Advanced Reactors (Workshop on Regulatory Challenges for Future Nuclear Power Plants) - June 4, 2001
UNITED STATES OF AMERICA
NUCLEAR REGULATORY COMMISSION
+ + + + +
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS
(ACRS)
SUBCOMMITTEE ON ADVANCED REACTORS
Monday,
June 4, 20001
Rockville, Maryland
The Subcommittee met at the Nuclear Regulatory
Commission, Two White Flint North, Auditorium, 11545
Rockville Pile, at 9:00 a.m., Thomas S. Kress,
Chairman, presiding.
COMMITTEE MEMBERS:
THOMAS S. KRESS
GEORGE APOSTOLAKIS
MARIO V. BONACA
F. PETER FORD
GRAHAM M. LEITCH
DANA A. POWERS
WILLIAM J. SHACK
JOHN D. SIEBER
ROBERT E. UHRIG
GRAHAM B. WALLIS
B. JOHN GARRICK
A-G-E-N-D-A
INTRODUCTION
Tom Kress. . . . . . . . . . . . . . . . . . 4
KEYNOTE ADDRESS
Commissioner Nils J. Diaz. . . . . . . . . . 8
DOE PRESENTATIONS
Overview and Introduction to Generation IV
Initiative
W. Magwood . . . . . . . . . . . . . .29
Generation IV Goals and Roadmap Effort
R. Versluis. . . . . . . . . . . . . .53
Near-Term Deployment Efforts
R. Miller. . . . . . . . . . . . . . .73
Generation IV Concepts
R. Versluis. . . . . . . . . . . . . .80
Next Steps Generation III +/IV
S. Johnson . . . . . . . . . . . . . 100
GENERATION IV DESIGN CONCEPTS
Pebble Bed Modular Reactor
W. Sproat, Exelon. . . . . . . . . . 116
J. Slabber, PBMR, Pty. . . . . . . . 119
International Reactor Innovative and Secure
M. Carelli . . . . . . . . . . . . . 171
General Atomic-Gas Turbine Modular
L. Parme . . . . . . . . . . . . . . 204
General Electric Advanced Liquid Metal
Reactor and ESBWR Designs
A. Rao . . . . . . . . . . . . . . . 237
NRC PRESENTATIONS
NRC Response to 2/13/2001 SRM on
Evaluation of NRC Licensing
Infrastructure . . . . . . . . . . . . . . 260
M. Gamberoni, T. Kenyon,
E. Benner,
A. Rae
Planned RES Activities
J. Flack, S. Rubin . . . . . . . . . 279
PANEL DISCUSSION ON INDUSTRY AND NRC LICENSING
INFRASTRUCTURE NEEDED FOR GENERATION IV
REACTORS . . . . . . . . . . . . . . . . . . . . 310
CLOSING REMARKS AND RECESS . . . . . . . . . . . 339
P-R-O-C-E-E-D-I-N-G-S
9:02 a.m.
DR. KRESS: I don't have a gavel to
convene this meeting, but I'll pretend I have, so the
meeting will now please come to order.
This is the first day of the meeting of
the ACRS Subcommittee on Advance Reactors.
I'm Thomas Kress, the Chairman of this
Subcommittee.
Subcommittee members in attendance are
ACRS Chairman George Apostolakis, Mario V. Bonaca,
Graham Leitch, Dana Powers, William Shack, Jack
Sieber, Robert Uhrig and Graham Wallis.
Also attending is ACNW Chairman John
Garrick.
The purpose of this meeting is to discuss
matters related to regulatory challenges for future
nuclear power plants. The Subcommittee will gather
information, analyze relevant issues and facts and
formulate proposed positions and actions, as
appropriate, for deliberation by the full committee.
Michael T. Markley is the cognizant ACRS
staff engineer for this meeting.
The rules for participating in today's
meeting have been announced as part of the notice to
this meeting, previously published in the Federal
Register on May 10, 2001.
A transcript of the meeting is being kept
and will be made available as stated in the Federal
Register notice.
We have received no written comments or
requests for time to make oral statements from members
of the public regarding today's meeting.
So that we can effectively manage the time
and allow for a maximum member, presenter and public
participation in sharing, the Subcommittee has set
down some rules of engagement, I guess we can call it,
or the following protocols. Please pay attention to
these.
Number one, the presenters should be
allowed to make their presentations without
substantial interruptions. Questions from the
audience and stakeholders will be entertained at the
end of presentation sessions, not the individual
presentation. So keep your questions in mind, you may
even want to write them down.
Members of the public and audience should
use question cards that we have supposedly provided to
you. The ACRS staff facilitator Mike Markley will
collect these and group them as practical and read
them into the record, and refer questions and comment
to questions to presenters and/or panel participants
as appropriate.
It may not be possible to respond to all
questions and comments, however all questions and
comments will be listed in the meeting proceedings
following the workshop.
Opportunities for direct audience
participation will be provided during panel discussion
sessions each day. Microphones have been arranged for
convenience of the audience during this meeting. So
it is requested that speakers identify themselves and
speak up with sufficient clarity and volume so they
can be readily heard.
I would like to remind speakers and the
audience that we set down some things that we want the
audience and the speakers and the presenters to
address. And I'd like to repeat what these are so
that we can focus correctly in this meeting.
One, we want to describe the design and
key safety features and status of the development of
the design for the various concepts.
We want to provide the planned license
application and deployment schedules, if available.
We want to identify licensing challenges
and opportunities as compared to Gen II reactors.
I think that's the major thing we want to get
out of this meeting, is to identify the licensing
challenges.
We want to discuss planned approach to
licensing, construction and operation as compared to
that currently used for Gen II reactors.
And this is another important element,
what changes are needed in the current NRC and
industry licensing infrastructure? Do the schedules
adequately support the planned Gen IV license
applications and employments. That's the licensing
schedule.
And a general comment, what if any
additional initiatives are needed.
So, with that as a statement of what we're
after here, I'll turn to the microphone over to our
Chairman Dr. Apostolakis.
DR. APOSTOLAKIS: I'm very pleased to
introduce our keynote speaker for this workshop,
Commission Nils Diaz. Dr. Diaz was serving as a
Commissioner of the U.S. Nuclear Regulatory Commission
in August 1996. Prior to that time Dr. Diaz had a
distinguished career in nuclear and radiological
engineering as a scientist, engineer, researcher,
consultant and entrepreneur.
In the research and development arena,
Commissioner Diaz worked for mundane light water
reactor safety and advanced designs to more complex
space power and propulsion systems and on the
conceptual design and testing of futurist reactors
like the UF-6, UF-4 and uranium metal fueled reactors
for the Strategic Defense Initiative.
Commissioner Diaz?
COMMISSIONER DIAZ: Thank you. I think I'm
going to stand.
Well, good morning. That last part of the
introduction was just to kind of let you know that,
you know, although some of these new reactors might
sound advanced, there were other monsters around that
were a little more difficult to work with.
I am reminded of the time that we actually
work with a reactor in which we only had to have it
working for minutes. How is that we only had to have
that reactor on and for three minutes? So somebody
finally said let's make things simple. Let's make
things very simple. Let's do away with everything
else. We just take uranium metal and start inject
into this reactor, it will be vaporized and we'll have
a uranium vapor reactor which will run and the core
was perfectly fine. It would run, very well for three
or four minutes. There was no problem. Looked over
all the core calculations, and looked at everything
else and everything was fine. It will actually
probably run.
There was minor detail, one of these
practical little details. It was the nozzle to inject
the reactor fuel, which of course the reactor was
liquid at the time. And no matter where we put it, it
will have a density of about, oh say, neutral blocks
of 10 to the 18 neutrals per square centimeters per
second, which power density will vilify the nozzle,
the fuel before it gets to the reactor.
So, those were the problems, and those
real problems.
I'm very really very, very pleased to be
talking with you today. This is an issue that, of
course, is very important to the country and it is
particularly appropriate that the Advisory Committee
on Reactor Safeguards is hosting this meeting at this
time.
The discussion on nuclear power has now
fully entered the national debate on the future of
America's energy supply and nuclear safety is going to
be a priority on everybody's agenda. The Commission
relies on ACRS for expert advice, safety of reactors
existing or submitted for licensing. The
recommendations of the Committee will be of particular
value to the Commission as we deliberate the
licensing.
I will be presenting my individual views
today. They do not necessarily represent the views of
my fellow Commissioners or the Agency.
I want to premise my remarks from a few
selected quotes from a "couple" of speeches during my
tenure as a Commissioner, just to set the tone from
where I'm really going to.
So let me start with a quote that I
believe is of extreme value.
"There is no credible regulator without a
credible industry. And there is no
credible industry without a credible
regulator."
"It is essential for the regulator to be
cognizant of the technology. It is
essential for the industry and
technologists to be cognizant of the
regulations."
"Regulations need to result in a benefit
or they will result in a loss." There is
no reason to be any regulations unless
they will benefit society.
"My goal is to ensure the paths are
clearly marked." That has been really
kind of what I've tried to do during my
years. "A path that is clear of
obstacles and unnecessary impediments,
with well defined processes, will provide
regulatory predictability, equity and
fairness."
Again, another one: "We are learning how
to define adequate protection in more
precise terms, and to define it in terms
that make sense to the American people."
And finally, "We have learned from our
mistakes and we are bound not to repeat
them." This last point, I hope that you
prove me right.
At the 2001 United States NRC Regulatory
Information Conference, I said "We might be asked, as
would other government agencies and the private
sector, to sharpen our skills, and improve our
efficiency to meet the needs of the country." We have
been asked. It is worthwhile to try to understand why
the President and the Vice President of the United
States have brought nuclear power generation center-
stage in the debate of the energy policy of our
country.
Shown in the next figure it's a
compilation of important aspects of the debate,
summarizing what has changed in 20 years. All of
these issues are known to you, both economically from
the regulatory side. Everything that had to do with
productivity, all of those things have actually
changed. A few things have remained the same. For
example, it is important to national security that we
have a stable generating base that will anchor the
electrical generation in this country. But many of
the other things have changed as the bottom line
changed from low predictability to good
predictability. It is our job to change it from good
to high.
The NRC has been changing to meet the
challenge of what must be changed and to strengthen
what must be conserved. I submit to you that we have
changed for the better, especially the last three
years, and that improvements in regulatory
effectiveness and efficiency are changing from goals
into reality. But it has not been easy, as many of
you know, and there are still lessons to be learned.
I must say, though, that there is one
change that I believe speaks louder than words for the
NRC staff and the agency as a whole: Priority is now
placed on what should be done better rather than on
what was done wrong. And this is a major cultural
change.
This cultural change is needed to enable
the consideration of newer, better and enduring ways
to exercise the mandate entrusted to the NRC by the
people of this country: To license and regulate the
peaceful uses of nuclear energy, with adequate
assurance of public health and safety.
I believe that we are now capable of
meeting the regulatory challenges that we face today
regarding advanced nuclear plants. The improve
industry performance over the past decade has enabled
the NRC to initiate and implement reforms that are
progressively more safety-focused. Furthermore, it
allowed the industry to concentrate resources on the
issues important to safety which provided a sharper
focus to regulatory improvements. Safety and overall
performance, including productivity, became supporters
of each other, with the clear and unmistakable proviso
that safety is first.
For existing nuclear power plants, the
list of profound regulatory changes and
accomplishments, many done under the mantle of the so-
called risk-informed regulation, would occupy the rest
of this meeting. Skip them. But five of them stand
out: The revised rules on changes, tests, and
experiments, the 50.59, the new risk-informed
maintenance rule; the revised reactor oversight
process; new guidance on the use of PRA in risk-
informed decision-making (Regulatory Guide 1.174); and
the revised license renewal process.
The list is growing. About two weeks ago,
the Commission approved COMNJD-01-0001 instructing the
staff to give priority to power uprates, bring it up
the priority list, make it a real purpose of the
Agency and allocate appropriate resources, streamline
the NRC power uprate review process to ensure that it
is conducted in the most effective and efficient
manner. All of these and most of the other regulatory
improvements conform to the Commission's decision to
focus attention on real safety. The resulting
improvements in rules, regulations and processes,
including changes to the hearing process and enhanced
stakeholders participation, are assuring the nation
that a fair, equitable and safety-driven process is
being used.
I mentioned risk-informed regulation, and
I can see Chairman Apostolakis a little more lively in
here, as an important component of the changes NRC
regulatory structure. And I firmly believe it is an
important point. I want to be sure you know what I
mean, what I personally mean when I use the term risk-
informed regulation, so I'm going to present you with
my own personal definition of it:
Risk-informed regulation is an integral,
increasingly quantitative approach to
regulatory decision-making that
incorporates deterministic, experiential
and probablistic components to focus on
issues important to safety, which avoids
unnecessary burden to society.
And I think you know most of these things.
I really want to focus on why I am extremely attracted
to risk-informed regulation, and it's the last
sentence, which avoids unnecessary burden to society.
And I firmly believe that that is the test.
The definition can also be used for risk-
informed operations, risk-informed maintenance, risk-
informed engineering, risk-informed design, whatever
you want to.
For new license applications, much
groundwork has been done, and a lot of it is useful to
address today's issues. Going back in history in the
statement of considerations for 10 CFR Part 52, the
Commission stated that the intent of the regulation
was to achieve the early resolution of licensing
issues and enhance the safety and reliability of
nuclear power plants. Nothing wrong with that.
The Commission then sought nuclear power
plant standardization and the enhanced safety and
licensing reform which a standardization could make
possible. In addition, 10 CFR Part 52 process
provides for the early resolution of safety and
environmental issues in licensing proceedings.
The statement of considerations for 10 CFR
Part 52 goes on to say, and it's a very interesting
statement "The Commission is not out to secure,
single-handedly, the viability of the [nuclear]
industry or to shut the general public out." In
essence, it's continuing to quote "The future of
nuclear power depends not only on the licensing
process but also on economic trends and events, the
safety and reliability of the plants, political
fortunes, and much else. The Commission's intent with
this rulemaking is to have a sensible and a stable
procedural framework in place for the consideration of
future designs, and to make it possible to resolve
safety and environmental issues before plants are
built, rather than after."
In February of this year, the Commission
directed the staff in COMJSM-00-0003 to assess its
technical, licensing, and inspection capabilities and
identify enhancements, if any, that would be necessary
to ensure that the agency can effectively carry out
its responsibilities associated with an early site
permit application, a license application and the
construction of a new power plant.
In addition, the Commission directed the
staff to critically assess the regulatory
infrastructure supporting both 10 CFR Parts 50 and 52
with particular emphasis on early identification of
regulatory issues and potential process improvements.
The focus of these efforts is to ensure that the NRC
is ready for potential applications for early site
permits and new nuclear power plants.
I repeat, the purpose of these efforts is
to ensure that the NRC is ready for potential
applications for early site permits to certify designs
or designs to be certified, and that the NRC does not
become an impediment should society decide that
additional nuclear plants are needed to meet the
energy demands of the country.
In this case, let me assure you that the
Commission I'm sure will be interested on necessary
safety-focused regulations, definitely yes.
Unnecessary, not safety-focused regulations, no. The
staff is working hard to carry out this direction and
I am sure you will hear about some of our efforts over
the next two days.
Risking being repetitive, I'm going to re-
start at the beginning, and I know that I sound
strange, but it's really at the very beginning.
The U.S. Nuclear Regulatory Commission has
a three-pronged mandate:
Protect the common defense and security.
To protect public health and safety, and
To protect the environment.
by the licensing and regulation of peaceful uses of
atomic energy. I have long advocated that an adequate
and reliable energy supply is an important component
of our national security. An important component of
our national security. And I firmly believe that this
three-prong approach is going to endure the test of
time because it is good, and because it is balanced.
Within that mandate, within that three-
prong mandate I am an advocate of change, functioning
under the rule of law. As we face the regulatory
challenges that are sure to be posed by the
certification and licensing of new designs, a series
of all too familiar requirements will have to be met,
regardless of the licensing path chosen. And this,
you know them well:
Public involvement
Safety reviews
Independent ACRS review
Environmental review
Public hearings
NRC oversight
I am convinced, and I have white hairs to
prove it, by practical experience that the present
pathway for potential licensing success of certified
or certifiable new reactor applications is Part 52,
and I will tell you why.
First, it exists; and this is not the
minor issue the fact that it's here and available, and
is in the books.
Second, it contains the requirements for
assurance of safety and the processes for their
implementation.
And lastly, it can be upgraded to meet
technological advances that require new licensing
paths, without compromising safety.
Windows of opportunity can be opened, yet
the price is always the same: Reasonable assurance of
public health and safety. A new technology, with
different design basis phenomenology. In other words,
things like single phase coolant that we are talking
about, could present the need for a different pathway.
Yet, it would have to face the same requirements
listed above. What could be different is the manner
in which some of these requirements are addressed.
There is definitely room for innovation and
improvement, within the safety envelope that has to be
provided for assurance of public health and safety.
I am also convinced that the NRC and all
stakeholders need to apply a common criteria to the
tasks at hand. Every success path, whatever direction
you're coming, however you define success should
follow this simple criteria: Every path, every step
has to be disciplined, meaningful and scrutable.
Allow me to consider widely different
roles.
The NRC has the statutory responsibility
for conducting licensing and regulation in a
predictable, fair, equitable and efficient manner to
ensure safety. Every step of these processes of the
licensing and the oversight has to be disciplined, has
to be meaningful and has to be scrutable.
Applicants need to satisfy the technical,
financial, and marketplace requirements, and meet the
NRC and other regulatory requirements. Every step
that is taken has to be disciplined, meaningful and
scrutable.
I have no doubt that there will be
objections and opposition and the law of the land will
respect them and give them full consideration. The
objections will have to be disciplined, meaningful and
scrutable.
These common criteria are necessary, but
they are not sufficient as you all know.It is
indispensable that what we have learned, and it is
much what we have learned, be incorporated into the
science, engineering and technology supporting any new
reactors; they have to be as good as the state-of-the-
art permits.
Let me take a chance and depart from my
statement. There is no doubt that we're all creative,
we're all innovative, we like to do things better.
But this is the time that will not take too many
errors. This is the time in which we need to be
patient and we need to exercise what we know in a
disciplined manner to make sure that errors are
avoided. Okay?
Things that we do will have to be upscale.
And everything applicants do will have to be on
budget. Anything else is not good enough.
Whatever we do with the technology, we
have to match it with the regulatory processes. They
have to be as good as the state-of-the-art permits.
I happen to believe that risk-information can be a
contributor to disciplined, meaningful and scrutable
processes and to the underlying science and
technology.
Someone once wrote a phrase framing how to
achieve high performance expectations, which is where
we are right now, and it may be appropriate then to
just pause a moment and think that a lot of us need to
promise to think only the best, to work only for the
best, and to expect only the best.
Thank you very much.
DR. KRESS: At this time I think we are
collecting some written questions. Is that true,
Mike?
MR. MARKLEY: We're working on it, Dr.
Kress. At this time we don't have any.
I think we could entertain oral questions
from the audience at this time while collecting these
written ones. They don't have to be written. So, if
anyone has a burning question they'd like to ask
Commissioner Diaz, please feel free to do so. Use
this microphone or this one over here, please.
Please identify yourself.
MR. QUINN: Commissioner Diaz, it's Ted
Quinn.
The question I have that the combined
operating license part of Part 52 is unproven. We
haven't run through that yet, as well as early plant
siting. Can you define how the Commission can help
the staff to provide, to make this a more stable
process as we go through it so that the financial
community will help us to get these through?
COMMISSIONER DIAZ: It's a very good
point. We have it, it's there. We've been looking at
it for some time, but it's not been tested. The issue
is how do we make sure that it works the way it should
be, effectively and efficiently.
I think we learned a lot at the license
renewal process. And I believe that what I have
learned the last few years is that Commission
involvement is very, very, very, very necessary in
this step. That we cannot let a lot of these things
go a lot of the time to perfection.
I will use one of the first phrases I used
in a meeting down there that the enemy of the good is
the better and the enemy of the better is the best.
And, therefore, we are going to have to be in very
close contact with the staff. And I believe the
Commission will actually take an important role in
making sure that the processes are timely.
In this respect what we have done is many
other things the last 3« years, is we have maintained
our doors open. We have allowed stakeholders from all
different areas to come and visit and let us sometimes
close this little gap that exists, it is vital
information to us how stakeholders, whether they're
industry or there are other, you know, groups that
have an interest in the proceedings, let us know how
things are going. And that has worked very well. It
keeps the Commission informed early. Sometimes, you
know, the staff protects the Commission and shields us
from knowing the little problems that are happening.
And sometimes that is fine. It's really, you know, I
appreciate it. But there are times in which we need
to know ahead of time.
And I think this process should be very
similar as far as the Commission is -- really on top
of it all the time.
DR. KRESS: Other questions? Do the
members of the ACRS wish to ask a question of
Commissioner Diaz.
DR. POWERS: Dr. Kress, I'd like to phase
the issue of nuclear waste, which comes up repeatedly
in connection with all the discussions of nuclear
power, especially as we go to looking at maybe an
increased use of nuclear power.
Are we making any progress on this nuclear
waste issue? Is there something that the NRC can do
or is this totally in the hands of the Department of
Energy?
COMMISSIONER DIAZ: I think the NRC has
done as much as it can do. We have engaged in the
process all the way. And we have tried to make sure
that everybody understands that we believe there is
the science and technology that offers a better
pathway that ensures public health and safety.
I think the decisions right now are
practically at final stages. I cannot comment on them.
I think that, you know, we are going to do what we do
best; we're going to take whatever the country decides
in the Congress of the United States and the
President, and EPA and we're going to work with them.
We're going to try to make it, you know, an inspective
process. And that is what we do best.
You know, whatever is coming down, we're
going to use it. And if an application is submitted,
we're going to try to license working through a
process, and that process if not assured. We're going
to have to look at it every step of the way. And,
hopefully, you know, the Department of Energy will do
a good job and will allow us to do a provision of it.
And we will like to ensure that the process is open to
the public. We need to make sure that this is
disciplined, meaningful and scrutable.
DR. POWERS: Not to get off point or
anything.
DR. KRESS: I have a question, Mr. Diaz.
With some of the new reactor concepts, I see one of
the hard places regulatory challenges to be in the
area of defense and death, which is you know a
general guiding principle for regulation.
Do you think the concept of defense and
death is sufficiently rigorously defined to quiet some
of the newer reactor concepts or will we have to
rethink what we think defense and death is?
COMMISSIONER DIAZ: This is a setup.
DR. KRESS: I'm sorry about that.
COMMISSIONER DIAZ: I think, you know,
those of us who work in reactor science know what
defense and death really is and what are its
limitations. I think we have actually reached the
limitations of defense and death, and that it is time
to move forward and use it in the best possible
manner, but complimented with everything else that we
can to make sure that we don't make cumbersome, you
know, design requirements or cumbersome regulatory
requirements. And I go back to that definition, the
end of the definition and risk-informed regulation,
which avoids unreasonable burden. And that's what we
have to do, because the burden eventually will be in
the top, you know. The logical thing the burden will
be on whoever it is, the burden is eventually in the
people of the United States.
So, I believe that we need to relook and
resharpen our focus. I know the ACRS has been working
on this, and I share a lot of your views.
DR. APOSTOLAKIS: Well, this is related I
think to the use of risk-information in licensing and
regulations. And we hear that the agency may, in
fact, receive license application in the very near
future. Do you believe, Commissioner, that the
regulatory system is ready to review such a license
application or does it require some fundamental
changes, which will take time, of course?
COMMISSIONER DIAZ: This is setup number
two.
Knowing we think we're ready, but we count
on the ACRS to make us ready.
DR. APOSTOLAKIS: I am speechless.
COMMISSIONER DIAZ: We will work hard at
it. And you guys are going to need to come and pitch
in. I think everybody is getting their attention
focused on how can we move in this area, what is that
we know sufficiently that will provide within that
envelop that I keep referring to provide the
protection of all the processes. And I think there
are hard decisions to be made, and I'm not kidding
that we can revoke our problems.
DR. KRESS: Any other questions?
Mike, are there written questions that we
could entertain?
MR. MARKLEY: No, we have no written
questions at this time.
DR. KRESS: Okay. With that, I'd like to
personally thank once again Commissioner Diaz for an
excellent keynote speak.
As a matter of fact, we're a little bit
ahead of time. But at this time I would like to go
ahead with our scheduled break. Let's keep it to
about 20 minutes, and return about 10:00.
(Whereupon, at 9:30 a.m. a recess until
10:01 a.m..
DR. KRESS: Let's get started again,
please.
Based on our experience so far, I'm going
to go out on a limb and change the mode of operation
just a little and do away with the cards as an
experiment and allow questions to be entertained after
each presenter makes his presentation, so it'll be
fresh in your mind what you just heard, and you can
give all the questions at each of the microphones. So
we'll try that and see if it works better. If it
doesn't work, we'll go back to the cards.
Now we'll turn to the spot on the agenda
in which we will hear extensively from DOE for Gen IV
and Gen III. And the first DOE speaker is listed as
Mr. Magwood, so I'll turn the floor over.
MR. MAGWOOD: Good morning.
Are you sure you can hear me? Are you
sure you want to hear me?
Well, good morning. I'm Bill Magwood, I'm
Director of DOE's Office of Nuclear Energy, Science
and Technology.
Thank you for scheduling a break in a time
that I was able to go to the restroom. I really
appreciate that. It will make the presentation a
little bit longer, but that's a good thing or a bad
thing; depends on what you think about what we have to
say.
First, in the way of introduction, and I
apologize. I'm a little behind on what the viewgraphs
look like. I know that I saw these about a week ago,
but since I've been out of town and then here I am.
So, I'll be sort of looking at these a little bit
fresh, I think.
Of course, I just got paged, and hopefully
it's not the Secretary's office. Okay. That can
wait.
Well, first, let me give you a little of
background about the Office of Nuclear Energy, Science
and Technology. Our program, as you know, has been
around since the beginning of the Atomic Energy
Commission back in the late 50s. And we're basically
the same program that's existed throughout the '60s,
'70s and '80s; the names have changed, the faces have
changed but basically we're the Nuclear R&D program of
the federal government. We're responsible for
advanced reactor technology development, fuel cycle
technology, medical isotopes, space reactors; the
whole range of federal involvement in nuclear R&D.
And over the last decade we've seen our
activities plummet to a really, quite frankly,
embarrassingly low level. Actually, in 1998 our
budget actually for nuclear energy research
development and development actually hit zero. And it
was kind of an embarrassing situation for us. We had
people coming in from Korea and Japan asking what's
going on, what does this mean. And it was very
difficult to explain to them well, you know, it's kind
of like being between jobs. You know, we're between
programs right now.
What we were doing during 1998, though,
was not sitting on our hands. What we were doing was
trying to understand what DOE's rule in nuclear R&D
really ought to be in the long term future.
In the past, DOE's program is
characterized largely by the creation of demonstration
reactors, very large, very expensive programs like the
integral fast reactor program, defense reactor
project, things like that. It was pretty clear that
we weren't going to be seeing hundreds of millions of
dollars anytime soon, so we were going to have to find
a smarter, more efficient way to do nuclear research.
What we came up with was a variety of
things. First, we recognized that we were going to
have to base our program much more on international
cooperation than in the past. In the past, DOE always
had been a large monolith to which other people tagged
on. The Japanese worked with us, the French worked
with us, other people worked with us, but DOE was much
more self-reliant and was more interested in
assimilating technology than it was in bringing
technology in. That had to change because of the
resource issue.
The other thing that we recognized was
we're going to have to bring in much more outside
perspective, much more of an outside peer review
approach. So that ultimately became our nuclear
energy research initiative, the NERI program which
some of you are familiar with.
But we also recognized that it was going
to require more of a cooperation with our stakeholders
such as NRC, which we're now working more closely with
than ever before, the industry, our Nuclear Energy
Compensation Program, entities like that. And also
focusing more on infrastructure, which is something I
think you're going to hear a little bit more about
over the course of the morning.
And one of the parts of research we have
been working on a great deal has been our university
research reactors and education program.
So our program over the last several years
has really changed dramatically from what it was, say,
five or ten years ago. In fact, I think a lot of
people looking at the program from that perspective
will probably be very surprised to see (1) how much
less money we have, but (2) but in the way we operate,
how different it is.
What we're going to be focusing on today
is what is the future for the nuclear research program
both in the federal government, but also more broadly
talk about that.
See the next slide, please.
One of the primary focuses that we've
enjoying over the last year or so has been Generation
IV systems. You're going to hear largely about that
I think this morning. I think that's the focus of
this presentation, and I'm going to explain to you
what that is.
Now, this proves this I haven't seen this
because I would never be giving you a talk with little
mailboxes on it. And I think these are pencils.
They're either pencils or ballistic missiles, I'm not
really sure which. Since we're a civilian program,
I'm going to assume they're pencils.
Generation IV energy systems are systems
that can be deployed by 2030. So, I'm going to
actually skip this chart and go to the next chart. I
think it's much more descriptive. Why don't you give
me the next chart. I think I'm right. Yes, okay,
much better.
Here's how we got to Generation IV.
Looking back in the past we had this first generation
of systems, such as the Dresden plant, the
Shippingport plant, the very first ventures in the
commercial scale of nuclear power production. These
lead to the most successful energy programs, I think,
in the history of the federal government in some ways;
today's nuclear power plants, Generation II nuclear
power plants. And these make up most of the plants in
operation in the world today. These are all the LWRs
in the United States and most of the LWRs throughout
the world, as you know, which are based on U.S.
technology.
The very successful program, obviously,
has not been entirely successful otherwise we would
still be building them, but nevertheless when you look
at the fact that 20 percent of our electricity comes
from these power plants, it's hard to say it's been
less than successful.
We did, however, need to do some
improvements. And as we learn more about how nuclear
power plants operate, we were able to design the next
generation of plants, Generation III plants, the
advanced light water reactors and the advanced BWR,
the System 80+, the AP600 that generation of nuclear
power plants. And this is also, I think, on the verge
of being very successful. They're already building
some of these plants overseas, obviously in Japan,
Taiwan, but also parts of the technology are beginning
to disseminate elsewhere in Korea.
So when we start to think about what the
future ought to be, the question really was where do
we go from here? Where do we go from the Generation
III reactors? Well, there's two steps. There's a
near-term step which we either consider to be just a
follow on to Generation III or we actually give a
little bit of an extra push and call it Generation
III+. And then we speak of Generation III+ we're
usually talking about slight enhancements to the
existing state-of-the-art nuclear power plants.
For example, the AP1000 versus the AP600
is considered to be a Generation III+. There are
others. I'll try not to get too specific about that
because you get in arguments about what's Generation
III+ versus Generation IV, and it's a pointless
exercise.
But part of our program is focused on
trying to move to this next step, deployment of the
state-of-the-art technologies possibly with some
enhancements in technology, Generation III and III+.
But the more exciting part of our program, I think, is
looking at Generation IV reactors. Generation IV,
quite frankly, is just characterized in very simple
ways: What comes next?
Now, we do have some more of a definition
then at this point, and I'll talk about that.
Let's go to the next slide.
What we've done so far is the Subcommittee
of our Nuclear Energy Research Advisory (NERAC) to
establish specific technology goals regarding these
future reactors. I think we're going to get some more
detail about this. But when NERAC brought this group
together in just October 2000, it's been a very, vary
active group ever since. Their job is to help us
develop a technology roadmap for Generation IV nuclear
power plants.
This technology roadmap is going to be
lead by a subcommittee of NERAC, which is composed of
people from U.S. industry, academia. And now there
are laboratory people helping them, but really the
core of the group is made up of academia and is co-
chaired by Neil Todreas at MIT and Sal Levy of GE.
And they provide a lot of leadership in trying to move
this process forward.
Let's take a look at the new viewgraph.
Okay. That helps.
The NERAC Subcommittee had as its first
action, and we gave it a very, very short term time to
do this, to draft these technology goals for the
direction for nuclear power plants. As I say, you're
going to hear more about this, but to give you an
example the technology goal for Generation IV is, one
of the goals, and it's my personal favorite states
that there should be no operating or accident
condition that required an off-site response to an
emergency. And that means eliminating the concern of
the public, basically, that the operation of nuclear
power plant would effect their lives. Whatever
happens to the plant stays on site. It becomes an on
site issue, but would not have an impact off site.
That's a technology goal.
Now, we had a lot of discussion about that
as a goal, obviously, because a lot of people say
"Well, you know, you can't ever promise it will never
be outside event. But, you know, we took a philosophy
that if it's a technology goal, you work towards that,
you see how close you get, you see where the
technology leads you. So, that's part of the process
and you'll hear more about this.
More to the point, these technology goals
aren't an end into themselves. They're used to drive
an R&D program. And what NERAC's next goal, and this
is where we are right now, was to take those
technology goals and formulate an R&D program based on
them. And how are we doing that?
Now, as you're about to hear what we've
done is we've reached out to a very, very large group
of people out to the international community. We have
-- let's skip over to the next one. I'm not going to
go on all these viewgraphs.
We've brought together something called
the Generation IV International Forum, which I expect
to be official by the end of this month. We're
working with eight other countries; Argentina, Brazil,
Canada, France, Japan, South Africa, South Korea and
the United Kingdom. We're working with these
countries to try to formulate what concepts, what
technologies can meet these very, very high level
technology goals that were set by NERAC. So the
Generation IV International Forum has worked with us
to identify approximately a 100 people all over the
world, most are in the U.S. but there's about 40
percent or so of them are actually international from
these various countries, but also including people
from the IAEA, people from the OECD/Nuclear Energy
Agency and people from the European Commission to help
look at all of the various concepts that are out
there, all the ideas that come from our NERI program,
for example, and put them through a very, very
extensive rigorous progress with the goal of arriving
at a small number of technology concepts about which
the international community including the U.S. can
rally about.
Our goal is that by the end of -- and I
don't know if the next one's got names or not, we'll
take a look. No, we'll skip that one. Okay, that'll
do.
Our goal -- work backwards on this chart.
Our goal is by September '02 to be in a position to
tell you what handful of concepts, we're aiming for
maybe about a half a dozen concepts, hopefully less.
But a half dozen is probably the most we can stand.
What small number of concepts would be acceptable
under the Generation IV technology goals and about
which you can write specific R&D plans.
Now NERAC's job will be to identify those
concepts and then write the R&D plans, and that will
constitute the technology roadmap.
This has already been a very ambitious
project. In fact, I think a lot of people when they
first heard about what we were going to try to do,
thought we would never be able to get this far. We'd
never be able to get so many countries to agree on a
process that would narrow so many concepts down over
such a short period of time. But so far, we've been
very successful.
We've been able to keep the Generation IV
International Forum together as a unit. In fact,
rather than having it fly apart, it's actually become
much more close knit, much more integrated than it was
when we started off. And we've actually agreed to a
charter that each of the countries will sign by the
end of this month. So we're very excited about that.
Now, in the nearer term, obviously,
because of the energy concerns we're experiencing in
this country, we do have to think about what can be
done this decade. Let me speak about the dates for a
moment.
One of the things that I said earlier was
that Generation IV concepts need to be deployable by
2030. That's not to say that if you can arrive at a
Generation IV concept it can be deployed next year
that we shouldn't go forward with it. But the limit,
the outer limit is 2030. That means that we don't
have a situation where we're competing with fusion to
be the long lead technology for the Star Trek
generation, okay? We want to make sure that where we
talk about real technologies things can be engineered
now and try to arrive as -- projects can be
demonstrated within a very, very reasonable of time.
So 2030 is the outer limit.
In the case of the near-term plans, the
Generation III+ technologies for example, we're
focused on things that can be done in about 2010.
Now, we're a little softer with that date because
there may be some things that are more arrival in
2012, say, versus 2010. So we're a little squashier
about that. About 2010 is the time frame we want to
see these new near-term technologies deployable.
Our goal is to make sure that we can
identify the technologies, the technology programs,
the institutional barriers that need to be resolved in
time to enable these plants to be built in the U.S. by
2010. And we are working very closely with the
industry on this. We have a task force under the
NERAC Subcommittee that's chaired, I believe, by Lou
Long of Southern Company. Is that correct? I think
it's Lou Long. Is there a co-chair? Tony McConnell.
Okay. And these folks are helping us on an industry
basis. In fact we've just come out with a CBD notice,
I believe and a Federal Register notice to solicit
input from the industry to identify what those
institutional barriers are, technology barriers are
and to put forward a plan to try to resolve all those
barriers in a time frame consistent with our 2010
date.
This one is a little ahead of the
Generation IV side. We expect to have that more
completed this September. And actually, most of it is
already done. We're really just about there. There's
a lot of things that need to be refined, but the
larger ideas are really in place. And by next year,
September '02 we'll have the entire Generation IV
roadmap.
So that's what we're pursuing at this
point. It's a very, as I said, ambitious activity.
It involves a huge number of people.
You're going to hear about how we've
organized this. Who's giving that? Is that you, Rob?
Rob is going to describe how we've organized this. It
looks like a spaghetti nightmare, but trust me; it
makes sense, it works.
Is that the last viewgraph? Okay.
With that, let me just summarize by saying
that the U.S. DOE has been gratified with the response
we've gotten from the international community and from
the industry, and from NRC and everyone else that's
worked with us on this. It's been a very important
activity.
And excuse me, John, for turning my back
to you. John here is helping us a lot with this, so
he's very familiar with what we're doing. And what
we're trying to do now is to bring all this home.
We've organized it, we've got participation from
everybody that we think we need participation from.
We're going to reach out a little bit more to
stakeholders over the next year, I think. But this is
really working and we're going to keep the work, and
we're looking forward to your thoughts as we go
forward.
And I appreciate the opportunity to talk
to you today, and I'd be happy to answer any
questions.
DR. KRESS: We'll entertain questions from
the audience or from the members, either one.
DR. APOSTOLAKIS: Dr. Magwood,if you had
to give us the two most important regulatory
challenges for meeting all these wonderful
initiatives, what would they be?
MR. MAGWOOD: That's a good question. I
think that the most -- I think I'll answer the
question a little more generic.
I think that it's extremely important the
NRC move as close to performance based risk-informed
regulation as possible. Because these technologies
are dissimilar in so many ways, and you're already
starting to see it. There's already a large
discussion going forward about the pebble bed reactor
versus light water reactor technology and how you
license those.
The only way to do that successful with
these different concepts floating around out there is
to move to a technology independent regulatory
approach. And unless you do that, you're going to
inhibit the development of these new technologies
because people will not have the confidence that NRC
can respond quickly enough to regulate these
technologies.
I know there's a lot of concern about how
long it's going to take to get regulations for the
pebble bed reactor. And we're working with General
Atomics at DOE with the development of their system,
and that presents similar challenges. So I think that
that larger issue is the one you have to deal with.
In the nearer term I think it's really
more a job of demonstrating the pieces are already out
there. But even as we look at these newer
technologies coming in before now, they present
issues, many that you are already very familiar with.
So I would say that pushing as fast as
possible towards a new regulatory regime that will
support new technologies in the next century is really
going to be -- should be a high priority.
DR. APOSTOLAKIS: In the next century?
MR. MAGWOOD: Well, in this century. I'm
sorry, I fell back. In this century. I'm sorry I fell
back.
DR. APOSTOLAKIS: Speaking of long term.
MR. MAGWOOD: Well, you know, it's
interesting one of the things I mentioned to the
international community -- I'll just sort of digress
for a moment.
One of the things that was very
challenging about pulling everyone into this early on
was that unlike the U.S., other countries know where
they want to be in 20 or 30 years. You know, the
Japanese have very specific plans of where they'd like
to be over the next 30 years. So, you know, getting
countries like Japan and France that know where they
want to go to agree to a process like this was
challenging, to say the least. But I think that the
fact that we're open-minded about where the answers
come out gives them confidence that, you know, that
their ideas may well fit into whatever comes out of
the end of this.
Also just for your gratification, one of
the things that we were very pleased about with the
international community was that they made very clear
that they believe that the U.S. was the only country
that pulled this together and that without the U.S. in
the middle of this bringing all these other countries
together, that there's no way you would ever be able
to arrive at what they believe, what many countries
believe the future really has in store for us which is
more common reactor designs international.
And so doing this on an international
basis is absolutely essential. Having the U.S. go off
and do this on its own would be a waste of everybody's
time and money. And so, you know, we've been very
pleased with the international response. But I think
that in the future we're going to see that the steps
that you take and the steps the NRC takes towards
regulating these new technologies will really set the
tone for the rest of the world. So it's very
important that we go about that in the right way.
DR. APOSTOLAKIS: Is NERAC going to give
us any ideas as to how we can have this regulatory
system that will not be technology specific?
MR. MAGWOOD: We've talked about whether
to get involved in that. And I think the main
conclusion was that we shouldn't because for two
reasons. First, it really is something that NRC needs
to deal with. You know, it's something that the NRC
has more experience with than we do and very few of
the people that we've been working with are very
comfortable going off to give NRC a lot of specific
advice.
And secondly, quite frankly, the time that
it would take to do that probably means that it would
require a different project than what we're currently
doing. That's not to say that we wouldn't have a
follow on step where we would try to move in that
direction. But for the near-term, I don't think
there's anything that NERAC's is going to add to where
NRC is going. We just need to encourage them to move
forward quickly with what they're doing.
In a longer term, it may make sense to
bring another group together to look at those long
term regulatory issues.
DR. APOSTOLAKIS: Thank you.
DR. KRESS: Other questions?
DR. POWERS: Well, it seems to me that if
you're going to encourage people to move to a
performance based regulatory system, that must mean
surely you're looking at performance indicators for
these new generation? Is that the case?
MR. MAGWOOD: I think the answer to that
is yes. If you look at our technology goals, and I
think you're going to get a rundown of that. Is that
going to be part of your presentation? You're going
to get a rundown of that.
You'll see a very high level version of
what those performance goals are. On a regulatory
space, you're talking about safety. You'll see some
indications where we think things should go, but not
to the level of detail because these technology goals
are very, very high level. You're not going to see a
low level of detail, but you will see an overall
vision.
DR. POWERS: High level and not very
specific doesn't make for useful regulation.
MR. MAGWOOD: That's a --
DR. POWERS: At some point somebody has to
come down and say if you want a performance based
system, you got to have performance indicators that
are used and monitored.
MR. MAGWOOD: But what I would say is that
what -- what we can provide as part of our process,
all these high level goals. These high level goals
will very quickly, depending on which technology
concept you're looking at, provide some framework that
NRC or someone else could use to begin to design a
regulatory approach. It's not really -- again, it
wasn't our intent to try to set this up to defeat the
NRC process. You know we clearly could to do that,
but that's not the intent here.
Our intent was to drive an R&D program,
not separate or instruct. Now, I'm willing to hear
some advice. You're an advisory group, so give us
some advice. We're part of the program.
If you think that we should follow on this
activity with an activity focused more to the
regulatory side, you know, I would be very happy to
work with Ashook and his group to try to put together
an appropriate advisory group that will do that.
Because I think it's important that it be done. And
if takes DOE involvement to get it started, I'm happy
to do that. But this isn't the activity to do it,
that's my biggest point.
DR. KRESS: Okay. Seeing no other -- oh,
there's one. Okay. Please identify yourself.
MR. LYMAN: Ed Lyman from the Nuclear
Control Institute.
Bill, I think there is public issues that
really have to be thought about before large expansion
in DOE's research budget has to be contemplated.
Because these days you have to really worry about
whether what looks like government subsidization of
one energy technology over another, how that will be
perceived, especially by small scale generators using
other competitive fossil fuel technology and stuff.
And in a deregulated environment that's going to be a
greater concern.
So, I was encouraged when these reports of
a task force on near-term deployment that recently
reported to NERAC discussed a cost sharing program
with industry for near-term deployment. I was
wondering if industry had actually made any firm
commitments in that regard, since would be a positive
step since I don't think they've put any money down so
far in these initiatives?
MR. MAGWOOD: First, it's important to
clarify, and I think you raised a good point. There's
two things really important to clarify.
First, in general, you know our office is
not in the business of corporate welfare. We're not
here to make technologies marketable that wouldn't
otherwise be marketable, you wouldn't otherwise
compete on it. In fact, our goals, and you'll hear
about it, for our Generation IV have a lot of built
into them about the need to be economically
competitive. That's a hallmark of what we're trying
to do.
And let me say for the record that there
should not be a new nuclear power plant that's not
economically competitive in this country. It
shouldn't be built because we're not going to
subsidize it and if industry is not willing to go off
and do it because they can make money, it shouldn't
happen. It shouldn't be done.
Now, regarding the specific point you
raised, I think that where we are right now -- well,
first it's important to recognize that this is a NERAC
advisory group, so we're not at the point where we're
making commitments on a policy basis on behalf of the
industry. We have asked certain experts in industry
along with academia and working with our national
laboratories to come together and make
recommendations. These recommendations will flow up
through the NERAC process and if it comes out the
other side, NERAC will make a recommendation to DOE
that we should go pursue a program in that vein.
But at that stage, if that were to happen,
we would be in a position to approach the industry and
say "Okay, your people were on this panel, here's the
recommendation that they made, Mr. CEO do you want to
buy into this?" And if they don't want to buy into
it, we don't have to do it. But, you know, it's a
recommendation. It's not a commitment on anyone's
part, especially ours.
You know, with my budget I couldn't commit
to anything they recommended at this point. So, it's
really a recommendation for the future.
The question we asked was if we were going
to solve these problems, how would we go about it?
And that's what these recommendations gives us. It
gives us a way of solving the problems.
It doesn't mean that we have to do it. It
doesn't mean the industry has to do it, but it gives
us a methodology.
So the answer to your question is no, no
one's made any commitments, nor would it be
appropriate to at this point in time.
DR. KRESS: Okay. With that, let's move
on to the next speaker. But before we do, the
question that George asked about what you may think is
the two or three most challenging, most difficult
regulatory challenges, each speaker might want to
consider that as a generic question and feel free to
volunteer an answer to it without it being asked.
The other item is, I don't have any
introductory information or remarks to make about each
speaker, so as was obvious with Mr. Magwood, so would
each speaker please introduce himself when he gets to
it.
So, with that, I'll turn it over to the
next speaker.
MR. VERSLUIS: Good morning, ladies and
gentlemen. My name is Rob Versluis. I'm the project
manager for the Generation IV roadmap.
Now that Bill Magwood has given you an
overview of Generation IV process, I'd like to focus
on the long term and in my talk summarize the roadmap
process and products that we expect.
The first objective of the Generation IV
roadmap is to identify and evaluate the most promising
advanced nuclear energy concepts. And we have three
years to do this. We started in October of last year.
And expect to be finished September next year.
An important role is played by the
advisory group. Bill has already mentioned it, the
NERAC Subcommittee. Actually, it's better known as
GRNS, Generation IV Roadmap NERAC Subcommittee,
although that's not actually their official name.
They are very much working with us and
directing or advising us on the direction for the
roadmap work.
The actual work is being done by several
working groups. The staff consists of about 50 U.S.
experts, about evenly divided between industry, labs
and academia. And recently we have received 40
volunteer experts from the GIF countries. That is a
very respectable participation from the international
community.
The second objective, and really the
product we are looking for from the roadmap, is the
R&D plan to support future commercialization of the
best concepts. And this completed roadmap will do two
things.
It will identify and evaluate concepts.
That is we intend to make a good start in calling out
the most promising concepts.
And secondly, it will formulate the R&D
tasks for the best concepts; that is to find a
sequencing and preliminary costs of the R&D tasks
required for commercialization.
We recognize that even after two years of
hard study there will be many questions left about the
viability of the most promising concepts. The R&D
defined by the roadmap is intended to both answer
questions of viability and show the real performance
capabilities of the selected concept.
And, of course, the final nuclear energy
system selection will involve industry and the
marketplace.
Like any planning activity we start with
formulating goals, which was actually done by GRNS.
And these goals strive to reflect energy needs for
mid-century, and we actually have the date of 2030 on
it, but obviously if these plans are going to be built
and deployed, they're going to be run for many years.
And so we've tried to envision mid-century
with its population growth, its growth in standard of
living, its world economy and its need for other
energy projects besides electricity, such as clean
water. This is reflected in the appearance of
sustainability goals alongside safety and economic
goals. And let me quickly take you through the goals.
In fact, this is all I'm going to show
about them, because Neal Todreas is tomorrow and his
talk will go in more detail about the goals.
There are three sustainability goals. One
that is concerned with the resource inputs, that is
fuel, materials, energy inputs in nuclear energy
system. Second with waste outputs. Waste streams of
all sorts. And the third is proliferation resistance
or nonproliferation.
Then there are three safety and
reliability goals. One on excellence, one on core
damage and one on emergency response.
And finally, there are two economics
goals: Life cycle cost and risk to capital.
These goals, in fact, provide the basis
for evaluating the technologies.
What do we really mean with a Generation
IV system? It is an entire energy production system,
including the nuclear fuel cycle front and back end,
the nuclear reactor, the power conversion equipment
and its connection to the distribution system. It
must recognize various energy products, electricity,
hydrogen, fresh water, process heat, district heat,
propulsion. And also the infrastructure for
manufacture and deployment of the plant.
Furthermore, we limit to systems that are
likely to be commercially viable by 2030. And also
the primary energy generators in the system must be
based on critical fission reactors. That means that
subcritical systems, accelerator driven system, would
have a secondary role in the fuel cycle, but the
primary energy generators should be critical systems.
The next slide shows the roadmap
organization. The central part shows the working
groups and the integrating functions. And I'll come
back to that in a minute.
On the left it shows the advisory
committee relating, of course, to DOE-NE in the
roadmap. And also the technical community, the left
bottom, from which both the GRNS and the roadmap draw
its resources; that is its staff. Further resources
are drawn then from the GIF countries on the right
hand side.
DOE-NE manages the program. This is where
Tom Miller, who will speak next, and I sit. And
underneath -- actually it shows the near-term
deployment group in orange underneath DOE-NE.
Then the next group that it shows is the
roadmap integration team, RIT. And look at those
abbreviations because they will come back in later
slides.
The RIT does what it says, it manages the
roadmap process and does the final integrating of the
roadmap itself. It is composed of two senior managers
from Argonne National Laboratory, two from Idaho
National Energy Environmental Laboratory and myself.
The next group shown is the evaluation
methods group, and this is the group that is charged
with defining the criteria and metrics by which they
evaluate the concepts on their ability to meet the
Generation IV goals. They actually start with the
goals and they translate them into criteria and
metrics, which is a long process, actually.
The actual work of identifying, describing
and evaluating the concepts is spread over the four
groups shown in the middle bottom. They are organized
by a coolant technology somewhat arbitrarily, but it
lines well up with people's expertise. And so there's
a group on water coolant, on gas, on liquid metals and
then there is none of the above where the non-
classical concepts are being evaluated and described.
In addition, we envision forming
technology crosscut groups. And that group, you know,
standing vertically there on the right is an example
of such a group. It draws actually from the same
people, from the same working groups, but it lines up
the experts in a certain technological area and it
puts them together to get a crosscut perspective over
all the concepts. And you can envision crosscut
groups like fuel cycles, risk and safety, materials,
power conversion and others, perhaps.
The fuel cycle group was formed early to
deal with the common fuel cycle issues for all of the
concepts, and also to define the fuel cycle framework
for the energy systems. And they have defined four
generic fuel cycles: The once through fuel cycle; a
single plutonium recycle; multiple plutonium recycle;
and a full actinide recycle. And they describe those
and provide a framework for the other groups to work
within.
They also analyze energy demand scenarios.
They're not making any new ones, they use the World
Energy Council's scenarios and they pick the three
scenarios of those to drive the thinking about
resources and build up.
This shows a high level overview of the
schedule for producing the roadmap.
Phase 1, the initial work is getting
organized and staffed. Phase II, the needs assessment
looks at the concepts and identifies the technology
gaps. Phase III, the response development defines the
needed R&D. And Phase IV, the implementation planning
actually finalizes the roadmap. And the slide also
shows the time frame when the activities take place
and about the product of the phases.
Let's step through the tasks. First the
goals and plans. First, we drive the technology goals
based on industry needs, and that has been done by the
GRNS and it's been reviewed and with some comments
endorsed by GIF. And it's captured in a technology
goals document.
Next, plan the activity. We published the
Roadmap Development Guide for use by the roadmap
participants that describes the overall approach, and
the working groups have been convened including
international participation.
The first time we convened all the working
groups was in February in Denver, and it only included
the U.S. participants and we described to them the
approach of the roadmap, the various responsibilities
of the groups and what's expected from them.
Then again, in Chicago we had the second
joint meeting of all the working groups. That was
last month in May. And that included all the
international participants. So we had, again, a
familiarization stage, but they also actually were
there to do work.
Then next we determine how to measure the
concepts against the goals. We developed a criteria
and metrics for each goal and then continue on to
develop the evaluation methodology. This is conducted
by the evaluations methods group with the feedback and
assistance from the roadmap integration team and the
GRNS.
This slide discusses how we're dealing
with the concepts. First, identify the concepts for
evaluation. We have now about 100 concepts and they
are drawn from the U.S. and a broad international
base. And they are now adopted by the technical
working groups and synthesized. When I say
synthesized, I mean that in many cases a concept was
not complete and needed to be synthesized with other
fuel cycle systems or parts of the fuel cycle system.
The concepts are also being grouped into
sets if they show sufficient similarity to increase
the productivity. To conceive a 100 concepts we're
going to have to package them up a little bit, and I
will talk about that later this morning.
Then the most promising concepts need to
be detailed better, so that's the next step. And the
TWGs are now interacting with the concept teams and
the advocates to get more information. They actively
study and compare the underlying technology. And they
are now getting ready for what's basically two
screening stages. The first screening is called
screening for potential and the EMG has developed
criteria, qualitative criteria for that. That initial
screening is pretty lenient and it's because it's been
based on limited information and we really don't want
to throw too many things out at this point.
And then a later evaluation next year
will be done next year.
Let me clarify what I mean with concept
and concept sets. Concept, as we use the word, is a
technical approach for a Generation IV system with
enough detail to allow evaluation against the goals,
but broad enough to allow for optional features and
trades. And a concept set is a logical grouping of
concepts that are similar enough to allow their common
evaluation.
In the second year we evaluate and
assemble. We evaluate the most viable concepts, we
compare the concept performance to the goals, and that
is really the finally screening. And then we identify
the technology gaps. And in this work the TWGs, the
technical working groups have the lead. And, of
course, the RIT and the EMG looks over their shoulders
and make sure that the criteria are being applied
consistently.
DOE has the approval function here, and we
will seek the endorsement of GIF.
And then the final stage is assemble the
roadmap to support the most promising concept. That
means identifying the R&D needed to close the gaps
that have been identified in areas of crosscutting
technology, assemble a program plan with recommended
phases. And that will then contain the sequencing and
estimated costs of the R&D tasks. And the groups
write here their final reports. The RIT takes the
input and integrates this into the roadmap. Again,
the DOE has an approval function and will seek the
endorsement of GIF.
This slide is another cut at the schedule
from the perspective of the screening and down
selection. A lot of work is actually going into
taking these goals, translating them into criteria and
metrics and applying them in these screenings. And,
as you see, the screening for potential is coming up
in July, 2001. Then there is an eight to nine month
period before we do the final screening, which will be
more strict and based on further developed and have
more sophisticated criteria and perhaps in some cases,
quantitive metrics.
After the roadmap completion, planning
becomes more uncertain as you go further into the
future because it involves things such as government
policy, budget, market, et cetera. But we have
indicated there sort of a base scenario that includes
the terms of viability and performance R&D. And we
have made provision for further down selection using
more quantitive metrics to show if the potential can
really be realized.
At some point we envision to hand off to
industry based on their reading of the markets.
That concludes my presentation.
DR. KRESS: Thank you. Questions anyone?
DR. POWERS: Yes, I have a question that
comes to mind when I see these plans for Generation IV
reactors. My good friends at the Nuclear Energy
Institute regularly provide me metrics on the
performance of the current generation of plants in a
variety of areas, including resources, safety and
economics. And they show excellent performance, just
outstanding performance in the last ten years going
along.
In all this roadmapping exercise, do you
carry along some representative of the current
generation plants as a comparison so you can see if
you're really going to accomplish anything with these
new plants?
MR. VERSLUIS: Well, it's a good question
because the initial screenings are really not much
more than comparing in a number of different areas
with the Generation III technology. So, they are
qualitative comparisons, and that's how we approach
it, is comparing it with the Generation III
technology.
DR. POWERS: See, now the Generation III
is like the --
MR. VERSLUIS: The fast light water
reactor.
DR. POWERS: The 600 or the 80+ or
something like that?
MR. VERSLUIS: Yes.
DR. POWERS: We don't have a whole lot of
performance and data on those Generation III plants
the way we do with the existing plants?
MR. VERSLUIS: We think at this point with
the amount of data that we have on the various
concepts, there is no need to be very, very precise
about these things. What the schedule, the last slide
really showed is that we need to do a certain amount
of viability research where we get a better handle on
how to measure, how we can measure the various
indicators before we can do a more sophisticated
screening.
DR. GARRICK: Rob, it might be important
to point out, too, that GRNS has put a lot of emphasis
on the total energy system concept, and that has kind
of evolved. When we first got together, that wasn't
so much an emphasis. And when you think about
performance indicators, you've also got to think about
the scope that we're addressing this time, namely the
total energy system.
So, it would seem that if we're going to
go in the direction of performance indicators that are
compatible with risk-informed performance based
regulatory practice, we'll be talking about probably
a different structure and at least a more range of
indicators that we've perhaps ever seen before. Is
that not correct?
MR. VERSLUIS: Yes. I thank you for
pointing that out. For example, the base case we're
comparing with, of course, has a once through fuel
cycle. We have various criteria that have to do with
the waste and use of fuel, but particular the waste
forms that can be achieved by other fuel cycles.
So, you're very right that we are not just
looking at the reactor, but the entire system from
soup to nuts, so to speak.
DR. APOSTOLAKIS: If we go to slide 3, you
had the word "excellence" under "safety and
reliability goals." What exactly does that mean?
That you don't want excellence on the other goals or
that this is something special here?
MR. VERSLUIS: Actually, it is something
special. And I would like almost to defer to Neil who
is going to be discussing those tomorrow. But I can
say that there is a strong feeling among the GRNS that
one of the important issues in improving the
technology and also making it safer is practices of
excellence in operations, maintenance, design. And as
such, they have made a specific goal with that title
and it translates into criteria and metrics having to
do with safety to the public during normal operations,
frequent occurrences all out -- throughout the fuel
cycle, not only the reactor but also the other fuel
cycle facilities. And so there's a number of metrics
that have been defined to implement this goal of
excellence.
MR. JOHNSON: Mr. Chairman, if I could add
to that response? I believe your question actually
ties very well into Dr. Powers' question regarding the
current operating fleet of reactors and the experience
and lessons learned from that, and how that's going to
feed into the process.
The goal of excellence truly is looking
at, you know, what are the best practices. You know,
what has led to the success in the current fleet of
operating reactors and making sure that the new
generation reactors, you know, meet or exceed that
level of operational and maintainability excellence.
So I think that is the intent of those goals.
DR. APOSTOLAKIS: Now when you say
reliability goals, I mean are they goals the way we
understand them, numerical goals for reliability? For
safety I understand it, but reliability?
MR. VERSLUIS: That's where we would like
to end up, but reliability you can't really put a
metric of reliability together until you know the
design pretty well.
DR. APOSTOLAKIS: Sure.
MR. VERSLUIS: And so early on we are
really looking at very general indicators that might
lead to reliability, but it's not -- as I remember
well, it's actually not a screen for potential
criteria. It doesn't come into play until later.
DR. APOSTOLAKIS: And a last comment, if
I may.
On the third column, "Economics Goals," it
says "risk to capital." That's a very interesting
idea. I mean, do you envision at some point in the
future that we will have a probablistic risk
assessment for a proposed design that in addition to
end states that involve various levels of damage to
the core, we'll also have other end states that refer
to economic losses? I mean, that would be a very
exciting thing to do, actually.
MR. VERSLUIS: Well, I don't know if we
need new methodologies along that probablistic risk
assessment line. But, yes, there are now ways of
assessing risk for a certain project and what we want
to indicate here is that nuclear energy systems when
investors look at them, the risk to their capital
should be comparable with other projects.
DR. APOSTOLAKIS: Which is intimately tied
to the second column, right, "Safety and Reliability
Goals"?
MR. VERSLUIS: Yes. Well, actually, many
of the other goals, of course, have an economics
impact. Definitely, yes.
DR. KRESS: I know you wanted to leave
something for Neil Todreas, but under that "Safety and
Reliability Goals" you have emergency response. Could
I read that as no emergency response?
MR. VERSLUIS: The goal is in fact to
eliminate the emergency response. And this may be a
good time to reiterate what Bill said. These are
goals that drive R&D programs. They are not
regulatory criteria. In fact, we take pains to point
out that it may not be possible to reach all these
goals, but we will be evaluating the concepts on how
well they get there on a scale from, you know, zero to
the goal; how close they get and across how many
goals.
MR. LEITCH: I'm trying to better
understand the level of effort that's going on. These
50 U.S. experts and 40 experts internationally, are
they involved full-time or only at times of these
meetings that you refer to? In other words, between
meetings what are they doing? Are they back home
working on this full-time or is this just part-time?
MR. VERSLUIS: We didn't mean anyone to be
working on it full-time, but they are expected to work
on these issues between meetings or the work wouldn't
get done.
DR. APOSTOLAKIS: It's pretty much like
the ACRS, I guess.
MR. VERSLUIS: Yes, right.
Roughly speaking we expect people to spend
some 20 percent of their time on the roadmap and in
the chairs, the co-chairs of these groups some more
time.
The international participants, again,
they're expected to do the same thing but they are
funded by their own organizations. Nevertheless,
there is a lot of work to be done here, which they all
recognize, and there is a real sense of wanting to do
this correctly. So, we are probably getting a little
more than we are paying for.
MR. LEITCH: And these individuals are
sponsored by their parent organization, either
industry or academia or labs? In other words, DOE's
responsibility is the oversight and management of this
program?
MR. VERSLUIS: For the U.S. participants
we contracted most of the individuals and our total
budget is $4« million for this year.
MR. LEITCH: Who do you see as the
customer of this activity?
MR. VERSLUIS: Well, the customer at this
point is DOE, because we are looking for guidance on
our R&D program in the long term. And we also are
looking for a well-reasoned, a well-organized plan
that allows us to discuss our needs with Congress and
with other agencies.
But ultimately, and this is one of the
reasons we have gotten the utilities -- I'm sorry, the
industry, owner operators and vendors involved very
early on, because we feel that they're ultimately the
customers for these efforts. And as I ended up my
talk, I said we need to be able to define a hand off
to industry at some point.
At this point I would say DOE is the
customer.
MR. LEITCH: Okay. Thank you.
DR. KRESS: With that, I think I'll stop
the questions and move on to the next speaker to keep
us on time. The next speaker is Mr. Thomas Miller.
MR. MILLER: Thank you. My name is Tom
Miller. I am in the Office of Technology and
International Cooperation. I'm responsible for the
near-term deployment working group of the Gen IV
roadmap effort. I'm also the project manager for NERI
and the INERI programs.
Very early on in the Gen IV roadmap effort
we realized that the effort in the near-term was going
to determine a lot of what happens out in the future
2020/2030 time frame. We didn't have a nuclear
component, a new nuclear component in the 2010, the
2020 time frame there probably wouldn't be something
beyond that. So we looked at what it was going to
take to have new nuclear plant deployment in the U.S.
by the year 2010. We picked that target date, and as
Bill said we're a little bit flexible on that date,
but that was our target date with the intention of
having new plant orders by 2005.
And the intention was to have not only
plant operational, but to see what it would take to
have multiple plants in operation by 2010. And by
that you can see some differences of how you may
approach things if you have multiple plants being
built.
The participants, and it's a multi-
industry oriented organization because of the near-
term effort, we have nuclear utilities; the major
utilities that are involved in the nuclear power
generation to date and those that are looking to the
future in nuclear power are participating.
The reactor vendors, national labs Argonne
and INEEL. We have academia through Penn State
University participating. Industry is also
participating through EPRI. And we have participation
of our NERAC committee on our panel.
Early on we identified two deliverables
that we felt were important. One was a working group
set of recommendations early that we called the near-
term actions for new plant deployment. That near-term
actions was intended to offer DOE some recommendations
based on the experience of the group itself without
any outside input, and it was intended to offer up
recommendations that could be used by the Energy
Policy Committee by the Vice President and DOE and the
lobbyists in helping support the department's budgets
in FY '02 and '03.
The longer term product of this group was
a near-term deployment roadmap that's targeted for
September of this year.
In the near-term actions the things that
came out of our group were recommendations involving
early site permit demonstration, combined
construction/operating license demonstration,
certification of the 1000+ MWe ALWR and confirmatory
testing and code validation of advanced reactors using
new technology. In effect, support code validation
and testing requirements that industry might not be
able to do for the gas reactors.
Supporting this effort we issued a request
for information to the general community with targeted
directions to specific groups. This RFI was issued in
April with a request to have material back in May,
with a one month turn around. As it turns out with
most RFIs, we're still having some information come
in.
The RFI was issued to the public through
the CBD. We gave a directed submittal to the members
of the NEI New Plant Task Force, directly to the
reactor vendors to facilitate getting a response back
in this one month time frame.
What we were asking for was to identify
the design specific generic institutional regulatory
barriers to new plant deployment, identify the gaps
associated with those. And in the RFI we broke it
down in various sections that looked at reactor
specific design issues and site related activities and
generic barriers.
We received responses from 12
organizations, and right now those are being reviewed
by the panel.
The RFI requested these designs, the
reactor designs to meet six specific criteria. And
these were intended to assure that they could meet the
2010 time frame, and it was intended to weed out other
designs that might have fallen more under the Gen IV
category rather than in this near-term deployment.
You all have these in the handout, and I
don't intend to read through them, but they were
focused on things dealing with: How the reactor
vendor planned to gain regulatory acceptance; did he
have an infrastructure that would support the
deployment of his design; what was his plan for
commercialization of the design; if he had a
particular utility that was interested in or not; if
not, how was he going to get it into the marketplace;
if there was work to be done and there was a need for
government level support, what is the cost-share, how
would they want to implement that and what are the
specific activities; they had to demonstrate economic
competitiveness to assure that they could compete in
the marketplace that was there within the next 10
years. And one of the most interesting was that they
had to rely on the existing fuel infrastructure.
Then we also addressed generic gaps. And
in the RFI we identified specific gaps that we, as a
group, knew already existed and asked the respondees
to rank those generic gaps and identify additional
ones. And in ranking those generic gaps, we also
asked them to identify what they believed were
solutions and appropriate levels of funding to reach
those solutions.
The responses we got in the design area
are on the slide. Typical that we expected from
Washington and GE responses. We got responsible on
gas reactors from Exelon/PBMR and General Atomics.
And one we had not expected, but showed up, was from
Framatome, the SW 1000.
At this point of time the group is
evaluating these designs. We're conducting a two
level review, one based on the six criteria and then
we're going to do a summary level design review of
each design and look at it from that perspective.
As expected, the generic gap responses
that came back pretty much matched what the working
group believed as necessary, but there were some
additional ones that were identified.
The three first ones involve parts of
demonstrating Part 52 licensing requirements.
Identification now shows up with the risk-informed
regulation for future design certification. And there
was a specifics identifying emergency planning and
plant security issues.
The last six were identified by
organizations that were not the reactor vendors or
your typical utility, but were other inputs we
received from the national laboratories and other
concerned nuclear industry groups, and they provide
some input for the group to consider.
Brought up earlier was the idea of
economic risk and risk assessment tool, and in fact
one of those was identified in our group.
As I want to state right now, we're on a
track to issue this report in September. The working
group is split off in teams right now. They're
diligently looking at these designs. Our next meeting
is the end of June, and we'll be having an assessment
by each of the design review teams given to the
working group, and in addition having the reactor
vendors come in and demonstrate to the working group
how they meet each one of these criteria.
And at this point in time I will conclude,
because there really is no further information I have
to give the committee.
Thank you.
DR. KRESS: Thank you.
Questions?
MR. WALLIS: I have a question. A lot of
your criteria is the credible plan for gaining
regulatory acceptance. Now, presently there's an
infrastructure for doing this. Response to things
like regulatory guides and standard review plans and
so on. In the absence of those from the NRC side, how
are you going to have a credible plan for gaining
acceptance?
MR. MILLER: This criteria was focused
towards those industry groups, utilities or vendors
that were going to come in with a new reactor design
and they had to show how they were going to try and
either meet Part 50, Part 52 and have a design that
was either accepted by the NRC or design certified and
ready to be built and operational by 2010.
From the experience we've seen with the
ALWR program, there is a timely process. We're asking
these vendors to come in and tell us how they had
planned to get through that process.
DR. POWERS: One of the frustrations, I
think, the agency has when it confronts new designs or
anything new with the regulations is that the
applications tend to come in piecemeal and whatnot.
There's some effort here to have more comprehensive,
better quality applications coming in?
MR. MILLER: We're not addressing that.
MR. LEITCH: One of the significant
activities that you list is design certification of a
1000 megawatt ALWR. Does that suggest a predeposition
to large reactors versus smaller modular designs?
MR. MILLER: No, that's not a
predeposition. That is one of the responses we got
back. We also got feedback from the GT-MHR from
General Atomics, which is a small design, the pebble
bed reactor design, which is a small design. There
was also a response back from Westinghouse for the AP
600. So, I don't see a predisposition to larger
plants.
DR. KRESS: If there are no more
questions, we'll follow on to the next item on the
agenda, which is Mr. Johnson. Mr., Mr. Versluis
again.
MR. VERSLUIS: Yes, that's me again. Yes.
Thank you.
I'm going to talk a little bit about the
Generation IV concepts that we have received. And I'm
going to take you on a whirlwind tour and scare you a
little, probably, in the regulatory area.
We felt that we needed to take a good look
at all concepts that could show promise, particularly
since we have built in a good period of R&D, we really
want to look at concepts with the proper amount of R&D
and can meet the goals or can advance very much
through the goals. And we started also with a request
for information in March. That request closed
sometime last month, a few things have still been
dribbling in. It was published in the Commerce
Business Daily, the Federal Register and was also
distributed very widely in the international
community.
We now have about a 100 responses, and I'm
going to be talking about the key features and the
statistics, and basically you're getting this hot from
the press without much digestion because we just got
them in. But I'll talk about grouping and then the
current activities.
This is the definition we've already gone
through, so next.
We received totally 94 concepts, but we
also had internally generated some of the concepts and
not all of these here were full energy concepts. So
we figure we have about a 100 total, and this shows
the breakdown by different coolant technologies, by
country and by organization type. And I will leave
this for you to pursue through at your convenience and
go to the next slide.
And this shows the variety of concepts
that were received. Going to the water group, and
these were reported by the water group, the variables
that they recognized in looking at these concepts are:
The coolant, light, heavy water; phase and conditions;
thermal, epi-thermal and fast spectrum; primary system
layout - there were a number of integral PWR types but
also conventional; the fuel cycle - uranium and
thorium once-through various recycles; the thermal
output and particularly also the maturity of concepts,
different.
Some of the crosscutting R&D issues that
they immediately identified for all of these are high
temperature materials, modular manufacturing
technologies, internal control rods and I&C issues.
That doesn't mean that these are the only ones, but
those jumped out when I first looked at them.
In the gas group the variables they
recognized are the reactor concepts and the
applications of fission heart. And within the reactor
concepts there were the gas turbine modular gas cooled
reactors, PBMRs, fluidized bed reactors and a gas
cooled fast reactor.
And there was a great variety of the
applications, the energy products for which the
fission heat could be used: Electricity generation,
both direct and indirect cycle; various process heat
applications as well as district heating and
desalination.
They recognized different fuel forms and
fuel cycles with uranium, thorium and uranium
plutonium. There are good plutonium burners, the gas
reactors, so there were a number of concepts that
focused on that.
And their generic R&D issues are: The
fuel fabrication quality assurance; fuel performance -
integrity and fission product retention; lifetime
temperature and irradiation behavior of graphite
structures; high temperature materials and equipment;
and, passive heat decay removal for fast-spectrum
concepts. Fast-spectrum concepts have less of a
thermal capacity because many of the lighter elements
have to be removed.
The liquid metal coolant, the variables
are: the size - large/monolithic designs, modular
designs, transportable designs - and targeted clients.
And I think I'm not sure what they meant through that,
but I think it means a transportable reactors that you
can take to less developed areas of the world with
less stable grids and less of an infrastructure.
Different coolants, sodium, lead and lead
alloys.
Fuel type, oxide, metal, nitride,
composites meaning the entire spectrum that you can
think of.
Primary system layout, look and pool.
BOP options and energy products also
there.
Energy conversion options that include
some pretty advanced things like Mtech, the thermal
electric conversion and other high technology MHD was
also in there. And fuel recycle technology, aqueous
and dry recycling.
Now in the non-classical concepts we may
have to ask assistance from Commissioner Diaz because
so many different things came in and he has a lot of
experience with some pretty way out designs.
The focus of this group is on adequately
defined concepts with significant potential, and the
variables there are: The cooling approach; the
coolant itself, molten salt, organic; the fuel phase,
solid, liquid, gas and vapor; electricity generation
technology conversion including a direct fission-
fragment energy conversion; alternative energy
products or services; and also the fuel cycle.
The crosscut issues that they identified
are: Modular deployable; hydrogen production and very
high temperature systems; advanced fuels and fuel
management techniques; and energy conversion systems,
especially non-Rankine.
Now, I'd like to say something about the
grouping, because that's really the first step of our
work is to look at this entire group and organize
them, and get them ready for the first screening.
All the TWGs, all the working groups have
taken a first cut at the grouping them into concept
sets that share a technology base and a design
approach. And rational for the grouping is, first of
all, the efficient division of the analysis effort,
but also the streamlined evaluation process and an
avoidance of premature down-selection at this point
when there's so little information available about
some of these concepts and we run the risk of throwing
out the baby with the bath water.
For the water group we found we have three
PWR loop type reactors. These are, in fact, the sets.
Three PWR loop reactors, a set of three. Integral
primary system PWRs, six. Integral BWRs, six.
Pressure tube reactors, three. High conversion cores,
11. Three supercritical water reactors and then 14
advanced fuel cycle concepts of various types, you can
read.
The gas group there were five pebble bed
modular reactor concepts. Five prismatic modular
reactor concepts. One very high temperature reactor
operating at ~15003øC. Five fast-spectrum reactor
concepts, and four others including fluidized bed and
moving ignition zone concepts.
The liquid metal group looked at four
major categories and concepts: Medium-to-large oxide-
fueled systems of which there were six; eight medium-
sized metal-fueled systems; eight medium-sized Pb/Pb-
Bi systems; and six small-sized Pb/Pb-Bi systems.
They're also examining three supporting
technology areas: oxide, metal and nitride fuels;
different coolants; and different fuel cycle
approaches.
And in the non-classical group, as you can
see, they were not real successful in creating a lot
of economy here with the grouping, but there are some.
There are two eutectic metallic fuel
types, four molten salt fuel concepts, a gas core
reactor, a molten salt coiled/solid fuel reactor, an
organic cooled reactor, a solid conduction/heat pipe
reactor and two fission product direct conversion
systems.
Okay. I hope this didn't scare you too
much.
The current activities now with the
concepts in the working groups is to analyze these
candidate concepts for performance potential relative
to the technology goals and to start working and
identifying the technology gaps.
And this fiscal year a report will be
prepared to describe these concepts and we have laid
out a format for that. We want all the concepts to be
described in a similar manner. The R&D needs will be
covered in that report. And the results of the
initial screening for potential evaluations.
And that's where we are.
DR. KRESS: Questions?
DR. SHACK: One of the things I noticed
this morning in the whole discussion of the Generation
IV thing was that the word "severe accident" never
appeared anywhere. Do you envision that as being a
technology need that will have to be addressed in the
R&D program?
MR. VERSLUIS: Yes. One of the goals, the
second safety and reliability goal has to do with core
damage. And then the third goal has to do with the
emergency response. So in both of these goals severe
accidents are an issue.
And the second goal will assume the
performance of a PRA. And the third goal will have to
involve all the severe accident that could lead to a
release off-site.
Does that answer your question?
DR. SHACK: I guess so. You know, I guess
my question is are you going to handle it by
essentially your PRA argument that core damage is so
unlikely that I don't have to address a severe
accident, per se? Or do you really envision a need,
for example, to determine source terms for some of
these reactor concepts?
MR. VERSLUIS: Well, for those concepts
that are selected that make it through the early
stages of the screening, there will have to be a
better description of source term and the various
scenarios leading to the source terms, yes. But early
on, as you can see by this wide variety of concepts,
we're going to have to use surrogates and indicators
with potential and severe accidents.
And we are looking at physics parameters,
at heat capacity at the typical things that you would
look at to determine whether or not it's likely to --
and what the passive severe accident would be.
DR. FORD: We've been told earlier on that
risk-informed regulation is going to be a part of your
strategy, and yet we're looking at a whole lot of new
systems here for which we have no experience at all in
terms of time dependent degradation. So as you're
going through your screening process, does the time
needed for R&D to resolve those questions, does that
enter into your timing, your decision making?
MR. VERSLUIS: Yes, it does. And
certainly we hope or we intend but in early on in
particular to focus on those issues where there's a
large amount of uncertainty and try to reduce that
uncertainty. That's how we will focus what we call
the viability R&D, so that we have a better idea of
what the potential is to really meet --
DR. FORD: And have you also taken into
current the question of manpower capable of doing that
research?
MR. VERSLUIS: Well, there will of course
be as part of the roadmap an estimate of required
manpower, resources and infrastructure. But we are
certainly aware that there is a lot of work needed
there and a lot of investment needs to be made. I
should probably let Bill Magwood talk to this issue,
because this is wider than just the Generation IV.
You want to say anything about that?
MR. MAGWOOD: Well, I think it's always
important to think between time and maybe the
distinction wasn't made as cleanly. But when Tom was
talking about the near-term deployment, we're aiming
for systems, and I think you can tell from the types
of technologies Rob was talking about, that on Tom's
side will be deployable before 2010. And then the
case that Rob was talking about, we're talking about
systems that will be deployable by 2030.
So, clearly once we make a selection of
the concepts that should be pursued, the roadmap will
lay out what the R&D programs should look like. And
that actually is a little -- to some degree. You know,
rather than simply saying we need to maintain a
healthy university system, we need to maintain a
healthy infrastructure to make sure that we'll be able
to develop advanced concepts, we'll be able to point
to the technology roadmap and say we can't do that
because the infrastructure doesn't look like the
following, we don't have the kinds of professionals
available.
One really good example in the United
States, and I think some of you are aware of this, is
that we're in pretty poor shape when it comes to
nuclear chemists. There just aren't very many left
and a lot of them are retiring. And the universities
aren't putting out any more nuclear chemists. So, you
know, as we get into some of these areas, especially
molten salt reactors and things like that, you know,
you're going to have to know that you have nuclear
chemists available to go off and do this research over
the next, you know, ten or 20 years.
So clearly the roadmap itself will become
a vehicle for us to get a better handle on the kinds
of requirements we need. Right now it's very
speculative, it's very high level, there aren't a lot
of specifics.
For example, NERAC has rolled out a long
term R&D plan to cover the wide area, but it doesn't
focus on specific concepts. This will do that.
So I think that there's time to respond to the
need.
But Rob was right, the much bigger issue
is support.
MR. WALLIS: When you were listing all
these concepts, it reminded me of the '50s and '60s
when there was a blooming of dozens of concepts,
rather like these ones and only two or three survived.
So, there's a sort of a redoing about this and I'm
trying to think about what is it that's going to make
a difference this time? Are there some breakthroughs
in technology or are there some changes in criteria,
or something which will make a difference this time
around?
MR. VERSLUIS: Well, I think you answered
your question partially yourself. There are indeed
new materials.
I also think that there has been an new
recognition among policymakers and the public that
we'd better start some planning for our energy future
and issues like sustainability, climate issues they
now play a much bigger role than they did 40 years ago
when we designed the first round of technologies.
But, yes, in fact when you look at the
technologies that have been submitted, many of them
are really not new. But it is time to look at them
with the eyes of today, or actually the eyes of mid-
century and the need for hydrogen production and the
need for clean water, and the need for other energy
products.
And in addition to that, of course, there
is the change in the market structure. There is
deregulation of the energy markets. There is the
internationalization of the vendors as well as the
owner operators.
So, really the environment for judging
these technologies has truly changed and it is worth
looking at them again.
DR. BONACA: Yes, going back to the
question of severe accidents, we call today severe
accidents those accidents which were not considered as
part of the original design basis of the plans. Are
you going to have designs that address all kind of
severe accidents, or something akin to what we had in
the past?
MR. VERSLUIS: There really is no doubt
among the roadmappers that the concepts that are
selected for the development as we get further into
the development and designs are becoming more
specified, that they have to be shown to be safe. I
mean, there's no way around that. And I'm not sure
how to answer your question other than, we're not
looking for cutting corners on safety. In fact, we
are hoping to make advances towards safety.
DR. BONACA: So essentially the design
basis of the plan will include consideration of severe
accidents?
MR. VERSLUIS: Yes.
DR. BONACA: What we call today severe
accidents?
DR. GARRICK: Rob, one of the things that
bothers me a little bit about this program is that if
I look at other programs like the Apollo program, the
atomic bomb program, et cetera, et cetera and ask what
was the real driver, where was the real cadre of
activity and creativity, and they of course had very
specific groups that constituted the think tank and
the nucleus of where everything kind of emanated from,
and I'm also thinking of the model that I think is a
very good one, the Lockheed Skunkworks. Here was a
small number of people that just generated immense
breakthroughs in terms of solving these kinds of
problems. I don't see that here.
I see a lot of review groups and I see a
lot of proposals from different organizations, but I
don't see -- and I don't know what this has to do with
regulatory challenge, but it might because they should
be part of that team, too. But I don't see the kind
of inspiration and drive that comes from a Von Brun
group that is putting together the rockets that are
going to get us to the moon. And yet the time
constant here is much longer than any of those
programs.
How is this all gelled together in terms
of a first rate group of people that we really look to
make it happen? Maybe Bill has to answer that one, I
don't know.
MR. VERSLUIS: Well, let me take a first
crack at it. I mean, I'm not sure I understand --
DR. GARRICK: I'm looking for the core
group.
MR. VERSLUIS: Right. What I wanted to do
at least is to point out that we're not only working
with the U.S. expertise, one of the things that Bill
has insisted in, and he's very right about that, is to
expand this into the world, and particularly into the
nuclear community with credible programs. The people
like the Japanese and the French that bring a lot of
resources and expertise to the table that we are just
kind of hanging on to.
So, I think that looking at taking a wider
view, there is a lot of resource or a lot of
capability available.
You were saying how can you focus it to --
DR. GARRICK: Right. Right. Where is the
Robert Oppenheimer group? Where's the Skunkworks
group? Where's the group that really is the driver?
MR. VERSLUIS: Well, they need money, and
this is -- and Bill can correct me if I'm not
representing this correctly, but this is a way to in
a fairly transparent manner make a strategic plan
where you start with all the concepts that you can
find and you narrow down to the most promising ones,
and then you focus your R&D on those.
So, perhaps the answer to your question is
we will get a focused effort, we will get a -- I don't
know if it's a small group, we hope it is, with enough
resources there to do the R&D that needs to be done.
But it will be focused and it will be done on a small
number of promising concepts.
MR. JOHNSON: John, if I could take a
shoot at answering your question. With all respect,
I'm not sure the analogy is an appropriate one because
those former federal programs were really single
objective oriented in terms of creating the bomb,
putting a man on the moon. What we're talking about
here is developing the enabling technologies and
getting those technologies to a point for a hand-off
to industry and industry to make a decision on whether
to take those technologies and commercialize them and
apply them. We're not advocating the United States
get into -- the federal government embark on a reactor
design deployment mission here.
DR. GARRICK: Yes, and I'm not even saying
it has to be the federal government. Because, you
know, the Skunkworks model was not necessarily a
government program. But, yes, go ahead.
MR. JOHNSON: Oh, I was finished, John.
DR. GARRICK: Okay.
DR. KRESS: Seeing no other questions,
let's move on to --
DR. APOSTOLAKIS: Just a minor comment.
DR. KRESS: Oh, okay. Comments,
questions.
DR. APOSTOLAKIS: I wonder whether for the
new concepts we should also rethink the terminology
that we've been using, which is of course water
reactor driven. There was a discussion on severe
accidents a few minutes ago, and I don't know that we
really want to carry over this terminology and other
similar stuff.
So, I know this is a detail at this point,
I mean you're thinking about much bigger things. But
it seems to me that's something to have in the back of
our minds, whether we want to continue using some of
the terminology of the past, especially since one of
the earlier goals that were stated was public
acceptance.
MR. VERSLUIS: I think it's something that
we should think about. We really haven't delved into
severe accidents much at this point, and it may well
be a good time to review the terms. Thank you.
DR. KRESS: Yes. That's a concept that
comes about because we have been used to design basis
accidents. And in order to separate the two, we'd
call them severe accidents. And it almost seems like
an arbitrary separation.
I don't know. My question is are you
going to try to fit -- well, I guess it may be
premature to ask this, but fit the licensing of this
into a design basis concept to fit it into the current
regulations or are you going to try to develop PRAs
that are sufficiently acceptable that you couldn't go
completely a risk-informed route? I guess that's my
question: Are we going to stick the design basis
concept?
DR. BONACA: The reason why I think is
important, however, is that we're still having to deal
with credibility of an accident. What is the most
limited credible accident. I mean when the current
design basis was defined, is because it was believed
that that was the most credible accident, the most
limiting ones. And so in good faith people put limit
to the -- and that yet is going to be challenging in
the course of --
DR. KRESS: There's a whole issue of how
do you go about defining design basis accidents.
DR. BONACA: Exactly.
DR. APOSTOLAKIS: Yes, it's very
interesting because the first paper on risk in 1967 by
Reg Farmer raised the same question; is it logical to
consider to have a distinction between credible and
incredible accidents.
DR. POWERS: And I think we have found the
limitations on the maximum credible accident kind of
concept. I was fairly excited when one of the
speakers said that the approach was that once they had
refined down their list of viable concepts down to a
more trackable few, that they would then look more
carefully at the source driven. It seems to me that's
where you'd look rather than the accident scenarios.
And I think this is a place where we need to come back
and revisit what we discussed in the past on frequency
consequence curves, which is actually coming back to
your man Farmer a long time ago that this may be a
much more valuable direction for us to take than the
classic level one, two, three kinds of approaches and
design basis accidents versus beyond design basis
accidents.
I mean, it's a much better continuum to
look at rather than these categorizations.
DR. APOSTOLAKIS: So you were excited
earlier, Dana, and now I'm excited.
DR. POWERS: Well, we actually find some
use for those probablistic things that you do, but
we'll get into some really good metallurgy stuff here
in a little bit.
DR. KRESS: With that, I'd like to move on
to the next speaker, please. Mr. Johnson, you're
next.
MR. JOHNSON: Yes. Thank you.
Good morning. My name is Shane Johnson,
and I'm the Associate Director for Technology and
International Cooperation for the Office of Nuclear
Energy at the Department of Energy. And what I'm
going to do briefly is just try to summarize what you
have heard over the last hour and 45 minutes from our
discussion this morning. And that is, where do we go
from here?
You've heard us talking about our
Generation IV activities, our Generation IV activities
being defined as both the near-term deployment
activities as well as our technology roadmap
development.
Before I embark on summarizing that, I
would just like to say to get back to a question that
the Chairman put early on relative to the regulatory
challenges. And that is we have recognized that in
both our near-term and our longer term activities that
there is an inherent regulatory facet to the programs.
For example, these two activities, both
our near-term deployment as well as our longer term
Generation IV technology roadmap, while we have got
them linked to under a single program, they are
somewhat as you've heard significantly different in
terms of their objectives and the time frames.
Our near-term deployment group really is
focused on identifying regulatory and institutional
barriers that exist in the United States for
deployment of new nuclear assets. And we have also
approached that in looking in terms of technologies
that require no or little further development. So our
near-term deployment activities are really focused at
the regulatory environment in the United States and
has very little in terms of a focus on technology
development.
Our Generation IV technology roadmap is
really just the opposite end of the spectrum of that,
and that is we're looking at in terms of the
Generation IV technologies is truly technology
development. Looking at technologies that are,
hopefully, stretching our current knowledge of reactor
design and operation. But simultaneous with that,
while we don't want to lose sight of regulatory
implications, again it's a technology development
program and the regulatory aspects of deploying that
technology are going to come, again, in the future.
The Department, as the Committee well
knows, is the federal government's technology agency
as opposed to the NRC, which is its regulatory body.
But in our activities we have been
working, in both the near-term activities and our
longer term Gen IV activities, with the agency. We
have been working with the Office of Research here,
Ashok Thadani and his staff, in both the near-term
deployment activities as well as our Generation IV
technology activities and having a representative from
the Office of Research involved especially with our
Generation IV International Forum. John Flack, one of
Ashok's staff here, has had the privilege of trotting
around the globe with us as we engage the
international community in the Generation IV
technology arena.
Quickly to summarize, first I'd just like
to address those things on the near-term deployment
activities, as Tom Miller went over earlier. And that
is our goal in our near-term activities is to complete
our near-term deployment report by September of this
year. The report will identify primarily generic
issues that the government could pursue in a cost
share cooperative basis with industry to establish an
environment that will enable industry to step out and
make informed decisions on the deployment of new
nuclear assets in the United States. Those issues as
it appears right now primarily are going to be related
to early site permitting, going through that untested
process, as well as the combined construction and
operating license process.
We are also working with the NRC in
helping them to get started in the development of
generic advance gas reactor regulatory framework,
because as everyone knows it's an area that needs some
work and there are organizations in the industry who
are coming forward and having those discussions with
NRC, so it is a responsibility of the federal
government to be prepared to address these technology
concerns. And we're glad to be working cooperatively
with the NRC in aiding them as they develop these
generic reactor technology regulatory framework.
With respect to our Generation IV
technology roadmap really our near-term actions, as
Robert Versluis has summarized, is to take the almost
100 concepts and to go through a systematic evaluation
of those concepts and identify those concepts which
are most promising which to the extent at which we are
able to make such an evaluation at this time, meet the
technology goals that have been established by our
nuclear energy research advisory committee as well as
our Generation IV International Forum. And after
identifying those most promising concepts, is to put
together the comprehensive research and development
plan that will, hopefully, lead to the development of
these technologies and bring them to a point at which
time in the future they can be handed off to industry
for further and eventual commercialization.
And with that, I believe our discussion on
the Generation IV activities is complete.
Mr. Chairman.
DR. KRESS: Questions for the speaker or
any of the previous speakers? I guess we must be
hungry. Ah, there's one. Please identify yourself.
MR. LYMAN: Ed Lyman again, Nuclear
Control Institute.
I just have to follow up from my earlier
question, because I think what we've just heard is a
list of activities which I don't think it's
appropriate for the government to be funding. These
are activities which are associated with providing a
regulatory climate or easing licensing advanced
reactors. And I think in today's context, that's a
cost that really should be born by the applicants.
Licensing is expensive, but that is part
of the package for trying to develop a new nuclear
reactor and market it. And so I think it raises real
questions whether DOE should be involved in trying to
facilitate or come up with ways of easing the site
permits and other regulatory activities.
I'm also concerned about DOE proposing a
licensing framework for reactors and then a way of
meeting those licensing criteria. I think there
really has to be a separation maintained between the
licensing standards and the actual applicant. Because
otherwise these criteria could be gerry-rigged to
justify or to facilitate the particular reactor you're
pushing.
MR. MAGWOOD: Again, Ed raises an
important point and I think it requires a little bit
of distinction drawn.
What we're doing, Ed, and for everyone
else who had concern about this, is we're focusing on
generic issues, and this is something that DOE has
done basically throughout history.
For example, in the case of gas reactors
there are some very generic issues related to the
implementation of gas reactor technology in the United
States whether it's a pebble bed or GT-MHR or
something else, you have to deal with, for example --
and this is something that we've had a lot of very
important discussions about. If in the case of a case
reactor you're relying very heavily on the quality of
the fuel, how does one go about thinking about fuel
manufacturing in concert with the design of a power
plant? You can't separate it as easily as you can in
the light water reactor. That's a very, very broad
generic technology issue.
And I think it's entirely appropriate for
DOE to be involved in that.
What we will not be involved in are the
specific -- and NRC, by the way I'll point this out,
NRC's Office of General Counsel has been very, very
diligent about keeping both NRC and DOE straight about
this issue.
We will not contribute to the specific
design related regulatory activities NRC will be
participating in with the vendors. There will be a
separate activity that will probably be coming on in
the next year or so. We expect that Exelon, or
whoever, will come to the NRC and will be obligated to
pay for those activities. We don't anticipate being
involved in that.
But the generic activities are things that
we think the government ought to be involved in and
should be involved in. And I'll be happy to talk with
you more about that later, but I think it's entirely
appropriate what we're doing as long as you stay on
this generic level. I think there has to be a
distinction.
DR. UHRIG: There's a number of rather
exotic materials involved in the various concepts that
have been talked about this morning. Is there any
consideration or any time being spent looking at the
availability of these? Even something as common as
hydrogen -- I mean helium, excuse me, there's a
limited amount of that unless you want to produce it
artificially. And I just wondered if this is an issue
that's going to be brought into the consideration?
MR. MAGWOOD: That's a really good
question, and something that I've actually started to
worry about myself. The answer to the question is no,
we haven't done this stage. And the reason we haven't
is because we haven't reached this 2002 target of
narrowing down the number of options. When we know
what concepts we're really going to spend our energies
on, we're going to really have to deal with those
materials issues.
And I can't talk too much about this, but
we are expecting in the next few weeks to really
strengthen our materials activities within the DOE
infrastructure and start to have more focus on these
issues. Because I think they're too disperse right
now. We need to really focus our energies there, and
we're going to be doing that very soon. We'll make
some announcements about that.
But your question is really good one, and
we're worried about it but it's too early for us to
really go a whole lot further.
MR. UHRIG: I guess my point was that this
could e an issue that would eliminate an otherwise
attractive concept.
MR. MAGWOOD: Well, that's a really good
point. I mean, for example if we don't have enough
helium for the helium cooled reactors --
MR. UHRIG: I think that's not a major
issue, but it's something that certainly should be
looked at.
MR. MAGWOOD: Yes. I think it's something
that will have to be looked at in concert with the
evaluations that NERAC is doing. I mean, I'm not
aware of any major materials limitations. If someone
has some exotic material that, you know, it's just not
available, I expect that will become one of the
technology issues. And if it is such an issue that
you simply can't rely on being able to build numbers
of plants, I would expect it would be kicked out on
that basis. So, I think that's something we ought to
take back to the group and make sure they're conscious
of that. So, I appreciate that thought.
But so far I've never heard of any exotic
material that would simply eliminate a concept being
considered.
MR. UHRIG: Thank you.
MR. FEINROTH: My name is Herbert
Feinroth.
As I listened to some questions from the
ACRS and also the DOE presentation I sort of see a
different -- there's a gap between what the DOE is
focusing on, which is the entire fuel cycle not just
the reactor and their interest is in these goals that
they've described to achieve safety and public health
for the entire fuel cycle. Whereas the ACRS is
focused, I believe, in the past and I think still on
reactors only. And it seems to me that this is more
of an observation than a question, because I don't the
question has an answer that the regulators need to
look at the whole fuel cycle as well and not just the
reactor as they provide advice or input to the DOE in
their section process.
The gentleman asked about the source term.
Well, the source term of importance to public health
is not just what's in the reactor, but what gets
transported, but gets recycled, what gets sent to a
repository. So I think the context that DOE is
looking at this is correct. And I think the regulatory
agency needs to figure out how to address the
imbalance, the public health from the different parts
of the fuel cycle. And my concern is the ACRS just
looks at the reactor.
I don't know if anybody has a response to
that, but I think that's an issue that needs to be
addressed by the regulatory agency.
DR. POWERS: Well, we'll comment quickly
that we do have the Chairman of the Advisory Committee
on Nuclear Waste look at the waste portion of it. And
that ACRS does also look at the fuel fabrication part
of the problem as well, though we probably haven't
focused on it very much in the discussion today
because the fuel cycle has only been mentioned briefly
here as being changed.
DR. KRESS: I think the questioner had a
good point. I did want to point out that the ACNW
also focuses on regulations related to sensitive
materials and materials applications.
Perhaps ACRS could do a little more on the
fuel cycle parts, but our conception, at least our
feeling is, the real risk part of the thing is in the
reactor or perhaps in the fuel fabrication.
George, did you want to say anything?
DR. APOSTOLAKIS: And we also have joint
committees with the ACNW when the issues warrant it.
But it's certainly a good thought.
MR. CLEMENTS: Yes, I'm Tom Clements with
the Nuclear Control Institute.
I was a little confused during the DOE
presentation about the relationship between the
roadmap and the review you're doing and what's
happening with the Exelon pebble bed reactor. From
what I hear, depending on what happens in South
Africa, they plan to start construction in 2004 and
have a reactor operating in this country 2006. It
sounds to me like you're behind the curve on what's
happening with that reactor. Are you going to ask
them to slow down their decision process in pursuing
this with NRC? You're behind the curve on what
they're doing here on the ground with the NRC or do
you assume that you're going to include this reactor
in your roadmap? I'm just confused about the
relationship between what you're doing and the pebble
bed.
MR. MAGWOOD: The pebble bed, that's a
good question because I saw something and I thought
someone would ask that question.
The pebble bed reactor that Rob spoke to,
he spoke to a class of PBMRs, those are not
necessarily , in fact may not really all be the
reactor that Exelon is interested in and is now being
discussed in South Africa. That specific design is
being discussed as part of the near-term deployment
activities. And, as I've mentioned, those activities
are largely complete and will be final -- scheduled to
be final through the NERAC process in September, and
include largely institutional issues that are being
raised by NERAC that are fully in concert with the
schedule that PBMR corporation is on.
And, in fact, there are representatives of
Exelon on some of the working groups that are
providing information about the schedule and trying to
keep everything in concert.
So that PBMR is slated for near-term
deployment as opposed to being in the longer term
Generation IV activities. And that's simply because of
the fact that it's of near-term interest to a utility
and, therefore, it's appropriate that we look at it as
something to be deployed by 2010. And whether it
actually gets deployed by 2010 or not is up to Exelon
and others.
MR. QUINN: It's Ted Quinn.
Bill or Shane, we've read the Vice
President's report -- or the President's report and it
addresses investment in new technologies for
renewables, for coal for example, and some of the 105
recommendations address advance nuclear. Can you
advise in FY '02 and beyond how those recommendations
will come into DOE planning?
MR. MAGWOOD: No. To expand on no, nein.
Let me just say that, obviously, certainly and our
international partners are all very pleased with the
outcomes that were in the Vice President's review and
have every hope that eventually there'll be more
resources devoted to nuclear research and development
by the government. Certainly there would have to be
to do any of the things that we've talked about today.
What will happen in specific fiscal years,
2002 in particular, I simply don't have an answer for
you. I think that as the government continues digest
results of the review, we'll begin to talk more in
terms of what do we have to do to actually implement
those things, and those discussions have already
started moving.
But I wouldn't expect to hear any specific
implementation announcements other than what you may
have already heard from the Secretary. I think he
made some announcements recently about specific things
in non-nuclear aspects. But on the nuclear aspects
it's going to take a while to adjust it, move on it
and to formulate those implementation activities.
So I would expect that over the course of
the next few months those would start to come out.
DR. KRESS: With that, I'd like to thank
all of our speakers this morning for getting us off to
an excellent start. And remind everyone that we have
some good things this afternoon on specific designs
and some of the regulatory activities that are
underway to get ready for this, and some very
interesting panel discussions on regulatory
challenges.
With that, I'll recess for lunch and ask
people to be back at 1:00 please.
(Whereupon, at 12:07 p.m. the Subcommittee
was recessed, to reconvene at 1:00 p.m.)
A-F-T-E-R-N-O-O-N S-E-S-S-I-O-N
(1:01 p.m.)
DR. KRESS: Let's get started again,
please, for the afternoon part of this Subcommittee on
Advanced Reactors.
Earlier when I mentioned into the record
the ACRS members present, I was remiss in not pointing
out that Dr. Peter Ford is also here as an ACRS
member. So I apologize and get that read into the
record.
We are the point now where we're going to
talk about Gen IV design concepts, and we're starting
out with representatives from Exelon. As I mentioned
earlier, I don't have introductory material for
people, so you have to introduce yourself. And so
I'll just turn it over to you.
MR. LEITCH: Dr. Kress, I'd like to
declare that I have an organizational conflict, so
I'll recuse myself from the discussions of the pebble
bed.
DR. KRESS: Yes. Yes. We need to do that
because this is a Subcommittee meeting. Thank you
very much.
MR. SPROAT: Mr. Chairman and fellow
members of the ACRS, thank you for your invitation
today for Exelon and PBMR to come to give you a
briefing on the pebble bed modular reactor project
currently underway in South Africa.
My name is Ward Sproat. I'm the Vice
President of Exelon Generation in charge of
international projects, and I represent Exelon's
interests on the board of directors of PBMR, the joint
venture in South Africa.
Today's presentation is going to cover
three areas. One is I'm going to give you a brief
introduction and project update about where the
project stands.
Second, I'm going to introduce by co-
presenter, Dr. Johan Slabber from PBMR Pty in South
Africa, who arrived yesterday afternoon with several
of his colleagues, and he'll be talking about the
design philosophy of the PBMR.
And then finally, I'm going to come back
on and talk about the licensing issues that we see
trying to license the PBMR here in the U.S.
Well, I'll keep talking and we'll move
forward.
Let me just start off with giving you a
project overview about where the PBMR project stands.
There's been a lot in the press, obviously, about the
project some of which is correct, some of which is not
correct. And I want to make sure that the ACRS has a
full understanding of where the project and where the
Exelon stands regarding this technology.
The project is completing the preliminary
design stages in South Africa at this point in time.
And we are currently finalizing what is called the
detailed feasibility report. That report is being
generated by the project team in South Africa as well
as several contractors, as well as with us, the
members of the joint venture. And that feasibility
report will be completed sometime probably this
summer, at which time then all of the investors in the
joint venture will make their own individual decisions
regarding whether or not to proceed to the next phase
of the project.
The next phase of the project is to move
forward with the detail design and the construction of
a demonstration PBMR in Republic of South Africa near
Capetown on the site of the Kuberg Nuclear Station.
The other investors in the project at this
stage of the game, besides ourselves, are BNFL,
British Nuclear Fuels Limited, SCOM, which is the
electric utility in South Africa and the Industrial
Development Corporation of South Africa.
So each of those investors will, in turn,
make their own decisions about whether or not to
proceed with the project, as well as the South African
government needs to make their decision regarding
whether or not they'll approve the instruction and
operation of the plant in South Africa.
Assuming all of those decisions are
favorable, which is not an assured outcome by any
stretch of the imagination at this stage of the game,
but assuming they are favorable, then construction
would start on that demonstration PBMR in South Africa
probably in late 2002 and would then take
approximately 36 months to complete construction with
then a one year start up test program in South Africa.
That's the program in South Africa. As
far as Exelon's decision making process and Exelon's
involvement, clearly we are pointing to make a
decision as to whether or not to continue to proceed
as a member of the joint venture in South Africa by
the end of this year. We'll make that decision
primarily based on economics; do we think that Exelon
can make money operating these reactors in a
deregulated electric utility market in the U.S. And
if so, then obviously we would have to require board
of director approval to proceed that way, but it would
be our intent to try and make a decision on whether or
not to proceed with the joint venture in South Africa
by the end of the year.
We probably also make a decision sometime
in that time frame, whether or not it's the end of the
year or early next year, to begin the licensing
process in this country for the first set of PBMRs
here in the U.S. And I'll talk a little bit later
when I come back on about what some of the obstacles
and challenges would be if we decide to move forward
with that. But that decision, I think, would also be
made sometime around the end of the year, nearly next
year as to whether or not to begin the actual
licensing process for the PBMR.
So, with that that's the current state of
both the project in South Africa and Exelon's
involvement in the project.
With that, I'd like to introduce Dr. Johan
Slabber, who arrived yesterday from the Republic of
South Africa along with several of his colleagues.
Hopefully, we have the right people here to answer
some of your questions as we go through this. And
I'll let Dr. Slabber introduce himself and explain his
background.
DR. SLABBER: Thanks, Ward.
Mr. Chairman, ladies and gentlemen. This
is a very nice privilege for me to be able to speak to
you. And I would like to give you some preliminary
information, and then go deeper into the design and
the important things regarding the safety as well
licenseability status.
Something about myself. My name is S-L-A,
although it is pronounced in South Africa as Slabber.
In America, if you pronounce it it sounds like
Slobber, and that I don't mind. You can say Slabber
or Slobber or Slabber.
Something about my background. I was
graduated as an electrical engineer with a physics
degree. And I did my Ph.D in mechanical engineering,
but between those two times, graduations, I spent some
nice years in Oak Ridge and I was fortunate to be able
to have attended the last -- in the U.S. So I am
really indebted to the U.S. for really wetting my
appetite for nuclear technology.
I also spent a short time, brief time, at
IAEA in safeguards. So in the matter of nuclear
nonproliferation, I am also in a position to highlight
to you the attributes regarding that aspect of our
plan.
The design actually started evolving when
I was an employee. I was General Manager Reactor
Technology at the South African Atomic Energy
Corporation. But at that stage the Board of Directors
said the climate is wrong, the money isn't there, so
please let's not look at something although it might
be very promising. So that was the point when I
departed the Atomic Energy Corporation to a systems
engineering company who still today is involved in the
project.
This, what I'm going to present to you,
was actually developed from the initial concept of a
direct cycle turbine generating electricity.
What we have as the philosophy and we,
right from the outset, have set as goals inherent
safety features employing passive means. It must be
modular in size because in South Africa we've got a
relatively small grid, but we want high efficiency.
And the possibility to eventually supply fresh water
for South Africa, it's a semi-desert country. So in
25 years we might run out of water, so that was the
focus for the first initial design.
And you will see on the screen there the
three bullets which are actually some of the
cornerstones of our initial ideas.
Employ passive and active engineered
features, but I would like to qualify this. Because
active might sound funny in this context. Active
there should be seen in the context of keeping the
facility, the reactor, operating within the normal
boundaries. In other words, supply cooling, supply
ventilation, et cetera. But the passive is to keep it
within the limitations which does not lead to
radiation release.
The second bullet is rather saying what it
is, just that you can mitigate but that you do not
have cliff edge effects like suddenly you've got time
built into your system.
And then the third bullet actually
supports that, reduce dependence on operator actions.
Can I have the next slide? This is,
unfortunately, you must see this drawing as it stands
at the moment drawn on unigraphics and modeled. But
just to show you the width is 25 meters. The length
is 50 and the height is 50 and 25 is below grade. But
what I would like you to concentrate at this stage on,
and it will become clear when we evolve from this,
that we have the reactive vessel sitting in an area --
and it's not very clear here -- which is we call the
reactive cavity. And we've got the power conversion
sitting in a volume called the PCU area and this total
strengthened section around the reactor and the PCU we
call the citadel which, in fact, is containing, acting
as a containment around all those high pressure
radioactive components. But I'll come back to that
later.
Can I have the next slide, please? This
is the complete stuff taken away. What we have here
is the reactor vessel of 20 meters high and 6.8 meters
diameter. And we have the PCU, and I think I must just
explain slightly the workings.
This was the initial concept of changing
from a single-shaft turbo generator to a multi-shaft
turbo generator employing a high-pressure turbine,
turbo compressor, a low-pressure turbo compressor, a
turbo generator.
And in the reactor cavity, which we have
the reactor cavity cooling system and then below grade
we have the spent fuel tanks which can house --
contain the fuel for 40 years of operation, 35
effective years of operations.
It is also designed to store the fuel for
another 40 years during the formal decommissioning
phase.
The fresh fuel is in the fresh fuel
building, and that area we've got the so-called helium
inventory control system which employs -- which uses
the helium to increase the thermal hydraulic power
taken up by the gas in the reactor. And due to
coupling of the heat processor co-efficient and the
negative temperature co-efficient, the neutronics is
just about following the request for semi-hydraulic
power.
And then we've got the fuel handling
system, which is loading spherical fuel into an
angular core in the reactor and graphite spheres into
the central and a central reflector. So the core
itself consists of an angular pebble bed core with a
central column of graphite spheres. And this was
necessitated because no control rods -- the design
objective was not to have control rods in the core
itself, but to have a system where the reactor physics
of the core pushes out the flux towards the reflector
region for reactivity coupling.
So we've got the fuel handling system,
we've got fueling tubes as well as graphite tubes.
And we've got -- and we've got some separation of fuel
and graphite at the bottom. So this is the PCU. This
is the spent fuel, the fresh fuel and the helium
control system and the reactor cavity cooling system.
Just at this point we are also taking note
of the proliferation resistant aspects that needs to
be built in a facility like this. So the reactor
safety design principles is actually highlighted in
these three bullets.
An objective of the design, to start off
with, was to focus the design around existing proven
German spherical fuel fabrication and testing
technology. That was a go, that was a given. No
deviation from that.
And then in the design apart from the
microsphere providing the primary barrier, multiple
fission product barriers to the environment, to the
public outside. And this is not really a safety
issue, but we put it under these, and I highlighted it
in the previous slide.
Can everybody hear me? Okay.
The fuel itself is a 6 centimeter diameter
graphite sphere with containing in the fueled region,
which is 50 millimeter diameter, 15,000 microspheres
of -- it's got a core of UO2, it's got a porous region
around the microsphere which acts as a fission product
buffer, something like the buffer region in a LWR.
Then we've got three layers, pyrolytic carbon, high
density pyrolytic. The silicon carbide and then other
layer of pyrolytic carbon. And the diameter is just
under one millimeter.
So 15,000 of these in there and in there
the enrichment is 8.1 percent for the equilibrium core
and 4.9 percent for the burning core. And the amount
of material in that little ball is 9 grams heavy
metal.
And around the 50 millimeter diameter
sphere we've got a five millimeter unfueled section to
take care of abrasion and while this is moving through
the core so that you don't expose and allow
microspheres to come out.
The first bullet, next slide, to assure
fuel integrity. So, as I said the baseline as far as
proven technology German fuel and we have been given
the opportunity to access and purchase into the total
German database which they have developed for their
high temperature reactors. And it's been in the
process -- for South Africa. And we are actually
planning, and I'll come back to that a little bit
later, to replicate critical experiments and
qualification experiments and tests that were done in
the German program.
The next sub-bullet is because it's an
onload refueling system, you've got to good control
over excess reactivity added to the reactor core under
various conditions, and also to ensure under all
conditions normal operation as well as upset events
you assure removal, heat removal from the fuel by
means of passive means.
And prevention of chemical attack, which
is one of the events defined as one of the licensing
based events, and prevent excess of burnoff.
Now, in the development project we had to
structure the project very definite according to
certain rules. And for that we have developed the so-
called integrated design process in South Africa.
It's a PBMR integrated design process which embodies,
and we call it the PIDP, the upfront evaluation of any
structure system or component, SSC, in its role to
mitigate or to cause events leading to the release of
radioactivity. And those components are then
evaluated and classified according to a scheme which
is in line with our national nuclear regulator, the
NMR, prescriptions of failure frequencies versus
consequences.
And we have the three regimes that we are
using in the development of this facility. Events
having a frequency higher than 10 to the minus 2, in
other words one in a 100 years, we call the
anticipated operational occurrences.
And the events lying between -- into the
minus 2 and into the minus 6 is the licensing base
events. And then the occurrences with a lower
frequency than ten to the minus 6, those are the
extreme events or the unlikely event.
So what do we do to design a facility in
these regimes? The two, the first ones, the ten minus
two and -- plus ten minus 2 and between 10 minus 2 and
10 minus 6 we design for all those events. Below 10
minus 6 we analyze for and see what the consequence
are.
DR. APOSTOLAKIS: I have a question. I
don't understand what you mean by event. Do you mean
a sequence or do you mean what we call initiating
events?
DR. SLABBER: Yes. Yes. I was explaining
the integrated design process, so I interrupted myself
just to say what we're focusing at. But it is a
sequence. It is initiating event that can lead to a
sequence, that can lead to a --
DR. APOSTOLAKIS: So the 10 to the minus
2 refers to this initiator or the whole sequence?
DR. SLABBER: It's the initiator.
DR. APOSTOLAKIS: Now, given that the
concept of an event is not really well defined --
DR. SLABBER: Yes.
DR. APOSTOLAKIS: -- I can imagine an
event that has a frequency of 10 to the minus 4,
therefore I have to design for it, as you said, but
then I can break it up into a 100 little pieces each
one having a frequency of 10 to the minus 6, so now I
don't have to design for it. So, how do you avoid
this kind of -- I'm sure you don't it that way.
DR. SLABBER: Oh, no, we don't do it. But
we're looking at the logic also. In other words,
there are some enveloping frequencies which is also
the initiator plus the consequence, the total chain in
looking at all the events in between.
I wouldn't be able to completely reply to
your question because it's in the process of being
done at the moment, but a similar philosophy.
DR. APOSTOLAKIS: But it seems to me when
you have to go with the cumulative frequency at some
point?
DR. SLABBER: Yes.
DR. APOSTOLAKIS: Because just where you
consider in sequences, you know, this is not a well
defined concept.
DR. SLABBER: But in any case, thank you
for that comment.
DR. KRESS: As you will notice, we've
departed from our usual procedure and we'll allow
questions that interrupt the speakers. It's just the
ACRS can't seem to avoid -- control himself.
DR. APOSTOLAKIS: It was pain from this
morning.
DR. SLABBER: In any case, after we have
now identified these events, we can for that specific
SSC identify a preliminary classification. And then
for that specific SSC, we also classify the various
loads that it will achieve during its operational and
upset lifetime, and we develop a loading catalog. And
using the classification which drives the quality
assurance requirements as well as the loading catalog
and the codes and standards to which the SSC will have
to be developed and designed, we call that suite of
documents; the design rules for that specific SSC.
The QA requirements, the loading catalog,
the classification and the codes and standards, and
maybe some other additional things which must -- could
come into play like safeguard issues, et cetera. And
those are the suite of documents which are the design
rules. And then from there, there might be some
situations to improve the SSC design, so we can go
back to square one. Typically if the failure
frequency is too high.
So in the total development of the reactor
we have given priority to looking at the fuel first of
all. Next slide.
So we look at the fuel quality here and
the fuel design which we have chosen has been proven
internationally. And another feature that we also
embody in the design is that we do not want to develop
new material. We will be sticking at qualified
materials for all the structure systems and
components. This is one component which we have
decided we will, as far as practically possible and I
agree there will be a question that how do you prove
equivalence on such an important issue. This will be
done by laboratory tests, PBMR specific tests and
irradiation tests, as well as maybe taking part in an
international irradiation program.
And this is actually what is said here in
this sub-bullet. The fuel qualification program will
follow and the fuel performance testing program and
the fuel fabrication quality assurance program which
is still at the moment already starting to be based.
DR. KRESS: The performance testing
program.
DR. SLABBER: Yes?
DR. KRESS: Excuse me for interrupting.
Is that under irradiation conditions?
DR. SLABBER: Yes. Yes.
DR. KRESS: So you do this in a reactor?
DR. SLABBER: In a reactor. We will do it
stepwise and it will be going beyond the design basis
burnup of 80 megawatts, which is presently the design
target. But it will be irradiated beyond that.
DR. KRESS: Did you say there were 15,000
of these pellets in the --
DR. SLABBER: 15,000 microspheres in one-
sixth centimeter fuel sphere.
DR. KRESS: And how many of those
centimeter --
DR. SLABBER: Pardon?
DR. KRESS: How many of those 6 centimeter
spheres are in the core?
DR. SLABBER: 330,000. So there's a total
of 4.8 to the nine small pressure boundaries, primary
pressure boundaries in the core.
Then in the facility, in the reactor there
will be an operational fuel integrity assurance
surveillance program which will monitor operational
release in the primary coolant and to compare it with
predicted value.
Next slide, please. One of the other
bullets which we've seen is the first one was fuel
quality and control of excess reactivity. The reactor
is designed to be load following, and to be able to do
load following we will use the inventory, called
helium inventory control system to pressurize the
helium in the primary circuit so that your heat pickup
in the core and the heat deposition in the bell
conversion unit is in-phase.
Now, to enable you to load follow one
needs to also to some extent -- Xenon buildup fission
products developed or Xenon developed during the
operational cycle. If you reduce your neutronic
power, the poison increase. So you've got to cater
for during load following operations for a certain
amount of reactivity that could be added by means of
the control rods. And we have limited that amount to
1.3 delta k effective. In other words, 1.3 niles and
this was chosen so that in the event of a stepping out
of a control rod without anything checking it, you can
add in a random fashion 1.3 delta k to the reactivity.
And this is a value of power that will limit you
inherently to a temperature, a maximum fuel
temperature below the maximum defined limit. I'll
come back to that.
We have also provided a measure to design
the system so that for all credible pressurization
events and reactivity events, if there are anything
which will raise the power suddenly, like a control
rod injection, the core geometry is always maintained,
even in a depressurization event where you could have
for a short time a pressure differential across the
core barrel.
The core is also, although it's tall it's
quite a long core. 8.5 meters high and 3.7 meters
diameter with a central column. Although it's tall,
it's still within the window which precludes Xenon
oscillations. In other words, a critical area at the
top uncontrolled and a subcritical area and swinging
of the flux. So the geometry precludes Xenon
oscillations.
And then due to the nature of the reactor
physics of the core, we've got a very high negative
temperature coefficient of reactivity. It's minus 4.5
times 10 to the minus 5, delta k over 33 centimeters.
And then we are designing an inherently
safe critically safe spent and used fuel tank.
Next slide. The material properties in
the core at end of life, and this is now talking about
thermal volatility and emissivity is all assumed to be
at the risk point and in a static condition with no
forced cooling. These material properties are
sufficient that the heat can be taken away from the
core into the outer side where it's taken away by this
passive heat sink provided by the reactor cavity
cooling system for an extended period.
The reactor cavity and its structures will
maintain its geometry. In other words, during a safe
shutdown earthquake, the reactor vessel will stay in
tact. It will stay or so be cooled. The reactor
cavity cooling system will still function. And this
goes for that third bullet there, the reactor cavity
including its structures will maintain geometry during
all credible events.
DR. KRESS: Does this heat removal depend
on having the helium in place pressure, or how does it
work --
DR. SLABBER: Can I explain the reactor
cavity cooling system?
DR. KRESS: Oh, sure.
DR. SLABBER: Yes. The reactor cavity
cooling system consists of three independent cooling
tanks. The ultimate heat sink is the C or air coolers
on the roof of the reactor for all three tanks. It
consists of two loops each. In other words, the
primary coolant flows through a heat exchanger which
then dumps its heat into the ultimate heat sink. So
there's an intermediate loop.
The cooling system consists of three tanks
of 50 percent in the cavity surrounding the reactor
vessel. Each tank is 60 centimeters diameter and
covers the total length of the reactor core plus an
area about 2 meters, 2« meters above the reactor
vessel.
The sequence of events could be seen now
during a loss of cooling event in that if for instance
something goes wrong in the primary cooling, because
primary cooling is done by means of the primary -- the
conversion unit. The turbo compressor is running
because it's a break in cycle, it's in a bootstrap
operation; they must be running to circulate. We've
got a -- what we call a starter blower system which
must bootstrap the breaking cycle to start off with.
So if something should go wrong and we
should lose this cooling loop in the primary circuit
to cool the core, because heat rejection is done in
the intercooler and precooler at the turbo compressor;
If that heat rejection mode is lost, then we've got a
core conditioning system which can run parallel to
that. And that is forced convection. That's active
component. But should that all fall away, then the
reactor cavity cooling system will be capable of
handling the decay heat coming from the core exactly
after shutdown, in other words 1.3 megawatts of heat
which could be dissipated to the reactor cavity
cooling system.
The reactor cavity cooling system has got
a few layers. It's an active system consisting of
these loops, the primary loop, the secondly loop which
is backed up with the cooler on the roof, and then if
that fails, then we go into a boil-off mode and the
tanks will boil-off if it's not being replenished by
means of operator action. After a couple of days,
even, it will boil-off in something like four days.
We believe that operator intervention will
take place in that time. However, if that even fails
the concrete structures are sufficient to eventually
dissipate. Obviously, in such instances, the reactor
-- the concrete will be heated up to a value which we
are still determining at the moment and we're
engineering some methods, but we believe that we will
not damage the concrete unnecessarily.
Does that answer your question?
DR. GARRICK: Can I go back and ask a
question?
DR. SLABBER: Yes.
DR. GARRICK: Out of curiosity, on the
Xenon oscillation issue. I can see with this annular
design where you would have good neutron coupling in
the radial direction.
DR. SLABBER: Yes.
DR. GARRICK: But it is not so obvious in
the axial direction.
DR. SLABBER: We have looked in it because
for Xenon oscillations there is a reactor height which
takes you out of the safe region of oscillation.
DR. GARRICK: Yes.
DR. SLABBER: And we are still within that
limit.
DR. GARRICK: Okay. But there is a limit?
DR. SLABBER: There is a limit, yes.
DR. GARRICK: Yes, okay.
DR. SLABBER: Any more questions?
Next slide. Skip that one. I'll come
back to that.
In the German program, the licensing was
completed for the HDR model and Xenon's developed a
curve which they used to convince the regulators that
the reactor is safe from a release point of view, and
they generated this curve, and I must explain to you
because this curve you might see also in our safety
analysis report.
We do not, and I stress do not intend to
just follow this slavishly. And I would like to
explain this and, please, we must take note of the
importance of this. It is so important in Germany
that they have coined the word "the holy curve." And
they didn't want to deviate from this at all.
Now, what we've got on this axis, we've
got the failure fraction of practical and we've got
temperature here. And then we've got three lines
representing beginning of life, fresh fuel. We've got
a life cycle and end of life.
What they've done to develop this curve,
they took 212 microspheres to get good statistics and
they did, on fresh fuel, they did a burn leech test.
In other words, as code fuel freshly produced they
just measured the unclad uranium friction by means of
a leeching test to see which of these microspheres are
cracked. And they found it to be 6 times 10 to the
minus 4. That is a very important baseline for them.
That is why, yes -- to the minus 5.
What they've done is that they took that
and then they irradiated all those 212 out of the same
batch, although it was the same batch, they took 212,
they took another batch 212; they irradiated it and
they didn't find any failed particles, zero. So they
were faced with a dilemma how to now extrapolate from
that result what is the end of life of failure
fraction. And what they then did, they applied
Poisson statistics for zero failures at the 95 percent
confidence. And they found that to be 2 times 10 to
the minus 4. And then they slapped on that some
conservatisms and they added that to the original 6
times to the minus 5 and they came up with that 2.6
times 10 to the minus 4.
And then what they did, they wanted to do
the same at 1600. They assumed that those values
stayed constant, because from a methodological -- the
graphical consideration is no reason for
disintegration of the cladding between those two
values. They extrapolated the same values and they
took a sample. And this is where we will be deviating
from their approach. They took only a sample of
65,000 and because of the statistics and they couldn't
find any broken particles after heating it up, so they
just used zero failure statistics and that pulled up
because of the uncertainty, the failure fraction to
that high values.
We in PBMR are planning to replicate this,
but we will be keeping the sample sizes constant. And
we expect that our fuel failure fraction will be
around 10 to the minus 4, and it will be relatively
constant up to 1600.
Can I have the next slide?
DR. KRESS: Excuse me, George. I was
surprised to see this as a failure fraction rather
than a failure fraction rate. Do you think there is
a rate involved here?
DR. SLABBER: Well, what is assumed, and
this is also our approach, is that we are not assuming
any rates the fusion constant, et cetera, because that
will put us in a maze of uncertainties.
DR. KRESS: Yes.
DR. SLABBER: We assume that if the fuel
reaches a specific temperature, the content is -- that
takes us away from proving experimentally that a
certain isotope like silver or cesium or strontium
defuses at a certain rate through the microsphere.
DR. KRESS: That's what General Atomics'
model does.
DR. SLABBER: That's right. And we
believe in South Africa that it puts you in a maze of
uncertainty, and we have done the analysis and we have
seen that with releases in a big depressurization
event, the containment performance -- and I'll come to
that a little bit later -- is sufficient.
DR. APOSTOLAKIS: So do I understand
correctly that these curves were produced from zero
failures?
DR. SLABBER: The rest. That was produced
from experimental we determined on means that were
done on the leech test. And everything was based on
that specific one.
We will be repeating this, but we will
allow us to be criticized at every point.
DR. POWERS: I guess what I don't quite
follow is that you're testing -- you're assuming that
just temperature is the variable. Does that mean that
you're not running these fuel particles through
operational events?
DR. SLABBER: Such as?
DR. POWERS: Shutdown, restart, abrasion?
DR. SLABBER: No. Abrasion we will be
testing in the fuel handling system, the diameter. And
if it goes below a certain value and that leaves you
a very big margin because thickness of the unclad --
of the unfueled section is five millimeters, we will
allow the diameter to go down to 58 --
DR. POWERS: What I'm asking you is there
no synergism between temperature, irradiation and fuel
motion as well as normal cycling operation on the cool
failure rates?
DR. SLABBER: I'm talking about fuel
failure rate in terms of microsphere failure rate.
DR. POWERS: Yes, I understand. I
understand.
DR. SLABBER: Yes. We believe it's
uncoupled.
DR. POWERS: And is there any
substantiation to that uncoupling?
DR. SLABBER: Substantiation for
uncoupling?
DR. POWERS: Yes. I mean, what I'm really
trying to understand is why is it the temperature is
the only variable to consider here?
DR. SLABBER: It has found that in the
German test that the temperature is the driver of the
cracking if there is something. And the
manufacturing-- the pressure, though, the ramp rate
because the temperature gradient through microsphere
integration has not been considered, and it was
believed that it's uncoupled.
Next slide, please.
The previous -- sorry. The previous
slide.
This is a artifact which we have developed
from German literature showing, and this is the -- if
you noted at this stage that it's showing the
tendency, what happens beyond 1600 without saying that
this is what we expect, because this was extrapolated
back from releases. Real releases back to failure
fraction.
Now what is happening here at 1600, the
silicon carbide coating on the microsphere slowly
starts thinning due to reactions with fission
products. And you get this slight increase in failure
fraction -- I say you're going this way now -- they
look back from a release rate. And then there is a
gradual increase, and then at 2200 degrees Centigrade
there is quite a gross dissociation of the silicon
carbide microsphere coating. And that's the reason
for this rapid increase.
So this is silicon carbide thinning and
degrading, and this is actually the disintegration.
This is the reason why I brought this, because this
will also be part of the testing.
Skip the next slide.
Our conversion unit is interfaced in the
coolers with an auxiliary cooling system which
interface directly in the coolers with a helium
coolant. What happens is that the pressure in the
primary system is always higher than the coolant in
the auxiliary cooling system. So if there should be
a leak, the water should leak out into the auxiliary
system and there are instruments that detect any leak.
If we are doing maintenance on the reactor and the
system is depressurized, then there is the only
interface with the primary -- with the core is by
means of the core conditioning system, which has got
a very limited volume of water circulating through the
heat exchangers. And then the primary coolant system
is always monitored from a radiation point of view to
see if there is any contaminants like fission
products, especially in this case, moisture and air.
The physical design of the core itself is
such that in the event of even a beyond licensing
based event, that the establishment of a established
flow regime of air through the core is not feasible,
but this is being modeled by means of CFD at the
moment. We believe we think our difficulty could be
to -- and this is also time dependent, and it's got a
temperature limit beyond -- 400 degree average
temperature, it is not possible even with gross
ingress of air. So, it's really an event which gives
you time.
Next slide, please.
The physical core design for the
prevention of excess burn-up, because a fuel has got
a limit and licensed to a limit of burn-up. And we
will be licensing our fuel for 80,000 megawatts and
it's got that limit. And we will -- the core is
designed that a ball could not be trapped like in the
German reactor program, there were certain of these
spheres that were trapped somewhere in the core,
pressed into the graphite for some long time. It did
not give rise to a rise in activity, but our core
design is such that the flow is so well defined that
we do not expect that.
And then we've got on-line spectrometric,
gamma spectrometric measurements because we need to
evaluate -- is it fuel or is graphite. And if it is
fuel, by means of gamma spectrometry we determine if
the burn-up has reached the limit or can it be
recycled.
So, these are those attributes.
Next slide, please.
So if we now look at the barriers to the
environment from the kernels, we have beyond -- before
that we've got the UO2 kernel which provides some
degree. I say some, but we do not take credit. And
then we've got the three -- the pyrolytic graphite,
we've got the silicon carbide and we've got the other
pyrolytic carbide. But credit is only -- we're
looking from a qualification point of view only at the
silicon carbide. We listed those three layers because
that is the reality.
And then we've got our high integrity
primary pressure boundary, and we are learning and
we're using information from the light water reactor
people which has developed materials, steels, et
cetera, and we try not to deviate from that developed
and evaluated envelop.
So we will be using pressurized water
reactor reactive pressure vessel and the pressure
boundary we will take note of developments.
And coming to the containment, which we
have been defining in the past, and it's a debateable
question, as confinement but we are using the term
containment. But at this stage let me just explain to
you what is happening during a event when release
takes place.
You get a rupture of the primary pressure
boundary, and we've got 10 millimeter breaks analyzed,
we've got 65 millimeter because that is the size of
the fueling tube. And then we've got big breaks like
the control rods or the bottom unloading shute or the
PCU pipes. And we've got graded pressure releases.
And for each of these we've got a system, a pressure
release system by means of ruptured panels which
release from specific cavities in the containment to
a pressure relief stack which automatically opens and
closes again after this puff goes out. And then it's
got a backup which could be closed if it does not
close automatically by an operator.
And then if there's an excessive event
like a 10 minus 6 and lower event like the rupture of
the big manifold pipes, in addition to this pressure
relief, there are -- if we think back to the first
slides of the building that can lift up above the PCU
and release into a big plenum. And then if the
pressure is still in excess, panels will blow out, but
remember, of the wall. But remember this is an
analyzed event.
The containment is designed to relief
through the pressure relief stack and be closed
automatically with operator backup. So we define for
the performance requirement that we need this
containment has a high leakage vented containment,
because we've got also the HVAC. And the HVAC is also
automatically closed off during such a
depressurization and could be opened again later to
filter light releases at a low pressure.
So we've got a concrete structure which is
a citadel, but actually it is high-leakage vented
containment. We've got a filtered vent path for later
releases. And we've got hold up of fission products
in plate out in the system, et cetera, which is not
lifted out. And the auto-close blowout panels. And
then we -- by means of this HVAC later releases from
these particles if there are any additional.
Thank you.
Just coming back to the nonproliferation
aspects.
Mr. Chairman, I'm sorry, I'm taking a
little bit longer.
There is a number of attributes. It's a
closed -- it's an on-load fueling system. The IAEA
can install flow monitors to see where fuel is and
track the fuel movement. And the burn-up is 80,000
megawattage per ton which gives a plutonium mix which
is very unfavorable for weapons manufacture. And then
the fuel produced during the operational time of the
reactor is all stored in the facility under the
surveillance of IAEA.
DR. POWERS: It seems like it's a design
that's well suited for producing 239 because of the
on-line fueling/defueling at the facility.
DR. SLABBER: Yes, it is. But if you look
at the amount for even the first cycle, you must
divert about 212,000 spheres continuously out of your
system to produce a favorable mix. So what we have
during a ten cycle, which is from a -- point of view,
the optimum at the moment we're thinking about five,
it gives some problems -- not problems, but a higher
flux higher up in the core. At discharge the mix is
66 percent 239 and compared to either --
DR. POWERS: Change your cycle. Lots of
239 --
DR. SLABBER: You need only one force of
fuel sphere to give you a very small -- and you've got
to take them all out into the diversion path. And
this is not difficult to detect.
MR. SPROAT: Thank you, Johan.
What I'd like to do is just briefly close
and address the issues. So now you understand a
little bit about the technology itself and the
preliminary design of the PBMR itself, what about
getting it licensed here in the U.S.?
As part of Exelon's decision making
process, we are currently evaluating and doing a
license ability assessment on the PBMR. And I want to
talk about very quickly the key issues that we see
both technical and nontechnical.
And on the technical side, obviously right
now most of the regulations existing in the U.S. are
focused on light water reactors. And if we were to
come in today with an application for this technology,
the NRC reviewers would sit there and they'd use what
we call the "two finger approach;" one finger on the
regulations and one finger on the submittal and say
"Okay, how did you meet this, how did you meet that?"
In some cases that'll be very appropriate and in some
cases it won't be appropriate at all given differences
and uniqueness of this technology.
So, working with the NRC staff over the
next 18 to 24 months, we hope to develop a regulatory
framework that they can use and that we can use to
design against, they can to review against so that
we've got a credible regulatory framework that we can
try and license the PBMR with if we go forward.
The second area is fuel qualification and
testing. Johan talked about that. The key thing about
the fuel is that, you know, this isn't new. You know,
trico-coated practical fuel was used back as early as
1967 in the dragon reactor in the U.K. So there's a
great body of information out there. We need to be
able to tap that and use it as part of our licensing
basis and not have to reinvent the wheel.
But the other aspect of this is the first
fuel loads for the PBMRs in the U.S., if we do go
forward, would come from South Africa. So the role of
the NRC in reviewing that fuel plant down there and
licensing it or not licensing it but certifying the
end product for use in a U.S. reactor is a whole area
that we really haven't explored yet and will need to
be addressed.
DR. KRESS: When you talk about fuel
quality, are you talking about that fraction of
particles fail versus temperature curve?
MR. SPROAT: Yes. Knowing how the fuel
will react under various conditions that's consistent
with the safety case for the reactor licensed in this
country.
DR. KRESS: Does that include any trapped
uranium that might get trapped in the --
MR. SPROAT: Yes, obviously the test
program takes a look at what the -- not only what the
failed fuel fraction is, but also the trapped uranium
that's on the outside of the particles as a result of
manufacturing process.
DR. KRESS: You have a goal for how many
particles can be failed within the core before you
violate 10 CFR 100 --
MR. SPROAT: I'm not sure we're that far
along in the analysis at this stage of the game.
DR. KRESS: Okay.
MR. SPROAT: Clearly an issue that we're
going to have to wrestle with the staff, once we
decide ourselves how we think the appropriate way of
addressing it, is what's the source term? Is it
mechanically mechanistically determined source term
or deterministically determined source term --
DR. KRESS: Well, it's the answer obvious
there?
MR. SPROAT: Pardon?
DR. KRESS: Isn't the answer obvious
there?
MR. SPROAT: No, the answer's not obvious.
I know what we would like to do, but the issue of how
good are your goods analyzing your diffusion
coefficients and being able to provide an analytic
framework for migration of fission products from the
core to the environment is going to be a challenge.
It's going to be a challenge.
Obviously, containment performance
requirements, Johan talked about the containment
design and whether or not a zero leakage or a LWR type
containment would be required versus moderate to high
leakage filtered containment would be required is
obviously an issue that's going to be discussed at
some length.
DR. KRESS: And that would be linked to
the fuel quality?
MR. SPROAT: Absolutely, and to the source
term.
The issue of the various computer codes
that are being used in South Africa to design this
plant, how they're verified and validated and how
they're benchmarked against the other existing codes
will be an extensive effort associated with that.
The PRA itself that's being developed in
South Africa that we're advising them on, it's kind of
interesting. You know, if you have -- what's your
endstate if core melt isn't a valid endstate for your
reactor? And what is your endstate? What are you
initiators and how do you determine your uncertainties
of your various accident sequences?
DR. KRESS: Your endstate is quantity of
fission products. Frequency of fission products.
MR. SPROAT: It might be. But the point
is that we're exploring some new ground here and,
obviously, there'll be some discussions with staff
about how we go and do that.
The regulatory treatment of nonsafety
systems and how we classify the SSCs, the safety
system components, will really be a key issue.
And then finally, an issue that I lumped
in the technical area, but it's a real practical issue
is there aren't a lot of people left in the U.S. in
the NRC, in the national labs or in DOE that have gas
reactor experience and understanding. And so,
obviously, I think you've gotten a sense as we go
forward with this, if we submit an application having
people who understand the technology, understand the
science and can provide good independent review of the
submittal is going to be a real challenge.
On the last slide I have is the
nontechnical, what I'll call the legal licensing
challenges. And I personally believe we have a very
good chance at satisfactorily resolving a number of
the technical issues that I showed on the previous
slide. I'm not as confident about some of these,
because some of these are potential deal breakers for
moving forward with merchant nuclear power plants in
this country. And that's what we're talking about
here; this is not a power plant or nuclear plant
that's going to go into a rate base somewhere. This
is a merchant plant where the shareholders are going
to take the risk of building and operating this plant
and whether or not it makes money in the deregulated
marketplace is solely dependent on the technology and
the company that runs it.
So, the first issue up here is Price
Anderson. The current law and the way it's currently
interpreted by the NRC is that each reactor in the
country is assessed a retrospective premium of $90
million per reactor in the case of an accident
anywhere in the U.S. associated with any reactor.
Well, if I've got a 2200 megawatt light water reactor
plant, like our Limrick plant, that means my
retrospective premium at risk due to a reactor
accident somewhere in the U.S. is $180 million
retrospective premium associated with that plant.
If I have the same capacity of pebble bed
modular reactors under today's law, my retrospective
premium would be $1.8 billion for that same amount of
capacity. Even I would have difficulties selling our
board of directors to take that kind of a risk
associated with that kind of retrospective premium
associated with an accident from a reactor that we
don't own or operate. So that's got to be addressed
somehow.
The second issue up there is the NRC
operational fees. Right now the operational fees are
approximately $3 million per reactor. Again, say at
our Limrick plant, that means about $6 million a year
for the two reactors. The same size for 2200
megawatts, you're talking about $60 million a year in
NRC licensing fees for a 2200 megawatt set of string
of PBMRs. Really excuse the economics of a merchant
nuclear plant significantly.
The decommissioning trust fund is another
issue that's clearly going to have to be addressed.
The law gives a number of different alternatives, but
those alternatives have presupposed that generally the
plant is going to be operated by a regulated utility
and that in the rate base in which the plant is based
rate, you have a set aside income stream that goes and
funds the decommissioning trust fund. In our case
that won't be the case. These plants won't be in a
rate base. How we fund the decommissioning trust
fund, how much we have to put up front and what we can
put into a sinking fund needs to be resolved. The law
is not clear on that at this point in time.
Clearly, Part 52 licensing process which
is, we think, the right way to go is untested at this
point in time. Nobody's actually done it. So the
staff will be learning, the applicants will be
learning, and how we actually work our way through
that and how long it takes is going to be a key
challenge for us.
And then finally, I have up there up the
potential number of exemptions. As I talked about
earlier, there is no gas reactor licensing framework.
And if there's not when we go with an application, the
staff might decide that a number of the things we're
asking for are very appropriate to license this plant,
but will require exemptions from the existing
regulatory framework. And, obviously, it would be
undesirable to all of us to have the first advanced
reactor in place with a significant number of
exemptions. It just doesn't work.
So, those are the key issues and
challenges we see on the licensing side, both from the
technical side and the legal side. And, as I said, we
are considering all that and now we'll go into our
decision making process as to whether or not to
proceed with both the venture in South Africa and the
licensing process here in the U.S. by sometime around
the end of the year.
DR. KRESS: These appear to me like mostly
policy issues rather than technical ones related to
the reactor design?
MR. SPROAT: A number of these will
require some policy statements and decisions by the
Commission itself, yes.
DR. KRESS: Very good. Is there any
discussion or questions for either of our two
speakers?
DR. APOSTOLAKIS: Yes, I have a question.
As I recall in one of your communications to the staff
in addressing these issues, the key legal licensing
issues, you proposed that a site with ten units be
considered as one reactor?
MR. SPROAT: One facility.
DR. APOSTOLAKIS: One facility.
Now, if this is accepted by the staff,
then should we also be applying the same idea to
various safety goals and say, assuming that the
concept of core damage makes sense here, that if the
goal is 10 to the minus 4 and that would apply to the
facility, so each unit then would have to ten to the
minus 5. And given the fact that you have ten of
them, you have some synergistic effects, maybe it'll
have to be even lower than ten to the minus 5.
MR. SPROAT: Well, synergistic effects is
not intuitively obvious to me that there are
synergistic effects when in fact the risk from one
reactor to the other. I'm not ready to concede that
point at this point.
DR. APOSTOLAKIS: Okay. Fine.
DR. KRESS: Some common mode.
DR. APOSTOLAKIS: Some common mode,
perhaps. Anyway, but I mean how about the thought
process here that you would apply stricter criteria --
DR. KRESS: Yes, instead of calling it
core melt, call it fission product release --
DR. APOSTOLAKIS: Call it something else.
Yes, fission product release.
If we treat 10 PBMRs as one facility with
respect to these five bullets that you showed us,
shouldn't we be doing the same when it came to risk
and treat it as one facility and apply the goals to
the facility, in which case of course we will have
much lower goals for each individual unit?
MR. SPROAT: Well, we certainly haven't
done that for two and three unit light water reactors.
So, I hesitate to do that for a smaller, supposedly
safer reactor.
DR. APOSTOLAKIS: Well, safer of course is
something that you would approve of.
MR. SPROAT: Sure.
DR. APOSTOLAKIS: But for a two unit
reactor there are some PRAs where they look at these
things. But a factor of two in the goals really
doesn't mean anything. But when you talk about ten,
a factor of ten, then you're beginning to see some
difference.
So it seems to me that if we are to apply
this idea to the five legal licensing challenges you
mentioned, maybe we ought to think about doing the
same thing to the goals. Now, you don't have to
answer right now, but --
MR. SPROAT: I would probably disagree
with that, but that's okay.
DR. POWERS: Explain why you would
disagree other than the fact that you wouldn't like
the numbers when they came out.
MR. SPROAT: No. What would the basis be
for doing that? For example, in airline travel
there's a certain risk associated with flying on an
airplane. Now, the fact that there are increasing
numbers of airplanes in the air doesn't necessarily
mean that your risk of being killed on an airplane has
proportionally increased.
DR. APOSTOLAKIS: The societal risk has.
DR. POWERS: Right.
DR. APOSTOLAKIS: The individual risk has
not.
DR. KRESS: You don't fly the same number
of people on the airplanes. What you have is a site
with a given fixed population around it, for example.
And that population is exposed to either one module or
ten modules who could fail independently of each
other, and in fact that's probably the assumption.
But the risk of being on that site and associated with
those reactors is, in my mind, ten times when you have
ten modules over one module.
DR. POWERS: Tom, isn't it even higher
than that because you've got a mode failure with the--
DR. KRESS: Yes. And then if there's
common mode failures, it's even higher.
DR. POWERS: Especially if you go up --
DR. KRESS: And that would be the
reasoning behind --
DR. POWERS: -- to a centralized control
room?
DR. KRESS: Yes. So you treat it as one
reactor, but in order to accommodate the ten of them
you have to do something to one end; you either up the
frequency by ten or the lower safety goal by --
MR. SPROAT: Well, then clearly you have
to take into account in that kind of an analysis the
concept of coincident events happening in multiple
units at the same time.
DR. KRESS: No, no, that's not --
DR. POWERS: It's just common mode failure
is what we are talking about here.
DR. KRESS: But that's not what I had in
mind.
MR. SPROAT: Assuming there is a common
mode failure that --
DR. POWERS: But that's not what we're
saying.
DR. KRESS: Yes, but that's not what we're
saying. I mean, that's another issue, coincidence
events and common mode failures. No, I'm not just
talking about an independent frequency of something
happening to one or something happen to the other
independently.
DR. SHACK: Of course, now he does get
something back because he probably has a smaller
source term.
DR. KRESS: Oh, I think that's a -- for
this concept, that's --
DR. APOSTOLAKIS: I didn't say anything
about the assessment.
DR. KRESS: Yes. He said --
DR. APOSTOLAKIS: I'm just talking about
the goals.
DR. KRESS: I'm sure they could meet the
ten times or the ten percent --
DR. APOSTOLAKIS: You don't use a facility
of ten PBMRs only on these things. I mean, and the
goals have to be reflected.
MR. PARME: George, I might add in the
mid-80s submittal on the MHTGR where there were
multiple reactors coupled to a common steam plant, it
was viewed as a plant and we took the safety goals and
the release limits that we were analyzing it and
considered multiple reactors. And, in fact, if you
look back in the mid-80s submittal you'll see there is
at least one event that has all four MHTGR models
leaking simultaneously without cooling. And it was
handled that way.
It's not quite the case where his reactors
are truly independent, but we did consider the four
modules to be a plant consistent with your thinking.
What you would do with truly independent modules, I
guess, is something that one might want to think of.
DR. APOSTOLAKIS: If we decide, for
example, that as we were saying earlier that the
appropriate way to look to formulate the goals here
would be through frequency consequence curves, then it
seems to me that you would have one such curve or a
family of curves for the facility.
DR. GARRICK: Yes. Well, why wouldn't you
have a CCDF for the facility?
DR. APOSTOLAKIS: For the facility, that's
what I'm saying.
DR. GARRICK: And every time you add a
module, you get a new CCDR.
DR. KRESS: Yes, absolutely.
DR. APOSTOLAKIS: Yes.
DR. GARRICK: Yes.
DR. APOSTOLAKIS: But the goal would be
one. And then what you do under it, you know,
assuming you're acceptable is your business.
DR. GARRICK: Right.
DR. APOSTOLAKIS: Anyway, that's just a
point.
DR. KRESS: But it's a thought.
MR. SPROAT: Understood.
DR. KRESS: Other questions? Okay. Please
use the microphone and identify yourself for the
record.
MR. GUNTER: Paul Gunter, Nuclear
Information Resource Service.
Obviously fuel integrity is a big question
here. And what I would like to get a little better
idea of, is have you looked at the THTR that was a 300
megawatt PBMR in Germany? I believe there was an
event there on May 4, 1986. And I'd like to know what
your assessment is of the fuel failure mechanism that
occurred there?
DR. SLABBER: I do not have at this stage
information about that specific occurrence. But what
I can tell you is that due to the uniqueness of the
THTR core where they had control rods and shutdown
rods of this size pushing vertically into effect
pebble bed during shutdown, that caused some of the
pebbles themselves to break, although no evidence was
ever found that they found loose coated particles
somewhere in the fueling system. But that gave rise
to a bigger than normal fuel sphere breakage, the
specific design itself.
MR. GUNTER: It was the graphite that
broke apart or was it the pyrolytic coating that
broke?
DR. SLABBER: It was the graphite, the
matrix that kept all these coated particles in a
configuration.
MR. GUNTER: Right. So just for my
understanding from what I've been able to ascertain is
that the fission products are to be retained inside
the pyrolytic coating, though?
DR. SLABBER: Inside the silicon carbide.
MR. GUNTER: Right. So if there was a --
so it would seem like there was some kind of failure
mechanism on that pyrolytic coating as well. I mean,
was the coating crushed as well as the graphite
sphere?
MR. SPROAT: What we know from that event,
and I haven't gotten all the details of the German
government review, is that as Johan said that the
pebble bed that's in the THTR in Germany had its
control rods inserted directly into the pebble core.
That broke a number of pebbles. So and then when they
tried to come out through the bottom for the fuel
handling system; if the ball's round, then it goes
through the system really well. If it's broken into
pieces, it gets stuck. And evidently what the German
operators did is they found they had some broken and
stuck particles -- not particles, but pieces of the
fuel spheres in the handling system that got stuck,
and they had to clear them out of there.
MR. GUNTER: And that was done with back
pressure of helium or --
MR. SPROAT: Well, I know that back
pressure of helium is one of the methods they used to
clear some of that fuel handling system, but they also
I think in that case you're referring to is they used
some mechanical force where they tried to either hit
things with either hammers or with rams to free that
piece. And it appears what happened in that case is
that a number of the little particles from that
mechanical impact were ruptured, and that released
some of the fission products from inside the spheres.
But it was basically mechanical damage to the fuel
particles itself due to operator interaction.
MR. GUNTER: Okay. If I could ask one
more question. It's also my understanding that the
Germans abandoned the technology because of problems
with quality control on unused fuel. Have you looked
into that as to what the failure mechanism was for the
unused fuel?
DR. SLABBER: The only records we have is
that the German program would have continued, but
there was some other political pressure to terminate
any further investigations. But the database that we
have access to do not address any of such problems
that you're highlighting now. In fact, they have
still available for evaluation some of their unused
fuel spheres and we intend to do some pre-irradiation
evaluation of those spheres.
MR. GUNTER: Of course, if there was
evidence of damage to unused fuel, you would be
interested in seeing that
DR. SLABBER: Of course, yes.
MR. GUNTER: Thank you.
DR. SLABBER: Can I just make another
comment. The design, the German design which had the
control rods in the bed directly in the core was one
of the reasons why pebble bed design deviates totally
from that design. And the decision was made, control
rods only in the reflector, sides reflector.
DR. KRESS: Okay. I'd like to move it on
because we are running behind now, and move to the
next topic, which is, I believe, the IRIS by
Westinghouse representatives.
MR. CARELLI: Good afternoon. I'm Michael
Carelli from Westinghouse Science & Technology
Department. And among the many things we do is the
leading edge support of the business units, also
heavily involved in Generational IV reactors, and
especially on IRIS.
Now, I have to tell you a couple of things
before I start. And the first one is you have in the
passouts some viewgraphs that aren't exactly right.
Last week I at IA meeting in Cairo and I
was trying to do very much control. This presentation
is terribly efficient. But we have the right package,
and if you need it you see me and I'll get you a copy.
And with that, I think my time is up now,
right?
DR. KRESS: Yes.
MR. CARELLI: Okay. Nice meeting you.
Okay. IRIS. Can I have the next one?
IRIS is International Reactor Innovative and Secure
and the key word there is international, and you'll
see in a second why.
If I can have the next, please? I'll try
to move fast as I can.
Is the new kid on the block. We've been
in business for about 18 months, so what you see is
about we started at the end of '99 this work, and so
in trying to compress in about a half of hour the work
we've done on a new design, I had to skip a bunch of
items. And I'll be happy to answer and expand them
during the session this evening. So right now I try
to kind of streamline on the key things and then hit
the issues, because for this new reactor thing that's
what you want to hear most.
So I'm going to have a brief overview; our
team, the funding, the objectives. I'm going to tell
you about a few designs. It's plural, it's not a
typo. It's few designs, plural. And then the
configuration of the integral vessel. And I'm going
to spend quite some time -- well, "quite some time"
relatively speaking on the safety design because I
think that's kind of a trademark of IRIS. They
approach the safety we have together with the
maintenance optimization. These are the two things
IRIS, I believe, does different. And then, as I say,
I hope to spend some time talking about the issues.
Let's move to the next one, please.
Overview, keep going. I have a bunch of fillers. At
least you know where we are.
Okay. This is a capsule on IRIS, just to
give you a kind of best view what the reactor is.
What you have on the right is an earlier version, it's
100 megawatt electric that we designed until around
December of last year. It's an integral system.
Integral means everything is inside the vessel; steam
generators, clamps, pressurizers -- pressurizer,
singular, is inside the vessel. Is integral, integral
configuration. And it has a lot of advantages. It is
really an excellent configuration for safety and we're
going to touch on that, as a straight bell core, no
shuffling to refueling. You put the fuel in, take it
out at the end of life.
And we have two designs for five years, an
ATS lifetime. And you'll see in a second, in a couple
of seconds.
It utilize LWR technology. In the new
engineering burnt is a proven technology. This is a
key point when you look at development schedule, this
is a new engineering. We are not demonstrating a new
technology. Also the integral configuration for the
light water reactor is not the first time. There is
a surface ship in Germany has been running along the
seas with an integral reactor like this, and of course
all of you know the submarines, they are running on
that. And also there's been experience -- on integral
reactors.
Safety is and most action initiators are
handled by design. And I'm going to go into safety by
design issue and what we do -- we do on that.
Potentially the cost, is the cost
competitive with that options both in nuclear and non-
nuclear and the development, the construction, the
deploying and everything from the very beginning is by
international team. This, by no means, suggest
Westinghouse -- this international team that is
designing IRIS.
And we are projecting the first module
deployment in the 2010-2015 time frame. 2010 is kind
of widely optimistic, 2015 is probably conservative.
And this morning you heard about 2020/2012, and this
is about the time I think we are targeting.
The way IRIS started was in answer to the
Generation IV RFI that we had from DOE. And basically
we were trying to look at satisfying the goals of
safety and unsafety, sustained development.
What you have on the left are the various
design features of IRIS, and you can read. And
basically what we found that those design features,
the way we started the design, was they were to
satisfy safety and to satisfy the waste minimization
issue. And then we found that every single one also
has a positive effect on economics.
Next slide. Thank you. So I said every
single one does end up on the positive column of
economics. So at that point to say, gosh, you know,
we had quite a good design for commercialization.
And, please, the next one. And that
basically what happens. And what happened was that we
started building a consortium of organizations where
they're interested in joining IRIS. So the first
thing we did was to have a colorful logo, and then
after that we went to work.
Next one please.
DR. KRESS: Is that Latin?
MR. CARELLI: Yes, that's Latin. From the
Italian, what do you expect? This is a Latin motto,
and I think even the translation has to do with
nonproliferation. Believe it or not.
So what we did, we had the initial team
was from Westinghouse, two U.S. universities,
California Berkeley and MIT, and from Milan. We
wanted the work published. We started having phone
calls other people wanted to join. And what you have
here is chronologically. This is the organizations
that joined IRIS in time.
At the beginning, it was mostly
development. Then what we did recently in the last
few months, we added an organization as a supplier
site because we had the design that is moving very
well along. Now who is going to fabricate, who is
going to be the manufacturer and so forth? So we have
additions to the size -- which from the very
beginning, an addition like Ansaldo, Spain and Brazil
to do the components.
And now what you see now is that we have
also team members from developing countries. IRIS is
very attractive for developing countries and, in fact,
I'm coming back from Cairo and had a very, very good
reception from developing countries. It's 100 to 300
megawatts and it doesn't clog up the -- of developing
countries like 1,000 megawatts does, so this is quite
attractive. Next please.
Now, you heard the question this morning,
John. It said what is a dedicated enthusiastic team?
Yes, you have. You have a dedicated enthusiastic team
that's designing IRIS and it's very enthusiastic that
this is the money we're getting from the UE. This is
the money over three years from Westinghouse,
California Berkeley, and MIT. This is the money that
the other participants are putting in on their own.
This is in kind contributions. People they're putting
to work. They're working on. Right now we're running
around this. So that's enthusiastic when you put out
that type of effort. Next.
Okay. One of the questions was what's the
schedule? The schedule was at the end of the first
year, this is the end of our first year of life, we
wanted to assess the key technical and economic
issues. Basically rather than going through the old
thing, we just pick up the key issues and resolve
them, and we have done that. Right now we're filling
in the blanks. We're doing the conceptual design and
the preliminary cost estimate and at this point is the
end of the NERI grant in 2002. At that time, we're
going to have the preliminary design completed, the
preliminary cost estimate completed. Sometime in
between now and then there is the pre-application
submitted to NRC. We're in the preliminary stage now,
we have been talking with the staff a few weeks ago.
I'm talking with you now and it's in the process.
I put a question mark because really I
can't say it's going to be July, August or so. But
it's going to be definitely soon that we're going to
talk.
Now here is where lightning is going to
strike. At the end of the first three years, the
consortium is going to sit around the table and say,
okay, now we have a design, we have a market, are we
going to proceed with commercialization? Right now
every indication is that the answer is yes but at that
point then it goes on a quantum step in terms of
effort. It's no longer $8 million or $12 million.
It's going to be quite a lot more. So if that happens
-- right now, of course, we're not doing this for the
fun of it. We are working assuming that is going to
happen.
Then our schedule calls for a complete SAR
by 2005, design certification by 2007 and first-of-a-
kind deployment beyond this. And I'm going to have
some discussion on these dates at the end. Next
please.
DR. KRESS: Would your SAR follow the SAR
process that we use now for light water reactors?
MR. CARELLI: Yes. When the issue is
safety, I think it should be simplified. Should be a
simplified SAR. We'll see.
Okay. Here now the cores. Originally we
worked on this. The proliferation resistance -- the
idea is you have a core and you put in no shaft and no
refueling. The host country doesn't have access to
the fuel. The longer you keep there, the more
proliferation resistance you have. So we found that
eight years we could have burn-up around 70,000 -
80,000 and we worked two designs with UO2 and MOX
interchangeable so essentially with the same IRIS
design exactly the same, you can put whatever fuel
core you want. So that's what we have done.
But then what you have, you have IRIS
requires eight percent enrichment. Right now we don't
have a licensed eight percent production facility and
we don't have the database for the burn-up and so
forth of the eight percent. What we said at that
point, we say why do we want to complicate the life
and let's say the first core with a five years design,
same thing straight through for five years, same
principle, nothing different. But this is 4.95
percent enrichment.
Our facility in Columbia can fabricate it
to model as exactly the same design and the same
configuration as the PWR assembly. So if you say that
you can't recognize the difference between a regular
PWR and an IRIS assembly.
It's well within the state of the art
because the average burn-up we're projecting is around
45,000. So at this point with this we have taken out
completely any licensing issue because this is a PWR
assembly. The only thing we are doing different,
instead of shuffling every three months, we let it
cook for five years. That's it.
At the same time, we are going to look at
this and we have here our university team members that
keep working on this and we're going for the licensing
extension while we're working on and eventually we ask
for licensing for this in the time frame of 2015 to
2020. So right now I want to say this is the IRIS
core. That's what we're focusing now. Next please.
The configuration. This is the 300 mega
version, 335 actually. You see here is the steam
generator and this is different from the pass outs
because in the last couple of weeks we changed the
pumps. What we have now, we have a pump which is
called a spool pump is inside the vessel and there is
no penetration. The only thing it takes is a couple
of inch line for the power, and that's about it. It's
already inside the vessel, high inertia and actually
I was told this morning there has been examples of
this with insulation. It doesn't even need cooling.
Now, the point is why we didn't have this
in regular reactors. Why this coming out of the
woodwork for the first time? And there is an answer.
This pump works with 18 PSI head and in present loop
reactors you never have an 18 PSI. In IRIS with the
very open core and the open configuration we have, our
pressure drop is less than 18 PSI. So in IRIS we can
take advantage of this thing, eliminate the
proliferation device and all of the stuff associated
with the pumps, LOCA -- and so forth is all gone
because we have a design that can take advantage of
this. I think IRIS take advantage.
These here are internal shields. What you
have here, you have here the core, here the steam
generators and you have a design rate of nothing. If
you put shields which doesn't cost much, just a bunch
of plates maybe with some boron carbide or even steel,
whatever. Next slides, please.
This is what you have. It's a gift of the
integral configuration. You get busy for free. The
rate outside the vessel is this. Is nothing. You can
touch the vessel. The vessel is cold. It has two
advantages. One, if you had to send the workers in
the containment, you don't need to put scuba diving on
them. They can go in there in t-shirt because there
is no radiation outside the vessel. The other thing
is simply -- decommissioning because you take out the
fuel and everything inside the vessel remains there
and the so the vessel is like a sarcophagus and this
is especially important if you want to deploy IRIS in
developing countries. You take IRIS in. At the end
of life, you take it back. And there is no
decommissioning, no cost left in the host country.
Next.
DR. KRESS: When you change out the core,
do you also change out the steam generator?
MR. CARELLI: No. I'm coming to the steam
generators. The steam generators, what we have, we
have this nice lady which you can see, but is in
Italian, and this is a picture at Ansaldo. Ansaldo
built the helical steam generators for Super Phoenix
and they tested the steam generators and this in fact
is a huge steam generator. I think it's a 20 megawatt
-- steam generator. They tested it. In next slide I
have what they tested. But what I wanted to give you
here because the steam generator, the perception is we
have so much trouble with steam generators now. This
crazy guy wants to put the steam generator inside the
reactor and this makes even worse. And there are
things you have to think.
First of all, if you put a steam generator
inside, now the primary fluid is outside the tubes so
the tubes are in compression instead of traction. And
so now you don't have any more of the tensile
distress, corrosion, so forth. Our IRIS doesn't have
a bottom so the chemistry is much better. Okay. The
other thing is you don't have -- so the bottom of the
deposit of the steam generators is the bottom of the
vessel. So there are a bunch of things that the steam
generator has a different environment in an integral
reactor versus a loop reactor. So don't think I have
all the problems of the loop, I am compounding them.
This is a different animal. We're talking different
animals.
Now, what they did in Ansaldo, they tested
the steam generators. Next slide, please. First of
all, there is experience with Super Phoenix and the
MFBR experience. In terms of LWR, as I said before,
the auto-on was running on helical steam generators.
The one you just saw in the picture before. So they
fabricated, tested, they confirmed the performance
with all the performance we have and by some stroke of
luck, our device is such that we can put eight steam
generators practically identical to the models Ansaldo
has fabricated. So now we have one thing and that
thing is important. What we have now, we have eight
steam generators for 300 megawatts. So we're not
talking redundancy. That's exactly what we want to do
because the steam generator have a very critical
safety function and you are going to see in a second
what it is.
Next. That's the safety by design. Next,
please. Okay. Now on the safety by design. Just
doing a little bit of background. The way we see on
the philosophy. You take a Generation II. You have
an accident and you have cope with active means, like
you have a loss of coolant accident and you dump --
emergency coolant system and make up water, all that.
On Generation III you do the same thing like you do
with passive means. So inertia is going to help you.
But still you are doing something to handle the
consequences. On Generation IV what we looked at is
rather than coping with the consequences, since we
have this new geometry, let's take advantage and
prevent the accidents through safety by design. Next,
please.
And that's basically what we've done. We
spent quite a long time looking at the integral
configuration and saying how can we exploit this? How
can we exploit the IRIS characteristics which is the
integral configuration long-life core to eliminate the
accidents from occurring. Number 1. Two lessen the
consequences and three, decrease their probability.
Next.
DR. APOSTOLAKIS: If you physically
eliminate the accidents, aren't you decreasing their
probability?
MR. CARELLI: No. No, no. Yes.
DR. APOSTOLAKIS: Are there different
accidents?
MR. CARELLI: That's different. Go back.
Could you please go back.
DR. APOSTOLAKIS: I understand.
MR. CARELLI: What I'm saying is that of
course the first thing you eliminate. You do that.
Fine. End of the story. Second, if you can not do
that, you decrease -- you lessen the consequences.
Fine. If you can not do that either, you decrease the
probability. So this is a kind of -- Next.
What we did, this is the one, we're not
passing out -- kind of messed it up and I'm not going
to this in detail because otherwise you're here until
midnight. But what we have on this column is
essentially the design characteristics. These are all
the design characteristics of IRIS. Just look at the
geometry, long-life core, all this stuff. Then I say
here what is the safety implication of this design
characteristic? Okay. I can't read it. This is --
and what happens here?
Now, the first thing is the most obvious.
You don't have the large LOCAs and it doesn't take
much -- you don't have any piping going from the
vessel to the steam generators, so no piping, no large
piping, no large LOCA. That's obvious. Everybody
does that.
But then we went to other steps and one
thing that we worked on, and I think this is something
that is interesting, is the small LOCAs. I still have
the two inch pipe break, could have, and historically
the large LOCA has never been a problem. All the
problems came from the small LOCAs. Next. Sorry.
Before doing that, out of that table we said, okay,
what happens now to the Class IV accidents that were
handled for AP600? And we look with the IRIS approach
of safety by design and we can eliminate the LOCAs.
We can eliminate the range of the actual accident if
we put the control rods and CDRMs inside the reactor
because then you have nothing to shoot out.
And all the others really, because of the
combination of the integral configuration, the steam
generators in compression, all this stuff, could be
reclassified as a Class III.
The only one we have left is the refueling
accidents. It's still a Class IV but the probability
is between one-third to one-fifth less. So that's
what I'm saying here. First you say you eliminate,
then you lessen the consequences and, for this, you
lessen the probability. So essentially out of eight
Class IV accidents of AP600, with IRIS you're left
with one and even that one with less probability.
Next.
DR. KRESS: But you're only going to
handle this fuel once every eight years.
MR. CARELLI: Yes.
DR. KRESS: Doesn't that give you an
additional margin, rather than just this one-third and
one-fifth lower probability. The time gives you much
less risk due to fuel handling because you're not
doing it as often.
MR. CARELLI: Yes. The other thing, too,
and as I said, I didn't want it to stretch, but the
other thing, too, when you're fuel handling, you start
moving things around. You move this assembly from
here to there and you drop one or drop the other one
and so forth. In the case of IRIS, you don't move
anything. You take the old tank and the block -- not
the full tank. We try not to move each assembly at a
time because they are pressure resistant, we like to
have them in big chunks. So you can count. Big
chunks.
So then you don't move one assembly at a
time. You move chunks. So as you said, you're
absolutely right. Reduced probability even more. I
think I had a very good story so I didn't want to
really stretch it any further. But it is.
On the containment. This is the best
part. The containment we have, first of all, it
performs a containment function like every good
natural containment. But we're doing an additional
thing. Since we have the containment, we make the
containment working together with the vessel to
essentially eliminate the other LOCAs, the small
LOCAs. So the small to medium LOCAs in IRIS are gone.
Now, how that comes. If you think why you
have a LOCA? You have the vessel and you have a break
and you have high pressure here, low pressure here,
and that differential pressure drives the coolant
across to the hole. Right. Now, if I decrease the
pressure in the vessel and I increase the pressure in
containment, I have a zero delta P and nothing comes
out. And that's exactly what you can do in IRIS.
First of all on the containment. We can increase the
pressure because we have a smaller containment. It's
about half the size of AP600 which gives a factor of
two on tensile stress. It's vertical which gives
another factor of two.
So now we have a factor of four. So for
the same thickness, for the same stress, you can have
four times the pressure in IRIS that you have in
AP600. Increase the pressure in the containment.
In the vessel what you have, you have,
first of all, a larger volume which means less
pressure. Also you have heat removal from the steam
generated inside the vessel which means a lower
temperature. So higher volume, lower temperature
means lower pressure. And that's exactly what
happens. If I can have the next one.
These are the pictures of the
containments. These are pictures inside the
containment. This is IRIS containment for 100
megawatts, this is the IRIS containment for 300.
Three hundred is about the maximum size you can have
with IRIS. You're not going to see an IRIS of 500
because there is a point where the thermodynamics
breaks and 300 is about the largest size you can go.
Next.
DR. KRESS: The trade-off on having the
smaller more compact stronger containment is you have
to pay more attention to the normal leakage rates
through penetrations?
MR. CARELLI: In the containment?
DR. KRESS: Yes.
MR. CARELLI: Yes. That's what we have to
look at. And again, it's high pressure containment.
Yes, that's something you have to look at. But the
economics is terrific because you have much smaller --
and besides, besides the economics, with our
containment, it chokes off the LOCA. That's a key
thing.
What we have done to prove that, we have
performed an IRIS with different break size, different
elevations, and this is no water make-up, no safety
injection, and we ran three codes. That's the beauty
of having an international team. We ran one at
Gothic, at Westinghouse, one by POLIMI, Milan and we
provided code and there was one at University of Pisa,
FUMO. All three codes predicted the same results.
Next one.
This is the pressure differential across
the vessel. What happens is after the first quick
build-down, for about an hour in the early part of the
transient the pressure in the containment is higher
than the pressure in the vessel because I'm removing
heat like hell inside the vessel while the containment
is cooled by air. And so essentially containment
temperature goes up. So essentially the pressure and
containment is higher than the pressure in the vessel
and actually the steam condenses and is pushed back
through the break. This is kind of quick. Okay. I'm
not counting on this but it's kind of quick for the
100 megawatt, actually for a part of the transient.
You have a coolant going back into the vessel.
But the bottom line is the next one. This
one shows that after two and a half days this is the
level of the water in the core with a 4" break, 12 1/2
meters high, which is the worst place where you can
have a break, and we didn't do anything. No core
make-up, no emergency coolant system, nothing. In
fact, IRIS does not have an emergency core cooling
system. What we have in IRIS, we have a bunch of
tanks which are used as pressure pools because you
have to keep essentially the pressure in the
containment up to a point and those, if necessary, can
be used for core make-up. But this analysis was done
without a core make-up.
So the 72 hours essentially for the LOCA
in IRIS, it goes and you do nothing. So I think we
have a very good study in terms of LOCA. So for all
practical purposes, LOCA for IRIS are gone. The next
one.
This is very important because there is
people still that doesn't know what are the advantages
of an integral reactor. This is a quote that I took
from Nucleonics Week, actually was in the article two
weeks ago. It was the lead article. Second one was
a presentation of IRIS for NRC. Basically they're
saying that the pebble bed can meet its challenge on
having all these things missing but you can not do
that for LWR. The point here is not to compare IRIS
with pebble bed. It's comparing the LWR.
What the perception is, with LWR you can
not take a loss of coolant, a loss of residual heat
removal system, and also measures the core cooling
system. That is true until you know IRIS.
In case of IRIS, IRIS can do that because
the loss of coolant accident is resolved by the safety
of the design. Large LOCAs do not happen, small LOCAs
are taken care essentially with no consequence. For
the residual heat removal system, we have a three
independent diverse system. We have the steam
generators, we have the residual heat removal
interchangers and we have the containment because the
containment is coupled thermodynamically with the
vessel so removing the heat from the containment
essentially goes on removing for the vessel. And the
containment is cooled both by air and water, depending
on the size.
In the case of the emergency core cooling,
core cooling is not needed. We don't have any CCS.
What you really want is, anyhow, the gravity make-up
is available. So that shows that really IRIS is a new
breed of a light water reactor with a much, much
better safety. It's a new dimension. Next, please.
Maintenance is the next thing. In the
case of IRIS, since we're refueling every four years
or so or five years, eight years, it doesn't make much
sense to stop and make maintenance refueling every
three months. Economically it doesn't make any sense.
Besides, it provides access to third world country
proliferation resistance. So what we looked at is to
say let's have maintenance shut down synchronized with
the refueling which means every four years, every 48
months. Next.
This was work done by our team member from
MIT and basically this is the philosophy on the
surveillance. "Defer if practical, perform on-line
when possible and eliminate by design where
necessary." Next one.
Essentially, what we look at is be
accessible on-line or do not require any off-line
maintenance and the first thing is have high
reliability. So this is the beauty of doing the
design now from scratch. We're designing all our
components to have on-line maintenance or a
reliability that exceeds the 48 months. That is built
in our design. It's not done after. We're doing that
now. Next, please.
In the case of the MIT work, a couple of
years ago they looked, actually, it was five years
ago. They looked at PWR and BWR to extend it to 48
months and this is 18 month cycle. These are the on-
line, off-line. What they did, they say let's go to
48 months. What happens is you're increasing the
number of on-line. These are the ones off-line that
can be extended beyond the 48 months. And they had 54
they couldn't handle. So 54 could not be handled for
regular PWR in either way, either on-line maintenance
or extended off-line.
When we look at IRIS, these are regular
loop of PWR. Now let's do for IRIS. Fifty four
became seven. So we now have seven items of
maintenance out of 4,000. We have seven items. If we
resolve them, we have maintenance every 48 months.
And we are working on that. We have several members
of the team are working on that. Next.
This is the one I really wanted to talk
because I gave a very brief rundown. I cut out a lot
of stuff. I'd be happy to answer all the questions
either now or later on. But this is our approach.
The first one is important. We do not need a
prototype. When people say, when are you going to
have the IRIS prototype, I hit the roof. I don't need
a prototype. A prototype is for new technology. A
prototype of a ship import or the prototype for the
leaking matter reactors. IRIS does not break any new
technology. it's light water reactor technology, it's
only good engineering. All you need is good testing,
not a prototype.
So what we have in IRIS is a first-of-a-
kind and, again, we believe that around 2010 or soon
after we can deploy the first of a kind. Future
improvements can be implemented in Nth-of-a-kind.
What we have with IRIS is not a static design. A
module doesn't cost that much. You're talking a
couple of hundred millions or so we're not talking
billions. So we can easily put improvements in next
modules. For example, the extended core reloads will
be in a second or third module. Next.
So you ask, what are licensing challenges
and opportunities versus the Gen IV reactors? First
of all, the first fuel core is well within state of
the art. So we have no challenge whatsoever. It's
just a regular PWR. The reloads and higher enrichment
fuel and they have to be handled through a licensing
extension. We're talking post-2015. So it's not an
issue now. That will come later. IRIS does have a
containment and this containment, in addition to the
classic function, is thermal-hydraulically coupled and
chokes off the LOCA. You've seen that.
The safety by design eliminates some
accident scenarios like the LOCAs, if we have internal
CRDMs and diminish the consequences of others. So
here is a chance for significant streamlining. When
I say the SSAR, simplified safety analysis, I hope I
don't have to go through -- of LOCAs because that's a
waste of time. So that's something that we have to
discuss. How can we simplify because some things do
not happen?
And here is a risk informed regulation.
Commissioner Diaz said this morning one thing that it
just hit me. He said it was deterministic,
experimental and probablistic. But the first word was
deterministic. Deterministically, our accidents for
LOCAs is zero deterministically. So we are starting
with IRIS, we are starting from a very strong basis.
So if we take the safety by design basis of IRIS and
we put on top the risk informed regulation, I think we
have a very, very good safety study which means that
with improved safety we can improve the licensing
position and we can really have that zero emission or
so that we are talking for Generation IV. It was a
lofty goal for 2030.
I believe with IRIS that goal is in the
next 10 years so when we are able to build one because
with this, I think we have a very good chance to go
with no evacuation of the staff.
And here is one question. Our maintenance
is every 48 months rather than 18 months. There are
some regulations that are tied into 18 months. So we
say are there regulatory changes necessary to
accommodate extended maintenance? That's just a
question. I don't think it's a measured thing. And
there are things that was already mentioned before
with the PMBR. We had modules with common parts like
control room and so forth which, of course, have no
intention to be the one control room for each module.
So we've got to have one room for several modules and
so those are things that has to be addressed. Next,
please.
The other question you had was what is
approach to licensing, construction and operation
versus Gen II? First of all for licensing. We do not
see at this time any unique major changes. It's
simplification, streamlining. We don't see any major
changes. There is, however, one thing. The testing
to confirm IRIS unique traits. For example, the
safety by design and the LOCA is great, is based on
first principle. We have three codes independently
producing the same results but we want to have
testing. We want to have experimentally confirmed
data. We do not have to have prototypic testing.
That doesn't make any sense.
We can do scale testing and properly
scaled testing with the proper parameters and so forth
and look at the parameters. That's something that has
to be done as soon as possible because that takes
time. That's a long lead item.
So the safety of the design, the integral
components like the stem generators and some of those
have already been tested, maybe some of the tests have
to be done for the IRIS conditions. But most of the
tests have been already done.
The maintenance optimization, the
inspections. Again, we have the components in the
core for 48 months or so where inspections are
required.
In terms of construction, IRIS is modular.
It's modular fabrication. It's modular assembly. So
it's a different ball game from the Generation II.
You have big items on-site and so forth. Bechtel is
one of our team members and Bechtel has the most
advanced of the EPC tools and we're going to take
advantage of Bechtel EPC for doing our construction.
We've already been talking. Bechtel is already
planning on putting that to full speed on IRIS.
Here is one thing that's interesting. It
is the multiple parallel suppliers. What we have with
IRIS, we have several suppliers all over the world.
For example, out steam generators can be fabricated by
Ansaldo, by Ansel, by MHI. Three different countries.
So what we have here, we have redundancy
of suppliers and something that obviously is an
advantage. If properly managed, it's definitely an
advantage. We have a staggered module construction.
Cost-wise, it makes a lot of sense. What we did --
economically for three IRIS modules and three years
stagger it. Basically, when we started building the
third one, the first one already is producing
electricity and has return. So with the module
reactor you can do that. It's nothing different. No
pebble beds to sit in, any modular design is a logical
thing to do. We stagger it.
In terms of operation, we have an extended
cycle length with a straight burn and we have the
maintenance no sooner than 48 months. That is
different, of course, from Gen II. And we have
refueling shutdowns. Right now it's five years.
Eventually after the reloads we can push up to eight
to 10 years.
These things combined means there's a
reduced number of plant personnel. We're not going to
have 1,000 people at IRIS. No way. You're probably
talking one-tenth of that. So it really has quite an
effect on O&M costs. And we have a multiple modules
operation which again is different from Gen II. And
I'm not talking a twin you may think a part of three,
five or more IRISes.
Next, please. Now what about the
schedule? This was your question. Okay. The two key
dates for the 2005 SAR. A little more important is
the 2007, 2007 is an ambitious objective.
Now how can we meet that? Several things
have to happen. First of all, the lead testing we are
to initiate by early next year. The testing takes
time. If we don't start at least the planning, the
analysis, all of that by early 2002, essentially this
date is going to slide to 101 because obviously we can
not have signed certification until we have the test
results. In testing, you can't accelerate this up to
a point. So this is one key thing.
The second key item is the consortium at
the end of next year is to decide yes, give the
blessing and go ahead with commercialization. The
third thing is a continuous NRC interaction. Having
an SAR by 2005 means that we interface with NRC and
ACRS from beginning in a few months continuously. So
when we plop the SAR on your table, you already know
what it is. It's not something, good reading when you
go to bed for the first time.
That way it only takes two years, 2005 and
2007. If this you see for the first time, no way you
can do it in two years. We'll see each other in five
years. We had that experience with AP600, so we're
learning from experience. So what we want to do, this
is critical, to have an interaction immediately and
continuously. And achieving the deployment, of
course, is the date that you saw this morning to have
a U.S. generator interested by 2005. So those are the
things. Next one.
So in conclusion, IRIS was designed for
Generation IV. Modularity and flexibility addresses
utility needs. Our first customer was DOE. At the
same time we have something that is also commercial,
as I went through. Enhanced safety through safety by
design is a trademark of IRIS. All integral reactors
have that. I think we are the one that really look
and took advantage and I'm sure that what we have done
will be now in other integral reactors because it just
comes out of the geometry. Just comes out of that.
It's physics. It's not clever design. This is
physics.
It's proven LWR technology and again, I
can't stress enough. We have to start testing in 2002
on selected high priority testing. Our first test
will be the coupling of diversity containment just to
show what you what are the predictions. That after
two and a half days, you're core is still under two
meters of water. I believe this is it. Thanks for
your attention.
DR. KRESS: I will entertain a couple of
burning questions if you have any since we're running
really behind.
MR. LEITCH: The reactor vessel in the
drawing looks as though it's large enough to
facilitate internal control rod drives.
MR. CARELLI: Absolutely. Thanks. When
I look at that geometry, it is a waste of a prime
estate to have that room inside of steam generators
full of control drives. The internal CRDMs are set
for integral reactor. Absolutely.
MR. LEITCH: Just let me understand. The
CRDMs are going to be internal? Has that decision
been made?
MR. CARELLI: The CRDMs, yes. I want to
have CRDMs internal. That geometry shows the CRDMs as
regular CRDMs.
MR. LEITCH: Okay.
MR. CARELLI: Because the CRDMS, there are
essentially two designs now. One is electromagnetic
driven internal CRDMs dome by the Japanese. MHI is
the one that's been testing for 10 years and again,
MHI is one of our team members. The other one is
hydraulically a controlled rods. And that is a
solution chosen by the Argentinean, by Curum, chosen
by the Chinese and actually they have a reactor in
Beijing that is running right now, is operating with
internal CRDMs.
So both of them and the Japanese are
planning the internal CRDMs for their MRX vessel. So
both of them are not a far fetch. There's been a
reactor already operating or being designed. What,
right now, I do not know is which one is best or
better. There are two. So I have to decide which
one.
MR. LEITCH: But if they're external, you
haven't eliminated the rod ejection problem.
MR. CARELLI: Absolutely.
MR. LEITCH: If they're internal, you have
introduced some new technology.
MR. CARELLI: Yes. You're absolutely
right. There's a fine line between a deployment by
2010 and 2012 or internal CRDMs. The point again, the
point is we're not starting from scratch. It has been
done. There has been 10 years, 15 years work on that.
What I need is about one or two years to look at
critically, make a decision. At that point, we'll see
how long does it take to implement. Can we make for
2012 or not? That will be the decision.
MR. LEITCH: Okay.
MR. CARELLI: But eventually IRIS is going
and Curum has it, the smartest thinking about for the
integral reactor is a shame to have regular rods.
MR. LEITCH: Thanks.
DR. KRESS: I think we'd better move it on
now. Mr. Carelli will be available for answering
other questions if you have them I think tomorrow.
He'll be here tomorrow. So let's move to the next
speaker which is General Atomics.
MR. PARME: My name is Larry Parme. I
think most of you are new. I don't recognize you.
Perhaps a few I do. But I've been working on gas
cooled reactors for about 25 years, primarily at
General Atomics but I've spent time in Germany and
have worked on pebble bed reactors as well, the THTR
in particular, and also have worked with the Japanese
in the early stages of their high temperature test
reactor.
What I'd like to do over the course of the
next 45 minutes, and if I can make it slightly
shorter--
DR. KRESS: Please do.
MR. PARME: I will try. Next slide,
please. I'll talk about the design description on the
gas turbine modular helium reactor, some background to
it, and then go to the key safety features, talk about
the licensing approach and then the design status and
deployment schedule.
As far as challenges we face in licensing,
I'll point these out as we talk about the safety
features and the licensing approach, and there are
several challenges though I believe most of those that
affect the GTMHR have already been brought up. Next
slide.
The U.S. and European technology, and I
don't have it listed here but I should probably also
mention the Japanese as well. But primarily the U.S.
and European technology gives us almost four decades
of experience which the MHTGR is based.
One of the things mentioned in the earlier
experimental and demonstration plants built in the
U.K., Germany, the U.S. and the THTR, all of these
when they were built, the vision of the future was
scaling up gas cooled reactor technology in the same
direction that water reactors had gone. That is, to
very large, high temperature gas cooled reactors.
Particularly we in the late '70s had PSARs prepared
for Fulton and Delmarva. The Germans were looking in
the same direction and Framatome themselves were
looking in that direction.
But about that time, that is the end of
the '70s going into the '80s, the same technology that
had been developed out of these various reactors, we
had a change in paradigm and took a second look at the
design and decided that rather than scale up to --
Fulton might have been -- I believe it was about a
3,000 megawatt thermal plant and you can figure out
the electric power would have been just under 40
percent efficient. Rather than go that way, we saw a
different way to optimize the characteristics of the
gas cooled reactor and in the U.S. we developed the
modular high temperature gas cooled reactor.
This is a steam cycle plant, the same as
these demonstrations plant and the same as the large
HTGR would have been, but much smaller. The MHTGR
design was developed to early and preliminary design
in the mid '80s when we developed a preliminary safety
information document and a risk assessment on the
design and went for a pre-application review with NRC
and also presented the design to the ACRS.
GT-MHR is an extension of that.
Basically, it builds on the technology of the MHTGR.
I can say there was an equivalent German design, I
believe. Doctor Slabber mentioned it. The HTR module
of Germany. But the U.S. design was a 350 megawatt
core. What we've done is taken that, enlarged the
core somewhat and replaced the steam generator with a
direct cycle gas turbine, a Braten cycle loop in the
other vessel. But it just builds on where we were in
the mid '80s. Next slide.
You can look through your slides and you
can read some of the writing yourself. I want to
point out some of the main features. I guess what
I'll do is you've heard about gas cooled reactors
direct cycle turbines, and I'll try to point out what
differences are between this and the PBMHR.
First of all, a reactor size is worth
noting. It's 600 megawatts thermal. We'll talk more
about that size. Electrical output is 285 megawatts.
The entire primary system, that is the reactor and the
turbine equipment, are all located within a below
grade silo. This silo or reactor building will
contain fission products or other releases, but it is
not a pressure retaining structure. It is designed,
if you pressurize it with your helium, to vent that
helium out and, in so doing, what you do is --later
when I talk briefly about some of the accidents -- is
you eliminate the driving force that could exist to
later carry off fission products when they do come out
of this reactor during accidents.
The other thing I wanted to point out, and
I have to apologize for the lack of detail here to
show it, but within the silo and around the reactor is
a reactor cavity cooling system. You've heard about
the concept on the PVMR. The idea is similar here.
The vessel is un-insulated and any heat radiates off
the vessel rather than heating the concrete structures
here is carried off to the environment.
On the GT-MHR the design of this system
could be water or air or current reference design.
It's an air-cooled system. It's naturally
circulating. It operates all the time. Heat loads
during normal operation are actually higher than the
heat loads during accidents. But you can continuously
monitor it and you know it's working normal operation.
Next slide. Could you use the
transparency I have, blow this up a little bit. I can
see the power point slide better. Why don't you go
back to that. The colors that are sharper there
helps. Taking a look at the overall design, I think
the first thing you notice about the GT-MHR is the
whole power conversion system is integrated into one
large vessel. All of the rotating machinery is
located on a single shaft. That includes the exciter,
the generator, the turbine and high pressure and low
pressure compressors. The shaft is for taking it
apart and doing maintenance. The shaft is separable
at this point below the generator so you don't have to
lift the entire assembly at once. But it's on a
common shaft. Surrounding the rotating machinery then
is the heat exchangers.
Up above there is a compact, high
efficiency recuperator and below that a pre-cooler and
an inner cooler. It's an inner cooled cycle.
Connecting the power conversion system to the reactor
is a small vessel with an inner duct for carrying the
hot gas from the reactor to the power conversion
system and then returning the cold gas back to the
reactor. I have a plan view of the reactor and I'll
show you that in a moment, which will give you a
better idea, but reactor is basically an assembly, a
10 block high core with reflector above and below
built of large, hexagonal graphite block identical to
throe used at the Ft. St. Vrain.
One feature that I wanted to bring up is
not for decay heat removal in a safety sense but for
the convenience of maintenance and operation, the GT-
MHR like a steam cycle MHTGR in the '80s, has a shut
down cooling system, a small circulator and heat
exchanger located in this vessel that allows us to
keep force circulation on the reactor core if one is
doing maintenance or repair on the power conversion
system.
Next slide, please. The annular core is
a key design feature of the U.S. designs, and a couple
of things to note. First of all, the biggest single
thing for the annular core, what is it doing for us?
Why do we do it? It keeps us as we have upped the
power from first 200 to 250 to 350, then 450 and
finally 600 megawatts, it allows us to keep the
surface to volume ratio or the surface area of the
vessel, the outside edge of the core. That ratio to
the power develop constant. It also assures us a
relatively small conduction path between the inner
most heat producing rings and the vessel.
A couple of other things to note on the
design is there are two sets of control rods. There's
a set of start-up control rods which from here I can't
read but they should be located just in the inner ring
of active core. These are pulled out before
operation. They're not used. They stay out. They're
not used in scram. However, the normal operating
control rods are located in the reflector. They're
not in the active core. There's also 18 channels for
reserve shut-down materials and the reserve shut-down
material is just to divert shut-down mean similar to
what's been used in Ft. St. Vrain and also there's a
parallel in the pebble bed reactor and it's just
material. It's pellets, boronated carbon that can be
dropped in the core.
I want to mention a couple of other
things. You'll notice there are a core barrel holding
the core here. With that there's riser channels. The
gas that returns to the reactor is not swept up the
side of the reactor. It's not against the reactor
wall. The reactor wall is exposed to it but in fact
the return gas comes up this channel and is then put
into the upper plenum. There is a desire to keep that
away from the core. The return gas is just over 900
degrees Fahrenheit. It is a high temperature vessel.
It does not use LWR materials. A nine chrome vessel.
Yes, nine chrome does need to be qualified for ASME
but the data is available.
Next slide. Shouldn't be any surprise
here. Key to both the economics and the safety of the
GT-MHR is coated particle fuel. I hope I can go
through this quickly, but I'm going to go over it
because it is so key to the gas-cooled reactor.
You've heard about the coated particle fuel, whether
it be uranium oxycarbide or UO2 fuel laced in a buffer
and then multiple layers surrounding it. I want to
emphasize. These little particles are really tough
things. They'll stand up to internal temperature
pressures of about 2,000 PSI. You've heard about the
temperature capabilities. I remind you. The case of
our reactor, those particles about the size of a grain
of salt or sugar are compacted with graphite pitch and
then that's baked and formed into rods. The rods are
placed into alternate holes in these fuel elements and
then the fuel elements are stacked up to make the
core.
Next slide, please. Just a couple of
words on the overall cycle. I mentioned it's a gas
turbine cycle. Exit temperature from the reactor is
850 degrees Centigrade. About 1,560 degrees
Fahrenheit. It's quite hot. With the fuel, we're
able to use these temperatures and it's quite
beneficial in the Braten cycle. The temperature and
the pressure is dropped by about a factor of two going
through the turbine. The turbine is a 600 megawatt
turbine. About 300 megawatts is going to the
generator to produce electricity. Roughly 300
megawatts is going down to the turbo machinery to
bring the pressure back up.
When the gas exits the turbine, it's still
rather warm. About 900 degrees Fahrenheit. Rather
than send that to a heat sink or try to compress it at
that temperature, it's passed through the recuperator.
At the recuperator we bring the temperature down to
just about 250 degrees Fahrenheit. At that point it
passes through a precooler where it's brought down to
room temperature. At that point we can more
efficiently compress the gas. You go through the
first stage of compression where not only do we raise
the pressure but we also heat the gas. Again, to keep
the efficiency of compression down, we take the
temperature back down in the intercooler, pass it
through the high pressure compressor and bring it back
up to the core inlet temperature of just 1,000 PSI.
At that point, we take the gas back, pick
up the heat that we took out of the turbine exit gas,
not waste it, and then pass it back through to the
reactor. Notice that when I've come down here I've
picked up the 300 megawatts that I passed down the
shaft. You're looking at the heat balance here.
There's 300 megawatts that's lost out the heat sinks,
300 to the compressors and the turbine.
Moving on to the safety, the next
viewgraph. I wanted to emphasize again the
fundamental change in design philosophy that came
about for these modular reactors in the early '80s.
If you look at the history of gas reactors built in
the U.S., be at Peach Bottom, Ft. St. Vrain, or the
large HTRs that were in the design stage, you'll
notice one thing in common with all of them. They
have an L over view ratio of about one. It's
efficient neutronically. It's also felt to be
efficient economically and keeping the vessel down and
cost down.
The penalty that was being paid as these
things were scaled up is you can see that the maximum
core temperature and a loss of cooling, loss of
coolant accident is you've got ever rising fuel
temperatures to the point where Fulton peak
temperatures predicted were just under 4,000 degrees
Centigrade. What we've done is we scrapped the idea
of trying to gain the economics in that scaling.
Instead, if you look at what the modular reactor is,
you see a very long thin core and then if you think
about the annular core, too, you'll realize just how
much the geometry has changed and, in fact, the
economic penalty that could be paid.
However, what the thought is with a design
where we're assured that regardless of the accident or
the accident conditions that keeps the fuel below the
temperatures at which you'll get gross fuel failure.
The idea was to gain the economics, keep the costs of
the plant down by simplifying the safety systems, the
complexity of plant operation, making it simple.
Next slide. I think you may have seen the
same figure cast somewhat similarly, but it's a
summary of tests that have been run in primarily the
U.S. and Germany. There's also some Japanese test
data in my figure. What you see is all the test data
on these TRISO coated particles show that for
temperatures below 2,000 degrees Centigrade, there's
just no experience of these things failing at those
temperatures. The question was asked earlier, what
about the ups and downs, the transients in normal
operation? The test data have looked at Ft. St. Vrain
fuel. Going up and down in temperature here has no
effect on failing. Repeated cyclings at low
temperatures do not affect these results.
We have established, and I notice PBMR has
established similar goals. For a design goal but not
actually a safety limit per se, but as a design goal,
we've elected to keep the accident temperatures below
1,600 degrees Centigrade. But I want to make it clear
that 1,600 degrees Centigrade is not a magic
temperature. You don't go to 1,601 or 1,650 or even
1,800 degrees Centigrade and these particles to burst
or anything like that. There's a time and temperature
effect that occurs as you start going to higher
temperatures. The time is not very long when you get
up to temperatures well in excess of 2,000 degrees C.
But below 2,000 degrees Centigrade, it's a time and
temperature effect with degradation of the silicon
carbide.
You notice the maximum peak temperature is
well below that 1,600 degrees and, in fact, the
average core temperature is below 1,000 degrees C.
during normal operation. Next slide.
Just summarizing where the design takes
us. You can look for what I would consider to be
worst case accident. You're starting with a maximum
temperature of 1,200 degrees Centigrade and if you
assume we lose the coolant circulation, we don't have
a lot of redundancy in coolant circulation. If you
lose that, there's a sudden drop in the maximum
temperature and that's just the drop in the profile
you get from fuel at power where there's a heat flux
going out to the coolant. You had a quick drop in the
maximum temperature and then there's a slow rise as
the fuel heats back up. You get natural circulation
within the blocks. You redistribute the heat. You
eventually heat the vessel back up and you reach a
point at which you just are radiating the vessel to
the cavity cooling system.
If you postulate that in addition to the
loss of force cooling that you also lose all the
coolant, same effect occurs. First, the fuel
temperature drops. Then it slowly rises and then over
a period of days it continues to rise in the center,
but you reach a point at which the heat is just
conducted through the graphite blocks booting the
reflector. There's radiation across the gaps to the
core barrel in the vessel, and then that heat is
radiated again to the reactor cavity cooling system.
Even if you assume that the reactor cavity
cooling system fails, the effect on core temperatures
is rather minimum, at least for a period of days. The
vessel gets hotter, the surrounding structures get
hotter, and I'm not claiming that loss of that cavity
cooling system is something I'd want to deal with on
a design basis event, but the fuel temperature is
relatively insensitive to it as you heat up the
structures that surround the vessel.
Next figure. In summary, the real safety
approach on the GT-MHR is keeping the fission products
right within the particles. Worse case fuel
temperatures are limited by the design features of gas
cooled reactor and really the properties that we've
got, the low power density, the low thermal rating per
module, the annular core and then passive heat removal
to outside the vessel.
Finally, and something I didn't bring up.
Okay. I'm sure that any number of reactors can shut
down without rod motion. All I'm mentioning is that
the thing has a negative temperature coefficient,
like any other commercial reactor in -- I hope -- the
world today. But there's something special about
this. In the gas cooled reactor, there is such a
large margin between the normal operating temperature
of 1,000 degrees Centigrade average core temperature
and the point at which the fuel starts to fail that we
really have the ability to utilize that negative
temperature coefficient and, in fact, if you just flip
back to the preceding viewgraph, at least up until
about 35 hours, at which point you start to get xenon
decay, the effect of inserting the rods or not
inserting the rods is not noticeable on the graph.
The transients are exactly the same. The maximum
temperatures. In fact, all temperatures are the same.
The reactor just shuts itself down.
If you could flip two forward. I want to
talk briefly about the licensing approach. I think
this is something that we and PBMR share in common, a
concern with the licensing approach. I tried to make
the point that we've taken a fundamental change in the
whole design philosophy. The large HTGR, the PSAR we
are preparing for Fulton and Delmarva, the licensing
at Ft. St. Vrain follow the framework that was used
for water reactors and then rarely with just some
exceptions and it was small.
But this approach is so different that
going through the list of general design criteria or
all the precedents for LWR is frustrating, it's
counter-productive and there is no guarantee that it
is either necessary or that it's sufficient and picks
up the important things for the GT-MHR.
In the mid '80s on the MHTR, our steam
cycle plant that I referred to, with DOE sponsorship,
both in the design and the licensing approach, we
started with a clean sheet of paper. The approach
used. It says PRA. I want to make it clear. It was
PRA techniques. Yes, we had a risk assessment of the
plant, too. But it was using risk assessment
techniques to systematically study what was important
in the plant, what were the safety functions? What
safety functions were needed to satisfy what goal?
And reconstructed the licensing bases. This approach
underwent pre-application review by the NRC and was
also viewed by ACRS.
Some of the main points of it were, first
of all, we looked and revisited. What are the
criteria, the safety goals, top level regulatory
criteria that we're striving to meet in the first
place? I'll come back to that topic in a moment
because it's key to be able to go through the rest of
the steps.
In addition, what we did, even though this
was using PRA techniques, we wanted to come up with
bases that were familiar to the NRC, things like
licensing bases events or design bases events, if you
will, equipment safety classification, the design
conditions that go with our safety equipment, and then
design criteria, if you will. And I'll talk about
these in a moment. But rederive them for the MHTGR.
Next slide.
Top level regulatory criteria. When you
go, if you're a gas cooled reactor person, when you go
to the body of regulatory guidance there is, it's
confusing, it's frustrating, in fact. We went back
and looked at the various statements and tried to find
things that really said how safe is safe enough?
Somebody doesn't like the term safe enough. Choose
your own, but we're trying to find some benchmarks to
work for. We looked for direct statements of
acceptable consequences or risk to the public or the
environment. We tried to find statements that were
quantifiable. We needed something that we could say.
Hey, either we were that good and we were that good
with margin, and it should be statements that were
independent of the plant design. Don't tell me that
I need an emergency core cooling system to back this
up. It doesn't help me much and it doesn't mean much
to my reactor.
These are not all the top level criteria
that we uncovered in the '80s, but they were the
limiting criteria as far as the design of the plant.
I'll come back to these criteria in a couple of
moments. Next slide.
Also, having gone through this evaluation
of the plant and starting with our clean sheet of
paper, we had gotten a handle on the safety functions
that were important to the gas cooled reactor. We
understood what criteria we were trying to meet and
then we developed licensing basis events that were
basically off normal or accident events used for
demonstrating design compliance with these criteria.
What we were doing is we were looking at the safety
functions, we were looking a range of phenomena and a
full range of frequency and trying to find what were
challenges to our safety functions that would
challenge staying within the regulatory criteria and
then defining using our PRA entries, if you will, the
types of challenges you could have and construct these
events. This was done and something that would be
very similar, do a water reactor. You could almost
look at them after the fact as deterministic events.
After that, we collectively analyzed in
the PRA all those events to show compliance with the
safety goals. The licensing basis events encompassed
anticipated operational occurrences, design basis
events and then something we call emergency planning
basis events and we'll come to that in just a moment.
Next slide.
I think this figure gives you a better
idea of what I'm talking about. What we did is I have
a frequency versus consequence, and this is whole body
gama dose, plot and what we did is plot the various
criteria we saw. We said 10 CFR Appendix I. That
applies to anticipated releases so we should said it
should apply to basically a frequency corresponding
down to once in a plant life time. So we said once in
40 years. That was our design life time. Then we
said 10 CFR 1000. Those are your design basis events.
We presented arguments why the reasonable range for
that is perhaps between once in a plant life time and
down to 10-4 per year.
Also practice said that for higher
frequency events rather than the full 25 rem of 10 CFR
100, some fraction of 10 CFR 100 is more important so
I believe I have 10 percent of 10 CFR 100 there.
Finally, for lower frequency events, we said the
guiding regulations are the safety goal but you'll see
something else here. The protective action guides for
sheltering the public, and you'll see that plotted
there and it really makes 10 CFR 100 safety goals non-
issues.
We were trying in the '80s and I expect we
would do the same thing in a future application to set
our emergency planning zone at the exclusion area
boundary. So a design criteria for us was to show
that there would be no doses even for rare events,
emergency planning basis events, that would exceed the
protection action guides. So that's the lines here,
the criteria, that's these frequency ranges we had
proposed. Finally you see, using the PRA, how we had
defined these events. These are not quite all the
events.
The only other thing I want to point out
so you understand our use of PRA and our what I would
say is a risk informed decision but still putting it
in an appearance that looks somewhat deterministic.
You notice all these accidents here and they actually
have zero dose. Those are not just the next order of
magnitude down. One of the key things in the risk
assessment that was done for the modular reactor was
done early in preliminary design and we were trying to
set our licensing basis with it, so it wasn't just a
matter of quantifying those event sequences that led
to releases. We assessed every phenomenological
challenge of importance and defined as events not only
those that had the highest releases but those that
represented unique phenomenological challenges to our
safety functions, and we felt that was an important
part of putting the framework together that the NRC
could live with.
Next slide. There's a viewgraph floating
around, if anybody is interested, that goes much
further than this but it didn't show up on the screen.
I thought there's no point in putting it up. But for
safety-related systems, looking again at what should
be safety-related, we said it seems from practice that
in general what's done is safety-related items in
water reactors are those items that are required for
your design basis events. Those items that are
necessary to show that you meet 10 CFR 100. We took
the same approach with this start of our safety
functions and then building down further we derived
those items in the GR-MHR which we claimed were
safety-related and would be subject to the same rules
as safety-related components in other reactor types.
Next slide.
So I've been talking about something that
was done in the mid '80s. How does this apply to the
GR-MHR? Well, the process is absolutely generic and
should be directly applicable to the GT-MHR. Our plan
is to pick up where we left off before. The prior
application of this to the MHTGR did not show any
great sensitivity to what happened in the steam cycle,
the power conversion equipment there. I wouldn't
expect a lot of changes when we apply this method to
the GT-MHR but there might be some differences in the
licensing basis events and perhaps safety-related
equipment.
Specifically, there's a potential for new
initiating events because of the large and higher
energy rotating equipment that we have within the
primary coolant. Certainly recognize that. There's
some potential for different consequences because of
the higher core rating. Even though it stays within
1,600 degrees Centigrade, the same maximum
temperatures the MHTGR had, it's nearly twice as
large.
Finally, water ingress events in the MHTGR
were a primary contributor to release. In that
assessment, we would expect that our licensing basis
events involving water would be very unlikely and
probably be much less risk important. Next slide.
The GT-MHR is now being developed in an
international program. This is being done in Russia,
primarily centered in Nishni Novagrad under U.S. and
Russian federation agreement and for the purpose of
destroying weapons grade plutonium. Program is
sponsored jointly by the U.S. DOE and Minatom, but
it's also supported by Japan and -- that should be
France rather than the entirety of the European Union.
The conceptual design is completed and we
expect to have preliminary design complete by early
2002. I was just in St. Petersburg a couple of months
back and it's quite impressive. A dollar goes a long
way in Russia. There is a large staff, and they're
moving along aggressively. Next slide.
The program is set to design, construct
and operate a prototype module by 2009 in Thomps. We
would also in Russia design, construct and license a
plutonium fuel fabrication facility in Russia. The
first four module plant would be up and operating by
2015 with a total plutonium consumption of 250
kilograms a year.
Just as a point of interest about GT-MHR
in Russia. Fuel contains no fertile material. It's
pure plutonium, weapons grade plutonium. This is not
like burning plutonium with MOX or anything. There's
no fertile material to make more plutonium, so it
destroys it and in a burn up you get better than say
on the order of 90 percent or better plutonium 239
consumption. Next slide.
Obviously, plutonium 239 and plutonium
cores are not of interest here in the U.S. to our
commercial program. So how does this international
program relate to the commercial reactor that I'm
talking about? It's basically designing a uranium
fuel core in the U.S. to replace the Russian plutonium
design. Next viewgraph. That's really the big
picture, but there are a few other things. We are
working with potential users of the technology to
define the requirements appropriate to the U.S. We
would anticipate doing the safety analysis and, of
course, the licensing submittal would be done out of
the U.S. but we would imagine doing the safety
analysis ourselves, even though we may well build on
analysis done by the team in Russia. Any performance
assessments would also be done here in the U.S.
Construction could begin with an aggressive schedule
in as little as five years here in the U.S. Next
slide.
I have a schedule here that hopefully you
can read at your place. It doesn't look too clear up
on the board. It relates the two programs. I'm going
to have to move away from the microphone. I hope you
can still hear me because I can't read it from there.
I think the key thing to note here is the relationship
between the two programs. Right now the intent is
that the Russian program sets and covers the cost of
design but in more than design, it especially gets
much of the component testing we want done.
Construction license is looked for in
Russia in about 2005 and the first prototype is built,
completed 2009. If you look down at the U.S., we're
talking about -- and this is the aggressive schedule--
but we've looked at it and bellevue that we can have
the construction and start up by just about a year.
Much of the safety analysis was already done in the
early '90s. Actually a 600 megawatt core was analyzed
by General Atomics in San Diego. So we're really not
starting from scratch. Much of the work was done in
'92, '93, '94 time frame. Putting that together and
putting it together with information we would get from
the Russians leading to a first plant by the end of
the decade.
Particularly vague in this is the question
of construction, combined operating and construction
license and credifiction. The goal here is clearly to
get a certification for the design. The current
thinking though is the application and that's key to
the program -- but that the application up front would
be for a combined operating and licensing license with
the eventual goal of design certification, but that is
one of the things we're looking to discuss in the pre-
application discussions with the Commission staff.
The other thing we're very interested in and is
unique to this program and we wish to discuss with the
staff is the question and possible pitfalls of
bringing what was once U.S. technology back to the
U.S. from Russia and one of the things we need to
watch for. Clearly, the more we can bring back from
the Russian Federation, the more smooth the path for
this program. I will say the Russians are not off
working on their own. The program is managed by DOE
and they are very interested in potential market
applications and are looking at, if not using, U.S.
codes and standards in the design of the components
and are continually asking us about U.S. safety
regulations so that this could go back.
Last slide. In summary, GT-MHR is rooted
in several, almost four decades of international
technology and it builds directly out of the 1980s
MHTGR experience. It represents an optimization of
characteristics inherent to gas cooled reactors or at
least high temperature gas reactors going for both
high thermal efficiency with the Braten cycle, the
ability of an all refractory core to go to throe kind
of temperatures, but also uses those characteristics
to have, I believe, simple, easily understood, assured
safety. And finally, international program facilitates
near-term deployment of this.
DR. KRESS: Thank you. I think I'll
exercise the prerogative of the chairman and ask the
first question. For light water reactors, the safety
goal that you have of 5 X 10-7 for early fatalities.
You hear statement like well, that's for light water
reactors because we can live with that number because
we have some idea of what the uncertainty is in the
determination of it. But because those uncertainties
are pretty big, we hear statements like well, we're
going to not let you do that all with preventing the
core damage. We're going to make you have a
containment because of uncertainties. There's no
quantification in my mind of what that uncertainty
level is where you no longer have to have a
containment. How are you going to deal with that
concept in the regulatory arena?
MR. PARME: I've heard that. I've heard
those kind of questions multiple times. In the '80s,
what we submitted first of all is we argued that the
goal of the NRC should be to assure the safety of the
public, environment if that be also the case, but the
criteria for the top level regulatory criteria and
going and giving me a criteria on core melt or core
damage is not really telling me anything about how
safe you want the public. I will admit they didn't
full accept that response, but in the case of the high
temperature gas cooled reactor, I'd come back in a
second. Perhaps it's not such a concern if something
like that were imposed on me.
In all of the accidents -- and some of the
accidents I plotted up there. You'll notice all of
throe things are less than a rem and typically they're
on the order of tens of millirems. Some of those
things include assuming that in the steam cycle plant
we had lost all electric power on one module, took a
break in a steam generator, lost our forced cooling,
started pumping steam from one module back to the
others for hours on end with nobody taking action.
Those are still the kind of doses we got. There's no
damage to the core.
However, I will add, we mistakenly in the
mid '80s said, what do you mean by core damage?
There's no damage. The graphite will stand up to
5,000 degrees Fahrenheit or more before it starts to
sublime. It won't be damaged. There's nothing here
you can get temperatures like that. Well then they
started redefining it as a dose over 100 millirem or
something like that.
I think the argument is tell me how safe
you want me to be. If Generation IV or if these newer
reactors are supposed to be quantitatively safer --
DR. KRESS: If I tell you how safe I want
you to be at some confidence level, will you be able
to give me the uncertainties in your determinations?
MR. PARME: I can certainly try it. In
fact, the submittal I will give them, the accidents we
submitted to NRC on MHTGR were not quote
"conventionally conservative analysis." They were run
statistically and we used Monte Carlo methods to give
them. I think we said what do you want? They didn't
know. We gave them 95th percentile confidence on the
results we give them. If you want more confidence
than that, I can do it. Most of these accidents are
simple enough to analyze that I can actually --
DR. KRESS: That's the problem. I don't
know what confidence I want. I don't know if anybody
does.
MR. PARME: I don't know but I think we
can perhaps talk and work to what amounts. At this
point in time, what would give you reasonable
confidence? And this whole method I went through
quickly but it does include -- classified events and
meeting the goals. Confidence in the answers.
DR. KRESS: I'm quite pleased to see your
frequency consequence curves because some of us on the
ACRS think that's a good way to go, particularly when
you don't have core melts.
The other question I wanted to ask you
that may come up, I don't know. Chernobyl had a lot
of graphite and it apparently burned. You have an air
cooled cavity where you're encouraging natural
convection. Is there an issue there?
MR. PARME: Let me say a couple of words.
In the NRC interactions we had in the '80s, we did do
some analysis of broken vessels, failed vessels, and
air ingress. First of all, reactor grade graphite in
the U.S., H451 for pebble bed modular reactor. I'm
not sure what the grade is but typically the German
graphites. They will not burn in the sense of a self-
sustaining chain reaction. Coal has --
DR. POWERS: Why do you say that?
MR. PARME: I will say that exactly as
follows. Coal will burn, charcoal will burn because
of its impurities. Reactor grade graphite -- and
there's been tests done at Oak Ridge where an
oxyacetylene torch was placed on the graphite.
DR. POWERS: It's a totally ridiculous
test. You're talking of the difference between a
point ignition and a homogeneous ignition.
MR. PARME: Okay. In the case where we
analyzed air going into the core, and here I'll speak
only of the blocks, the reaction rate is driven by
temperature that is held up by decay heat. The heat
generated from oxidation of the graphite was about--
and it's been 10 years -- but on the order of 10 to 20
percent of the total heat generated was -- in fact, 10
percent or less was due to oxidation. Also the
reaction then becomes oxygen-limited as the air passes
up the channels. We did an analysis assuming a vessel
failure in that cross vessel that connects the two
vessels and then assumed that the silo was open and
you could get air in that. What you would get was air
coming in the hot duct, going up through the core,
down through the vessel and out the return duct.
We did the analysis for about 24 hours and
I think we did it beyond that but, once again, I'd
have to go back and look at the calculations, though
it is in Appendix G, I believe it is, to the
preliminary safety information document that was
submitted. I think you see there's no increase in
particle failures, but what you do is you are getting
releases. They're pretty substantial because they're
a driving force and the releases you're seeing and the
doses that come with it are due to picking up the
contaminants that are within the graphite. As you
oxidize the graphite, there are contaminants there.
They were -- I want to be careful about quoting the
doses. I rather doubt that they stayed within the
protection action guides for that accident. However,
they were well within the limits of 10 CFR 100.
My comment on combustion was implying just
primarily that the reaction is driven by decay heat.
It's not as if you had a charcoal pile there. But you
will oxidize. There's no question you will oxidize
graphite.
Incidentally, in the large HTGR, the
approach to that, if you got a break and the primary
cooling system got air in the system, it's a coolant.
What you do is if you've got a circulator, you turn
the circulator on and you cool the core with air.
Once the core temperature is down, it will not oxidize
so you just run the circulator. That was the design
approach for the large HTGRs. If you had a circulator
running, that's how you do it. You just turn the
circulator on, blow the air around and cool it off.
DR. POWERS: I'll also comment that you
need to be very careful about reaction kinetics and
graphite. They are catalyzed and they catalyze by the
impurities he speaks of. One of the most effective
catalysts that I know of, by the way, is cesium.
MR. PARME: It is effective. You're quite
right about that. Fortunately, while dose-wise it's
a major contributor, a fairly small amount of it
that's in the graphite, but you are correct. It's a
very capable catalyst.
DR. KRESS: I think with that, even though
we're running considerably behind, that I'll take a 15
minute break. So please be back at 4:15.
(Off the record at 3:59 p.m. for an 18
minute break.)
DR. KRESS: Can we resume our meeting,
please. I think we're on the agenda where we're going
to hear a presentation on the advance liquid metal
reactor ESBWR from General Electric.
I would like to note for the record that
our member Peter Ford, who shortly was an employee of
General Electric, has a conflict of interest on this
subject and this is a formality we have to do for the
record. With that noted, I'll turn it over to our
next speaker.
DR. RAO: My name is Atam Rao. When I
joined General Electric Company after doing my Ph.D.
at Berkeley 27 years ago, they said that nuclear was
going to come back in five years. Still waiting for
that. I hope when it comes back there'll be nothing
but a slew of ESBWR orders followed by B.S. prism as
we run out of fuel with the light water reactors.
Next slide, please.
ESBWR is a design that is based on the
SBWR which was a 600 megawatt design and the ABWR. It
basically uses a lot of the components from the ABWR.
It's a natural circulation reactor. It's got a lot of
the ABWR components but a lot less of them. It's got
passive safety systems which were reviewed by the NRC
for the SBWR program. We have done a significant
optimization of the building and the structures to
improve the overall economics and the construction
time. It's been an eight year international design
and technology program, and the goal of that program
was to improve the overall performance, safety and the
economics. We did stop the SBWR program because at
that time we realized that it would not meet the
market conditions of overall economics.
The major regulatory issues are right here
on this first chart for the ESBWR. How much use can
be made of the ESBWR review done by the NRC? We've
done an eight year testing program. Is that enough?
We've done an eight year testing program before that
for the SBWR. So there's an extensive test program
which has been reviewed by the NRC.
However, I'm not going to tell you how
long it's going to take to license this plant. A lot
of the previous speakers did tell you that. In fact,
that is our biggest question at GE. We know that our
experience with the last round of certifications was
that it took eight years. I think the AWBR took 10
years. And the question is really how high is the
hurdle and will the bar be being raised every time as
you go along.
We believe for this plant design we have
done all the testing. The design and the technology
is complete. How long it'll take to get it through
the certification hurdle is still an open question.
The next charts shows that General
Electric Company had a steady program of evolving the
designs, improving the reactor designs. All the
actual designs started from the initial submarine
reactors and we have been simplifying the design.
It's interesting to see that a lot of the advanced
designs that were presented earlier are either called
integral design or direct cycle designs. We've had
that for quite some time. Those were Generation I
reactors for the boiling water reactor.
The one that I would like to mention is
the ABWR. The plant is licensed, designed and
operating. When it comes to regulatory challenges, we
still believe that the issue of COL and ITAACS is an
issue that needs to be addressed. Very generic to all
of the plants, whether they come up for application in
the U.S. The ABWR, we believe, hopefully will be the
first in line to go through that process. The ESBWR
evolved as we further simplified the ABWR. Next
chart.
We also had an evolution of the buildings.
There is not enough time to, like Rodney Dangerfield,
I guess, if you're from California, you get little
respect. You're last. You only get half the time to
present each one of your reactor designs, but that's
okay. They are so simple, it doesn't need much time.
The ESBWR design has evolved over the years. We have
evolved containment building also. The ESBWR followed
from the ABWR, the SBWR and we had an earlier design
of the ESBWR and now we are in the process of changing
the building design.
The next chart is direct cycle, boiling
water reactor. You pull the control rods, water
starts boiling and turns that steam turbine. Fairly
simple design. Couldn't get any simpler than that.
Next chart, please.
This shows a comparison of some of the key
parameters, just to put it in perspective. I have
shown the SBWR in the middle there and the ESBWR on
the right, the ABWR on the left. It's basically the
same power level as the ABWR, like I mentioned. In
fact, one of the reasons we chose that power level was
we wanted to keep the components the same, the reactor
vessel is the same diameter. We wanted to make sure
we came up with a practical design. Our emphasis is
on something that's practical that commercially
viable. It is an -- circulation reactor so the fuel
height is three meters compared to the 3.7 meters for
a traditional boiling water reactor, and we have about
10 percent more fuel bundles, about 1,000 bundles.
We have reduced the number of control rod
drives which are an expensive component of the design,
and the bottom line is that last item bullet there
which talks about the building size. The cubic meters
from megawatt electric. Like I mentioned earlier, the
ESBWR is the ABWR, just less components. And that
shows up in that final number. What we have is any
less systems which results in an overall smaller
reactor building and containment. Next slide, please.
Like I mentioned, ESBWR is a program
that's an extensive program. In fact, it's been going
on. We have not talked about it much publicly. It
had four elements. One was the overall requirements,
design, the technology and what we were doing relative
to licensing. The requirements were based on utility
requirements. We've had a utility steering committee
running this program for the last eight years. We
have been making major changes in the overall design
to improve the economics, improve the margins and
improve the performance. We've had an extensive
technology program with a lot of testing. We extended
technology beyond that.
For the SBWR there was a major test
program called TEPSS and this one NACUSP and TEMPEST
is ongoing and basically the reports that were
produced for the SBWR program as a result of the
additional testing done in support of the ESBWR. The
ongoing program, Phase 3, is a program where we are
improving the overall plant margins, completing some
of the testing and completing the technology reports.
Next phase would be the safety analysis report,
SAR preparation and, like I mentioned earlier, the
thing that we can define accurately at GE is how long
it takes to produce it, how long it takes to review
it. Next slide, please.
The ESBWR design is based on the SBWR.
Shown on that chart is the SBWR safety analysis
report. So there's a lot of paper that's been
produced, a lot of design that's been done, and it's
also using a lot of the ABWR components. Next chart.
It's a natural circulation reactor which
is standard BWR technology. It's really hard to
imagine an integral vessel where you pull the control
rods out and the steam is produced at the top. It's
hard to imagine anything much simpler than natural
circulation BWR vessel. 7.1 meter vessel. It's about
27 meters tall. Next chart, please.
The safety systems are inside the
containment. The safety systems are fairly simple.
Up on the top right hand corner, the blue is what we
call the water make-up system. It's 1,000 cubic
meters, fairly small. You don't need much water.
You've got a standard suppression pool. You can see
the quenchers from the safety relief valves filling up
there in green. It is raised off the base mat. It's
the same size as a standard boiling water reactor.
The interesting thing about this design is
that all the safety systems are inside the containment
and the decay heat removal heat exchangers are setting
on the top off the drive wheel above that pool up
there. Next chart.
This shows what we've done over the last
eight years, a comparison of the reactor and
containment building of the 600 megawatt SBWR and the
1360 megawatt ESBWR. You can see that the buildings
got much smaller. We have done significant
optimization of the building and the systems. Next
chart, please.
ESBWR design philosophy compared to the
SBWR has been to increase the margins. Even though we
doubled the thermal power, the overall margin, both
flow -- next chart, please. What we did was we also
did an extensive test program. In the handouts are
actually more charts than I'm using in my
presentation. There are about twice as many. They
give a lot more detail on the background of the
additional testing that was done.
What I mentioned earlier was the overall
design philosophy has been to increase the performance
margins. On this chart out here is shown some key
typical parameters for the plant performance. The
natural circulation flow rate, whether or not the
safety relief valves open following a transient,
whether minimum water level is falling in accident and
what the containment pressure is following an
accident. And generally the results show that ESBWR
performance has been improved over the SBWR design.
So even though we went up in power level, we were able
to increase the margins which was a significant
improvement of the overall design of the passive
plant.
Next chart, please. People have been
using terms like minimizing initiating events. What
we've done in this basic design is that the ESBWR has
no safety relief valve opening following a reactor
isolation, for example. This shows the reactor
pressure following a reactor isolation. Next chart.
We have adopted passive safety systems,
not as a religion. Passive safety systems were
adopted only if they simplified the plant design.
It's interesting. The idea of the optimized plant
design would be where the plant systems and buildings
were set by normal operation and you got the safety
systems for free. When we looked at the cost of the
safety systems, we found that they are reduced so much
on the ESBWR compared to the total plant design that
we've essentially gotten it for free. So it seems
that it'll be not possible to optimize or reduce the
cost of a design like the ESBWR much further.
This shows a schematic of the safety
systems, and there's not enough time to go into how
the safety systems work, but let me just mention,
since some of you might have heard about the SBWR.
The safety systems are essentially the same as the
SBWR. Here's what I call the water make-up pools
which run the reactor vessel and when you depressurize
the reactor vessel. These are decay heat removal
condensers up on the top out here. This is for
removing the decay heat following a reactor isolation.
On the left side you find the passive containment
cooling system, heat exchangers similar to the SBWR.
The design is the same. The components are the same.
We are using the same basic design philosophy as we
had for the SBWR. So if someone were to ask me how
long would it take for the NRC to review this, my
guess is maybe a couple of weeks. As long as it takes
to read the reports because there is not anything
that's new and it's been backed up by additional
testing. Next chart, please.
This just shows another plot of the water
level following a loss of coolant accident. Again,
the key thing that I want to leave you with, the
message I want to leave you with is that this was the
SBWR. This is the top of the active fuel. This is
functional time. The water level above the top of the
active fuel. The ESBWR water level is higher than
that for the SBWR, so we have improved the margins so
it should be easier in the review process. Next
chart, please.
Extensive test program was done for the
SBWR. This shows some of the test facilities. This
is the depressurization valve. This was the ground
water-driven cooling system test facility, and it's
all real stuff. Parts of full size components were
tested. Next chart, please.
The decay heat removal, similar to the
SBWR design. No change in the overall philosophy.
Several diverse means of decay heat removal. Next
chart, please.
Again, this is where we did a lot of
extensive new testing. The SBWR and ESBWR Phase I
test programs are listed out here on the left side.
We have completed some additional testing in the
Phase 2 program which was completed in '99, and we are
doing some additional testing which should be
completed by the year 2002. Again, these are all
confirmatory testing and we don't believe there's
anything that's left out there. In fact, some of our
technology partners kept asking us to define
additional testing that could be done, and we just ran
out of ideas on anything that could be done. So we
don't think there's anyone who can think of anything
else that needs to be done, but we may be wrong. Next
chart, please.
This is a prototype of a vacuum breaker.
I just put these charts in there to show you that this
is a program where there's been hardware that's been
tested. Next chart, please.
Again, there's not enough time to go over
each one of these, but in your handouts there's a
description of some of the test programs that we used
to qualify the new features of the SBWR design. Next
chart, please.
The TEPSS program was a program that was
performed in Europe which was a three part program to
extend to the SBWR database to the ESBWR. What we
tested were some innovations that we made in the
design and also the different scale for the SBWR.
Next chart.
We have an ongoing design program to
improve the economics of the plan further and to
improve performance margins. That should warm the
hearts of regulators as we are improving both the
containment pressure margins and also addressing some
of the issues that some of our European utilities are
concerned about. But at the same time, we are fairly
practical. Our overall goal is to improve the
economics, and we hope to be reducing the cost of the
buildings by 30 percent more while increasing the
margins at the same time. Next chart, please.
We have ongoing technology programs also
which should be completed by 2002 and they should
provide further data for qualification of the computer
codes. And finally, I wanted to leave you with just
an overview just to whet your appetite for the ESBWR.
It is an eight year design program where we have
reduced the components in systems to further simplify
the design. We have reduced the structures in
buildings which we believe will simplify the design.
But our goal has always been to increase the margins.
As I showed you in some of the plots, we have
increased the margins.
The technology program basically shows
that what we've done is increase the margins over the
SBWR and we have qualified the computer codes for the
incremental changes that we made on the ESBWR.
Challenges for the coming year. This is the one, the
BC is the biggest challenge, is how do we cross the
regulatory mine field? We think we've done everything
that we could possibly do that would be needed for
getting this plant licensed, certified. We have the
experience with the SBWR and the experience with the
ABWR. We have two safety analysis reports sitting on
our desk. We have done the testing. The tests were
completed with our partners who were involved in the
SBWR program and we can not put a number on how long
it'll take, what effort it'll take, to complete
certification effort.
In summary, we've completed the extensive
technology program and we believe that the SBWR and
ABWR experience should ease the regulatory challenges.
Again, the number that I didn't have in the charts.
One of the reasons for embarking on the ESBWR program
was to improve the overall economics of the passive
plan compared to the SBWR design and we have increased
the power by a factor of two and have also improved
the economies by a factor of two which is sometimes
hard to do. Economies of scale don't let you do that,
but there are some innovations that we've done which
have allowed us to do that. So that's the ESBWR.
DR. APOSTOLAKIS: What are the most
dangerous mines in the mine field that you feel we
ought to be working on?
DR. RAO: Our experience on the last go
round was that the fact that it was -- I'll say again
-- it's a time and material effort. So there tends to
be no closure when you're having NRC review of the
licensing submittals, whether it's with the national
labs which are consultants to the NRC staff or the NRC
staff. So there is a minimum incentive for closure of
some of the items. That was our experience with the
SBWR in the past.
We don't think there are any technical
issues that are there because we've had -- I haven't
emphasized the international part of our meetings.
Typically we meet twice a year and have 30 or 40
people from national labs and people from all
different parts of industry. So we don't think
there's any technical issues. It's just bringing the
NRC staff up to the same state where we are. That's
one thing.
The other question is do the people who
reviewed the SBWR in the NRC staff, are they still
there? I think some of them are still there. That
would make it go faster. The process of someone else
coming up to the same level of understanding as those
who worked on it is, I think, one of the major
challenges we faced in the SBWR. I remember -- I
don't know whether it was Ivan Catton or someone on
the ACRS. It took several years before we got people
to appreciate how simple our passive containment
cooling system was, for example. It was actually not
a natural circulation system. It was a -- circulation
system. And so if the same members of the NRC staff
are not there, we might have to go through that same
process again.
So it's those kind of institutional
issues, I think, which will be a harder challenge for
us.
DR. POWERS: Is what you're saying that
you can't write this thing up so that people can
understand it clearly?
DR. RAO: No. I am just saying that
someone starting fresh sometimes has some preconceived
notions or concepts about systems work and it does
take some time for people to appreciate it. That's
just human nature. I think it takes time for people
to come up to speed. There is that learning curve.
DR. KRESS: I think the speaker will find
that the climate at NRC now is somewhat different, and
they are quite interested in closure and such things
in spite of the fact that you're from California.
You'll find them quite interested in not dragging out
reviews and getting them done in an efficient manner.
So you may be quite pleasantly surprised if you come
in with an application today.
DR. RAO: You might notice this is our
first coming out also. We have also sensed that there
may be a change and that's why we've been working on
this for quite some time and this is our first coming
out on this design.
DR. KRESS: In fact, your system looks
enough like reactors that NRC is used to that it
almost fits into the regulatory system as it now
exists and may be an easier task to get one of those
licensed.
With that, I'll ask if there are any
questions from the audience or from other members.
Everybody is anxious to get us moving on. Good.
DR. RAO: There is one other issue that I
wanted to mention that's mentioned out here.
Resources. It's still our position that in the near
term what we believe where the resources should be
focused, you know the NRC. It's getting the plants
that are already certified through the ITAACS and the
COL. I mean if there was a choice of where the NRC
spends its resources, that's where we would see
resources being spent. This would come after that.
And after there've been 100 ABWRs built in
the near term and 200 SWBRs after that, the answer to
what you do when the fuel -- next chart, please. What
happens when you run out of all the uranium? We have
something for you for that also. That's the S-PRISM.
It's a liquid metal reactor which is the next
presentation. Next chart, please.
DR. APOSTOLAKIS: How many did you say?
Two hundred?
DR. RAO: How many?
DR. APOSTOLAKIS: Yes. Did you just say
200? In the United States?
DR. RAO: No. I was just kidding. I
don't know how many it'll take before we start running
out of fuel, but this next chart addresses that
question right here. I think NEI said 50. Fifty by
2020. Isn't that right?
DR. APOSTOLAKIS: I have a more serious
problem. The safety goals are stated in terms of
rates per year and if you have 200 units in addition
to what we have now, I'm not sure that the goal should
stay the same, which is now creating a new problem, I
think.
DR. POWERS: George, if you doubled the
number of units that we had operating, it's a factor
of two. We know the safety goal so precisely the fact
two makes a difference one way or another.
DR. APOSTOLAKIS: A couple of 100 I can
live with but if it's a couple of hundred of this, a
couple of hundred of that, as you know, pretty soon--
DR. RAO: The actual numbers, you know, I
think the NEI goal was stated as 50 by 2020. In the
U.S. all plants. We'd like to see them all be ours
but we're realistic.
When you look at this chart of the fuel
availability, it's really interesting to see why we
need the fast reactor. We don't think it's needed
today, but it's a design that we've worked on at
General Electric for many years. Next chart, please.
Not only does it help in extending the
availability of the fuel cycle, it also reduces the
toxicity of the waste and the spent fuel. Next chart,
please.
I'm going to go through these fast. Okay.
Basically, it supports the geological repository
program and it reduces the environmental and diversion
risks, and that's why we think some time in the future
there will be the need for a reactor like the S-PRISM.
What I'm going to do is give you an overview. Next
chart, please.
What I'll give you is a brief overview of
the design and the safety approach. I'll also give
you a little bit on the description and how it's
competitive, the previous licensing interactions and
the planned approach to licensing the S-PRISM. Just
to put it in perspective. What's different about this
liquid metal reactor compared to the ones that have
seen the light of day earlier? This one, we believe,
is commercially attractive. Next chart, please.
The key features of the design. It's a
compact pool-type reactor with modules of about 300
megawatts electric. It's got a passive shut-down heat
removal system, a passive containment cooling system.
The nuclear safety envelope is limited to the nucleus
team supply and located in the reactor building.
We've also designed in seismic isolators so the
complete nucleus steam supply system. To achieve
conversion ratios less than or greater than one.
Next chart, please.
The design description. Next chart,
please. The power train is shown in this chart out
here. What you've got is a reactor module, the steam
generator, intermediate steam generator, and you've
got reactor vessel auxiliary cooling systems similar
to the cooling system that was mentioned for the gas-
cooled reactors where you have air cooling of the
reactor vessel.
The power conversion system is high grade
industrial standard and it's like any of the typical
plants which don't have direct cycle. Next chart,
please.
Next chart shows some of the key design
parameters. It's 1,000 megawatt thermal reactor
module and the power block consists of two reactor
modules. Its gross net electrical output is about 800
megawatts electric. And the overall plant could be
put together as different modules and you could end up
with about 2,200 megawatts electric, depending on the
number of modules you put together.
The next chart shows a picture. On the
left hand side is the reactor module out there. It's
an integral design. That's a new word that I'm
picking up. It's sort of fairly standard for several
liquid metal reactor designs. This is the reactor
module out there. This is what are the passive vessel
cooling systems and this is the intermediate heat
exchanger on the left side there.
The number of fuel assemblies in the next
chart shows it's 138 fuel assemblies and it's fairly
standard fuel for the liquid metal reactor. Moving on
to the next chart, what I was going to show you was
some of the numbers and the reason for considering the
S-PRISM compared to some of the earlier designs of the
liquid metal reactors. Next chart, please.
What it shows is that earlier designs were
what we call monolithic plants and this is a modular
plant. What it shows is that the cost is
significantly improved, partly because of the learning
curve. Skip the next chart, please. And skip the
next one, also. And put that one up.
This shows a comparison of the Clints
River -- reactor which is a 350 megawatt electric
plant. This shows the footprint. That was followed
by an ALMR plant which was 311 megawatts and, since
then, GE has worked on the design we call the S-PRISM
which is a 760 megawatt electric plant. What it
basically shows is significantly smaller. Produces
twice as much power as Clints River and it's a lot
simpler.
Next chart, please. This design has had
previous interactions and what I show you on the next
chart is what the design and licensing history has
been of this liquid metal reactor. GE PRISM program
was GE funded in the years 1981 to 1984. That was
followed by a DOE program of about $100 million where
the PRISM design was developed and the ALMR program
was one of the designs that came out of that effort.
Finally, when that program was completed, GE continued
developing the liquid metal reactor design and
developed the S-PRISM. What we have out here is a
multi-year program. For almost 20 years we've been
working on this design. Spent $100 million.
And what we have is, on the next chart,
the ALMR which formed the basis for the S-PRISM was
reviewed by the NRC in '93-'94. There was a pre-
application safety evaluation of the ALMR. It
included the staff for the ACRS agreement concludes
that no obvious impediment to licensing PRISM design
have been identified. So what we believe is that the
design out here where, again in your handouts, there's
almost a 50 page handout which goes into a lot more
detail of the design which there wasn't enough time to
cover out here. The design is fairly well advanced
and the approach for licensing the plant is shown in
the next couple of charts.
Next chart, please. Land approach to
licensing the S-PRISM would be shown on the next chart
which is basically a detail design, construction and
prototype testing. This shows the schedule for that.
It is a fairly long schedule which would take up to
about 15 years, but again, as we mentioned earlier,
the need for this basically arises once we start using
a lot more waste or using up a lot more of the
uranium.
So basically in the next chart, the key
issues in a safety review would be looking at the
containment, looking at the core energy potential,
analysis of design basis, team generator leaks, ESA,
nuclear methods, hydraulic methods, validation of the
fuel database and, of course, efficient product
treatment and disposal. There has been extensive
experience with sodium-cooled fast reactors and -- are
expected. But the key issue has always been
commercial viability. We believe this design, when
you look at the compactness and the overall design of
this design, we don't think there's much that's not
known in terms of the overall physics. The main thing
is to build it, test it and test out a prototype and
make sure it operates as planned.
What I'd shown earlier was the overall licensing
approach to getting one of these plants through the
licensing process.
And the last chart is component
verification and prototype testing. This shows the
basic approach that would be needed for licensing this
kind of a plant for testing of a prototype reactor
module. Thank you.
DR. KRESS: Questions, anyone? Comments
or speeches? No speeches. Seeing none, let's move on
then to what might prove very interesting. Some of
the NRC reactions to all this and activities they have
ongoing. So I'll turn it over to whoever on NRC wants
to carry the ball.
MS. GAMBERONI: I'll begin. Good
afternoon. I'm Marsha Gamberoni, the acting Section
Chief in the Future Licensing Organization. You might
have heard the acronym FLOW in NRR. We've a panel of
project managers here today from FLOW to discuss the
issues in our May 1 response to the Commission's
February 13 SRM. The panel members include Nannette
Gilles, Tom Kenyon, Alan Rae and Eric Benner.
Our agenda this afternoon, if you can go
back to the previous slide, includes discussion of the
future licensing and inspection readiness easement,
early site permits, the construction inspection
program, status of the AP1000 review, and regulatory
infrastructure issues.
The next slide shows our organization. We
were established late March/early April of this year.
Majority of the group is on rotational assignments,
but we're currently working on permanent staffing.
Our SES manager, currently Richard Barrett, reports
directly to the Associate Director for Inspection and
Programs, Bill Borcher.
Close near term objectives are to identify
the steps needed to prepare for future licensing
reviews, to determine the necessary resources and
technical skills needed to perform these reviews and
to identify the areas for improvement so that the
reviews can be completed in a predictable time frame.
I'd like to mention that we're working closely with
two other organizations in the NRC, the Advanced
Reactor Group in Office of Research which you'll hear
from shortly, and also the Special Projects Branch in
the Fuel Cycle Safety and Safeguards Division in
NMSS.
I just wanted to mention two meetings that
we have upcoming before I turn the presentation over
to the project managers. We're meeting with the
Commission on July 19 on future licensing issues, and
we are also planning a workshop in late July on future
licensing issues.
MR. GILLES: My name is Nannette Gilles.
I'm what is commonly referred to as the FLIRA lead and
FLIRA stands for the Future Licensing and Inspection
Readiness Assessment. The staff was directed to
perform this assessment by the Commission in their
February 13th SRM, and we were asked to assess the
staff's technical, licensing and inspection
capabilities and identify any enhancements that would
be necessary to ensure the agency would be prepared
for any future licensing activities that would be
ongoing.
This assessment will evaluate a full range
of licensing scenarios. We will be looking at all of
the processes identified under 10 CFR Part 52, the
early site permit process design certification, the
combined license process. We will also be looking at
custom designs and also be addressing the reactivated
plant licensing scenario because we do know that there
has been some interest in that area.
The assessment will also look at the
staff's readiness to review applications and perform
inspections and specifically we are going to look at
staff capabilities, and we are in the process of
assessing critical skills needed to perform these
actions and which areas we may be lacking resources in
some of those skills. We are going to be looking at
schedules, external support from this committee and
from contractors and our external stakeholders, and we
will be looking at the regulatory infrastructure, both
at current rulemakings that are ongoing and we are be
planning for possible future rulemakings that will be
identified during this process. In addition, we'll be
looking at regulatory guidance.
We will be making recommendations in many
of these areas to the Commission, in the area of
staffing, training needed. Obviously there will be
training needed in some of the new technology areas.
We've been making recommendations with regard to
contractor supports, schedules, and again,
recommendations with regard to needed rulemakings and
updating for regulatory guidance documents and
inspection plans. And the schedule currently is that
we will complete this assessment and submit it to the
Commission by September 28th of this year.
I'll turn it over to Tom Kenyon for early
site permits.
MR. KENYON: My name is Tom Kenyon, and
I'm working as a project manager on our early site
permit efforts. Although 10 CFR Part 52 was
promulgated back in 1989, the staff has not received
an application for an early site permit as yet.
However, talking to NEI and other industry
representatives recently, we expect to receive one by
mid 2002, which is why we're in the process of
preparing for that eventuality.
Subpart A of 10 CFR Part 52 allows an
applicant to obtain approval to build multiple classes
of nuclear plants on a particular site, independent of
a specific plant review. And so that allows the
applicant to bank the site for future use for 10 to 20
years. This reduces the licensing uncertainty by
resolving site specific issues early on in the process
before the applicant has to commit large amounts of
resources for the effort.
An early site permit review consists of
three separate reviews. The first is site safety.
Another review is in the area of environmental
protection and the third is in emergency preparedness.
When the staff performs a site safety review, we look
at site characteristics that are specific to the site
such as the seismology in the area, the hydrology,
meteorology, and the population demographics. The
staff looks at these site characteristics to determine
whether or not any of them would preclude building a
nuclear plant on the site.
Then staff also performs its environmental
review. They perform it in accordance with 10 CFR
Part 51 and the requirements of the National
Environmental Policy Act of 1969. NEPA requires that
all federal agencies use a systematic approach to
consider environmental impacts of certain decision
making proceedings. In this case, building a nuclear
plant on the site. So the staff looks at the
potential environmental impacts of constructing and
operating a plant there so it can make an informed
decision as to whether or not it is acceptable from an
environmental standpoint to build the plant.
The staff reviews the emergency
preparedness to look for potential physical
impediments at the site to see if there's anything
that would make it difficult or impossible to develop
and implement an acceptable emergency plan. They're
going to be looking at things such as the population
in the area, ingress and egress routes to the site,
support capabilities and facilities in the area, and
any other things that could affect the emergency plan.
Staff will be working with Federal
Emergency Management Agency and other federal, state
and local authorities to make sure that the emergency
preparedness submittal is acceptable. The staff will
be interacting with the public in the form of public
meetings at certain stages of our review and the
public will be given the opportunity to participate in
the hearing on the application.
Subpart A 10 CFR Part 52 is the regulation
governing the reviews of our early site permits. We
have a regulatory infrastructure in place now to do
these reviews. We have regulatory guides. We have a
standard review plan. We have a recently revised
environmental standard review plan, and we have other
guidance to support our review. We've been talking
with industry representatives and other stakeholders
about the upcoming applications.
We've recently had a couple of meetings
with the NEI Early Site Permit Task Force to discuss
regulatory issues as well as guidance questions, and
we've been told, as I said earlier, that the first
application is expected to come in mid 2002 with two
more coming in 2003 and, despite what the slide says,
there's only one expected in 2004. I apologize for
the misprint. So staff right now is in the process of
preparing for these expected reviews by looking at
resources and skill requirements. We're going to be
looking at what kind of training is necessary to make
sure the staff is ready for the application review.
Next slide, please. The second topic I
was going to discuss is our construction inspection
program. In order to prepare for the actual
construction of the plants, staff is reactivating
earlier efforts that it had in revising its
construction inspection program. The staff was
revising the program to incorporate lessons learned
from our construction inspection activities back in
the 1970s and '80s and also to incorporate any changes
that are needed to support inspections of plants
licensed under 10 CFR Part 52.
The staff has been looking to see what
needs to be done to enhance the program, and we're
going to be doing such things like ensuring that
there's a continuous NRC presence at the site during
the construction of the plant. We're going to make
sure there's a better match of inspector expertise to
the construction activities that are underway and,
very importantly, we're going to be making sure that
the acceptance criteria is more clearly defined for
what the staff is to be inspecting to.
Another issue that's going to be
incorporated involves developing procedures for
inspecting plant components and modules that are built
at fabrication sites that are off site from the
facility and then, after they're constructed, they'll
be brought in and installed at the site. And of
course, we're going to be developing a training
program to train the next generation of nuclear
inspectors.
Most of our focus has been on looking at
the construction activities and inspection activities
of new plants that are going to be coming down the
pike over the next decade, but we recently met with
Entergy Northwest to talk about the feasibility of
reactivating the construction permit at their WNP-1
site in Washington state. They're in the process of
performing a feasibility study that's going to be
completed in August of this year, after which they're
going to make a decision whether or not it's
economically and practical to resume the construction
activities. Of course, the staff is going to have to
be prepared in the eventuality that they decide they
want to come back in and resume construction and so
we're going to have to have our construction
inspection procedures and training programs in place
in a time frame to support that kind of activity.
The last bullet is identification of an
industry concern regarding the inspections test
analysis and acceptance criteria that's required of
plants licensed under 10 CFR Part 52. There is a
concern as to whether or not the license applications
need to have an ITAAC on operational program such as
the quality assurance program and their security and
training program. The staff is currently in the
process of discussing this issue with the industry and
other stakeholders and we expect to resolve this issue
within the next several months.
That ends my discussion on the
construction inspection program.
MR. RAE: Good afternoon, everyone. My
name is Alan Rae. I'm the AP1000 project manager
within the Future Licensing Group. I'm actually from
Great Britain. I worked for the nuclear safety
regulator in Britain which is the Nuclear Installation
Dispatcher but I'm here working with NRC nine months.
In contrast to the bulk of this seminar
which has been about activities for the medium and
perhaps even looking forward towards the long term,
the AP1000 project is a current short term project.
The AP600 design certification was completed by NRC in
late 1999. What we're working on at the moment in
AP1000 is to look at how the design certification can
be translated into potential design certification for
the extended operation of the AP1000.
It was decided that this will be carried
out in three phases. Phase I is about complete and
was carried out under review by the staff at the end
of which a letter was issued identifying six key
issues that could impact the AP1000 certification. Of
these, four were taken forward into Phase II. They're
listed in the middle of the slide. The other two
issues which was decided would not be taken further at
the moment. First, the PRA that had been done for the
AP600 certification. Westinghouse felt that there
were no significant new issues there and they didn't
need any further advice from staff before making the
AP1000 application.
The second was the review of the key areas
of the design certification document, as it's known.
That is the case, the justification which underwrites
the AP600, looking at which were the main areas that
would have to be changed as this was taken forward to
AP1000.
Phase II scope then was four key issues.
Westinghouse is seeking further detail from the staff
on the applicability of the AP600 test program to the
AP1000 design, the analysis codes, the acceptability
of the use of what are called design acceptance
criteria. These are forward commitments given at the
time of design certification which will actually be
completed as part of the first of a kind or as part of
a subsequent program. And lastly, the applicability
of exemptions granted at the time of the certification
of AP600. For that, you can read the reconciliation
perhaps between the codes that existed at the time
when the design was developed and the certification
that was eventually given.
Of these, the major item was always going
to be the AP600 analysis codes and how these were
developed. Westinghouse presented a report on this
code development supplied to NRC in May. There's some
work been done by staff getting themselves
familiarized with the issues within that report.
There's a meeting later on this week at which
Westinghouse will present the contents of that code
report and hopefully dialogue on how we're going to
get the regulator assurance that's required to
complete this stage of the review.
Phase III of the AP600 review will be a
conventional design certification and it's expected
that Westinghouse will come forward with that in 2002.
Thank you.
MR. BENNER: And lastly, I'm Eric Benner,
the Regulatory Infrastructure lead for Future
Licensing Organization. My blanket statement on this
is what I'm about to discuss are known to-dos. These
are things that were either already being worked
before the creation of FLOW or have been brought to
our attention subsequent to the creation of FLOW. The
readiness assessment being performed by Ms. Gilles and
her group is doing a more thorough scrub of the
regulations to see what changes would be necessary to
support future licensing activities. So we'll have a
more detailed picture when that's complete.
The first item that we have going on is a
rulemaking to update 10 CFR Part 52. You've heard a
lot of references to 10 CFR Part 52. That was put in
place as an alternative licensing method and it
discusses combined licenses whereas the previous
licensing contained in Part 50 dealt with the
construction permit and operating license. 10 CFR
Part 52 discusses a combined license which really
wraps those two items together. It also makes
provisions for early site permits, which Mr. Kenyon
spoke of, and design ,certifications which is
basically when you take a design and certify it not to
license to operate but for someone to just manufacture
so that someone else could license it at a later time.
This rulemaking is basically to clean up
some loose ends after Part 52 is issued. After three
design certifications were done, there were some
lessons learned from that. That'll be incorporated.
There'll be some deletion from Part 50 of repetitive
appendices now that Part 52 is established. There
will also be some incorporation of general provisions,
licensing provisions, under Part 52 from part 50 that
again, on a look back, it seemed like the general
provision should carry forward.
Basically, where we're at now is there was
a preliminary letter that went out some time ago
asking for some comments on this, and the staff
intends to issue a proposed rule package in September
of this year.
There are also two other rulemakings
ongoing. They both involve some of the NRC's
environmental regulations. The first is a rulemaking
on alternative site reviews. Basically, 10 CFR Part
51 is how the NRC incorporates the National
Environmental Policy Act. One of the keystones of
that act is the assessment of alternatives to any
action that's being taken. The NRC has narrowed it
down to look at one of the alternatives that should be
looked at is, hey, you're planning on putting this
power plant at this site. What alternative sites
should you look at?
In the past, that was a little easier task
because you had utilities that had distinct service
areas. So the alternative sites could reasonably be
limited to that utility service area. Now with both
deregulation and consolidation, you get to a point
where you could look at alternative sites much more
broadly. So the staff is currently looking at how
that should be dealt with. That's very preliminary at
this point. We're anticipating an initiation of
rulemaking mid fiscal year 2002.
The last rulemaking is environmental
regulations. Tables S3 and S4 in Part 51. What these
tables basically list are ramifications of the nuclear
fuel cycle. It lists things like average effluence
for reactor, any land and resource uses, and there are
some comparisons for each of these aspects to coal
power plants.
Part of the changes that have to be done
are because all those tables, all the data in those
tables are referenced solely to light water reactors.
So obviously you've heard today about a lot of lot on
light water reactor technologies, so there could be
considerable work to be done there.
There's also going to need to be an
assessment done of the fact that some of these new
technologies use higher enrichment uranium, so all
these tables do have some bounding uranium enrichment
that it deals with. Again, at this point, that's
preliminary activity and, again, I think we're talking
about initiation of rulemaking some time next year.
Next slide, please.
Also at this time, we're not talking about
implementing any of this by rule change, but instead
some of it deals with interpretations of rules are the
NRC's financial-related regulations, specifically
anti-trust, decommissioning and modular plant
requirements. That's specifically to Price-Anderson.
That last one, basically the Price-Anderson Act talks
about retroactive liability and it imposes a financial
burden per facility and if you look at the modular
plant design, say you have 100 megawatt module,
currently if you just looked at how our regulations
are structured, we equate a reactor to a facility. So
you could have 100 megawatt module paying the same
amount as 1,000 megawatt light water reactor. There
is some assessment going on now as to what is truly
fair, and I can't presuppose what the answer will be
there, but we understand there are some concerns.
The anti-trust and the decommissioning
funding requirements. Some of throe questions again
come about because of deregulation of the electric
power industry. There's assessments as too -- again,
in the old days, the utility owned the plant, owned
the transmission lines and what not, so there were
more concerns about anti-trust. Now licensees are
coming and talking about making argument. The
merchant plant arguments say, hey, we're building one
of these plants to provide supply in the competitive
market. There should be no anti-trust issues there
when you're looking at that.
Some future activities that we have
earmarked, and I understand that some of this is going
to change. The Nuclear Energy Institute has talked
about a petition for rulemaking for a generic
regulatory framework performance-based, risk-informed,
a pretty large scope activity. I understand now that
the mechanism for that may change from a petition for
rulemaking just because there are restrictions on the
interfaces that the NRC can have with petitioners but
suffice it to say that that would be a large scale
activity as to how to risk inform the licensing
process.
The last thing on my slides is really just
a mechanistic thing. There's been a lot of talk now
about schedules and regulatory hurdles and mine
fields, I believe was the word. We understand that
rulemaking by its very nature can be a long process.
Some of these advance technologies don't fall nicely
into our current licensing schemes because they are
all geared towards light water reactors. The beauty
of the design certification process is long-term, that
the design gets incorporated into 10 CFR Part 52.
That's a very clean, open process, but it is time
consuming. It does take some time.
In the short term, we have licensed non-
light water reactor technology in the past. You've
heard of some of the examples. Fort St. Vrain and
what not. Basically the mechanism would be to use the
current regulations and for those areas where
regulation intent may not apply, there would be an
exemption granted if the argument was made and in
those areas where the regulations may not be
sufficient, then the NRC can use license conditions to
incorporate other requirements. So that's just kind
of plug for where we're at. That's the end of my
presentation.
MS. GAMBERONI: That concludes our
presentation.
DR. KRESS: Okay. I think we'll entertain
questions on this part of the presentation.
MR. LEITCH: Question about early site
permits. Where a site was approved for multiple
reactors and only one was built, does that other unit
have to go for an early site permit or is that site
for a potential second unit considered banked?
MR. KENYON: I'm not sure. Are you saying
under the old Part 50 licensing?
MR. LEITCH: Yes. In other words, they
had approval to build two units but only built one.
MR. KENYON: Under Part 50.
MR. LEITCH: Yes.
MR. KENYON: That's not really banked
under the Part 52 rule. What's occurred is that when
we license that plant, say we approved it for two
nuclear plants, that was licensed to a specific plant
design. I'll just pick on a BWR design, for instance.
Therefore, although the construction permit and the
license that they had would only allow them to build
the same plant on the site. So if they wanted to
build an ABWR there, they would have to come in for a
different permit.
MR. LEITCH: My question really was if
they wanted to resume their original intent.
MR. KENYON: To build the older design?
MR. LEITCH: Yes.
MR. KENYON: I'll defer to Mr. Jerry
Wilson who's our PAR 52 expert.
MR. LEITCH: I was specifically thinking,
I guess, of I think it's Perry.
MR. WILSON: Jerry Wilson, NRR staff.
Your question gets to whether or not the original
construction permit is still in effect. Assuming that
it was in effect, they could use that construction
permit and build another one of that design, although
the designs we're talking about are quite old at this
point and I'm not sure that anyone is interested in
doing that.
MR. LEITCH: Okay. Thank you.
DR. KRESS: Okay. Let's move on to the
presentation from NRC Research. We'll do a little
musical chairs here, I guess.
MR. FLACK: My name is John Flack. I am
the Acting Branch Chief in the Office of Research,
Regulatory Effectiveness and Human Factors Branch.
This branch will become the focal point of advanced
reactor activities in the Office Research. We have a
small group.
DR. APOSTOLAKIS: And human factors, you
said?
MR. FLACK: Yes, human factors. Did I
miss that one? We're in the process of transitioning
to pick up the advanced reactor work, so what I'll do
is I'll briefly go over the activities that are
ongoing now in the office and the more specific
activities with respect to the pebble bed Stu Rubin
will cover.
Historically, the office has been involved
in pre-application reviews that go back to the 1980s.
This was on the MHTGR, PRISM, SAFER. In many ways, it
enhanced the understanding of the concepts and really
set the stage for licensing applications. There's
really, I count up about five important areas and
features of the pre-application review and the
outputs. First, it all starts with promoting
regulatory effectiveness by identifying early safety,
policy, licensing issues, and then the basis for the
follow-on resolution of those issues.
It also provides important feedback to the
Commission and the stakeholders involved in
entertaining an application for the advanced design.
It also helps to generate Commission guidance on
regulatory approaches that differ, sometimes
substantially, from light water reactors. It
identifies infrastructure needs, in-house expertise,
and it also allows us to hold workshops and interface
with the ACRS, which is one of the important items on
our list. Again, the Advanced Reactor Group that's
being formed in the Office of Research is in the
Division of Safety Analysis and Regulatory
Effectiveness.
On the next chart. Advanced reactors have
greater reliance on new technology and that indicates
the needs for new safety licensing criteria as we move
toward risk-informed performance base initiatives.
The pre-applications give us the introduction, you may
say, to entertaining these new ideas. In an EDO memo
issued in November, 2000 the Commission articulated
the responsibilities of these advanced reactor reviews
and in the next three bullets that I have on the
viewgraph, NRR has the lead with research support for
the light water reactor, advanced reactor pre-
application initiatives, NMSS with the fuel cycle
transport and safeguards, and Research has the lead
for the non-light water reactor, advanced reactor,
pre-application initiatives with longer range new
technology initiatives that would essentially
establish the infrastructure for the follow-on
licensing application.
The memo also identified Research as
having the lead on the South Africa PBMR in
coordination with NRR to plan and implement work in
that area. Recent industry requests for pre-
applications are listed there. Westinghouse with the
AP1000 last year 5-4-00, Exelon with the pebble bed
came in December. The next two, General Atomics GT-
MHR. We've met with them and essentially responded to
them leaving the door open for follow-on discussions
on pre-applications. And then there's the
Westinghouse IRIS. We had a meeting with them on 4-6
of this year.
In addition to throe pre-application
interactions, there is the NEI risk informed framework
for advanced reactor licensings which we are waiting
the review. Next chart, please.
I'll briefly go through the PBMR. Stu
will focus more on the details of that review, but
basically we're engaged with Exelon on that review.
There was a plan developed that was put out in SECY-
01-007 but at the moment I'm not aware that it's
publicly available, but it will be any day now. Pre-
application work is under way and with again the
objective identifying issues, infrastructure needs and
framework for the PBMR licensing.
The GT-MHR. Again, we just met with them
and really we're just saying that the door is open.
WE're waiting for them to take the next step on that.
We're thinking about time frame 2002 for initiating a
pre-application review. Next slide, please.
IRIS is similar. This was a design
developed under DOE, an area program which I
understand you heard about earlier today. We met with
them on 5-7-01 and again we are expecting a pre-
application review, possibly in next fiscal year.
Generation IV is an area where we've been
observing. It's an international activity coordinated
by DOE. It's a longer term effort. We're thinking of
designs out to 30 years, but basically we've just been
gathering information and passing that on to the
Commission and staff to keep abreast of those ongoing
activities.
And the last activity that we're involved
in or anticipating being involved in is the NEI
developing proposal on the generic framework, of
course, that leading to the need for NRC to establish
an effective and efficient risk-informed and
performance-based licensing framework.
DR. APOSTOLAKIS: John, I'm a bit
confused. If someone comes to you using Part 52, is
there anything there that says that you need the risk-
informed performance-based system?
MR. FLACK: There's nothing in Part 52
that says that we need to have a risk-informed
performance-based licensing approach.
DR. APOSTOLAKIS: So they could approach
the licensing issue without using risk information.
Could they?
MR. FLACK: Yes, I would expect that would
be the case.
DR. APOSTOLAKIS: Is there anything that
gives you the authority to request risk information?
MR. FLACK: Other than the requirements on
the PRA. I think Jerry Wilson might be the one to
answer questions regarding the PRA under Part 52
requirements there.
MR. WILSON: Jerry Wilson, NRR. The Part
52 licensing process is just that. It's a licensing
process, and so it references back to parts 20, 50, 70
and 100 for the actual safety requirements. So
whether or not those safety requirements remain as
they are or change as a result of some risk-informed
process, it will use whatever is the requirement
that's currently in place.
DR. APOSTOLAKIS: I mean the slide said
need for NRC to establish an effective and efficient
risk-informed licensing framework.
MR. FLACK: That's an internal processing.
DR. APOSTOLAKIS: What if the industry
doesn't want to use risk information? What if they
just want to use existing regulations with exemptions
or changes and maybe they feel that going to a risk-
informed system adds an impediment because we have to
understand it and do it. It's new. And try to go
with the existing system and maybe a PRA would be an
assessment at the end if you guys request it but maybe
it will be a good idea not to bring it up at all. Why
is that the need?
MR. FLACK: I think it would be to their
advantage to come in that way. Stu.
MR. RUBIN: Stu Rubin, Office of Research.
I would point out that the Commission's advanced
reactor policy statement that was issued in the '80s
does allow, if not encourages, applicants or pre-
applicants for advanced reactor designs to submit
along with their designs proposals for new kinds of
regulatory frameworks, frameworks that are less
prescriptive than the current basis of looking to Part
50 and looking at exemptions.
So it is an option on the part of any
applicant to go with the existing framework or to
propose a new approach to licensing for their design.
So it is very much an option for them, and there's a
decision that needs to be made whether or not it's an
attractive option to try to plow new ground to develop
a new framework or go with existing framework which we
all know has significant burdens associated with it.
DR. POWERS: George, it seems to me that
the Commission has made it clear that when the staff
thinks they want information, they can ask for risk
information.
DR. APOSTOLAKIS: I think they have to
give some argument though that issues of adequate
protection are involved. Isn't that correct?
DR. POWERS: No. They have to give an
indication that there's substantial risk associated
with the idea, whatever concept is put forward.
DR. APOSTOLAKIS: Which comes close to
touching on adequate protection.
DR. POWERS: Shouldn't be terribly
difficult to come up with those ideas. It's an
interesting thing because risk has been notably absent
in our discussions today.
DR. APOSTOLAKIS: Yes. I mean we keep
talking about risk-informing the regulations and yet
major regulatory decisions right now are being made
without risk information. For example, license
renewal. I believe the power operators do not use
this information.
DR. POWERS: Within this context of
advanced reactor codes. I guess it surprised me how
little risk information has seemed to be involved in
those designs.
MR. FLACK: You seem to support the
bullet, the need for it.
DR. APOSTOLAKIS: No. I just was
wondering whether there's a real need. I think there
is a need.
MR. SHACK: This relates to the NEI
proposal. NEI sees the need. You can ask them why
they see a need.
DR. APOSTOLAKIS: But again, the NEI may
propose an option.
MR. FLACK: Moving right along, the last
slide that I am about to present is the --
DR. APOSTOLAKIS: John, before we go on.
How hard do you think it would be to satisfy this
need? Are we talking about a 10 year effort or are we
talking about maybe a year or two?
MR. FLACK: I think the need is to improve
it. Where you stop, I don't see there's any clear
cut-off where we'd have enough of it. I think it's
something that continues to grow and you develop.
Maybe more sometimes than another but I don't see any
specific cut-off on it.
DR. APOSTOLAKIS: Well, it depends. I
mean if one wants to get rid of their notion of design
basis accidents and use instead the PRA, then it's not
obvious how one would do that. So that would a very
ambitious task.
MR. FLACK: We use the PRA to pick the
design basis.
DR. APOSTOLAKIS: Well, that, too. That
would be -- okay. Fine. Thank you.
MR. FLACK: The last slide which I'll
present is on significant technology issues, and
obviously we could spend a lot of time looking at
these issues one by one. I just put it up to get a
feel for the kinds of areas that are highlighted and
need for NRC to really understand with confidence the
advanced reactor designs when pushing forth these
regulatory changes.
If there's no other questions, I'll turn
it over --
DR. APOSTOLAKIS: No, there is one. We
heard today from several speakers, I think, that
they're trying to reduce involvement of the humans.
Do you think that the human performance issue will be
as important here as the current reactors?
MR. FLACK: I've discussed this at length.
I don't know whether we can say it's going to be less
important. I mean it's going to be a different
environment which that human operates in, and one has
to understand that environment and what's changing in
that environment. So it's something that one has to
look at very carefully. So it's hard to say.
DR. POWERS: It seems to me that the
change is really entertaining and in the direction
that's most difficult for us because as they design
the plants to be less and less dependent on the human
operator intervening, seems to me we become more and
more worried about the fact that the operators are not
going to sit there and do nothing and they will
intervene and the potential for them to intervene
incorrectly in a system that's designed to operate
with rather minor low head forces operating on it.
So you get into the problem of errors of
commission that we are most incapable of addressing.
It's a subtle problem.
MR. FLACK: Yes. The environment changes
and you don't really have as much data as you wish
you'd had to go on.
I want to turn it over to Stu Rubin.
MR. RUBIN: Thanks, John. My name again
is Stuart Rubin. I'm a Senior Technical Advisor in
the Office of Research and I'm also the PBMR Project
Manager. First meeting with Exelon with on April 30
and our second meeting is scheduled for next week, so
we're just starting our review. Can I have the next
slide, please.
This next slide summarizes the objectives
for the pre-application review. First of all, the
objective is to evaluate the information that we're
going to be receiving from the applicant on their
design and their proposed new technologies and their
regulatory process and framework for planned
licensing. From that review we will identify where
the information and the proposals appear to meet our
expectations and needs for licensing of PBMR but we
also intend to identify where there are gaps, gaps in
the information on the design or design basis. gaps
in the technology basis or the demonstration of that
technology or the plans, therefore, and shortcoming
that may have existed in their proposals for a
licensing framework.
From those differences, we will endeavor
to lay out the guidance and requirements that the
staff and the Commission feel needs to be in place in
terms of additional information and additional actions
that will be needed to allow the design technology
and framework to be acceptable as a basis for
licensing.
The second objective is to develop an NRC
core technology capability and capacity to conduct an
actual licensing review. We are not doing a licensing
review. We're doing kind of a feasibility licensing
review. But should that feasibility prove positive
and there is a decision to move forward, then the
staff needs to be ready. So we will gain that
capability from this work that we're now embarking on
as well as additional training and the development of
contractor capabilities, et cetera. Next slide,
please.
This next slide identifies the significant
review guidance and references that will be used to
conduct the review. First of all, very important high
level guidance and expectations for such a review and,
for that matter, a licensing review are contained in
the Commission's policy statement on advanced reactors
as well as there is an additional NUREG document 1226
which provides additional staff implementing guidance
for that Commission policy.
In general, the policy encourages
innovative designs and innovative safety criteria but
you still need to satisfactory consider such
traditional aspects of our regulations, the
application of the Commission's philosophy on defense
and depth, safety goal policy, severe accident policy,
application of industry codes and standards.
Also in the case of innovative designs,
new technologies, demonstration testing, a prototype
plan is particularly encouraged. Additionally, we
will draw upon previous pre-application review
experience as well as a safety evaluation report, a
draft safety evaluation report, that was completed for
a similar advanced HTGR design that was proposed by
DOE in the mid 1980s. When one looks ta that design,
one sees that the passive design features and safety
characteristics of that plant are in many respects
quite similar to the PBMR design and safety
characteristics.
I would mention that kind of an underlying
foundation for this entire effort will be an emphasis
on traditional engineering and traditional design
analysis viewpoints. The quality of design,
conservatism of the design and analysis assumptions
and safety modules. Again, our key objective is to
identify the key issues that need to be addressed at
the licensing stage.
Next slide, please. This next slide is
intended to convey the broad scope that we have
planned for the review. For example, in the fuels
area we plan to carefully at the experience base and
the analysis basis for the fuel design and to assess
the fabrication processes and manufacturing plans for
the production fuel. We also plan to look at the
operating experience program and plan fuel performance
demonstration and testing programs, not only on
prototype fuel but that which would apply to fuel
manufactured in a production facility as well as
looking at plans for monitoring performance of the
fuel in reactor.
Just to mention a couple of others in the
nuclear design area, for example. Since the PBMR is
designed to have passive shut-down characteristics, we
intend to clearly assess how this will be demonstrated
and, among other things in the nuclear area, we'll
assess how well power distributions can be predicted
for the PBMR -- moving fuel pebbles. In the thermal
area, since the reactor there too is designed for
passive, in this case, accident decay heat removal,
we'll evaluate the effectiveness of these design
features and, among other things, assess the
capability to analyze temperature distributions during
events as well as there are plans for verifying these
tools including plans for using any prototype testing
to benchmark the codes.
Just to mention a few others. The full
scope testing plans that may be conducted we'll be
looking at extremely carefully to look at what is to
be included and what credits can be allowed by that
testing. The planned PRA and there is an expectation
that a PRA at some level will be provided for the
plant. Certainly we'll need to get that kind of
information in looking at any proposed framework for
determining regulatory requirements.
Another important area will be the
postulated events that will be applicable to the
design. Certainly if one puts in or takes out certain
events, it can affect the seriousness of the impact on
fuel behavior. Next slide, please.
This next slide summarizes the overall
process. My understanding is that we're not going to
get an up front design package or, you might say, a
preliminary safety analysis package from Exelon and so
our plans are to kind of roll out the review on a
month to month basis so a plan is to conduct monthly
meetings with Exelon and the purpose of each meeting
will be to allow the staff to get introduced to
different topics through presentations from Exelon and
subsequently to have that information provided
formally on the docket and then to have the staff
review that information and feed back its needs for
additional information.
Again, we had our first meeting on the
30th at which Exelon discussed its plans for
submitting formal proposals and basis for those
proposals to mitigate or to eliminate certain
requirements in the licensing process that they view
as burdensome to a potential PBMR licensing. Those
formal docketed proposals and bases have been
submitted and staff is now reviewing those.
With regard to the proposed framework for
determining regulatory requirements, that was
discussed. We do have a description of that framework
and the staff has developed its questions on that
first proposal and fed that back to Exelon and we'll
continue to dialogue at our next meeting which is next
week. Again, future meetings. We're going to discuss
traditional engineering design and design analysis
areas such as nuclear thermal design. We plan to
have meetings on fuel cycle safety and plant PRA,
classification of SSCs and the like. Prototype
testing is certainly going to be a major topic.
Again, we'll identify additional
information after each of these kick-off meetings, you
might say, that we'll have on a periodic basis and
then that information will be documented and we will
review that. So we will kind of continue our reviews
and at some point, in addition to these public
meetings, these meetings are intended to allow
stakeholder comments at the end of each topical area
so we can get some input from stakeholders on an
ongoing basis. But in addition to that, we also plan
to have a workshop that's specifically intended to
invite in stakeholder comments on any and all areas.
We also clearly will be meeting with the
ACRS and ACNW as we have completed our preliminary
assessments to obtain advice and input and ideas that
we need to consider before we go final and also as we
progress through these reviews, we will inform the
Commission in SECY papers of our findings and the
staff positions and recommendations in various areas
and then we'll feed back. Once we get Commission
feedback sa guidance, we'll notify DOE and Exelon as
to our positions and guidance in these various areas.
I would mention that as far as the
Commission is concerned, in those areas where we view
Commission policy decisions as necessary to establish
licensing requirements such as in the containment
design requirements or emergency planning requirements
or a number of licensing process issues and legal and
financial issues, the SECY paper will be a Commission
policy decision paper. The staff will present its
findings and recommendations and then we will obtain
Commission decisions and guidance and then, following
that, we'll be back to Exelon on the NRC's
requirements in these areas.
The next slide, please. This next slide
lists the technical resources and regulatory expertise
that the review will utilize. Our strategy basically
is to draw upon the best expertise that's available
within the agency in both power reactor licensing and
applicable HTGR design and technology expertise and to
supplement it where possible, where resources allow,
with additional outside expertise and experience. In
each area, we intend to form a group of one to several
part-time staff who will review that area and, if
possible, to supplement it with contractor support.
For example, in the assessment of Exelon's
risk-informed framework for making licensing decisions
or establishing licensing requirements, we formed a
review group of research staff and NRR staff as well
as OGC staff and we do have contractor support
identified familiar with risk-informing processes here
in the agency.
I should point out that some members of
the staff who will be working on this review also
participated in the previous pre-application review of
the DOE-sponsored modular HTGR in the late '80s. We
also have the benefit of a rather complete draft
safety evaluation on that review and that provides
good resources as to the issues that one would want to
take a look at and kind of a template for going
through this review.
The design and operating experience of
Fort St. Vrain will also be factored into the review,
and we also plan to meet with NRC's foreign partners
with HTGR design and operating experience, especially
those with expertise and experience in coated fuel
particle design and fabrication, radiation and testing
experience and those who have design and possibly
operating experience with the passive design features
and safety characteristics.
Finally, in addition to Exelon input,
we'll endeavor to get stakeholder input from federal
workshop and to get ACRS and ACM input. Next slide.
This next slide lists some of the design
and technology in regulatory areas where we expect
there to be significant challenges in developing the
guidance and the requirements for licensing of PBMR.
A significant area will be the development of the
guidance on information and actions for adequately
demonstrating acceptable fuel performance and fuel
integrity and demonstrating fission product retention
capabilities over the life of the fuel and over the
life of the plant and over severe event conditions.
One of the key points in all of that, as
I mentioned, will be consideration of what are the
design basis events and, beyond design basis events,
that the fuel will need to be analyzed. Another area,
just to mention one, is the guidance and requirements
that the staff will look to develop for assuring
acceptable performance of the core graphite components
and reactor system pressure boundary metal components
at the operating temperatures and levels of neutron
flows are expected over the life of the plan. Again,
the effectiveness of the design features, the passive
design features, what kind of guidance we will need
for adequately demonstrating. That will be another
area that we'll be looking at.
Among the Commission policy issues, and
I've tried to identify those with asterisks, the needs
we believe will require a Commission policy decision
are, for example, the possible use of a mechanistic
approach to the source term. What are the postulated
design basis events and, beyond design basis events,
we need to postulate. The need for a leak tight
containment. Whether that's what will be required or
whether a confinement type structure with controlled
and filtered release would be acceptable. That's
clearly going to be a Commission policy decision.
And again, this question of using risk
information to determine licensing requirements. That
is new and we feel that that ultimately will require
a Commission policy decision. Next slide, please.
I'd like to review our scheduling plans
for the PBMR review. I would like to mention there
are a couple of corrections on this slide. First, the
third bullet should read "feedback on selected
processing issues" and the fourth bullet should read
"feedback on regulatory framework, financial issues
and remaining licensing process issues."
As I mentioned, we kicked off the review
on the 30th and we plan to complete the entire review
in 18 months which would put it out to around October
of next year. We're going to have monthly meetings
with Exelon. We intend to get written follow-up
documentation on what's presented and we plan to
periodically feedback, as I mentioned, to Exelon our
policy and positions on these topics. Again, we also
plan to meet with the ACRS before we do all that.
So in just going through these feedback
milestones, by this August or September time frame, we
will endeavor to provide Exelon, to the extent we can,
the staff's guidance and it's positions on the
licensing process questions involving the early site
permit proposal, combined license and design
certification for initial PBMR facilities. Also by
the end of this year, we will endeavor to provide
Commission policy decisions and guidance on the
proposed risk informed approach for making licensing
decisions and the legal and the financial issues and
the balance of the licensing process issues.
Within 12 months, we expect to feedback
non-Commission policy level positions involving the
technical and the regulatory and technology areas and
then finally by the fall of next year, we will intend
to provide the results of the Commission policy
decisions on these major design and technology issues
to the containment design requirements, emergency
planning, source term, et cetera. Next slide, please.
This is kind of a repeat of what John
talked about. Again, an objective and a by-product,
if you will, of this review is to develop the
infrastructure to effectively and efficiently conduct
an actual licensing review on a PBMR. These kinds of
development activities are fundamental to the role of
research in supporting the agency's review of advanced
reactor licensing. And so we plan to develop a
training course with the support of contractor in HTGR
technology. Our first class is hopefully going to
take place this fall. We will be developing
analytical tools for the analysis of designs such as
the PBMR.
Also, hopefully going to have as an
outcome a regulatory framework for conducting a
licensing review of PBMR and possibly one that
involves a risk-informed approach for making licensing
decisions. And the other thing is we will identify
where we might need independent testing and
experiments on things such as the fuel performance and
possibly the need for additional industry codes and
standards for designs such as the PBMR. That's all.
Thank you.
DR. KRESS: Thank you. Any questions?
DR. GARRICK: This is probably the
question that I was half asleep on when George asked
the question about the risk assessment. But you
mentioned that on the PBMR you're going to get a risk
assessment. What's the nature of that? Has that been
requested?
MR. RUBIN: We have urged Exelon to
provide as much information on the current risk
assessment that they've done for the plan to support
our review of this risk-informed framework for making
licensing decisions. I wouldn't call it a risk-
informed regulations framework as the extent of wholly
replacing Part 50 but we think we now understand that
this framework is not quite going to do that but will
through risk insights be able to identify systems
requirements for mitigation, prevention, the level of
redundancy in those systems, which systems should be
designated as safety significant and also things like
what are the special treatment requirements on the
system. But we're not talking about a regulations
framework which covers all of Part 50.
But to answer your question, we have asked
for that and we've also asked, to the extent possible,
that we get information on the design itself. We have
not yet, except for these kinds of viewgraphs that
we've seen today, gotten what I would call a
significant design description and principles of
operation document from Exelon. I think the staff
would very much like to get both a PRA and a design
description so we have a context for reviewing this
framework. It is on our schedule. We talked about
that. It's not now but it is later.
DR. GARRICK: The thought is that it seems
to me there's a possibility of a very much missed
opportunity here. If you're talking about gearing up
to license for advanced reactors, I can't imagine,
given the history of pushing for performance-based,
risk-informed approach here, of not being further
along than you apparently are in establishing an
infrastructure for doing that and, if there was ever
an opportunity and a place to start it, it would be
with the advanced reactors. I'm kind of shocked at
the words I'm hearing. Possibly, maybe, a list of 500
other items here, 400 of them would be in a good PRA.
I'm just kind of struck by this passiveness that comes
across, to me at least, with respect to getting
serious about practicing what you're preaching.
MR. FLACK: I agree with you. The PRA is
an important piece that we still need to get. A lot
of the underlining structure of that PRA is going to
be in a sense driven by the success criteria, as you
know, and the cost of fuels in this context is going
to be extremely important. So you're absolutely
right. We're ultimately going to have to put all this
in perspective, and we're sort of going into it step
by step. We had pushed the fuels issue up though
because a lot of -- you know, understanding that is
going to play out in PRA.
So I'm not too concerned that we don't
have it right at this moment because in a sense it's
going to take a while before I think they come up with
a good one. I mean they probably will give us one,
but I don't know how good it will be if we ask for it
right now anyway. So I don't think it's holding us up
any.
DR. GARRICK: Well, I made my point.
DR. WALLIS: Can I try to make a similar
point? I listened to NRR and RES. Both parts of the
agency are looking at what capabilities they need to
develop to respond to a new design like GMR. So
there's a tooling up. There's assembling expertise,
there's building up infrastructure and all kinds of
details. Seems to me that you're always going to be
playing a long game of catch up with industry unless
you have some other framework which is inherently more
adaptable to any new technology and it seems to me
that this framework has to be more based on risk
information. It has to have a structure which puts
risk in the forefront. Otherwise, you're going to be
going through and building up a tremendous amount of
deterministic type stuff which is then particular to
every design, and it's going to take too long.
MR. RUBIN: Yes. I would absolutely agree
that the time is now right to move forward quickly, as
quickly as we can to develop this kind of a framework.
Eighteen months ago, if someone were to propose what
we're talking about now, you'd get a yawn from them
because we did not know that there were such an
interest that was going to be around the corner. But
now that it's here, we agree that it's --
MR. THADANI: Stu, if you don't mind,
pardon me for interrupting you. But I think we need
to recognize that Part 52 for design certification
requires the applicant conduct a probablistic risk
assessment to provide that information to the agency
to learn what the insights are to utilize those
insights in the design. The only difference would be
that under Part 52 it does, as Jerry Wilson said
earlier, it does take you back to Part 50, Part 20 and
so on. Now what we're talking about is an opportunity
to really start with a clean sheet of paper and to
build in risk insights up front. But anyone coming in
under Part 52 design certification would be required
by regulations to conduct a PRA. There are a whole
host of other issues. Maybe we'll get into these
issues later on during panel discussion. But I think
there should be no misunderstanding what the
Commission's expectations are.
DR. APOSTOLAKIS: But the PRA the way
things are now could probably be one input to an
integrated decision making process, would it not?
MR. THADANI: Again, it depends on what
level of design information you have and the quality
and robustness of the PRA. You could establish, it
seems to me, a conceptual approach which would use
probablistic thinking and then you could get into some
design specific considerations driven by the level of
information available. How far you can satisfy some
conceptual set of requirements. We're not there.
One of the points I wanted to also say was
we need to understand that while we talk about this
small group that John Flack mentioned, we're just
getting started and we're very sensitive to make sure
before we go too far, we have Commission approval
before we expend any significant resources. So all
you're hearing is reporting to you on some of the
meetings that have taken place and not really
intensive thinking that is necessary. We will go
through that process once the Commission does approve
what John was talking about under SECY-0070.
So all these questions and issues you're
raising I believe will be part of the process that
we'll go through. The most significant being I think
most of us are in agreement with what's being said.
We want to try and maximize risk-informed thinking up
front, clean sheet of paper kind of approach, rather
than be overly influenced by existing structure.
DR. APOSTOLAKIS: Maybe we're getting into
the panel debate here but I must say that I second
Dana's observation earlier that we've heard very
little about PRA today, and I'm under the impression
that there is a gap between the staff's thinking and
the industry's thinking. I mean most of the industry
people who made presentations said, and we will do a
PRA, whereas here we are saying we want the risk-
informed and performance-based system and so on, so
I'm not sure that the industry and DOE appreciate how
important risk-oriented thinking is in both the design
and licensing of these reactors.
I'm sure they will say no, they do realize
it, they do know and so on, but it didn't come across
from the presentations. I'm talking about
quantitative risk assessment. Don't tell me that
we're thinking about safety and we're designing
against that.
MR. PARME: No, absolutely not. I want to
make it clear. You were out of the room at the time,
but we made it very, very clear that our intent on GT-
MHR is to pick up where we left off in the mid '80s
and I spent some time going through exactly that using
risk assessment techniques and a risk assessment to
build up our safety case. We believe that had to be
done for a new reactor type and was the direction we
planned on going. I understand you're busy and may
have been out, but I want to make it clear that
industry agrees with you completely.
DR. APOSTOLAKIS: I'm happy to be
corrected. Thank you.
DR. KRESS: It sounds like we're almost in
a panel discussion. I'd like to take a five minute
break before we do the actual panel discussion to give
us time to do some musical chairs and reorient. So
five minutes.
(Off the record for a nine minute break at
6:16 p.m.)
DR. KRESS: Let's please come back to
order. This is the time to ask questions and to make
comments and get your points in. We don't have a
particular protocol. I don't think we're going to
have each member make preliminary comments. I'll just
open it up for questions and let anybody who wants to.
MR. THADANI: Since we're talking about
the PRA, it seems to me that the way we talk about PRA
right now is being mentioned in a way that -- because
first of all, it seems to me we are looking at these
new designs with old criteria. They were talking
about new PRA -- design and using some of the criteria
here to get -- additional burden and I feel that
unless we -- try to set a different kind of
performance measures, for example -- we're going to
simply -- requirements which may not be necessary.
DR. KRESS: Does anybody on the panel want
to respond to that?
DR. BONACA: Certainly the Commission has
been very clear, I think, in articulating its
philosophy and moving more and more towards risk-
informing regulations even for the operating reactors.
So it's very clear that when we're going to these new
advanced designs, you're exactly right that risk-
informed thinking has to come in up front, recognizing
some limitations. One has to be careful that one
understands what the uncertainties might be. We have
a tremendous opportunity now to start with that
thinking up front such that it can then identify
potential areas where we need additional information.
For these new technologies, I would expect
we would put together a number of panels to look at
phenomenon, see what the important phenomena are,
identify those, rank then and rank them understanding
what the risk implications might be. And it seems to
me that would be a good way to define not only the
kind of testing programs that would be appropriate but
also to make sure that the tools, the analytical tools
that we have are robust enough to give us that
analysis capability which can then be turned around
back again trying to understand what the risk
implications are.
So I would expect we would go through that
process. Clearly, it's a policy issue. You heard
earlier about potential petition coming in from NEI.
I don't think they are thinking petition option any
more, but I'm not certain. But we are as part of our
plan that we've been talking about that we've sent to
the Commission, this is one of the issues and I would
fully expect support. That's the way we would
proceed.
DR. BONACA: The reason why, just to
complete the thought process, my sense, from what I've
seen and we're going to have maybe an SAR coming in
with Chapter 15 with all the traditional analysis
coming in. Okay. That's the understanding I got from
the presentation.
MR. THADANI: I think we are open, up
front to what I described as conceptual model pretty
much will have to take into account more than the
Commission's safety goals because the surrogates that
we use from Commission safety goals have two points
essentially: core damage frequency and large early
release. Clearly, we need the whole spectrum which
means you do have to have the whole sort of CCDF, the
complimentary cumulative destruction function. If you
start out that way, the questions that we would then
face would be is that the level at which you can say
that's technology neutral safety -- so to speak. And
then if you were to go design specific considerations,
is that when you come up with general design criteria
or something else?
It is at that point that that information,
seems to me, ought to help us come to grips with what
are the design basis events. They need to be driven
by this safety philosophy that has to be let out up
front and which, in my view, is more than what the
current safety goal policy statement says.
MR. PARME: Let me add, in response to
your question, whether it's a burden. Going back to
the DOE submittal of the 1980s. The PRA that we used
at that time was not a significant addition to our
task. In fact, it was the forerunning analysis. The
PSID, preliminary safety information document, which
accompanied the PRA and had deterministic analysis,
was pulled out of the PRA. The PRA gave us the
uncertainties and the understanding of this up front.
Obviously, two documents cost more than one but, in
fact, having started -- and in fact, I can recall in
1982 working with the Germans, having evolved our PRA
with our design and the first cut being I think it was
a 25 page memo and having evolved that through the
early '80s as we had the design, it was not a large
incremental cost on the thing.
The only thing that became a burden was
having gone to the Commission and having a rationale
for why we did all these things and then to have the
Commission come back. It was a good interaction but
when the Commission came back at times and you got a
response, we don't agree, and the reasons were often
there was no point to discuss why they didn't agree
with what we had done. That was frustrating. That
was a burden and that cost more money than doing the
PRA.
DR. POWERS: Ashok, you bring up
phenomenology and I'm delighted that you did because
I don't think it's possible to do technology
independent regulation. Sooner or later you have to
get down to how the system really works. I think
that's going to raise a real headache for the NRC
because you don't have the wealth of phenomenological
information about these new designs that you have for
your existing designs. Seems to me that indeed
frequency consequence curves look like an appropriate
approach to go. That means you have to go to
something like a level 2 type analyses and you're
going to have to make a decision along that way at
which point you have to do your own confirmatory
experimentation, your own confirmatory codes.
It looks to me like in the past we've done
that on a catch as catch can basis, but if there are
indeed going to be these multiple kinds of designs
coming to you for at least consideration of licensing
if not actual certification kinds of applications,
we'd better start putting in some sort of a process by
which we can make these confirmatory experimentation
and analysis decisions in predictable kinds of
fashions. That just seems like a priority that the
ACRS and your organization needs to start kicking
around outside of the more formal structures because
it's going to be necessary in spades. You're going to
have lots and lots of head knocking taking place where
licensees presenting test results that say, gee, I
present you these results because I have assumed that
coated particles failure only depends on temperature.
And that's a fine assumption to make but you're going
to want validation of that.
The question is do you get that validation
or does the licensee get validation? It's a question
that's going to have to be answered some place.
MR. THADANI: I agree. First of all, I
think it's very clear -- and I brought this report
just to really make a point I think fits in nicely
with what you said. This is work we did on AP600 in
cooperation with Jerry in Japan. It was at ROSA
facility and I can tell you it was extensive
involvement. I think we did 20 separate experiments.
Some of the work that was done here led to actually
changes in design and impacted schedule in a positive
way because we were able to use this information to
respond to many of the ACRS questions, as a matter of
fact.
My own opinion on NRC's need to do
independent testing comes from the fundamental view
that you get deep understanding by doing things, not
just by reviewing other people's work. That's a
fundamental point. Second, there are some areas in
the fringes which are not necessarily required by
regulations requirements. I personally think it's
appropriate for a public health and safety agency to
sort of poke and probe at the fringes. Try to
understand where the thresholds might be. That would
be independent testing.
In terms of confirmatory work, it's clear
to me that there are some very crucial areas. Fuel or
fuel cladding may be very crucial from the metal
things to safety. It's the most important barrier
we're talking about. I think it's appropriate for the
agency to do some independent confirmatory testing,
even if the industry were doing some testing in that
area. It's amazing sometimes how much you learn by
conducting such testing. How certain issues come to
surface that really get you to go into a fairly
challenging dialogue sometimes as to how one would
proceed.
Analytical tools. Historically we have
really gained a great deal by our ability to do
independent analyses. And so I personally again am
very much in support of making sure we have those
analytical tools that we can employ and when we get
results, try to see if there are differences and sort
of hone in on what they key issues might be.
So basically I do agree with you but
that's why I think PIRTS are going to be very
important for us to know where should we focus really
our attention in this area?
DR. POWERS: I think the program you've
carried out in high burn up fuel has shown you that
the PIRT technology has applications for getting your
staff up to speed beyond the thermal hydraulic area.
At some point we're going to have to come down to
pretty hard and fast decisions on where to
investigate. I think you're right. Fuel is going to
be a head ache here because we just lack the kinds of
experience with this kind of fuel that we're going to
have to have to feel comfortable.
DR. KRESS: I partially think the time
frames are such that to get the kind of data you want
on particularly these coated particle fuels, that is
a difficult task because we're talking about a fuel
that's radiated to some burn up level and get
appropriate statistics for 15,000 per thing, it has to
be put in a reactor, it has to be run through the
temperature transient that you're dealing with and
you're looking for two things. You're looking for
fuel quality in the first place and then you're
looking for what do the transients do to the fission
product release and what sort of model can you put on
that fission product release to get a source term out
of it?
I just don't think we have the time to do
confirmatory research in that area. So I think NRC is
going to have to decide on how they're going to deal
with those particular issues. I think they'll have to
rely in this case on existing data and existing
fission product release models and existing analytical
tools.
DR. POWERS: Stun me if you could, Tom.
I mean we've got basically models based on chemical
diffusion and poor diffusion in a situation where
thermal diffusion is going to be dominant.
DR. KRESS: Exactly.
DR. POWERS: I just don't think you can.
I think you're going to have to do tests and it's the
classic story of --
DR. KRESS: I'm not even sure we have the
reactors to radiate these things.
DR. POWERS: It's the classic story of
planting trees. The best time to plant a tree is 20
years ago. The second best time is right now.
MR. SPROAT: Let me just say in this whole
area of particle fuel testing, there's no doubt in my
mind that the application of particle fuel and pebble
bed application if we go forward here in the U.S.
clearly will have to have a well-documented fuel
testing qualification program that answers some of
these questions. However, there is significant data,
both operational data and test data, that exists on
particle fuel including naval reactors, and I would
severely question the need to go back and replicate
and duplicate at great expense and great delay all of
that information. I think it's incumbent on both us
as the applicant and I think it's incumbent on the
regulator to be able to go back, extract the relevant
data out of the existing vast bodies of data,
determine where the gaps are and focus the additional
testing on those gaps and not reinvent the wheel.
DR. KRESS: Is the naval reactor --
MR. SPROAT: To some extent, yes.
Absolutely.
DR. KRESS: -- How do you see the role of
a prototype test in this respect in terms of
validating the codes and the assumptions that go into
it?
MR. SPROAT: As we took a look at trying
to license the PBMR here in the U.S. Clearly, I think
I said in my presentation, we can't go for
certification first in this country. We have to go
for a COL first. We fully expect that as we go
through the licensing review process here with the
NRC, there will be a number of technical issues that
will be unresolved or open as we go through the review
process which will need to be resolved during the
start-up test program of the demonstration plant in
South Africa.
It's one of the great advantages we have
with the program, at least as it's currently
envisioned, which is with the demonstration plant in
South Africa leading whatever we do here in the U.S.
We'll be able to utilize that demonstration reactor to
reduce significantly a number of the uncertainties
associated with the codes, with the codes, the fuel
performance, that type of thing.
So what we would like to do ideally is to
get far enough through the review process with the
staff here so that the key unresolved issues are
identified and then we can jointly figure out with the
staff and with the South African project how the South
African start-up test program needs to be modified
with the appropriate acceptance criteria so that the
appropriate testing is done during that one year
start-up test program that's in the schedule for the
South African reactor and put those issues to bed
before the license is issued for here. We think
that's a reasonable approach.
DR. GARRICK: Has this data that you refer
to been documented and peer reviewed, et cetera?
MR. SPROAT: I'm not a fuel expert, and I
personally have not reviewed the fuel data. But the
Germans spent over several billion Marks on particle
fuel testing and the ABR. They had their experience
in the THTR. Obviously, in the U.K. gas reactor
program, particle fuel was also tested there and
utilized, and we have the naval reactor programs here
in the U.S. and over in the U.K.
In addition particle fuel is currently
being fabricated in China, Japan, Russia. I mean
there is a significant amount of international data on
this fuel. Now, does it all necessarily envelope the
exact operating conditions of the PBMR as we're
designing it? Personally, I'm not sure and clearly,
if we were to go forward with the licensing process,
we do need to make sure that it's appropriately
enveloped, see where the gaps are and design the
testing qualification programs to cover that. But I
think we'd be amiss if we walked out of here today and
left the subcommittee with an impression that this
particle fuel stuff is all new and there's not a lot
of information about it because that's not the case.
DR. FORD: I'd love to hear the opinion of
the panel about the whole question of materials
degradation, time-dependent degradation, especially
with a risk-informed regulatory environment we're
going into. I heard no one talk at all about it.
Every one of the designs that we've been talking about
in other countries, Southern Korea to the advanced gas
reactors in Britain and light water reactors in this
country, of course, have all undergone cracking or
embrittlement problems of some type or other. You
mentioned the -- chrome situation. For the IRIS, I
didn't see anything at all in that design to say that
you would minimize the frequency of cracking events.
You may influence the impact them but not the
frequency. Could someone address this?
MR. SPROAT: Let me start off and just
talk about the PBMR materials. Clearly, one of the
areas we've looked at very closely in our involvement
in the project is materials because you're looking at
core outlet temperatures of 900 degrees Centigrade.
The ABR in Germany ran the bulk of its career at 950
degrees C. core outlet temperature. If you're
familiar with gas reactor technology at all, clearly,
you know that graphite aging under irradiation and
temperature is a an issue and how graphite reacts
under long-term irradiation where it first shrinks and
then re-expands is a phenomenon that's known but it's
very much specific graphite material dependent.
So my answer to your concern is, #1, that
it's absolutely a valid concern. #2, that it needs to
be addressed in detail during the detail design and it
needs to be addressed via the appropriate materials
testing qualification program during the design phase
and the development phase of the particular technology
that you're talking about. We've been working with
the South Africans to try and make sure that their
thoughts about what needs to be done in their
materials testing development program coincides with
ours, based on what we know are issues we'll have to
look at. As part of our application if and when we
come in, we would have a materials test and
development program in there.
Right now, just to give you an idea,
graphite is clearly one area. Some sort of carbon
carbon composite insulation material that we use in
the hot duct piping is clearly another area. Fuel
we've already talked about. The material we'll use in
the high pressure compressor blading for the turbo
compressors is another. But again, we're in that
preliminary design stage where those issues and the
limiting conditions for each of those key materials is
just now being identified, developed and a mitigation
strategy put together for them.
MR. PARME: Let me add to that. Forty
five minutes is kind of tough to cover all the
subjects when you describe a design, but if you pull
up the plan view of the prismatic block core, you'll
see that both replaceable and permanent reflector
elements are noted in there from the experience
through the '70s and '80s and radiation experience
with graphite type of age and radiation and who's
changed the block is known, and that's designed for.
Right now in our program in Russia, one of
the primary things it's looking at is overhaul of the
turbines. We're well aware the turbines will not last
the life of the plant. In fact, nowhere near that.
And it's designed to come out. It's designed to be
serviced and currently we're looking at various
alloys, alloy possibilities for the blade but also the
possibility of whether we should go to turbo machinery
replacement or is it possible -- mind you, these
turbines, there's some plate out of activity on them,
especially the turbine itself -- whether we can go in
there though and change the blading out. So there are
a number of these things being looked at but, as I
say, I wasn't the materials expert. They sent me, the
systems engineer and safety. They said that's what
they'll want to hear about. But these things are
being looked at as the design proceeds and certainly
I think the industry experience says you need to look
at that up front.
MR. CARELLI: You asked about the IRIS.
Again, IRIS is the youngest design here and, very
honestly, I didn't look at the materials because right
now this is not a top priority. In the case of the
light water reactor, we rely on what it is the body of
the light water reactor. There are two things with
IRIS -- light water reactor and the first one is our
power rating is much lower. We are talking probably
half of the power rating of LWR. Actually, we'll do
even in AP600. So a neutral environment is more
benign.
The other thing is what I showed you
earlier, the capability of putting internal shields.
For example, the vessel. We don't want to put numbers
but the vessel in IRIS should last a lot longer than
the vessel we have in the present LWRs because
basically there is no radiation in the vessel. So
there is no question that the materials is an issue
and, in the case of IRIS, will be especially an issue
on what is new. Like the steam generators, the pumps
that are going inside the reactor. Those are the ones
we'll be focusing on. We already started already
looking for the steam generators. In the case of the
pump, I mentioned the spool pump we have.
The only reason we've been holding on
putting that as a reference design is because of
materials issue of the bearings at high temperature.
So definitely we're going to look into that. Again,
it is the kind of thing that we can not look at other
materials once we have a design. Our first emphasis
is to have a design. Now we have a design and we're
going to look at the materials.
One thing we've done, for example, for the
extended life time core, the one that reloads, the
cladding most probably is going to be a stainless
steel. So we've been looking at those issues.
MR. THADANI: I just wanted to make sure.
John Flack gave us some idea of the issues. High
temperature material issues are amongst the top
issues, particularly when we are talking about getting
temperatures of 900 C. to 1,000 C. Not only
degradation, aging would be an issue, but we're also
going to be looking for some other kinds of challenges
such as thermal shock external to the vessel, for
example. What are the potential impacts of things of
that sort when you have material at such high
temperatures? So it's going to get a fair amount of
attention from us as well.
DR. FORD: I guess as a follow-up
question, Doctor Thadani, you weren't here when I
asked the question this morning. That's all very well
and good, but you've got a severe weight limiting step
with the number of people who can do this job
adequately in the time that you have. I think you've
got a major problem. We all have a major problem in
that particular area.
MR. THADANI: It's a challenging task, I
agree.
MR. RAE: Let me add my two bits to it.
The devil's in the details. At least we at G.E.
believe that materials are a big issue and we have
tried to keep the design within the range of all the
experience base that we have right now. We have a
second line of approach which is to make sure that the
internals are removable, so we are making the internal
designs such that they are easily removable in case
whatever you taught us we didn't learn properly.
Finally, on the sodium reactor.
Unfortunately, I can't answer that question. That's
a little further out in time.
DR. KRESS: I hope I made it clear that
people in the audience are welcome to enter into this
debate also if they want to make a burning comment or
question.
I have a question for you, Ashok. You
mentioned one possibility for frequency consequence
curves could cover most of the regulatory objectives
and I'm confident you can derive the end points for
those using the safety goals. I'm not sure you can
get slopes, but you can get the end points.
The question I have is in view of the
advance reactor policy statement which has an
expectation, I think, of a better level of safety,
what safety goals are we talking about? Are we
talking about the ones in the utility requirements
document or the ones we have now that we use in 1.174?
MR. THADANI: Remember, 1.174 is only
looking at deltas.
DR. KRESS: No, it looks at -- also. But
it's debatable.
MR. THADANI: Yes. I go too far. But I
think I learned from experience, as we all do. When
the EPI requirements document was submitted to NRC, it
had some objectives for designers. One of the
objectives in that was that the core damage frequency
shall be equal to or less than 10-5 per reactor year
of mean value. Let me be clear. And so on. At least
at that time, the guidance we got from the Commission
was very clear that it was driven by the statements in
advanced reactor policy statement.
The view was the Commission expects these
new designs to be safer. Expects these new designs to
be safer. But that doesn't mean that we should
establish requirements that make them safer. Their
view was that we should not go beyond what the
Commission safety goal policy statement says. That's
the only background I have to go on at this stage.
Now we're embarking on some really quite
significantly different arena. At that time, the
Commission's decision, I'm sure, was driven by
understanding what the margins were and what the
various levels of defense that were provided. I think
we will have to go back to the Commission. We'll have
to go to Commission regardless. It's very clear to me
that the one end point of the safety goals is not
enough to develop risk-informed -- that's just not
enough.
So we'll have to go back to the Commission
and seek their guidance on how much farther we can go.
At this stage, I can only tell you what we've been
told up to now.
DR. KRESS: In that same respect, take,
for example, the modular pebble bed reactor. They,
I'm sure, show they can meet something like the early
fatality safety goal with lots of margin. The
question I have there though is -- and they could
probably meet some sort of frequency consequence curve
that you might establish to cover the full regulatory
set of objectives. The question I have is how in that
arena, how would you deal with defense in depth?
Where does defense in depth come into play when you're
asking someone to just meet a frequency consequence
curve?
MR. THADANI: That's why I said that you
can establish in a conceptual sense that you can't
really answer these questions you're raising about
defense in depth until you get to a specific design
and until you understand where the uncertainties are
to make some decisions.
DR. KRESS: You would relate it to the
uncertainties in the --
MR. THADANI: It seems to me that's the
most logical.
DR. KRESS: I certainly --
DR. APOSTOLAKIS: In this respect, would
it be crazy to look at past history and say, boy, we
were surprised four times in the last 20 years and
we're going to be surprised again. The prudent thing
to do is to really require defense in depth in which
case, of course, extra measures of defense in depth,
in which case you reduce the significance of the PRA.
I wonder whether that's just an academic exercise or
it's something real? The reactor safety study under-
estimated significantly the importance of external
events and design end point study show that these were
very important. We were not paying much attention to
the human element until Three Mile Island.
So this feeling that we are dealing with
a new design, new concepts, we're doing the best we
can with the PRA, we'll use it to the maximum extent
we can. There's always this uncertainty about things,
metaphysical things that we don't know about. Would
it be prudent to add an extra layer there at the risk
of making the design uneconomical? I think that would
be a major issue, a major challenge, and I really
don't know how to handle it.
DR. GARRICK: But, George, you do agree,
do you not, that one way to address defense in depth
is in the way in which you express your confidence
about the parameters?
DR. APOSTOLAKIS: I do agree with that.
What I'm saying is that my confidence may not be what
the analysis shows. For light water reactors, it
really took us what? a good 20 years to reach a mature
representation in terms of risk matrix and so on. I
don't think that anyone expects that tomorrow there
will be a risk assessment for an LWR some place that
will come up with something fundamentally different
the way Indian Pain and Zion did or other studies
later. It's mature now. We have reviewed it
1,000,000 times. We understand it. We have a
significant experience and so on.
When you start with new concepts, I wonder
whether that kind of thinking should play a role. I
think that was the thinking in fact behind defense in
depth to begin with, that we could not quantify. I
guess I'm talking about something that you don't like,
John. Unquantified uncertainty.
DR. GARRICK: You're right, I don't like
it.
DR. APOSTOLAKIS: I know you don't like
that, but it's a fact that this thing is there.
MR. PARME: Let me suggest there is one
way of possibly -- I don't claim to have an answer.
It's a difficult question to answer, but one of the
things that we were thinking about. If you look at
the '80 submittal it basically says below 5 X 10-7.
There's nothing else bounding us. There was no reason
to analyze things below there except to sum up risk.
But one of our thoughts -- we had the same question.
Finish with conceptual design. You know there's a lot
of uncertainty in the work you did and it's new
design, too.
But I think one of the things that built
our confidence was we just took them all to the worse
case and made some simple assumptions at the bottom
and what we did then with the risk assessment though
is we could see what were, in a sense, not so much
from a frequency point of view but phenomenologically.
How bad could things go on us? We had that on the
table on paper. We had the calculations that showed
us. Once we understood that, we suddenly were not
quite as worried, have we missed a frequency here by
some amount? Have we misunderstood this? If the
worse case reactivity accidents were only so bad and
took three days before you really heated things up or
if pumping steam from the other nearby reactors for
several days into a scrammed reactor. I mean it's
absurd but we could see what happened. And it sort of
gave some feeling for what were the chances that we
have missed something important?
Of course, our argument to the NRC was
that's in the PRA. It's not frequency of concern. We
don't want to be judged against this. But my hope was
they could read the same document, too, and determine
how comfortable they were or were not with the
uncertainties that are bound to exist. As I say, I
don't think it's a complete answer but it was one of
the ways we tried to address it and I think it has
merit. Just understanding what's sitting there --
DR. APOSTOLAKIS: I agree. I agree. I
mean if that argument can eliminate all this
uncertainty that I'm talking about, then great.
DR. KRESS: That, in essence, is a kind of
uncertainty.
MR. PARME: It is. Yes.
DR. POWERS: I think that's something that
we do too little in this field is to go look and see
how bad things become if everything goes wrong. I
will remind people that a lot of defense in depth
comes about by asking the question, what if you're
wrong?
DR. APOSTOLAKIS: On the other hand, you
can't really push that argument too far because you
end up with traditional deterministic --
DR. POWERS: You and I have written a
paper in which we said don't push it too far.
DR. APOSTOLAKIS: Okay. Good.
DR. POWERS: Push it to the first level
and stop, as I recall.
DR. BONACA: I was curious about this.
This morning we heard a presentation from Doctor
Slabber in which you were mentioning, for example, on
fuel integrity, you are designing for anticipated
transients, 10 X 2-2 and then to the range of 10-2 X 2-6
for licensing basis events and beyond that is
analogous. Are you using PRA behind this analyses in
licensing efforts?
DR. SLABBER: Yes. To answer your
question, we are using generic values at the moment to
get into the ranges. What we do and then
deterministically we calculate the consequence and, in
general, it doesn't take you out of the range which is
prescribed by the licensing authority. So even if
you've got some error bands which are quite large, it
still, with this type of reactor, it keeps you way on
the low consequence level so it doesn't really impact.
But the question is yes, we're using generic--
DR. BONACA: And so you can use that PRA
as a basis for justifying your analysis that you
submit into the licensing area?
DR. SLABBER: Yes.
DR. KRESS: Ted, did you have a comment
you wanted to make? You've been standing there a
while.
MR. QUINN: Okay. I have a question.
It's Ted Quinn. It has to do with process. To set
the stage, a number of the vendors, the applicants
today, have discussed the importance of the pre-
application process. I'd just like to ask the ACRS or
panelists. The going forward part of the next year or
two as we look at it, in the pre-application process
Stu Rubin put up a list of items that are very
important, for example, to the PBMR. Any one of the
applicants could have that similar list. As you go
forward, they've also stated that the results of the
pre-application review are very critical to their
management or the process of going forward after this
is done because some of the key issues that are being
presented, some of which are technical and some are
policy, can get decided as part of this process.
Is it clear to you, the ACRS, that
sufficient information can be developed as part of
pre-application that the staff can review it, that the
ACRS can weigh in and that the Commission can approve
policy issues such as EPZ and definition of some of
the key issues as part of this so that the companies
can go back and go forward with a detailed design?
DR. KRESS: Anybody want to take that one?
I'll give my opinion. I've seen preliminary designs
for most of these reactors. I've seen safety analyses
for most of them and looked at some of the
competitional tools that they've had. I think the
answer is yes, that you can. I don't know. That's
just a personal opinion.
DR. GARRICK: I think that there's a model
for this with respect to Yucca Mountain. Why do you
laugh?
DR. POWERS: Doesn't sound like a
promising model.
DR. GARRICK: But a model from a process
standpoint. Your question was a process question, and
the question that is being tackled now with respect to
licensing Yucca Mountain, is there a sufficient basis
for there to be an application for a license? So
that's an inherent part of the process, to establish
that there is a basis for going forward with the
license application. And it's a very systematic,
deliberate and detailed process.
MR. THADANI: If I may. Certainly we
think we can do it in 18 months. I just want to be
sure that there's clear understanding of what it is
that we will deliver. It's sort of what I would call
some key technical issues or key policy issues. It
would a roadmap basically to lay out what will it
take, the kind of information, data, the need for
tools and so on, what will it take for us to resolve
throe issues? It's not that we have developed all the
information and resolved, clearly not. It's just that
laying out a roadmap as to what is it that we need so
there's a clear understanding like the PBMR, there's
a clear understanding of what the expectations are and
for Exelon then to make some decisions.
So I think it's a good process. It really
is. It not only helps Exelon. I think it helps us.
It helps our reviewers as well. Anyway, so I think
it's doable.
DR. KRESS: I think we're getting tired
and hungry. So I think at this point, unless someone
wants to make a final comment, I'll recess this
meeting until tomorrow morning. We start again
tomorrow in this same room I think at 8:30 instead of
9:00. So the same room tomorrow at 8:30. We stand
recessed.
(The committee recessed at 7:13 p.m. to
reconvene tomorrow at 8:30 a.m.)
Page Last Reviewed/Updated Tuesday, August 16, 2016