
soluids, something moving through the system that now
has a concentration grading between where it is in the
fracture and in the much lower concentration in the
DR. SHEWMON: But it is diffusing along
very fine crevices; is that right?
DR. CLARKE: Yes.
DR. SHEWMON: It is mechanical diffusion.
DR. CLARKE: Yes.
DR. CAMPBELL: Paul, I think the question
that you are asking -- and correct me if I am wrong --
is why do colloids carry stuff faster than on average,
DR. SHEWMON: And the answer that I am
getting is that they stay to the fractures pretty
DR. CAMPBELL: Because they tend to have
a negative charge and the surfaces of the minerals
tend to be negatively charged. So that through
something called anionic exclusion, anionic species
tend to be excluded from these very tiny pore spaces.
So they tend to stay in these larger pore
spaces where the flow rates are faster. The amount of
DR. STEINDLER: They don't stick to the
DR. CAMPBELL: Right, they don't stick to
the walls, and so you have a distribution of a flow
rate, and it tends to move the stuff attached to
colloids to the upper end of the distribution.
DR. CLARKE: Right. And if there aren't
any velocity radiants, then the velocity is higher.
DR. CAMPBELL: One of the things to keep
in mind about the migration of plutonium at a Nevada
test site was that the specific area was a place
called the Benum Test, and in one of the wells, they
were able to identify plutonium by its isotopic
signature as having come from that test.
It was about 1-1/2 kilometers from the
test site. This is in the saturated zone, and it is
well within the saturated zone. Actually, in a
portion of the Calico Hills saturated zone, the
amounts of what they call colloidal material were
And we are dealing with large
concentrations that are very low concentrations, and
it wasn't just plutonium. There were a number of
radionuclide, and what they did was that they filtered
the water and these particles were filtered out at
some sized fraction, which fell within the range of
what is called colloidal.
But it not only included plutonium, but
also cesium and some other stuff. And it was presumed
that these were essentially natural colloids that
these radionuclides had become attached to.
It has not been seen in a lot of the test
sites, and so it has never been clear why that
particular shot -- it was a big one. It was over a
megaton -- produced this effect. But it is there and
they did see radionuclides in this well that they
DR. MCCARTIN: And at one time I thought
there was still some debate as to whether this
occurred very shortly after the shock, and the
transport. You know, this is not a long term
transport problem, but this occurred very quickly
after the shock.
CHAIRMAN WYMER: Yes.
DR. MCCARTIN: But I know that there was
some discussion early on, but I haven't followed it
for a while. But no one -- well, they found it, and
it might have been there 40 years ago, but it was
DR. SHEWMON: They just hadn't looked in
DR. MCCARTIN: Yes.
DR. CAMPBELL: The group that does this at
Los Alamos has been monitoring wells all over the test
site for some period of time, and looking for
migration in that.
And I will add that the issue of transport
as a colloid is still open because of the way people
measure or attempt to measure colloids. You can
generate artifacts with that if you don't do a really
There is some work that has been done
actually by a group that I know about, because they
are actually oceanographers that are doing it, both at
Savannah River and Hanford, in which species that were
thought to be colloidal transported plutonium, was in
fact a transport of dissolved plutonium that was in a
more oxidized state.
So there are artifacts that can be
generated through the filtration processes that one
has to be very careful about. Sometimes what appears
to e colloidal transport isn't.
DR. SHEWMON: Thank you.
CHAIRMAN WYMER: Enough already.
DR. CAMPBELL: Those are some of the
CHAIRMAN WYMER: Jim, what else have you
DR. CLARKE: Maybe I can just transition
into the issues and sub-issues.
CHAIRMAN WYMER: I would add that I do
think that the whole question of colloids is one that
is going to be brought up, and it is going to be a
point in which the intervenors and citizens are going
to grab a hold of and say what about this, and so I
think it is an important issue.
DR. SHEWMON: But you don't mean to imply
that it isn't being dealt with?
CHAIRMAN WYMER: No, I do not mean to
imply that it is not being dealt with. I mean just to
stress the importance of it.
DR. CLARKE: Okay. The radionuclide
transport key technical issues, and there are four
sub-issues. I have done nothing on sub-issue number
four. So we will not be talking about that today.
But as far as the first three sub-issues,
the system has essentially been organized under porous
rock, and this would be floating through the rock
matrix, and the alluvial, which again would be treated
as a porous medium, and radionuclide through fractured
Again, Tim, maybe you can help me with
this, but as I understand it, porous rock is being
addressed in the unsaturated zone, and the saturated
zone is primarily being looked at, if not exclusively,
as fractured in the volcanic units.
DR. MCCARTIN: In volcanic.
DR. CLARKE: In volcanic, and of course
the alluvial after that should be treated as a porous
medium. So that is the way that these issues are
CHAIRMAN WYMER: Do you see any gaps in
DR. CLARKE: No. I think that covers the
system. You could organize it differently, but I
think that is everything. All of these issues are
what is called closed pending.
I believe we went into the meeting at
Berkley with the first three open. Were they all
DR. MCCARTIN: Yes.
DR. CLARKE: But in any event, they are
all closed pending on it. I thought I would just show
a few. It is going to be hard to see these, but I
DR. CAMPBELL: Everybody has a hard copy,
DR. CLARKE: Okay. Really, the only
reason I wanted to show these was just to give you a
feel for the kinds of things that come up in this
issue resolution. And it strikes me that they can be
pretty much be organized into requests for additional
documentation, requests for more justification.
And in some cases the data simply haven't
been developed yet, which is the case with the
alluvial, where there is an ongoing investigation to
not only determine transitions, but also to look at
the characteristics and other features of it as well.
So these tend to be the requests that come
out of that. For example, radionuclide transport
through porous rock, the first one, is provide the
basis for the proportion of fracture flow through the
Calico Hills non-welded vitric.
Provide analog radionuclide data from
tracer tests for Calico Hills at Busted Butte, which
Andy spoke to before. So in many cases the data are
there. They just need to be provided.
Provide the screening criteria for the
radionuclides selected for PA.
CHAIRMAN WYMER: This just pertains to
colloids apparently, number three.
DR. CLARKE: I thought it was more general
than that. Are these not two separate questions; one
is the list of the radionuclides that will be the
model, and the other is --
DR. CAMPBELL: Those radionuclides that
can be associated with colloids. So what DOE -- if I
can remember correctly, what DOE agreed to do was in
their inventory of fraction AMR they are going to
apply the basis for screening out particular
And then the AMR on waste form colloid-
associated concentration limits, they are going to
provide their argument for why they are only focusing
on a few key radionuclides, in terms of colloid
CHAIRMAN WYMER: Yes, that's what I read
it to say.
DR. CLARKE: And as you can see, there are
a number of issues on the alluvial, given the status
of that program, and to provide further justification
for the range of effective porosity in alluvium.
The other thing that I should say is that
the way these model predictions are done, at least on
the DOE model, is that the perimeters that drive the
flow of transport or transfer, and let's talk about
that, are handled either by what is called bounding.
In other words, there may be some
perimeters that have constant values for the region in
which the calculation is being performed, and then
there are a number of perimeters that are handled
So the distribution is set up for these
perimeters, and this is not an uncommon way to do
these predictions, and then the distribution of sample
in the process.
It strikes me that most of the perimeters
are handled statistically and certainly all of the
ones that we considered sensitive to those
Provide a detailed testing plan for
alluvial testing at the alluvial testing complex, and
again these are the kinds of questions that are being
asked and the agreements that are being made.
And I think this kind of speaks for itself.
DR. CAMPBELL: You certainly don't need to
go through each and every one of these agreements.
CHAIRMAN WYMER: I think one of the
significant things that comes out is that there are an
awful lot of requests for trivial data and for
documentation, which I think is sort of typical of the
approach that is used in these issue resolution
meetings. NRC is always saying show us the data, and
show us the documentation.
DR. CLARKE: It strikes me that while
other things do come up and get discussed from time to
time, at least these meetings are very focused and
very focused on the issues. I am not saying that is
either good or bad, but that is the nature of the
CHAIRMAN WYMER: Well, one thing that
bothers me a little bit about this aspect of the
process was that very often DOE will say, okay, there
is an AMR available that discusses that, or we will
give you one at the next meeting.
And that sort of leaves it hanging. You
aren't really dead sure that that AMR they referred to
has really got the stuff in there, and that kind of
bothers me. You have to sort of take it on faith.
DR. CLARKE: Well, as I mentioned before,
the issue of the documents being generated at least to
resolve these issues, and they are very focused on
doing that, just resolving these issues.
So information is brought in from whatever
sector it needs to be brought in from to address the
particular issue. One of my concerns, and really it
may be unfounded, but one of my concerns is that the
issues are fragmented.
There are a number of issues and a number
of sub-issues, and there are some critical interfaces.
At some point in the process, if it is not already
being done, I think there would be a great deal of
merit to pulling together more comprehensive -- and
what I would call technical basis documents.
And which would not only deal with flow in
the saturated zone, or radiated flow transfer, but
would deal with source terms, and other things that
need to be dealt with across an interface.
I don't see that now. It may be out there, but I
haven't seen it yet.
CHAIRMAN WYMER: We talked about that a
little bit, and that up to a point, that is handled in
the building materials. But things get so abstracted
at that level that you aren't exactly sure that things
really have been handled across the interface
DR. CLARKE: Also, I think it would be
helpful if it is not already being done, but the
people working on the radionuclide transport key
technical issues, to be up to speed on what is going
to go into the sub-surfaces as a result of near-field
processes, container lifetime --
CHAIRMAN WYMER: Yes.
DR. CLARKE: And it may be that you have
to get to the TSPA level to get the total treatment,
but I can't tell. But I think there is a lot of
synergy there, and a lot of good reasons to work
across that interest.
DR. CRAGNOLINO: I want to make a point.
This is precisely the idea what is going to be called
a degraded high -- and that means that all of the
integrated parts of the evaluation of a repository are
going to be linked together in different ways.
CHAIRMAN WYMER: But that's quite a ways
in the future is it not?
DR. CRAGNOLINO: No. It is going to be
issued in September. We are preparing the outline,
and trying to focus a way to integrate it in different
CHAIRMAN WYMER: One of the points that
the NRC has hit on time and again is with respect to
the total system performance assessment, because we
don't understand it. It is so big and so grandiose
that we can't wrap our minds around it.
We have not been emersed in the details
and so we don't have the background to bring to it,
and which you people are steeped in, and therefore,
what we have been saying time again and time again is
to simplify, simplify, and it is hard to simplify
something that is inherently complex.
But I guess I would say the same thing
about an integrated resolution document; that it has
to be understandable not only to the real experts in
the field, but to people who have to get a warm fuzzy
feeling when they read it.
And when they read it, feel that things
are all right here, and that I understand it and it
looks pretty good. It is a real challenge to do
something that way and still cover the technical
But if you don't do it, some of us are
just going to keep hammering on it, whatever that
DR. CLARKE: I guess the other thing that
I would suggest if it is not already out there or
being worked on, and in response to the concern that
Andy raised earlier, would be a blow-by-blow
comparison of the assumptions in each of these
different models, and the expectations as to how those
assumptions would affect the final outcome.
MR. BRADBURY: Let me give you an example.
This figure that you put up before on the use of
hydrochemistry and the flow path. It is fascinating,
because what it does is the lines, the flow lines,
essentially connect lines of equal concentration of
conservative constituents -- chloride, sulfate -- and
those are the ones that I consider conservative, and
there are other ones maybe, but maybe not.
And so they are saying that the
concentrations of these constituents remain constant
along these flow lines. That assumption then says,
well, forget about dissolution along the flow path.
It is a very big assumption; that they
must therefore for consistency sake carry that through
and include that assumption also in their performance
assessment, or they don't use hydrochemistry in this
way to delineate the flow path.
It is a very powerful assumption, and I am
not sure whether they have actually thought that far.
Well, let me put it this way. That definite changes
-- and that was surprising to me when it was pointed
out this way.
CHAIRMAN WYMER: Well, I picked that up
from the --
DR. CLARKE: Well, I ran the risk of using
it as an example of something else, and looking for a
good graphic that showed the flow paths as they are
currently understand. I know that is a controversial
DR. STEINDLER: But that is a conservative
DR. CLARKE: Right.
DR. STEINDLER: I mean, that is very
CHAIRMAN WYMER: No, highly, highly
DR. STEINDLER: The question is whether or
not the staff should hammer at DOE, the NRC staff
should hammer at DOE to justify what I think all
parties would agree is a very conservative assumption.
And that usually gets the guy right up out of the
MR. BRADBURY: Actually, I think it is
conservative, but what if these aren't the flow paths
then? What if there is dilution, and Steve Hanaver
raised this issue before. Normally, they assume that
there is this evolution of the composition of water as
it moves through the rock.
And so this is going against normal -- the
scientific community's normal assumptions, and so you
might have to think different. Well, if you have
different flow paths, how does that impact
DR. CLARKE: It is conservative from a
dose standpoint, but if they give you the wrong
MR. BRADBURY: Well, are the paths
perpendicular to these? I don't know if they are or
DR. SHEWMON: It doesn't make a
MR. BRADBURY: I don't know the answer to
DR. SHEWMON: I assume that the staff's
focus is what goes on at the 20 kilometer where some
guy is pumping water out of that well as fast as he
If that is not the focus of the staff,
then I must say that I have missed the point, and I
wonder what the regulations are. If that is the focus
of the staff, then anything that reduces -- and any
challenge to an assumption that would reduce that dose
can be argued to be irrelevant.
And therefore you can approach -- if it is
an issue resolution, you can approach it in another
CHAIRMAN WYMER: Yes, right.
DR. SHEWMON: I don't know whether DOE
argues that, and I don't know whether the legal folks
would allow that, but I would guess that is not a
DR. MCCARTIN: I think the answer to your
question is that if DOE has made a case that this is
a conservative assumption, and you believe that the
information that they presented supports that, you're
right. The issue is closed.
CHAIRMAN WYMER: There is no reason for us
to challenge it.
DR. MCCARTIN: In technetium, they are
using a retardation of zero, and we don't care. I
would argue that they are done.
CHAIRMAN WYMER: And from a very practical
point of view, that is exactly correct. It does not
satisfy scientifically, but it is okay.
DR. MCCARTIN: For us to make a decision
based on that approach, we are confident that we can
make a decision that will protect public health and
DR. STEINDLER: Exactly.
DR. CRAGNOLINO: May I raise a point? In
order to complete the response to that question, Dr.
Steindler, we have adopted that criteria by inserting
that DOE is conservative. There is no solid technical
basis for the assumption that conservatism --
CHAIRMAN WYMER: That is the point.
DR. CRAGNOLINO: And that is in the debate
currently for cladding, because DOE uses the criteria
for cladding that they consider is conservative, at
least the criteria of the solution of cladding by
fluoride, but assuming that they are claiming more,
and claiming that localized corrosion of cladding is
not possible in their package.
They assume let's use fluoride as a
surrogate, but the claim that that is conservative
because it is an assumption of localized corrosion due
to fluoride. But it is essentially controlled by the
ability of fluoride, and that is contradictory.
DR. STEINDLER: Well, really what you are
doing is that you are challenging the conservative
nature of the assumption. Fine. If you have some
mechanism of doing that, that makes sense, and you
have provided one particular case, say fine.
But if you don't have any reason to
challenge that assumption, and whether or not the
stuff actually runs down that flow path, or disperses
and reduces its concentration, are the only two
options that you have so far identified.
If somebody comes in out of the blue and
says, hey, guys, that's dead wrong. There is an
underground river that this stuff drops into and
whistles down to the guy's well, now you have got an
assumption, or a statement, or evidence that makes
this non-conservative. A different ball game. That's
all I guess I am saying.
DR. AHN: However, DOE agreed to analyze
the established --
CHAIRMAN WYMER: The reason they agreed to
it was because he followed it.
DR. AHN: That is one way of doing it.
CHAIRMAN WYMER: And what you wonder about
is what hasn't somebody thought of. We have Gustavo
in this area, but how about some of these other areas?
DR. CRAGNOLINO: I think this is a general
problem that we have to confront.
DR. AHN: We have a list that has been
identified containing the --
CHAIRMAN WYMER: Well, for example, the
kind of figure that exemplifies the point of what
happens with that is the effect of lead on Alloy 22,
and granted that things are way out of reason, the
conditions under which they ran these experiments, but
it was something that wasn't thought of. It was lying
DR. AHN: The current DOE position is to
reopen whenever we identify new things.
CHAIRMAN WYMER: I realize that, but it
bothers me because we are drawing on a limited pool,
with specific areas, and what are you going to do.
DR. MCCARTIN: One quick thing, because
this gets to one of the things that you were saying
about the transparency traceability, which is clearly
a big issue for us also. The challenge to write this
in a simple form is hard.
CHAIRMAN WYMER: It is a big challenge.
DR. MCCARTIN: And I don't know how much
of the TSPA-SR you have read. I mean, it is a fairly
thick document. And having read some of it, I think
that DOE has done a very good job of trying to pull
out and distill from all the AMRs that they reference
what are the key ideas.
And in addition, in terms of what have
they missed, I think they have given other evidence of
why I should believe this approach, and why this
approach is correct. They have cited other evidence
from analogs and other information throughout there.
And I have not read it cover-to-cover.
CHAIRMAN WYMER: I have not read it
DR. MCCARTIN: Some areas may be better
than others, but I guess for the committee it is
useful to read that. But having said that, I would
say that I have been doing nuclear waste for 20 years.
It takes me a long time even to read 10 pages of it.
I have to really think about is being
said. It is a slow process. That is a big damn
document, and even for someone who is -- well, I have
done nuclear waste as I said for a long time, and it
is a difficult thing to read through.
And I don't know in terms of -- well, I
think they have put a tremendous amount of effort in
information there. But anyone who thinks they can
read it quickly, I don't know if anyone would be able
to do that.
And therein lies the challenge. I don't
know if you can distill it more than that. I just
CHAIRMAN WYMER: Yes, that is the
DR. CLARKE: So this has the elements in
the document that I was describing.
DR. MCCARTIN: It will be interesting to
get different people's reactions, and I would say it
will take 2 to 4 months before some has read it from
cover to cover. Now here are my comments.
But tangentially I am impressed and
relatively happy about what they have attempted to do.
I am sure that there are areas where we have
CHAIRMAN WYMER: Well, it sure looks
formidable. I will tell you that.
DR. MCCARTIN: But there is a lot of good
information that they have distilled.
DR. CLARKE: It looks a lot smaller.
DR. STEINDLER: Gustavo, did you have
DR. CRAGNOLINO: Well, it was with respect
to the comment that you made about the connection
regarding led. And there was some discussion going on
this morning regarding oxidation energy --
DR. SHEWMON: Going back to what the
people from Nevada brought in, or Catholic University,
you are saying that is a high activation energy
process, and so below a hundred degrees C, or below 80
degrees C, it would go an awful lot slower?
DR. CRAGNOLINO: We don't want to take
this for granted at the present time without further
examination, but this is the way that you bound.
DR. SHEWMON: Ray, let me change the
subject completely if I can, but a general discussion.
I have something that maybe you wrote. I don't know.
CHAIRMAN WYMER: What is it?
DR. SHEWMON: It is the chemical
environment on the waste package. Anyway, it's here.
And it says that relative humidity, and when relative
humidity exceeds the critical concentration, 80
percent, we consider that corrosion is going to occur
on the waste package.
The last thing on the page says the
composition of the water contacting the waste package
will not change significantly because of chemical
interaction with it.
CHAIRMAN WYMER: That is a DOE statement.
DR. SHEWMON: Fine. But that is what
offends me, is that the gas that the water all comes
in through the vapor, and that keeps corroding, and
the corroding nature producing ions, and there is no
place for these to go.
But they can't change the composition of
the liquid, which is silly. It has to saturate all
the way. So it is conservative, but wrong.
CHAIRMAN WYMER: But it is silly, yes. Is
that what I put in dark print there?
DR. SHEWMON: Yes.
CHAIRMAN WYMER: I bolded that.
DR. SHEWMON: I hadn't come across it, and
maybe that is the way that the cookie crumbles in this
CHAIRMAN WYMER: Well, I don't think this
world is scientifically any different than the world
that you live in. But something has occurred to me,
and I don't know whether it is real or not.
But there is a continual update of these
documents, and there is a continual rewriting, and
they dig out more information, to a large extent
pushed by NRC for more supporting data and more
And they do this piece-wise, and I am not
sure how well or how often everything has gone back to
square one, and all these things are put together.
Now, this is an integrated thing, which
itself will be a transitory document, because there
will be a lot of stuff coming in after you write this
So, I am not sure whether after the pieces
of the puzzle are joined together like this from one
part of what happened to another part, and then they
get dislodged maybe by some new information.
DR. CRAGNOLINO: Well, let me make a point
since we are having a dialogue. An example was made
about corrosion, and the critical factor controlling
the life of a waste package containing Alloy 22
doesn't have any date.
Therefore, they put together a bunch of
experts like people that are in this room, and they
offer their distribution of corrosion rate. So they
have a group of people who have spread the rate of
Now, we have to recognize that even though
there are critical comments about the way that the
corrosion rates are measured, at least they are
reported and supported by current information.
It is our responsibility to be very
objective in analyzing this, and this is what allows
us to come to this agreement, because the issues are
much better defined now. And we can focus on very
certain narrow issues, but are they issues that allow
the program to move forward.
If we resolve these issues, we are in a
different stage, and we can say, well, this has a
certain impact, and we can move forward. But I think
that this is the type of situation that we have to
recognize and we have to be astute and apt in
identifying what are the problems, and not believing
that we are much more clever than the other side of
DR. DAM: I am Bill Dam from the NRC
Staff, and I wanted to respond to a few things that I
CHAIRMAN WYMER: Have you got a list there
about three pages long?
DR. DAM: Not too long, but in terms of
requesting more documentation, and also your statement
about colloids are a very important issue, I just
wanted to highlight to the committee working group
that in the information Jim handed out on page 7,
there is an agreement that we came up with at DOE, and
in number seven we said that they should provide
sensitivity studies to test the importance of colloid
transport parameters and models to performance for
unsaturated and saturated zones.
Basically what happened at the Busted
Butte test was that they weren't able to get their
microspheres, which are the articles that they were
using, they weren't able to move, and so now they
don't have any data for looking at colloids in the
natural unsaturated system.
So one of the things that we requested was
that they look into doing a test such as that Alcove
8/Niche 3, where they could inject microspheres, or
other colloidal material. We gave them the option of
maybe considering that.
We can't be prescriptive, but we just gave
them ideas on how to proceed, and then you can see
that we requested that information by this month.
DR. SHEWMON: Physically can you make
polystyrene particles that are submicron?
DR. DAM: Yes, they are using them in
CHAIRMAN WYMER: They are typically used
to measure deficiency of filters.
DR. DAM: So the point that I was trying
to make is that when we request more documentation,
often times we are trying to request information that
they maybe weren't planning to provide, or information
that will get them to do an additional analysis that
will be given to us in a future report.
And in this case it is going to be a
letter report to us right away to tell us if they are
going to be able to evaluate this technique.
Secondly, they still have not given us a very good
adequate justification for using the microspheres as
analogs for colloids, and you will see our agreement
CHAIRMAN WYMER: I am not crazy about the
idea either to tell you the truth.
DR. DAM: And it is interesting, because
that agreement, which deals with C-wells, which is in
the fractured saturated zone, also applies to their
current testing of alluvial tracer s, which is in the
alluvial material where they are using microspheres.
So there is a lot there in those
agreements that I just wanted to make the committee
aware of, and going back to the statement that
colloids are a very important issue, and it will be
brought up by intervenors and other people, we are
doing some things about that.
We have had discussions, and we had a
conference with the American Geophysical Union last
spring, where we discussed tracers and brought in
quite a few presenters to give talks about their work
And there is another session being
considered and proposed for the fall of 2001
specifically on colloids, and we are also getting in
speakers to come in to the office and talk to us about
bringing us up to speed from other sites, such as in
So we are trying hard to get up to speed,
both on the science and understanding the mechanisms,
and understand DOE's modeling approach. It is
interesting that we heard at the meeting that colloids
are the greatest uncertainty in TSPA. So it is
something that we are taking quite seriously.
DR. SHEWMON: I think you should.
CHAIRMAN WYMER: Yes.
DR. STEINDLER: Are they the greatest
CHAIRMAN WYMER: It is more of a
perception thing than it is a scientific thing.
DR. STEINDLER: So if the uncertainty is
never resolved, then it won't make all that much
difference; is that what you are telling him?
CHAIRMAN WYMER: Except to the
DR. STEINDLER: Except to the intervenors.
Well, but I mean --
DR. DAM: Well, I think that is important
to -- for instance, the Benum test that I mentioned,
we need to pin down the mechanisms for the transport,
and was it induced by the blast.
And the purpose of having these kinds of
meetings, technical meetings, is to separate the
perceptions from the science, and try to give what the
hard facts are.
DR. MCCARTIN: But DOE has analyzed
colloids, and it doesn't seem to be a significant
contributor relative to other things, like technetium,
DR. STEINDLER: And so I guess my question
continues to be if that is true, and I have no reason
to believe it is not, why spend resources trying to
fuss about colloids? It will take one great deal of
effort to take that Nevada test site information and
try either experimentally or by having another look at
existing data to try and unravel how that plutonium
traveled 1.3 kilometers in 30 years.
CHAIRMAN WYMER: And that is a valid
DR. STEINDLER: My question really is why
is the staff pushing for that?
DR. MCCARTIN: Well, here is a case -- I
mean, I don't know -- well, I will go with my memory,
and that DOE is the one who brought this up more than
we have. They brought up colloids as a problem that
they were looking at.
We actually don't have it in our PA model.
They brought it up and they put it in, and then they
are giving this information as to how to represent it.
Well, if you are going to bring it up, then --
CHAIRMAN WYMER: And you have to deal with
DR. DAM: And then there is TSPA, and it
does make a difference on it, in terms of dust.
Plutonium, colloids, do have an impact on dust.
CHAIRMAN WYMER: Some, small.
DR. DAM: It all is very small.
CHAIRMAN WYMER: And it is perceptible.
Why don't we turn our attention now just for the last
little time here on defense-in-depth and multiple
DR. CAMPBELL: It's your turn to be on the
hot sat, Tim.
CHAIRMAN WYMER: One of the sort of basic
questions that comes to my mind -- and I don't expect
anybody in this room to answer it, but how many
barriers constitute defense-in-depth? What is
expected? Are two enough?
DR. MCCARTIN: Absolutely.
CHAIRMAN WYMER: Well, there is the
DR. STEINDLER: Okay. Anything else you
want to know?
DR. MCCARTIN: I think basically that one
is engineered and one is --
CHAIRMAN WYMER: One is natural
DR. MCCARTIN: Yes, and I think the rule
is very explicit in terms of multiple barriers.
Defense-in-depth is really a broad philosophy for the
CHAIRMAN WYMER: It is a bigger issue.
DR. MCCARTIN: And we would argue that
Part 63 encompasses defense-in-depth. But in terms of
multiple barriers, they are required to demonstrate
that they have one engineered and one natural today.
Obviously drafted rules at the Commission
could change that, but if you looked at the proposed
rule, the intent was one natural and one engineering.
If they do more, fine.
CHAIRMAN WYMER: Provided that both or
those independently provide protection.
DR. MCCARTIN: Well, I am not sure what
you mean by independently provides protection. They
are not intended to be redundant.
CHAIRMAN WYMER: There is not much depth
if either one taken alone doesn't meet the standard.
DR. MCCARTIN: We have never said that it
is redundancy. There is nothing in the proposed rule
that says you need to meet our regulation with only
natural or only engineered.
The only statement made is that they both
have to -- and I will caveat it and put in this word,
is to have capability to either impede the movement of
water, or radionuclides.
CHAIRMAN WYMER: Then that is an
inadequate rule isn't it?
DR. MCCARTIN: Well, it depends on your
perspective. I will go back and check, but I don't
believe we got any questions to the effect or comments
to the question that the barriers should be redundant.
I could be wrong on that.
CHAIRMAN WYMER: Gee, somebody missed the
DR. MCCARTIN: We did not offer that, and
we tried to be fairly explicit that it was not
intended to be redundant barriers. Now, you may
disagree with that, and that's okay.
CHAIRMAN WYMER: Well, what do you think?
DR. MCCARTIN: I don't think redundancy is
required. I support what the proposed rule requires.
CHAIRMAN WYMER: So if one is scratched
and the other one doesn't meet the standard, it is
DR. MCCARTIN: Well, if one barrier was
removed, and --
CHAIRMAN WYMER: Or so diminished that it
doesn't do any good.
DR. MCCARTIN: Well, they have to both act
as barriers, okay? I mean, they have to have a
natural and engineered barrier, and they both have to
have the ability to act as that.
Because I will maintain that one of the
things that -- well, if you had a 10,000 year waste
package and a 10,000 year compliance period, that does
not mean that you are relying a hundred percent on the
Yes, you are getting a zero dose, and you
are getting a zero dose because nothing got out of the
waste package. But the natural system still has come
capability that didn't disappear because the waste
package didn't fail. And it has to provide something.
DR. DAM: No one barrier can have undue
DR. MCCARTIN: But if failure of a barrier
CHAIRMAN WYMER: And you only have two.
DR. MCCARTIN: -- in what I will call
"unacceptable doses," you would have a problem. But
unacceptable doses is not 25 in my mind.
CHAIRMAN WYMER: It is a hundred or 500,
depending. That's the rule.
DR. MCCARTIN: No, the rule does not
define what is an acceptable dose, and I think that is
left at the discretion of the commission. Some people
would say a rem is not an unacceptable dose.
But there is no specific number or time to
DR. SHEWMON: Is the drip shield
redundancy, or layers of defense on a waste package
that is already good for 10,000 years?
DR. MCCARTIN: It sure looks like
redundancy in terms of water.
DR. SHEWMON: I am just trying to get the
idea whether redundancy is two identical pumps, when
one will do it, and they don't have to be identical to
DR. MCCARTIN: Well, we have not claimed
that the repository has to be redundant, and in fact
the preamble to the proposed rule did a pretty good
job of -- there might have been a time when the
commission set up sub-system requirements in the old
rule, the waste package lifetime, and throw in travel
time, and release.
And there was a hope that these were
independent barriers, and I think that in 1980, yes,
that was a feeling. As time went on and you started
analyzing the system, the repository system more, I
think people realized that these really aren't
They aren't redundant, and there is --
well, unlike, say like a reactor, where you could put
in two pumps, and this one fails and this one will
kick in, we have got a waste package that is dependent
on the natural system.
The environment that it is in is certainly
related to its corrosion, and the same thing with the
drip shield. Now, the drip shield waste package, I
guess you can sort of look at it and say there is a
measure of redundancy between the two.
But the multiple barrier requirements is
not a requirement for redundancy.
DR. SHEWMON: Okay. You have answered the
CHAIRMAN WYMER: Isn't EPA's position that
you can't exceed 50 MR per year at the site boundary
and not pore for water?
DR. MCCARTIN: That is their proposal. It
is not final yet.
CHAIRMAN WYMER: If that is true, then it
is not up to the NRC to say, okay, it can be anything
we decide it is.
DR. MCCARTIN: Well, no. In terms of
compliance, and in terms of multiple barrier
requirements, let's say that DOE did an analysis,
where they -- right now under the proposed rule, you
need to identify the barriers, and you need to
describe their capability, and give the basis for
their capability. That is the multiple barrier
Now, let's say that DOE does an analysis
to give information to the commission as to how
barriers perform and will neutralize the waste package
and calculate the dose. Let's say it is 150
milligrams when they do that.
Right now there is no quantitative
requirement to say that it has to meet whatever the
dose limit is, whether it be 15 or 25. Here is what
happens when all the waste packages fail at T-zero.
Is that good enough?
Right now I think it is a subjective
decision for the Commission to look at, and that's
what I meant. There is not necessarily a quantitative
requirement in the proposed rule as to what -- well,
there are no numerical goals for what constitutes a
We did get criticism on that and primarily
Bob Buettner, who said how does DOE know they are
done. You need to give them something so that they
know that is a barrier.
CHAIRMAN WYMER: I think I agree with Bob.
MR. BRADBURY: Tim, my understanding is
that the amount of the contribution of a barrier
doesn't mean that you have to get it done at 25 or 15,
but you have to show that he dose was reduced by a
So you are saying it is up to DOE for it
to define whatever is defined as natural or barriers,
and show me the relative contribution of that
individual barrier, and then the TSPA, so me the
overall contribution of the combined engineered and
natural barriers keeping the dose below the dose
limit, which is 15 or 25.
So, for instance, you can see the natural
barrier alone knocks out all the short radionuclides.
So 99 percent go just on natural barriers. So it is
up to the engineered barriers to be designed to take
care of that one percent.
In doing so, it has to be so robust that
it also independently takes care of the other 99
percent. But the contribution is still there from the
natural barrier, number one, of knocking out the 99
And even with the remaining one percent
delay and all this other amplifying the benign
environment to design again for the engineered
CHAIRMAN WYMER: Well, there seemed to be
opportunities for quite a few barriers, chemical
barriers, to back up all these other barriers. And
there is a great depth of barriers possible.
DR. MCCARTIN: And DOE is required, I will
maintain in the rules, to where they have to identify
the barriers. In their performance assessment
calculation, they have to identify the barriers that
are contributing or have the potential to contribute
to a decreasing dose.
They can't, for example, say, well, we
will just count on the drip shield and our engineered,
and the alluvium as our natural.
CHAIRMAN WYMER: Those are two.
DR. MCCARTIN: Those are two, but even
though our waste package is lasting for 120,000 years,
we are not going to count that. Well, the fact that
the waste package lasts for 120,000 years is a
significant barrier, and they have to identify that.
So anything in their PA calculation that
has the potential to have a significant influence on
performance is barrier and they have to identify it.
Now, we don't require them and say, gee, we think you
are going to get a lot of retardation in the invert,
and include that.
But they don't have to, but if it is in
their Ph calculations, they have to identify those
things like that.
CHAIRMAN WYMER: So if they decided just
not to put in the drip shield, then they would fail?
DR. MCCARTIN: Sure. It is what they are
taking credit for in their PA calculation.
CHAIRMAN WYMER: Well, I can see why they
don't want to get into these chemical factors much
DR. STEINDLER: I do have to ask the
question reduce those from what? When somebody says
reduce, you have got to show that it reduces the dose
CHAIRMAN WYMER: From what it would be
DR. STEINDLER: From what it would be
without it, but if you are in a geologic disposal
area, it is difficult to eliminate the geology. I
mean, otherwise you are in the business of saying,
well, my waste package is sitting on top of the
ground. Things get pretty silly is what I guess I'm
DR. MCCARTIN: Well, we don't require that
type of calculation in the proposed rule, and it is
what is the capability and what is the basis. So I
would maintain for the geology that you could go to
the alluvium and look at Kds.
DR. STEINDLER: Then let me ask the
question differently, and I couldn't remember what he
answer is if there was an answer to it. Do you
DR. MCCARTIN: No.
DR. STEINDLER: You do not require
DR. MCCARTIN: No. In fact, we said the
barriers are not truly independent.
DR. STEINDLER: They don't have to be
DR. MCCARTIN: Well, we don't think they
are. They can't be, because, for example, the waste
package is totally dependent on the environment that
the natural system creates for it.
DR. STEINDLER: But that is a very broad
description. The waste package is pretty independent
from the Calico Hills, unless you believe that water
is going to rise.
DR. MCCARTIN: Yes.
DR. STEINDLER: So if you call the Calico
Hills one of a series of defense-in-depth barriers,
those are independent.
DR. MCCARTIN: Yes, but --
DR. STEINDLER: The Commission has
required in the area of functional criticality in the
case of at least facilities in the field cycle, three
independent separate events.
So the whole notion of nested safety is a
long term notion in the Commission's general
philosophy, unless they have changed them in the last
few years, and I haven't paid attention.
I would be startled if independence in
that sense is not a requirement. Otherwise, it
doesn't make a whole lot of sense frankly to require
a whole series of defense-in-depth, a set of nested
If I can knock them out with one event,
what have I got? I mean, the intervenors will cut you
to ribbons and should I think.
DR. MCCARTIN: Well, we did not try to
prescribe any type of independence, redundancy, or
anything to the multiple barriers other than looking
at engineered and natural, and just describe for us
the barriers that you have in your calculation.
DR. STEINDLER: I can remember when Mel
Napp gave us a lecture about a committee, and gave us
a lecture about the role of barriers. Basically, his
argument was you have got to have them, because who
knows, there may be something that goes wrong with one
that you haven't thought of.
And so our comeback was that we're smart,
and so is the staff, and they have thought of
everything. He didn't buy that. So in that sense
independence is a requirement if you haven't thought
of it. But that is an observation and I am not trying
to argue the issue one way or the other.
DR. MCCARTIN: I think the closest that we
have come to it is in the subpart on technical
criteria, and we talk of that you are looking at
multiple barriers to provide a measure of resilience
to the repository.
DR. STEINDLER: Yes.
DR. MCCARTIN: But there is no explicit
statement that they have to be independent.
DR. STEINDLER: Do you use the term of
DR. MCCARTIN: No.
DR. STEINDLER: Good, because there is a
big argument about whether that makes any sense at
all. It is a thousand year ground water travel time
turned out to be kind of laughable when you are
talking about what travels.
DR. MCCARTIN: Yes. Then we had a 300 to
a thousand year waste package lifetime, and if you
look at it now --
DR. STEINDLER: Well, there was a lot of
faith involved that geology in fact would do something
for you. And geology doesn't do quite as much for you
as you thought.
CHAIRMAN WYMER: And defense-in-depth
incorporates non-scientific things, too, if you really
explore what it means, you know. It could be part of
your organizational structure and the way that you
have got things set up.
DR. STEINDLER: Well, that wasn't allowed
I don't think. Defense-in-depth generally involved
technology, or technological criteria more than
DR. MCCARTIN: Well, I think since the
white paper came out on defense-in-depth from the
Commissioner, I think we have tried to look at Part 63
and how there are the elements of defense-in-depth,
and I think part of it, for example, from a mitigation
standpoint, there is a requirement that they have to
do post-closure monitoring.
DR. STEINDLER: Yes.
DR. MCCARTIN: And part of that is a
mitigation measure, and that you are going to put up
a system for perpetual care and monitoring of the site
by DOE and can we rely on it? No. But there are
certain things like that that have an element of the
DR. STEINDLER: It is my personal view
that the post-closure monitoring order on 300 to a
thousand year life package, and a thousand year ground
water travel time, in terms of ethicacy, and of giving
me warm and fuzzy feelings.
CHAIRMAN WYMER: Again, it is a question
of did you anticipate everything.
DR. MCCARTIN: Of course, we can embrace
CHAIRMAN WYMER: I think we are getting to
the end of the string here. Tomorrow morning we will
begin again at 8:30, and tomorrow morning will be
largely a bull session. We are just going to kick
things around, and probably try to decide on what the
format of the content of a letter might be, and what
kinds of things we should include.
We will not discuss specifically what we
are going to include, but exactly how we should
structure the letter, and what things we should cover.
DR. CAMPBELL: We do have some discussion,
a couple of facts. I don't intend to really go into
TSPA, because as some of you have seen, this is a lot.
And in fact what I have kind of pulled out
and talked about today are really things that I have
been pulling out of TSPA and maybe going into AMR. I
think we need to talk about a couple of effects.
One of the things that came up earlier was
how all of this discussion relates to the issue
resolution process and I sent them to you guys, but
you probably didn't drag them with you, and that is
the summary highlights of the three main tech
exchanges that impact what we are talking about.
One is the container life and source term,
and I am going to leave these with you guys just to
help you, Evolution of the Near-Field Environment, and
What these do is to at least give you --
you know, take a look at them and see if there is --
well, given your particular concerns or issues that,
one, has it been addressed by the staff, or two,
hasn't it been. We are adjourned.
(Whereupon, the meeting was adjourned at
4:39 p.m., to reconvene at 8:30 a.m., on Thursday,
February 22, 2001.)
Page Last Reviewed/Updated Monday, October 02, 2017