ML20058K323
| ML20058K323 | |
| Person / Time | |
|---|---|
| Issue date: | 06/29/1990 |
| From: | NRC ADVISORY COMMITTEE ON NUCLEAR WASTE (ACNW) |
| To: | |
| References | |
| NACNUCLE-T-0024, NACNUCLE-T-24, NUDOCS 9007030219 | |
| Download: ML20058K323 (176) | |
Text
"
~
,s,.6
'g OFFICIAL TRANSCRIPT OF PROCEEDINGS:
lg i
h t
L I
Ageng>:
Nuclear nesulatory commission.
t Advisory Committee on Nucicar Waste i. L
Title:
21st ACNW General Meeting Docket No.
t3
,o 1
f
LOCAT10N:
Bethesda, Maryland DATE:
Friday, June 29, 1990 PAGES:
1 - 121 L
6 ACNW OFFICE COPY - RETAIN FOR THE LIFE OF THE COMMITTEEr..
L L
ANN RIIEY & ASSOCIATES, LTD.
G V
1612 K St. N.W, Suite 300 Washington, D.C. 20006
( _.
(202) 293-3950
[
n...
~.
u e on. eo. m
,.m.m
' ' K.t 9
F p(
~
l 4
1
[ry,
' j, 2
s 3
i 4
PUBLIC NOTICE BY THE 1
1 5
UNITED STATES NUCLEAR' REGULATORY COMMISSION'S 6
ADVISORY COMMITTEE ON NUCLEAR WASTE i
7 8
DATE:
June 29, 1990 9
q 10
-l 11 12
. (/~'g.
13 The contents of this transcript of the
\\)
14 proceedings of the United States Nuclear Regulatory 15 Commission's Advisory Committee on Nuclear Waste, 16 (date)
June 29, 1990 L
17 as reported herein, are a record of the discussions recorded at 18
.the meeting held on the above date.
1 c
19 This transcript has not been reviewed, corrected
!l-20 or edited, and it may contain inaccuracies.
' 21-l 22 23 24
(
25 l
1
1 s
l l'
1 UNITED STATES OF AMERICA N;l 2
NUCLEAR REGULATORY COMMISSION 3
p 4
ADVISORY COMMITTEE ON NUCLEAR WASTE 5
21ST ACNW GENERAL MEETING 6
7 Nuclear Regulatory Commission 8
Room P-110 l
m L
9 7920 Norfolk Avenue 10 Bethesda, Maryland 11 i.
L 12 Friday, June 29, 1990
, /~s Ii
)
13 l
14 The above-entitled proceedings commenced at 8:30 15 o' clock a.m.,
pursuant to notice, Dade W. Moeller, Committee 16-Chairman, presiding.
17 18 PRESENT FOR THE ACNW SUBCOMMITTEE:
19 Martin J.
Steindler, Member 20 William J.
Hinze, Member 21 Harold J.
Larson, Cognizant Staff Member 22 Eugene E.
Voiland, ACNW Consultant 23 D. Orth, ACNW Consultant l; ("]
24 P.
Pomeroy, ACNW Consultant l
%,.)
25
--, - ; \\ c
+
+-t.,
?3/ t m0,13 t
e 4
' 'o 4. 2-
? Dm b,t. 9,,
3:,
'i
- 0
[f' l6,).
s 6'
m, q+m.
r
'M wl
-.a w, ~,
+
u, 8
m
+
+
JV y,..
v, q,m a.
c>
cs -
3.
,v -
_ - w
., p-e
- i w
1 i
lIn y
f.
)
G T l.,
)
q s
1p t
. M l.
- gs
,;ie n l;g: t'
%: A{p. 'i r-
'm.
1 6@!p'J.9th r.i j
3
.g' t
i 4
g 1"
JL 1 h -
's -
M,j@p:ff. }
.- a y
nc
?M gn 9/M1W
.: PARTICIPANT 8i!
i s
z;1,3r j
i
,.3,f ~ -
3 "y u..
1-
.gf5-
[ j=
y g
.-r
-8' '" y' )I,y n
iQa.
, ~
o o
)-]R V t.- {-
j./t
)
j i
a llg -
a t
"q' s
Wifi M, 13'
-H.?Schofer6 H.1Peterson m-
<1 dji...
~
,n..
o n
p
,[
J g isg.
~i.
d r
.a c
< V
<!4"
- C.JHornibrook-.
. L i F a i r o b e n t L:-
q m 3.v y; y.
j>
c'.
i + f,.
,. 1
.:e i, bi L
PL Robinsoni
'A. Upton:
ap 1
n -7 (5 n
y yy M & g-p q_r
(
47 3p=
y a.
r n
e AJO S: 0 6e J.JGreeves
,u I.q
< c r m, ).,.' ;
/, k. '
.i s
I I 8 a
-4
.(=rc
, r[
. ;:, f j
.. +.,
- n
,,u.
f
'vJ_'
.,:13
. d':
'i ! [: *'
- l][ (
e % ',k,-.[
1 f'8 3,
qiy,
tic-e'
'!j i$
-s
.., 'Mf 9
- d 3h 4
. < '; g o
5
+ ! (;
.y 4 1
v g3 39
> y')
f l10:
n
E
- i
't p 1 i.
,E 4
4 7 71+ lli c,it oy ;m y
- m. :A,v-
+
' hbf f
7 a
I
~
~
)
j!![. [K.,. !
y; y'14--
7 t,. {..
{
7 5 p,3
-^
W
,.l15 L.
k 3
y
,[
s,
k
,1h
,y l
ty i,s na
'.r
,<. +
t t'>
's4 P90 C17 J tl^ t
-- - % 3
- i_ $ =n
- w
..>e
.g h
J f.
c'
.18..
g 5 '
3 a
- 195 g
r: x 1
te e
s
\\)
i.n 2'0
.sk
+,
K210 3:
s
',?
ry ~
i'; 9 7.;
l ; -., ' i
.i
- 4
,1 l'..'
- 22 1
if fi l:.
i l
p-
,123 a
l{y
,y
}
- 24 e
m
.. y i
25,
, p-fI..
- i
?
4.)
m y
s n-m
~..~-m
_________________.__.________,,-.__m__.-_..,_.__._._..-.,_-m_.
4
~..,,, -
.-.m
1 1
3
-1 PROCEEDIMGS I
N' 2
[8:30 a.m.]
3 MR. MOELLER:
Good morning.
The meeting Will now J
4 come to order.
This is the second day of the 21st' meeting
~
5 of the Advisory Committee on Nuclear Waste.
I am Dade 6
Moeller, the Chairman of the Committee.
The other two 7
Committee members with us today are Martin Steindler and 8
William Hinze.
We have a team of consultants'with us; Paul J
9 Pomeroy, Donald Orth, and Gene Voiland.
'10 We have two items on our agenda for today.
One is 11 to hear a briefing on the iodine 129 source term for low-12 level waste disposal sites.
Secondly, we will hear a 3
m
( j' 13 briefing and update on the BEIR V Committee report, the 14 Committee on the Biological Effects of Ionizing Radiation of 15 the National Research Council.
The report is titled:
'16
" Health Effects of Exposure to Low-Levels of Ionizing 17 Radiation."
18 This meeting is being conducted in accordance with 19 the provisions of the Federal Advisory Committee Act and the 20 Government in the Sunshine Act.
Today's sessions are 1
21 entirely open to the public.
The designated Federal 22 Official for the initial portion of our meeting this morning 23 is Howard J.
Larson, seated to my right.
The rules for
('~N 24 participation in the meeting were announced as part of the 25 notice published in the Federal Register.
4 1
We have received no written comments from members l
't
'x /
2 of the public, nor have we received any requests from 3
members to'make oral statements at today's meeting.
If,
~
4 however, there is a member of the public here that has some 5
comments and significant contributions to make to our 6
discussions, simply make your desires known to Howard Larson R
7 and we will certainly accommodate you.
8 A transcript of portions of the meeting will be 9
kept, and it is requested that each speaker identify herself 10 and use one of the microphones and speak with sufficient 11 clarity and volume so that he or she can be readily heard.
12 We will move right into the meeting.
To repeat, the first Oy) 13 item.is the briefing by the EPRI and NUMARC representatives 1
14 on the Iodine 129 source term.
We have with us Pat Robinson 15 and Carol Hornibrook from the Electric Power Research 16 Institute.
Also joining them is Lynne Fairobent from 17 NUMARC.
18 We welcome all of you and we very much, in fact 19 for quite some months, we have looked forward to hearing 20 this presentation.
We know that there is a lot of doubt or 21 questions about the lodine source term.
In fact, there's a 22 lot of misinformation out there.
So, maybe today, you can 23 bring us up-to-date and make sure we know what is going on f)
24 and place in the record then some factual information for
%J 25 other people.
Welcome, and the floor is yours.
4 5
l 1' MS, FAIROBENT:
Thank you, Dr. Moeller.
I am s-)
.2 Lynne_Fairobent, Senior Project Manager with the Nuclear 3
Management.and Resources Council.
It is our pleasure to be 4-here today, as you said, to enter into the record and 5
- hopefully clarify some of the discussion you may have heard 6
as we have been trying to deal with the states that are 7
siting the next round of the low-level waste compacts'and 8
try.to straighten up for them and assist them in identifying 9
and quantifying their source term in order to do their 10 performance assessment.
11 The Electric Power Research Institute, as part of 12 their support role on technical issues to-the states,
()
13 started this project a little over a year ago.
NUMARC's 14 interest, of course, is that as we go forward and straighten 15 out the source term from the state problems we have to be 16 cognizant and aware of potential implications.that might 17-impact on the utility programs.
We provide the' regulatory 18 liaison and support work for the Nuclear Power Industry on 19 generic technical issues.
EPRI provides for the industry, 20 the technical research work with which to support our effort 21 in rulemaking.
22 Carol Hornibrook, who is a Project Manager with g
23 EPRI right now, is going to talk today on an introduction of g'%)
24 a review of the technical basis for 10 CFR 61 waste form L)
L 25 stability requirements.
Pat Robinson who, up until April of 1
6
-1 this year was the EPRI Project Manager, is on a sabbatical -
()
2
- a two year leave of absence attending Berkeley to obtain a 3
Masters in nuclear engineering.
She is still under 4
contract, however, with EPRI and is supporting some of the 5
efforts-in the works that were underway prior to the 6
sabbatical.
7 I will turn it over to Carol now to start with the 8
discussion.
We would be happy to field questions during any l
i 9
of the presentation as we go along, if you wish.
10 MR. MOELLER:
Thank you.
Carol, we have a 11 microphone up there, and Howard will hook it up.
12 MS. FAIROBENT:
While Carol is getting the
( s')
13 microphone on, I would like to give you her phone number, s_/
14 Dr. Moeller.
It is not on the cover slide.
It's area code 15 (415) 855-2022.
4 16 MR. MOELLER:
Thank you.
While Howard is 17 adjusting che microphone, Lynn, I appreciate your providing 18 the phone number.
It is so good to see three sets of 19 handouts where the names and phone numbers and the 20 organizational affiliation are provided.
If you could 21 somehow teach the NRC Staff to do likewise, it would be 22 great.
23 MS. FAIROBENT:
I was in a meeting recently where
-(~%
24 the staff was chastised and I was not going to get into that C
25 trap.
1 p
7 l~
MS. HORNIBROOK:
I would really like to thank you
($
[ (~ /
2 for inviting us.
Since I know that you are interested'in 3
waste form, I wanted to bring you up to speed a little bit 4
since you are interested in source terms -- bring you up to 5
speed a little bit on waste form work that we have been 6
doing in terms of how it controls dose, because'it is a very 7
nice lead into the source term discussion that Pat Robinson 8
will be giving you.
9
[ Slide.)
10 We are very interested in what the waste form 11 requirements were in 10 CFR Part 61 in some senses because 12 of the difficulty in terms of solidifying different wastes,
[
_13 the costs involved depending on whether it's a cement
[
14 solidification or HIC that is involved.
What we did was a 15 technical review as Lynne said of 10 CFR 61, the waste form 16 stability requirements, r
17 As many of you are. aware, the waste form stability 18 and site stability is kind of the cornerstone of 10 CFR 61.
19 The findings that we initially came up with, which I am sure 20 all of you are very famillar with is the fact that mobile 21 anions of carbon-14 and iodine-129 contribute essentially 22 all of the groundwater dose, at least when you run NRC's 23 impact code, this is what you find.
(
l
("%
24 What we find that it is related to, the doses are L.Y 25 controlled pretty much by their release rate of the
i 8
1 nuclides.
This is controlled to some degree by the
- ~
)
0 \\ #'
2 infiltration of the water volume that actually enters the 3
disposal unit.
The volume of water that actually enters the 4
disposal unit again, as we said, is really controlled by 5
site stability in' terms of the cap settlement and things of 6
this nature.
l 7
MR. MOELLER:
Excuse me.
You are looking then not 8
only at the source-term in terms-of the actual quantity of i
9 iodine-129 being placed in the disposal sites, but you are 10 then going beyond that and estimating how much of.it is 11 going to become released into the groundwater?
12 MS. HORNIBROOK:
Right.
We are using your impacts (q
e
)
13 code to-see just what kind of effect does the waste form P
11 4 have on controlling the dose to the public.
What we are 15 really interested in looking at is, are there alternatives 16 that could provide equal protection to the public.
17 MR. MOELLER:
Thank you.
18-MR. STEINDLER:
Before you leave that, that first 19 set of findings, that is entirely based on the assumption 20 that the code is correct.
Has there been any validation or 21 any experimental data on validating that code?
22 MS. HORNIBROOK:
What we did was run the actual 23 impacts code, and we did have some problems with some of the
[~ )
24 lines in the code itself.
What we did was go back to the
, V
'25 EIS and the FEIS and DEIS and look at the actual verbal
9 1
descriptions.
Then, we used those in running the actual N-2 code analysis.
Those numbers came out more in line with 3
what we expected.
4 MR. STEINDLER:
That's not quite my question.
My 5
question is, has anybody gone out in the field and dug a 6
hole, did an analysis of actual transport of carbon-14 and 7
iodine to find out where the model has any relationship to
-8 the real world?
9' MS. HORNIBROOK:
I am not familiar with that.
Pat 10 may know more, but let me just say how much I know.
I am 11 not familiar with whether anyone has done a specific look at 12 each of the steps.
I know EPR1 has funded some project work
(T 4
-( )
13 that looks at the uptake by the roots of plants of carbon-14 14, and we have found that we think that some of the.
15
. assumptions that NRC uses in that code are a little bit 16 conservative.
17 on the other hand, I would say that when you run 18 the iodine and carbon-14 numbers through the impacts,-presto 19 and cosmos codes -- presto is EPA's, cosmos is AECL, the 20 Canadian group code -- you get pretty much the same relative-21 concentrations.
They differ a little bit because there are 22 some slight modifications.
All in all, it seems that many 23 of the same assumptions have gone into these different code
(]
24 developments.
V' l
25 Pat, would you like to add something?
l
)
10-1 MR.-STEINDLER:
If one set of assumptions is 2
wrong, then they are all wrong.
l 3
MS. HORNIBROOK:
That could be.
4' MS. ROBINSON:
Perhaps let me just comment a
=!
5 little bit on it.
The actual performance assessment models j
I 6
have not been completely validated by measurement and, if 7
you take a moment to really give it some thought, with the j
8 long-lived-radionuclides and you are discussing the release j
9 and the transport from these long-lived radionuclides to the q
10 groundwater is a very long process.
11 don't really peak in the analysis until out too many 12 thousands of years, 4,000 or 5,000 years.
So it is, of
()
13 course, impossible to validate that transport part.
14 However, what is measured are the important 15 parameters that are used in the performance assessment 16-models such as the uptake factors for carbon-14 and some of f
17<
the leach rates from the. actual waste form.
Those are 18 actually measurements that are performed.
Did that help?
19 MR. STEINDLER:
Yes.
It doesn't solve the problem 20 but it helps, thank you.
MS. HORNIBROOK:
As I said, the mobile anions of 21 i
22 carbon-14 and iodine 129 -- I guess I already did this.
You 23 have no more questions, I hope, on this slide.
(
24
[ Slide.]
25 Our conclusions were that really based on these
lh
,f 11 1
long-lived -- carbon-24 it is almost 6,000 years and iodine O
2 129, 16 million years -- site stability therefore, those 3
features have to be essentially permanent.
We find that 4
waste form stability may enhance site stability by 5
decreacing the organic content per unit volume of the 6
trench.
By this we mean whatever volume that stabilization 7
material takes up, you have less organics in the trench 8
itself.
On a per volume basis you have a little bit less 9
organics to break down and create void spaces.
10 What we really believe is that it is highly likely 11 that the dominating features, which contribute to site 12 stability are package void volume, emplacement, backfilling,
()
13 compaction, and cover design. Finally, because iodine 129 14 and carbon-14 long half-life, we feel that waste stability 15 will have really no practical inipeci cr. retarding migration.
16 The same is really t: cue for the vault type cements, the 17 engineered barriers that people are_1csking at.
We feel 18 that these cement type structures will last about as long as 19 the wasts form.
' 20.
By the way, the NRC requirement is that it last 21 about 300 years for your waste form.
l l
22 MR. STEINDLER:
Is that what you mean by I
23 essantially permanent, 300 years?
24 MS. HORNIBROOK:
No.
What I think essentially 25 permanent has to do with the iodine 129 half-life whJch is i
12 1
16 million years.
As pat says, when you look at the 7s
'--]
2 modeling doses they peak really closer to the five million 3
or ten million year time zone.
It has to be essentially 4
permanent, the site, as far as its stability.
5 (Slide.)
6 The next area where we find these long-lived 7
nuclides having an impact is on the intruder and where, as 8
we also said in my opening remarks, is where NRC hoped the 9
waste form would provide some protection.
Waste form 10 stability, what we have concluded, provides intruder 11 protection only for the short-lived nuclides because of the 12 life of the waste form itself.
f3
(_,)
13 What we believe is that passive warning devices 14 could provide actually_the same or possibly greater level of 15 protection, depending on what they are made of, and that 16 waste form stability provides essentially no intruder i
17 protection for long-lived transuraaios.
18 Here again, the engineered barriers, because their 19 life is not any greater than about 300 years -- some people 20 feel that they may have 1,000 year cement and it really just 21 doesn't last long enough -- that this will just not provide-22 intruder protection because, as we said, the finite life of 23 these materials.
24 (Slide.)
25 Our conclusion, and this is also based on the 10
e r
i I
13 I
<3.
1 CFR 61, is that deeper burial is really significantly more
\\,]
2 effective in reducing intruder doses than waste form 3
stability.
When we ran the number of scenarios that they 4
had, it really showed that the deeper you disposed of the 5
waste the better off you were in terms of intruder 6
protection.
So, deeper burial of always greater than Class 7
A -- anything greater than class A would virtually eliminate c
i 8
the class B and C waste form stability requirements while 9
providing more effective intruder protection.
i 10 We are saying that anything greater than Class A, 11 we think you should bury it deeper in order to get that J
12 intruder protection.
13 MR. MOELLER:
Again, how deep?
14 MS. HORNIBROOK:
In 10 CFR 61, I believe it is 15 five meters that they say, five meter depth or-greater.
16 MR. HINZE What assumptions are made concerning 17 the intruders that permit the greater depth to give you i
18 longer protection; what assumptions do you make regarding 19 the intruder?
20 MS. HORNIBROOK:
What I believe they base it on is 21 the depth of foundations, things of this nature.
Waste 22 form, what they are assuming is, the intruder digs down and 23 hit something hard, they go back to the local town records
\\
24
[d and find that there was a disposal site there, so their i
25 contact time is only about six hours.
If it is still l
l
4 14 1
shallow and there is no waste form, they feel that people 2
might be digging around in it for around 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br />.
'~
3 These are assumptions that the impact model makes.
4 As far as the deeper disposal, as I say, I think it is just 1
5 based on what you would be digging for, to put in a small 6
foundation for home and something of this nature.
7 (Slide.]
8 Finally, this really leads to the waste form 9
stability testing criteria which have been under 10 development, actually, redevelopment in a sence.
The 11 assumption here is that vaste form that meets the test 12 criteria will provide adequate stabilization in the disposal
)
13 site.
Really, our questions with that is, is the waste form 14 stability testing criteria in terms of how we have reviewed 15 it, doesn't really relate on a one-to-one basis with 10 CPR 16 61 performance assessment and that it does not represent a 17 simulation or an accelerated testing of the actual disposal 18 environment in terms of this long term -- how long will it 19-hold up when it's in the actual disposal unit or whatever --
20 it really doesn't test that kind of thing.
21 Mostly what you have is, will it hold up for about 22 a year and then you know whether or not a rock has been 23 disposed of.
/O 24 (Slide.)
i h 25 In summary what we found is, waste form stability I.
15 i
l provides really no protection for groundwater dose, again,
,~
)
'/
2 those long carbon-14, iodine 129.
It provides intruder 3
protection for short-lived gamma nuclides.
It provides no 4
protection for long-lived transuranic nuclides in terms of 1
5 the intruder protection scenario, and can be effectively 6
replaced by deeper burial and passive warning devices.
7 The question is now, what can be done.
If waste 8
form stability is likely to have very little or no impact on 9
groundwater doses and, as we said, iodine and carbon-14 are 10 the source terms that ultimately control groundwater dose 11 what should be done or what can we do.
12 Just briefly, I have mentioned that we have looked n(,)
13 at this waste form question.
Carbon-14, I mentioned in 14 response to a question earlier, that we did some plant 15 uptake studies and we are going to continue to do more of 16 that.
We-are also looking at leach factor studies, which is 17 the rate at which carbon-14 will actually leave the waste 18 itself.
Leading into the discussion today, iodine 129 and 19 technetium predictive model validation is something else 20 that EPRI has gotten involved with.
21 What we have been sponsoring is actually the 22 validation of the iodine 129 and technetium fuel release 23 model for predicting disposal site inventory.
Battelle is f'J
\\
24 the group that actually collects the ion exchange test
)
w 25 columns that we use to validate the model that has been
16 l
1 developed, and the measurement data is then compared against s) 7
(
'~ #
2 the prsdictive model.
They are given the same information -
3
- the modeler is given the same information in terms of the 4
time we collected our sample in the reactor coolant sampling i
5 stream and then we compare the results against our
[
6 analytical results.
i 7
I think I would like Pat to give you a better 8
description of exactly what has occurred so far.
9 MR. STEINDLER:
I guess I have a little problem 10 trying to figure out where you are.
You were looking at the 11 release of iodine 129 and technetium from the fuel?
12 MS. ROBINSON:-
I will be happy to pick up that p(,)
13 discussion and give you a little bit more background 14 information on exactly what we are working towards.
15 Somewhere here is my introduction slide.
16 (Slide.)
17 I also would like to express my thanks for the 18 opportunity to come in and discuss with you some of the 19 specific issues related to siting new disposal sites in this 20 country.
As you know, we are into a very involved technical 21 development as well as a very involved political development 22 of these new sites.
Carol has presented a nice overview for 23 you on sort of the comprehensive scope of EPRI's research
}
24 program, not just locking at iodine 129 as a source term 25 issue but also looking at the other disposal site issues
17 1
that relate to siting including waste form and some of the
)
2 other long-lived radionuclide performance.
3 This morning what I am going to do is discuss with i
4 you a little bit more on the specifics of why we are focused 5
on iodine 129 this morning.
I will present to you a little 6
bit of background information as to what the issue is with 7
it as it relates to the nuclear power plants, as well as 8
then go into what solution we have for our alternative 9
method that we are proposing iodine 129 inventories for new 10 disposal sites, a little bit of the evaluation work that i
11 EPRI is conducting to validate this alternative methodology, 12 and then want to bring it home a little bit to you with a
,-(,)
13 specific application of how a compact would use this new 14 methodology and the validation data to aerive their source 15 term.
There is one compact that has this work currently in i
l 16 progress and have some very prelimir.ary results.
I think 1
17 you will find them very interesting.
18 Then, I will give a brief summary at that point 19 and turn it over to Lynne Fairobent, who will discuss l
l 20 reporting and compliance issues with the new method that we h
21 are proposing.
22
[ Slide.)
23 As Carol mentioned as you will recall from 61,
/~')
24 groundwater doses are inventory controlled and are not
%/
l 25 concentration controlled.
These are controlled by the I
l
18 1
mobile anion isotopes of iodine 129, carbon-14, technetium i
)
2 and tritium.
Inventories of technetium and iodine 129 have
~
3 the potential to limit the siting of new low-level waste 4
facilities.
5 Currently, from the preliminary performance 6
assessment that I have now soon in three compacts, iodine 7
120 contributes in the performance assessment greater than 8
90 percent of the total dose to the maximum individual.
It 9
is important that we are talking about individual maximum 10 doses and not population impacts.
11 What we want to ensure that we have whenever wo 12 are going through licensing actions, when we are addressing gm
(_,)
13 the public health and safety, is to ensure that we have as 14 accurately as possible defined what the radionuclide 15 inventory will be in these disposal sites.
Additionally, 16 because we realize that we must in essence prove our case to 17 the public, we want to ensure that we have the appropriate 18 level of technical conservatisms and that we can defend it 19 from a technical, theoretical basis.
These are our two 20 goals in defining the disposal site inventories.
21
[ Slide.)
22 I do need to give you just a little bit of 23 background on sort of how the inventories could be developed f^^
24 from the use of shipping manifests and explain a little bit 25 of some of the issues related to actually the practical
19 m
1 operational limitations at a nuclear power plant as well as t
i
\\
/
2 the techr. ology limitations for analyzing and characterizing
' ~ '
i 3
these wast es that come from a nuclear power plant, because 4
they really are the genesis of an iodine 129 estimate which 5
we believe is conservative by a factor of 100 to 10,000, 6
depending on what waste stream and what waste form.
7 I am not sure if you all are familiar with scaling 8
factors or how the process is really applied at a nuclear 9
power plant, but let me kind of brief you quickly on that.
10 Iodine 129 is not a gamma emittert therefore, we do not have 11 the capability -- the analytical capability on-site at a 12 nuclear power plant to measure readily in a day-to-day basis n(,)
13 iodine 129 concentrations as well as many other low energy 14 beta emitters.
15 We therefore do sampling and analysis that we send 16 off to commercial laboratories to provide us information on 17 concentrations of specific waste streams such as rad waste 18 resins, dry active waste, secondary ion exchange resins, one 19 of probably 20 different waste streams that come from 20 nuclear power plants.
21 From that analytical information that we receive 22 back from the laboratory, they have also given us 23 information on the gamma emitters which we can measure
[V) 24 readily.
We refer to the iodine in the carbon-14 as a i
25 difficult to measure radionuclides, mostly saying that they
20 1
are not gamma emitters and we have no instrumentation rx.
])
2 readily available to measure them on-site.
3 So, in order to be able to determine or infer --
4 is really what we do -- the iodine 129 concentration in a 5
waste package, we develop what we cal'1 a scaling factors.
6 The scaling factors are really ratios between a known of 7
something that we can measure which is cesium 137 in the t
8 case of fission products, and iodine 129 in the case of-9 activation products like carbon-14.
We really look to 10 correlate it with other activation products such as cobalt 11 60.
12 MR. MOELLER:
Excuse me, Pat.
Are there no other 13 easily measurable radionuclides, say fission products, that 14 would have chemistries similar to lodine and so forth that 15 you could have used a ratio?
16 MS. ROBINSON:
That's a very good question Dave, 17 because you realize that iodine performs as an anion even in 18 reactor coolant system, and cesium is a manovolant and 19 performs differently.
There are no other anionic species 20 with any reasonable half-life that we can index to.
21 Initially we thought that we could perhaps use some other 22 isotopes perhaps -- I think it was cerium 134 but its half-23 life is fairly short and somewhat lost in the background of
' O' 24 all the other gamma emitters.
It is very difficult to 1
G 25 resolve that analysis.
1 1
21
,cq There really are no other good key isotopes for l
2 fission products.
Because of the exact comment that you i
3 make, we call these scaling factors to differentiate from 4
what we might call correlations.
To me, a correlation truly_
1 5'
does have a theoretical base for why these isotopes should 6
have this same relationship.
In fact, these are empirical 7
relationships, and that is really sort of how we have j
i 8
evolved to having such over conservative numbers.
Without a l
9 fundamental theoretical base, empirical data is always just 10 sort of out there, and you never really never have a good 11 idea of good or bad it is because you have no basis for 12 comparison of a theoretical model.
()
13 Let me explain, since the implementation of 61,
[
14 the nuclear power utilities have really been using the best 15 available technology to provide the best information on 16 their shipping manifests as to the content of all-17 radionuclides required in 61 reporting which ultimately have 18 been used for disposal site inventory definition.
19 (Slide.)
20 If you take just a moment, I have put up industry 21 scaling factors here for iodine 129 and cesium 137.
I am 22 just going to talk about the PWR's here for a moment.
You 23 can take a moment in your leisure -- perhaps your plane y
trips home -- and take a look at the BWR numbers and kind of 24 25 logic your way through why they look at little over
22 1
conservative in some areas.
)
2 One thing that I do want to point out is that the 3
analytical capability -- the best available commercial 4
technology for iodine 129 analysis is liquid scintillation.
5 It has a fairly limited detection limit of ten to the minus 6
six microcurie per cc.
Really, Iodine 129 is released from 7
our reactors but in extremely small quantities.
When you 8
get it into waste streams that become very dilute, we start t
9 to bump up against the analytical sensitivity of the 10 instrumentation.
11 Now, we are in a quandary of -- as in the case of 12 dry active waste.
You would expect that it would not see a
,m
,()
13 ratio quite this high.
In fact, it is probably high by a 14 factor of over 1,000 -- conservative but high.
The reason 15 is that because not only is iodine 129 dilute in these waste 16 streams and the utilities are forced to use LLD values 17 because they have reporting requirements for iodine 129 at 18 the disposal facility, also cesium is dilute in these waste 19 streams.
20 So, you have a cesium number only that is being 21 rationed to a LLD, and consequently these scaling factors 22 are over conservative.
They are extremely over 1
23 conservative.
But, they are the limitations of the best
[' 'N, 24 available technology that we have.
The waste is extremely
%)
25 dilute in radionuclide content.
23 1
MR. MOELLER:
liow did these scaling factors come 7.-
t
).
2 about; were they imposed upon industry?
3 MS. ROBINSON:
No, they were not imposed upon the 4
industry but it was really the best methodology that could 5
be applied to do the very best job at trying to quantify 6
these radionuelides.
There was a considerable amount of 7
investigation done not only by the NRC but the utilities as 8
well, to look at methods of what can be done.
9 MR. MOELLER:
I guess when originally developed 10 there was not too much recognition that these scaling 11 factors might have a real later significance?
12 MS. ROBINSON:
Conservatism -- I will tell you, I
?m(,)
13 Dave, what really happened.
There was a lot of focus from 14 61 when 61 came out on waste classification, because that 15 dictated what trench it went into, how deep it was buried 16 and whether or not you had to have the waste form solidified 17 or not.
Iodine 129 does not control classification in 61.
18 It is a rare, rare situation that it does control 19 classification.
20 In fact, I can only think of two waste packages in 21 the whole country that have ever shifted from a class A to a 22 Class C classification because of iodine 129.
23 The focus were on those radionuclides that control
('~'}
24 classification such as the transuranics, strontium 90, the n-25 total cesium quantity, those isotopes.
A lot of that focus l
24 I
was put there.
At the time everybody was thinking well, i
\\
2 it's okay to use a conservative number for iodine 129.
Low q
3 and behold, five years later, we are trying to site new 4
disposal sites and people are trying to use shipping 5
manifest data that was really prepared from a standpoint of 6
waste classification and transportation as a definition of a 7
disposal site inventory.
j 8
Hopefully I am filling in the background for you, 9
a little bit of how we have this over conservatism.
l 10 MR. STEINDLER:
One other question, if I might.
l 11 MS. ROBINSON:
Certainly.
12 MR. STEINDLER:
You led me to believe that if your fh
(,)
13 detection limit is X, that it is necessary to assume that 14 that is the content of the waste package; is that right?
15 MS. ROBINSON:
Well, I did lead you to believe 1
l 16 that.
17 MR. STEINDLER:
Yes, you did.
18 MS. ROBINSON:
Where that comes from is that there 19 is not a reporting requirement in Part 61 for anything that 20 is less than one percent of the total radionuclide activity.
21 So, 61 does not require that you take an LLD number and make 22 it into a real number and report it.
23 However, the states that own the disposal sites
('~)'i 24 have imposed that requirement because they are trying to L.
25 prepare themselves for a performance assessment at site
r 25
'7-1 closure by summing up what they think the iodine 129
?
2 inventory would be.
So, they are really the ones that have 3
imposed the reporting requirement for an iodine 129 4
concentration number in a waste package.
5 MR. STEINDLER:
Somewhere in the South Carolina 6
regulations there is a statement that says if your detection 7
limit is whatever and your analytical procedure is such that 8
you don's see any iodine 129, you have to use the detection 9
limit concentration for your presumed inventory?
10 MS. ROBINSON:
I don't think that the state 11 regulations are quite that specific.
They just have a 12 reporting requirement.
Lynne, do you want to comment?
(_)
13 MR. STEINDLER:
That is what I am getting at.
If 14 the reporting requirement is there which says in effect you 15 have to tell us how much is there, the answer is zero.
16 MS. FAIROBENT:
Let me touch upon that maybe from 17 a different perspective, which is compliance and getting our 18 waste into say Barnwell for example, since you brought up 19 South Carolina.
20 I am not sure that it truly is a reporting 21 requirement from the State of South Carolina placed on the 22 utilities that are shipping it.
However, the disposal 23 facility operator, in order to accept that shipment of
( D 24 waste, requires a numerical value to be entered on the LJ 25 shipping manifest in order that he can then sum up what he i
i t 26 1
has accepted for disposal on the site for down the road to 7
(
)
2 do a site closure performance assessment.
3 In my mind it would not be a hard regulation in 4
the sense of a legal requirement in Part 20, Part 61, Part 5
71 for transportation or DOT regulations under 49 CFR 173.
6 It is a practical limit placed on all shippers to not 7
necessarily record an NDA value or a zero value when you 8
cannot conclusively show that there would be zero activity 9
of that isotope in the waste form package coming in on site.
10 So, I think that is where we are getting into this 11
- dilemma, 12 MR. STEINDLER:
I have trouble with that last
,)
13 statement of yours, you cannot conclusively show that there is 14 is zero.
You can't show anything below -- that is the limit 15 of detection --
16 MS. FAIROBENT:
Not under the commercially 17 available.
We can do mass spec analysis which Pat will get 18 into, which has led us into doing this iodine 129 validation 19 work.
The sensitivity of the detection capability in the 20 commercial. labs that do the analysis and from a practical 21 operational implementation on-site at the utility, the 22 instrument sensitivity isn't down below ten to the minus six 23 microcurie per gram.
'N 24 There is capability, however, limited in this
[D 25 country in mass spec analysis that would get us down to ten
v I
27 1
to the minus 13 microcurie per gram.
Pat is going to talk a i
7~y
(
)
2 little bit about that, I think, when we get further on and
\\'
3 the detection sensitivity cn the mass spec in actually l
f 4
identifying the quantity present.
5 MR. STEINDLER:
I am well aware of the analytical 6
problem and have a pretty good hand on what detection 7
capabilities exist.
What I am trying to find out is whether 8
this is a box that you folks have gotten yourself into on i
9 your own account or whether you were driven into it by 10 regulations. I must say that it isn't at all clear to me j
11 that you were driven into it by regulation.
12 MS. FAIROBENT:
I would say that we are not driven
(-(,)
13 into it by NRC.
Maybe John Greeves can clarify from NRC's 14 view as to why we may be in the box.
15 MR. MOELLER:
John, go ahead.
1 16 MR. GREEVES:
I hope that I can make this simple.
17 My understanding is that Part 61 requires you to report 18 iodine 129, period.
You have to on the manifest.
The 19 industry knows that there is some amount of iodine 129 in i
20 dry active waste.
So, the regulation requires you to report 21 it.
Your only ability to detect it is down to the lower 22 limits of detection so that the generator knows it is in 23' there and knows it is no greater than the LLD.
24 The way that the requirement is set up, he has to
(
25 write some number in there.
In a compliance mode, he can
4 28 1
write LLD and be in compliance.
He cannot write zero and be
<~
l i (s) 2 in compliance.
Lynne, is that simple --
3 MS. FAIROBENT:
Yes.
What I am saying is that the 4
site operators have taken it a step further in some cases i
5 and they don't like to see an LLD entered into the manifest L
6 when it comes into site, so they look for the numerical I
t l
7 equivalent value net to the LLD.
That is why you are l
8 getting numerical value on shipping manifests that then can 9
be summed as it has in some of the compacts to use as the 10 source term in siting the new facilities.
11 The manifests I think when they were originally 12 designed were never designed with the intent that the data
)
13 should be used as the source tera quantification for siting
(
14 a new facility.
It is the only legal record that is out 15 there.
Yes, we are boxed in, and it's a combination of 16 factors.
What we'are trying now to do is get out of that 17 box and still provide and maintain adequate protection of 18-public health and safety.
19 MR. STEINDLER:
I don't want to prolong the argument.
I think what John is indicating is that you have 20 21 to fill in a blank.
22 MS. FAIROBENT:
Right.
23 MS. ROBINSON:
Yes, that is correct.
)
24 MR. STEINDLER:
If in fact, instead of filling in 25 a blank with a number you write lower limit of detection is I
i l
29 1
not exceeded -- that apparently would satisfy Part 61?
7s I
's 2
MS. FAIROBENT:
Yes.
3 MR. STEINDLER:
Okay.
After that, I think just 4
simple statistics will tell you that if you write any number 5
and multiply it by a large enough volume of waste you can 6
get to any quantity of iodine 129 that you want.
Obviously, i
7 there is something wrong with the system.
8 MR. GREEVES:
In terms of accuracy, the staff has 9
been over this issue with EPRI, and we are interested in 10 this.
We haven't come to resolution on it.
It comes down 13
- the two points that Pat made earlier on; is accuracy and 1
12 t.sfensibility in these new site applications.
We are i
()
13 pursuing this issue.
14 MR. STEINDLER:
Let me just make a personal observation that scaling factors are neither accurate nor 15 i
i 16 defensible I would think.
The current solution doesn't 17 sound like one that you could back up too thoroughly unless, 18 in fact, you were willing to expend what apparently is a 19 reasonably large sum of money to go into mass spec data.
20 MS. ROBINSON:
Yes.
Fortunately, whenever I come
[
21 forward and talk about a problem I also like to have the 22 solution in hand, which we fortunately do in this case.
23 Later on in my presentation I will discuss an alternative
/~'}
24 methodology which I think will give us both an accurate and l
\\_/
25 very defensible source term.
l
t i.
30 1
MR. STEINDLER:
Great.
(
)
2 (Slide.)
3 MS. ROBINSON:
As you mentioned, perhaps you 4
peaked ahead to my next slide.
When you have a scaling 5
factor that is very conservative and now you happen to be 6
multiplying it by the largest volume, you even further drive 7
up an inventory number which is non-representative of what 8
it might be.
In fact, when you really give it some thought 9
to the process operations inside of a nuclear power plant, 10 Iodine 129 behaves very predictability in most situations as 11
.in anion and the clean up capability on the reactor coolant 12 side from the CVCS system in a reactor water clean up system p) of the primary resins, we believe that you will find-almost
's,
13 14 80 percent of the iodine 129 in primary resins and not in 15 DAW as scaling factors would imply.
16 (Slide.)
i l
17 We have touched on the causes of the over j
18 estimates.
I am going to go into it just a little bit more I
19 specifically than what we have and focus on the analytical 20 limitation.
As I mentioned, the best available analytical b
21 measurement techniques are not sensitive enough to measure 22 iodine 129.
I am saying iodine 129, but I also mean 23 technetium.
It is the same situation.
We could improve
[~)
24 both numbers in these very dilute waste streams.
In fact,
!V 25 between the work that the NRC has done and the work that the t
i 31 l
r 1
Electric Power Research Institute has done over the last
- )
2 five years in collecting and analyzing the industry database 3
on actual measurement results from these contract 4
laboratories.
5 You will find that out of 3,000 samples there are 6
only 70 measured values that are above an LLD on iodine 129, 7
and the rest are reported as LLD's, 8
There is a point I want to touch here on kind of 9
the practical operational limitations.
If we know that all 10 the iodine 129 should be on primary resins, then let's 11 analyze primary resins.
Well, the difficulty is that even 12 on the primary resins the iodine 129 concentration is so low (m
(m,)
13 that it cannot be resolved.
Additionally, we are on 14 extremely high radiation doses on primary resins, anywhere 15 from 20 to 400 per hour.
In fact, we would expend more man 16 ram exposure going in and trying to representatively sample 17 an extremely hot waste stream than anyone would receive from 18 doses of this waste at a disposal site.
(
19 Additionally, there is another practical 20 limitation.
That is, when you sample a primary resins that i
21 have been in a CVCS system for six months to 12 months, none 22 of the commercial laboratories can deal with that 23 concentration of gamma emitting, that radioactivity being
[)
24 emitted from those resins.
If you think about sort of N_-
l 25 practical sampling limitations, here we have let's say 100 I
1 32 i
1 cubic feet of primary resins that are 400 R per hour, that l
)
2 ability to get representative sampling of both anion and 3
canaan resins which load differently, which separate by 4
densities differently and are extremely radioactive, brings 5
us down to some cases actually counting beads.
6 That then brings into question about how 7
representative is your sampling.
Of course, it is not very.
8 (Slide.)
9 So, I think you have a good picture hopefully by 10 now of sort of the practical and operational as well as 11 analytical limitations plus the reporting requirement. I 12 think that what the NRC would like to see as well as the
(_,/
13 industry would like to see is, we can identify and report 14 concentrations of radionuclides and in this case 15 specifically iodine 129 as accurately as we possibly can, i
i 16 We have proposed as an alternative method to 17 sampling and analysis, a method which really goes back to 18 maybe days that you all might remember, and that is when wo 19 originally looked pre-1961 at what a performance assessment l
20 from a disposal site might be.
We have made some 21 assumptions about fuel performance failure, and we assumed 22 that it failed at some rate and released some calculated 23 inventory at some burn up and used that as the inventory
/~)
24 number.
%.)
25 This method is akin to that, but much more l
\\
33 1
sophisticated to that.
It involves the modeling of Iodine I
\\
2 129 and technetium release from the fuel, and it is assessed 3
on reactor coolant data that we take at our nuclear power 4
plants every other day.
Those are the short-lived Iodine 5
and the gamma emitters that come from reactor coolant liquid 6
samples which we do have the capability at our power plants 7
to measure readily, and they are required to maintain these 8
records for life of reactor.
So, we have a very good 9
historical look back through this data.
10 MR. MOELLER:
Excuse me.
What you are doing --
11 perhaps I am jumping ahead -- you are now using a surrogate 12 or another Iodine isotope which you know behaves the same as
,m 13 I-129.
14 MS. ROBINSON:
Exactly.
15 MR. MOELLER:
Okay.
16 MS. ROBINSON:
What we are going to do and what I 17 am going to explain in the methodology is how we are going 18 to determine the release rate of Iodine which we have known 19 relationships for and index it against Iodine 131, which 20 performs chemically similar not only in the reactor coolant 21 system and in the fuel but also on the primary resins.
22 (Slide.]
23 Just quickly, let me go through this methodology,
/'~'s 24 a little bit of background -- simple background.
I didn't Q.,)
25 realize this was quite so simple until I read that first l
34 1
line.
(
)
2 Iodine 129 is a fission product.
Sometimes I give
'~'
3 these presentations to people that don't have any background 4
in nuclear.
Of course, as a fission product, the source is 5
always a fuel.
6 What we know is that it is released from fuel 7
either inside the clad or from exposed fuel or tramped j
8 uranium, and that it is released by three different l
mechanisms at different rates because of the differei.ce in r
l 10 the mechanisms.
These mechanisms are recoil, diffusion ind j
11 knockout.
Fuels people will probably be a little more 7
l 12 familiar with them than I am, but I will explain how we l
/^\\
l V 13 determine the mechanism.
l 14 (S1ide.)
l l
15 This is just a pretty picture.
I t h ght I would l
16 use it just to show the fuel is sometimes inside the clad.
]
17 Here is a picture to show that sometimes it is outside the l
l 18 clad through defects.
This happens to be a defective fuel i
19 rod, a very small and tiny -- it looks like some crud 20 deposition like sort of a hot spot burn through on the fuel.
l 21 This, of coure.e, releases Iodine 129 into the 22 reactor coolant which ultimately ends up in the process 23 waste streams and ultimately in the disposal site.
f5 24 (Slide.]
O 25 This is a bit of a schematic of a rod defect i
f l
I
35 1
release of Iodine 129.
Cesium is being released also.
(\\)
2 Recoil diffusion through not only the fuel but into the gap 3
and knockout, and then through the gap into the coolant and 4
then exposed tramp fuel of the same three release 5
nechanisms.
r 6
What we tried to do is resolve these release 7
mechanisms since they release Iodine, all iodine at 8
different rates because of the different mechanisms, we are 9
trying to resolve these three mechanism release rates.
They 10 are dependent, the release rate itself is controlled by age, 11 burnup of the defective rod, production, of iodine in the 12 rod, the defect size, whether or not you have a pin hole g(,)
13 defect, and your more controlled by knockout mechanisms or 14 whether or not you have a very moderate defect where now you 15 have the ability for water to come in and in essence flush 16 out some fuel.
Then you start to see enhanced diffusion 17 because of the larger surface area into the coolant.
18 (Slide) 19 We do use, as I mentioned earlier, the five short-20 lived iodine; Iodine 131 through 135 as well as the cesium 21 134 and 137.
I am going to skip a few of the slides in 22 there, but I think it is no surprise that cecium 134 and 137 23 is used to age fuel.
There are a couple of other slides in (V)
24 there that show that the relationships between Iodine 129 25 and 131 by mechanism -- and these are relationships that are
36 1
derived from half-life dependency relationships that are 7
\\
)
2 known on these isotopes.
3 This happens to be an example.
Recoil is a 4
constant at 4.55 E to the minus two and the Iodine 129, 131 5
ratio diffusion of course is dependent on burn up and defect 6
as well as knockout.
The thing that I really want to show 7
here is that the release rate can be from minus ten to minus 8
eight, so it's a couple of order of magnitude.
That is 9
important to why it is important to resolve those equations.
10 Actually, that is how the model does establish.
11 It really has five unknowns and five equations, and we 12 iterate on those solutions until we find the solutions.
of
/N
(_)
13 course, whenever you use a mathematical solution technique i
14 there are some that may be negative numbers that have no 1
1 15 practical meaning in reality, so we have a decision analysis l
16 part of the code that resolves the contribution from each of t
17 these mechanisms.
18
[ Slide.)
19 This is examples that I have taken from old fuel 20 performance or reactor coolant data based on the fuel 21 performance of these plants.
I think that it is very 22 interesting, because it exactly points out how you can have 23 really different controlling release mechanisms in any
[}
~
24 reactor.
In fact, I will show you later how you can have
%j 25 even different controlling release mechanisms within the
t 7
37
,3 1
same plant, within the same cycle just as soon as you open 2
up the defect or you close up a defect, which occasionally 3
happens.
4 This happens to be SONGS-3, if you recall early on 5
in -- maybe in the last five years it was -- SONGS had some 6
Water chemistry problems that created a lot of fuel defects.
7 In fact, they had very large amounts of fuel.
You will see 8
not surprisingly that when fuel is outside the clad you are j
9 really controlled by the diffusion mechanism.
That was 10 really quite the case at San Onofre.
11 (Slide.)
12 Just to recap a little bit about what exactly do l
)
l
(_/
13 you do with this methodology.
I have given you a little bit l
14 of background on the theoretical aspects of it.
We do take 15 the reactor coolant data, and what we do with all that data 16
-- this is thousands and thousands of data point -- is 17 develop concentration and release rate profiles for Iodine 18 129.
We determine the fuel release mechanisms from the core 19 conditions and the fuel release param'ecers that I mentioned 20 carlier.
21 We know the fuel release parameters and then we 22 know the release rate, because we have the relationships of 23 Iodine 129 to 131 and because we are indexing to 131 at this
,< \\
(
a
-24 point and it starts to get easy.
You multiply the release v/
25 rate ratios times the actual 131 concentration and you
38 1
produce the Iodine 129 release rate.
From the Iodine 129 O
2 release rate profiles, how the heck do you get to this site 3
inventory?
4 We have chosen to use the past three fuel cycle 5
data to be representative of the range of fuel conditions 6
that any one reactor would have and use that as a basis for 7
projection.
The basis for a projection may differ depending 8
on where the actual release rate number from all the 9
reactors let's say in a compact is actually resolved to be.
10 It may be that we can use the average for a 30 year 11 projection or we can use the most conservative for a 30 year 12 orojection or something of that sort.
i 13 We determine the Iodine release rate profiles, and l
14 I am going to show you some graphs here in a minute over all I
15 three cycles.
Then, we integrate of course, under the curve 4
16 to determine the inventory in a release cycle.
We determine 17 that on a microcurie per megawatt day, so we know the power 18' production to calculate that Iodina 129 release.
19 (Slide.)
20 This is how the data looks from the Iodine 131 and 21 reactor coolant during a cycle.
You can see that it changes 22 of course, depending on what is happening with the fuel and a
23 even power levels.
This is an example of one plant that is
(}
24 currently being evaluated in one of the compacts.
It's a 25 PWR.
I am going to take a look at the next couple of l
1
i
-l 39 1
cycles.
This is perce1.t contribution by release mechanism.
O 2
You can see that in this cycle they had a very constant fuel 3
performance throughout their control by knockout being the 4
largest contributor to Iodine release, at about 60 percent.
5-(Slide.)
i 6
In cycle six, this is sort of a demonstrative 7
pictuv^ of how they had a defect and were being diffusion 8
contro. led.
I would perhaps suggest that something 9
happened.
It's not so much that the rate of diffusion may 10 have -- I think in this case the rate of diffusion did i
11 change.
It either closed up the defect a little bit, which
~
I 12 sometimes happens with crud getting deposited.
What is
)
13 important here is that you see a shift in the controlling 14 mechanism which implies that the rate of Iodine 129 release 15 is shifting.
16 MR. STEINDLER:
Are those data all based on your j
i 17 theoretical isotopic ratios?
18 MS. ROBINSON:
They are all based on --
19 MR. STEINDLER:
Is that the way you assign 20 fractions to a particular mechanism?
.21 MS. ROBINSON:
They are all based on the half-life 22 dependency relationships betwo.en the Iodinc, the Iodine 131 23 to 135.
(
24 MR. STEINDLER:
Okay.
25 MS. ROBINSON:
This is sort of a composite over l
______________1_____
4 40 s3.
1 four cycles, but this is the picture that we are looking at i
2:
to base a projection on'the four cycles.
You can see 3-
-dramatically, the effects of these four cycles.
If I knew a 4
little bit more about fuel performance, I could maybe even 5
tell you when they were opening and closing a defect.
These H
6 are cycle changes.
7 You can see, just in a decade span, that we go I
8 from E to the minus five down to a very good E to the minus 9
seven -- minus eight or somewhere in there -- a very wide 10 span on the release rate of Iodine.
11 MR. VOILAND:
What is the unit on the Iodine?
12 MS. ROBINSON:
I think that past graph was
(%
(_)
13 microcurie per second release.
What we are going to do 14 actually is -- we actually now calculate in microcurie per 15 mecaw9tt day release rate.
We take -- this is an example of i
16 the c.us PWR three, four, five, six and seven cycles.
These
.17 r o the ca' '11ated Iodine 129 release rates, microcurie per 18 L,m f Tv. Ic:y.
We know the megawatt days in that fuel cycle 19 i
wu ed.
.ly them together and get the actual Iodine 129 20 released over that cycle.
21 It is interesting I think to note, here you have 22
-two cycles that are very close in megawatts per day, but 23 very far apart in actual Iodine 129 release rate in fact, a 24 couple of orders of magnitude.
This is in fact, reflective 25 of the fuel performance during that period.
Cycle five,
41 7.q
-obviously, had some-defect that was releasing. iodine'in a-1
- !' -)
2 much larger rate.
Cycle six looks like it was a very tight 3
fuel performing cycle with little if any defects, perhaps 4
just the component from tramp uranium.
5-
[ Slide.)
6 This work is currently underway by the way.
These 7
aie preliminary results from compact three which has more 8
units than this.
I jagt happened to pick three BWR's to be 9
an example so that ycu can take a moment also in your plane 10 trip to compare from plant to plant and cycle to cycle to 11 see how these numbers will vary in performance.
Also, a 12 little bit of information on BWR's which I have sort of Q
13 ignored by not really.
14 When you look at fuel performance in the United 15 States, you will see that PWR's generally operate with one 16 to two defects, and in rare circumstances maybe there is 17 ten.
It is very difficult to know how many defects you 18 really have because the coolant really has an averaging 19 process of-it taking place.
BWR's on the other hand, have 20 had really excellent fuel performance over the last -- since 21 the early days of some of the original problems.
Over the 22 last five to ten years it's been very good with the 23 exception of one mechanism of failure which is corrosion
(-}
24 induced or localized corrosion occurrirrg from some copper q;
25 condensers that they have -- silk failures.
We call them
42 1
silk failures, 7_
d
)-
2 Generally, BWR's are very good fuel performers.
3 MR. MOELLER:
Excuse me a second, Pat.
You have 4
mentioned tramp uranium several times.
5 MS. ROBINSON:
Yes.
v 6
MR. MOELLER:
I recall when I worked at Oak Ridge 7
that the fuel people would roll the fuel -- it was a sand 8
wedge with aluminum cladding bonded right to the uranium and 9
the machinery or rollers when you have rolled uranium will 10 have contamination.
When you rolled the aluminum there was 11 tramp uranium on the outside.
12 Ilow do you get tramp uranium in a fuel -- on the eS
(,)
13 outside of a fuel rod?
14 MS. ROBINSON:
I think there is still an impurity 15 level -- at a very low levels that is a problem.
When I say 16 tramp, perhaps more accurately I should say exposed fuel.
17 It really can be a historical look of a past defect that 18 released and then deposited on core materials.
19 MR. MOELLER:
Fine, okay.
20 MS. ROBINSON:
I really should say release from 21 exposed fuels, which is a more accurate statement.
22 MR. MOELLER:
Okay, thank you.
]
23 MS. ROBINSON:
Let me just further tighten in with
(~}-
24 the situation on the compacts now, which kind of brings us
\\J 25 back to where we started, I hope.
I have information just t
43 gq from three compacts on what a preliminary Iodine 129 source l
\\')'
2 term would look like if you derived it from the shipping 3
manifest data which is a sampling analysis and scaling 4
factor, all the issues that-we discussed early on.
5 What they predict or what they have actually 6
derived from shipping manifests is somewhere from point one-7 or.04 to.26 curies per year, and it varies a little bit 8
because the number of reactors are different in each of 9
these compacts.
It's in the ballpark.
10 The sites that they are licensing are 30 year 11 sites primarily.
Some are looking at 60 and 50 year sites.
12 When you multiply these out you get an inventory that ranges f3 k) 13 between one to seven or eight curies of Iodine 129, all ms 14 derived from shipping manifest data.
When you now use the 15 methodology that we just. discussed, the very preliminary 16 data shows that you don't have curies and you have one-half 17 a millicurie.
18 Interestingly enough, it happens to be about in 19 line with some of our more rigorous waste characterization 20 studies that we have done.
Carol mentioned earlier that 21 Battelle is doing these characterizations for EPRI.
We had 22 some data that we conducted in the BRC owners group where we 23 went to absolutely heroic efforts to be able to find these
)
( j radionuclides and draw active waste, and that's how we know 24
~-
25 what the conservative is in active waste.
i 44 t
s 1
That included sorting waste into different drums Y
2 by activity, counting gamma scanning through a direct assay 3
-unit-that Battelle has, and sub-sampling that drum until we 4
matched up the gamma spectrums, and then taking 500 grams of I
5 booties and clothing and in some cases we even found a fin 6
from some diver who obviously had been inside a fuel pool 1
7 and dissolve those with acid dissolution and took it over to 8
the next step.
Then we did all the 61 analysis.
Battelle 9
has the best capability in the world for -- in this country 10 anyway, I don't know the world -- certainly in this country 11 a very strong capability to detect Iodine 129 down to ten E 12 to the minus twelfth microcurie per cc.
(,/
13 You can thank the Department of Defense, because 14 the Department of Defense put in about $5 million worth of 15 development and equipment to be able to develop those 16 procedures down to very low levels of Iodine 129 which, by 17 the way, is one of the chartered assignments of this 18 laboratory, is to continue to pursue radionuclide detection 19 at very low levels as well as improve it with very, very 20 high activity backgrounds.
21 (Slide.)
22 A bit of a summary of the advantages of the 23 approach before I stop into the validation.
One of the
( ))
24 things that I do want to highlight about this point is, it Am 25 is a conservative number and I will tell you why.
If you
45 j3 1
think about it, here we are calculating the total release of l'
\\'~'
2 Iodine 129 from the core, and we are assuming that the total 3
release goes in the disposal site in total.
In fact, we 4
know that there are gaseous effluent losses, and there were 5
some losses in the liquid systems that we are not 6
subtracting out for.
7 From a mass balance standpoint, we know that we 8
are always going to predict on the conservative side but not 9
over conservative.
We believe that it is much more 10 accurate.
In fact, we'have validation testing to show that 11 it is more accurate, and also, to validate that the code is 12 in fact capable of giving you accurate predictions.
It does
(_,)
13 use a very large reactor coolant database.
Actually, it's 14 really more than that.
15 Most utility sample reactor coolant every other 16 day as sort of the minimum and all that data is analyzed in 17 total over the last three cycles.
Additionally, it is 18 comparatively low cost to the cost of going in and doing
-19 rigorous sampling and analysis campaigns that you would have 20 some statistical strength behind your samples and your 21-analyses, and then you would have to use a detection 22 capability equivalent to Battelle so that you could really 23 get some real numbers.
That would be in the many, many
/
24 millions of dollars, probably greater than $15 million if we
\\s l 25 could really do that which really we can't because there is i
i
L 46 1-only one laboratory in this country that I know of that has 7s 5
)
1
/
I
'~'
2 the capability to detect these levels of Iodine.
l 3
Most importantly, and I really feel this quite 4
strongly, that this methodology can be independently l
5 validated using controlled plant testing and rigorous 6
measurement and analysis.
I am going to explain that.
I 7
Rarely do I ever feel that in low-level waste that we have l
8 the opportunity to validate many of the models that we have 1
including the transport models. Validation of transport 9
l l
10.
models are, to me, akin to doing accelerated corrosion tests 11 which I used to do.
You end up accelerating the tests so l
12 much that you are really disturbing some of the key O
i,)
13 parameters that have very large influence on that final 14 number.
15
[ Slide.]
l l
16 So in validation testing, EPRI is the primary 1
17 sponsor of the validation testing.
It was really very 18 fortunate that this data was available.
It was really 19 developed for another application, another investigation, 20 but it proves out to be very useful in validating out this 21 test.
I will explain a little bit about the testing 22' protocol, but I do want to just highlight here that EPRI is 23 doing ten units.
There is also some work being funded by a 24 ESEERCO and SERDA, and they are doing another three units of
(
)
25 this validation testing.
47 1
In the parenthesis not only do I have the plant
,,_s.)-
2 and type of plant which implies a different fuel, BNW fuel l
3 or GE fuel or Westinghouse fuel, in parentheticals there 4
were the estimated number of defects by the fuel people 5
during the testing period.
We are also trying to span from
]
6 no defects to as high a defects as we could find in the 7
country during this operating pe'iod which was at a BWR.
~
r 8
All of these testing -- actual plant testing is dor.e.
Not 9
all of the analysis is yet completed.
10 These are the few plants that are remaining for 11 this round of testing.
I think there will even be more L
12 validation testing as the compacts start to use this l )D) methodology which I think now there is one underway and I 13 14 expect that both the compact that has the disposal site in 3
l 15 Michigan as well the disposal site in North Carolina and 16 Illinois will employ the same methodology.
l 17 It is likely the NRC will first see the 18 application of this technology in license application i
19 submittals from these compact areas.
20 MR. MOELLER:
Excuse me.
Maybe you have told us, 21 but what basic differences do you find between BWR's and 22 PWR's?
23 MS. ROBINSON:
Mostly the fuel performance is much
/\\
24 better.
You will find low release --
L) 25 MR. MOELLER:
In the PWR.
i 48 1'
MS. ROBINSON:
From BWR's.
7-~s I
\\
i
'j
\\
2-MR. MOELLER:
Lower release.
3-MS. ROBINSON:
Yes.
4 MR. MOELLER:
Okay, thank you.
5 MS. h0BINSON:
Typically.
But.I could answer that 6
better after I see the rest of the research data.
7 MR. MOELLER:
But basically, there is no 8
fundamental difference?
9 MS. ROBINSON:
No.
There is no fundamental 10-difference in release mechanisms that affect it.-
That's 11 also the really nice thing, that about a good strong 12 fundamental theoretical modeling approach you don't rely.on l-7)
l A,,/ -
13 any empirical data that is really subject to those set of L
14 testing assumptions or conditions that were used.
I find a 15 lot of times when we have some empirical introduction into a i
16 theoretical'model -- I will give you an example.
i 17-The fuels people have really just tried to get to 18 see if they could predict number of defects in the fuel l-19 condition. Unfortunately, early on I think for I;
20 simplification of complex transport models, they eliminated L
g, 21 knockout and recoil as dominant mechanisms.
Recoil rarely 22 dominates, except for when you have absolutely tight fuel.
23 By eliminating thoce two mechanisms to sort of simplify the l-I ')
24 mathematics and the ability to use it on the computers of L._/
25
.the era of ten or 15 years ago, they really lose the
49 p~s 1
strength of the tool and end up using more empirical p
~#
2 information like escape rate coefficients and so forth.
3 Consequently they have a less stronger tool.
To 4
me, it's really kind of interesting to see how important it 5
really is to hold the fundamental models at all.
The i
6 validation testing -- what we do is, we go out to nuclear 7
power plants -- each one of them that are on the list -- we 8
bring with us tucked under our' arm our little test equipment 9
that I will show you a schematic of in a moment.
We 10 actually install this equipment directly on reactor coolant 11 sample lines and under controlled conditions, under known 12 flow rates and for a known amount of time which is generally p)
(_
13 two days -- we initially run 30 days but we found that.we i
14 collected too much activity and couldn't analyze the samples 15 of even do the chemistry that we needed to do to analyze the 16 samples except in a hot cell.
'17 We cut back the length of time over the~ tests.
18 This is what the equipment really looks like for a PWR.-
19 Since the CVCS systems are mixed bed resins where you bottle 20 those exact flow conditions that are in the CVCS systems in 21 these small columns, reactor coolant comes in and runs 22 through the column back into the drain.
Right now we are.
23 not using 30 day tests as I mentioned because it did get j )
24.
over 400 R per hour contact, and that is way too difficult Q
25 for any researcher to handle that column.
e 50 s, _
1 We run, as-I mentioned, the full conditions of the 0
2 CVCS systoms.
What is important and what point I want to-3 make is, we are eliminating the causes of the over estimates 4
from the waste stream basis because we are collecting an i
5 inventory.
We are going to analyze this inventory in total, 6
so we have eliminated the sampling potential and we are 7
using Battelle's analytical method, so we have eliminated 8
the limitations of the detection capability of commercial 9
laboratories.
10 MR. STEINDLER:
Is the efficiency of this mixed 11 bed. resin column that you have equal to that used in the 12 plant?-
A k) 13 MS. ROBINSON:
In this case for collecting the 14 inventory, we don't exhaust the resins.
We want to be sure i
15 that we are not breaking through iodine in the affluent that 16 we are collecting the total inventory in total.
On the BWR 17 side, because the reactor water clean up systems use resin, 18 we actually have a small millipore filter assembly that we 19 have pre-coded with the same resin mix we use in reactor 20 water cleanup and then we run these for two days.
21 One thing that I want to mention that I'm afraid 22 that people sometimes think this is an easy test and let's 23 just do this all the time.
There is a real radiological i
[')
24 protection issue here.
As the reactor coolant, when it
\\_/'
l 25 comes right out of the core, we have had very little time
51 j-sp_
1 for decay of some of the very high energy materials -- gamma l
l
'2 emitters as well as the beta emitters.
We have actually run
~
3 into a radiation protection issues because if you make a 4
mental mistake and you make an error with these samples, you 5
have the potential for somebody really picking up a 6
significant beta dose or gamma dose if they are not very 7-careful.
-8 So, I am not advocating that we use this 9
methodology on any kind of routine basis, because I think P
10 it's far too risky from a radiological protection standpoint 11 and I think it's unnecessary because the code will be 12 validated to the point that the NRC and any other regulator O-(_,/
13 is convinced that the code has the capability.to accurately 14 predict. It might be used perhaps in the future as sort of a 15 requalification or revalidation.on something that I see in 16 the three to five year time basis.
17 As I mentioned, the whole column is shipped off to 18
_Battelle.
They strip the inventory in total, they do-the 19 analysis and they give us what the inventory on that column l
L 20 was, the code -- which I haven't mentioned previously what 21 the name of it was.
It's down here in little numbers.
It's 22 3R-STATE, which is reactor radionuclide release status, is l
23 the name of the code.
l -[')
24 It is used to predict the total quantity loaded on l-%J j
25 that resin for the same time period for the same test l
52 j
1 conditions, i.e.,
coolant volume that had flowed through.
.,3 1
s 2
We take the coolant iodine analysis and run 3R-STAT to give 3
.us the microcurie'per second release rate of Iodine from 4
that core.
We know the time, the duration of the test and 5
the coolant volume so we can determine the inventory.
And j
6 then, we compare the results, of course.
7
[ Slide.]
i 8
This first test is the same PWR that we talked 9-about earlier. It just shows that we passed through 227 10 liters, ran a 48 hour5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> test and the reactor was at 100 11 percent power.
12
[ Slide.)
ry i,)
13 These are the results.
I included cesium here, s
14 because.I believe the tool, the 3R-STAT with maybe another 1
15
$100,000.00 of development work would have the capability to 16 predict all fission. products that are of interest, including 3
17 strontium 90 as well as transuranics.
I hope that somewhere 18 somebody has the ability.to fund that work, because I think 19 it will be very important.
20 Anyway, you can see the predicted versus the 21 measure -- these numbers are extremely close, I mean, they 22 are really phenomenal -- as well as the Iodine 129.
We do 23 see that the predicted is a more conservative value, but I ')I =
24 still in all, these numbers are terrific.
If we are within 25 a factor of two they are fantastic.
x s
53 1
MR. STEINDLER:
What was the input data for the
,,_q.
- '-) -
(
2 3R-STAT code?
3 MS. ROBINSON:
The input data is information on 4
reactor-specific design conditions which would be the 5
reactor coolant volume, the fate of value which is the 6
Iodine carry over fraction in the steam, the steam carry 7
over rate for BWR's, the composition of the fuel originally, 8-mixture of uranium, and -- let's see what are the other i,
9 conditions --
1.
l 10 MR. ORTH:
There's at least one more.
For 11 example, let me just -- for the edification of those who may l
12 not have followed everything, and I hope for my own because
{',/]-
1 _j 13 I have to verbalize as I go along.
14 The amount of Iodine 131 that is in there is a l
15 balance between that which is being produced and released 16 into the system, minus the amount on a steady state-that is 17 being removed by the clean up system, minus the decay.
18 MS. ROBINSON:
Right.
19 MR. ORTH:
So, it also has to be very specific 20 according to what the fraction of coolant that is going 21 through the clean up system.
For example, if nothing was 22 going through the clean up system then you would still 23 arrive at the constant release rate at a steady-state,
)
constant concentration of Iodine 131, even though the Iodine 24 25 129 was increasing all the time.
54 L,y l'
MS. ROBINSON:
You are right.
The reactor water i
)-
2 clean up flow rate or the CVCS flow rate is, in fact, the 3-parameter that is used.
Thank you. I can tell you were
.4 right with me.
5 MR. VOILAND:
What assumptions are made about the 6
homogeneity of Iodine in the fuel?
The cesium migrates out l
7 during the exposure in the rector so that it is on the 8
cladding -- the fuel interface.
What assumptions are made i
9 about that?
10 MS, ROBINSON:
Steady-state in the fuel, 11 primarily.
12 MR. VOILAND:
Homogeneity uniformly distributed or
( h(_,/.
l 13-migrate like the cesium.
14 MS. ROBINSON:
That we are at a steady-state.
15 There are some things that do happen in the fuel.
Not only 16 do you have the production and decay taking place and the 17 release mechanisms, but you actually have transport in the 18 gap which includes some of the Iodine plates out --
19 MR. VOILAND:
That's my question.
20 MS, ROBINSON:
We are assuming all those 21 conditions in the fuel are constant, that they are at 22 steady-state, which is also the other parameter which is t! e 23 reactor power.
We need to look at the data after 30 days of f ))
24 reactor start up, because it takes that long for the Iodine 25
-- all the Iodine to grow into steady-state decay -- does
t 55 js 1
that answer your question?
a i
2 MR. VOILAND:
No, it didn't.
Do you assume that 3
the Iodine migrates to the gap?
4 MS. ROBINSON:
Yes.
It comes from the fuel to the 5
gap and migrates from the gap to the coolant.
6 MR. MOELLER:
If you have a plant that is having 7
problems in unscheduled SCRAMS and so forth, how much does 8
that upset this?
I am assuming that you would prefer a 9
plant that is just humming along at steady-state.
10 MS. ROBINSON:
If they scram they come down.
It
-11 depends on how long --
12 MR. MOELLER:
It doesn't -- back to what Gene
(_/
13 Voiland was talking about -- if you had an excessive number 14 of scrams in bringing it back to power and so forth, doesn't 15 that tremendously or significantly affect the release rates 16 and so forth?
17 MS. ROBINSON:
It will tremendously stress the 1B fuel, of course.
19 MR. MOELLER:
Yes.
20 MS. ROBINSON:
If there were any defects in those 21 fuel at time of scram and coming back up to power, you may see a change in fact in the defect on the release rate.
22 1
l l-23 That is possible.
/~'%
(Q a
24 MR. MOELLER:
But your system then should, t-1 25 although that is occurring, your methodology should still l
.a l
56 i
1 give you the correct or a good estimate.
, ~3
]
'~'
2 MS. ROBINSON:
Yes.
Actually, I think -- the code 3
is capable of handling reactor power and coming back up, 4
But in reality -- if the one time period that we don't use-5 is the first 30 days of start up because the reactor Iodine 6
-- all of the Iodine'have not grown in.
If you think about 7
it,, that also -- we then assume for those 30 days that the l
8 Iodine release date was at day 30, 9
For the first 30 we assume that it is 10 conservative.
11 MR. MOELLER:
That is conservative.
12 MS. ROBINSON:
Yes.
[
(s,)\\
13 MR. MOELLER:
Okay.
14 MS. ROBINSON:
If you think about it, it's a 15 conservatism -- it is a very small fraction over the whole 16 cycle or over the whole operating history of a disposal 17 site.
18 I will be happy to address questions, if you do 19 think of more after Lynne's presentation.
I do think that 20 some discussion about future compliance and how the industry-21 is going to address that would be appropriate at this point.
22 MR. MOELLER:
I guess the question we obviously' 23 have is -- it is good that John Greeves is here -- you have
~
(d
\\
24 done this.
Now, what is the next step to see if the NRC 25 will accept it,
57 1
MS. RODINSON:'
Lynne has that answer.
I will turn O-1 2
the microphone over to here.
3 MR. MOELLER:
Lynne has that, okay.
4 MS. FAIROBENT:
I am not sure that I have the 5
complete answer Dr. Moeller, but let me tell you what we 6
have done and where we are with the NRC on this issue.
We 7
have met twice with the NRC in att information exchange i
8 meeting format between EPRI, the ifRC staff and NMSS.
9 What Pat did not mention !s that EPRI and the 10 utility industry does not own 3R-STAT.
3R-STAT is a i
11 proprietary code that was developed by Gene Vance of Vance 12 and Associates, and is owned completely at this time by.
)
13 Vance and Associates.
We are working however with Gene for 14 the release of some form of this code in order to have the i
15 utilities be able to purchase it and then be able to use 16 this code as an instrument or implement format.
I 17 EPRI is doing the validation work.
One of the 18 concerns that we had, and I guess from my view and from 19 NUMARC's view is that-it is fine for us to go out and 20 correct or clarify for the states, their source term 21 quantification problems for siting the new facilities.
22 However, we have to be a little bit sensitive as to what 23 that potential impact might be on utility operations or
(}
24 accusations that what the utilities currently are reporting 25 is not sufficient, does not meet the legal requ.irements
58 l
,r-.3 because we now have out in public forums and meetings saying 1
l !
JL 2
that we know the Iodine 129 numbers are over estimated by at 3
least a factor of 1,000 in some-cases.
4 We could have people making accusations that we.
i 5
are not reporting correct information.
We are.
We are L
6 doing what the regulations require.
One of the things that l
7 we are talking with NRC staff -- I think the next step from 8
the utility viewpoint -- will be some alternative compliance 9
and reporting mechanism in addition to what is in Part 61 10 today.
11 What we do know, and this touches a little bit 12 upon what I want to talk about, is that we do have to look
)
13 at some alternative viable options for waste streams other i
l 14 than DAW.
If you remember what Pat was saying, although the l
l 15 preliminary data leads us to believe the Iodine would all be i
l 16 in the DAW, we know that this is not correct.
We think we I'
17 can use current systems and scaling factors for DAW without l
18-any problem.
We have to look at it for the other waste 1~
l 19 streams that we are shipping off site.
20 (Slide.]
i 21 We talked a little bit earlier about what are we 1
22 required to list on the manifest today.
What we have to 23 look at is an alternative reporting mechanism to the l
[~'/)-
24 manifest requirements in the future.
L
\\_
25
[ Slide.]
l l
l
59 1
As I said, we think we can live with the current
-~(
T- ) -
2 compliance scheme of scaling factors for everything other 3
than the Iodine and technetium and the generic scaling 4
factors for DAW.
In very preliminary -- and I want to i
5 stress this -- very preliminary, not official utility 6
position or industry position, we have had some discussions 7'
with the NRC staff of where do we go now.
We correct'the 8
source term problem from the states.
The first time NRC 9
will probably get a license from one of the compacts that 10 has used this methodology will be the first time they_are 11 doing a technical assessment on the predictive code.
12 That, in our estimation, is a siting compact
( ))
t'
.13 problem.
It is not a utility industry problem specifically.
14 However, we also have to recognize since we are generating i
15 the isotope, we can't keep our head in the sand and just 16 leave then out there by themselves which is why we are 17 involved with this work in assisting them.
We also have to 18 look at what other mechanisms, once we get the source term i
19 identified for further compliance determination and 20 evaluation.
21 Pat alluded a little earlier that she did not i
22 believe that this test should be run every year, and I think i
23 I would have to agree.
Once we validated the code and we l
24 are all confident that the validation and the predictive 25 methodology of the code is working, what we have thrown out L
60 1
is that perhaps on an annual basis we would -- say January
,-(j 2
of the year do a predictive estimate of what we feel we 3
would release in the next year as far as the Iodine 129 4
release rates.
L 5
on a quarterly or semi-annual basis similar to 1
6 what we do with-liquid and gaseous effluents, go back and 7
refine that data and perhaps report that at that frequency 8
to the NRC and to the disposal site operator and state 9
regulatory agency where that is.
That is what we are 10 looking for.
11 Also, realizing or have reccgnized that just to do 12 that sort of reporting mechanism without any repetitive
/%
(,)
13 validation to ensure that the model is still working, that 14 everything we went into as far as coming out on the 25 conclusions based on the initial validation, we are still 16 holding up.
We would probably have to buy into'some sort of 17 repeat validation program.
We have not talked details with 18 NRC staff, we have not asked them to say yes, no, that they 19 might accept this.
20 I think that they are open to these discussions, 21 and we are all open to solving this problem down the road.
22 I think that what we would probably come in and suggest is 23 something on the order of maybe 20 percent of the power
N 24 (O
plants run a test maybe every three to five years so that 25 you would have some tests being run each year but it would
-f 4
61 1
not be the entire industry testing each year.
.fs.
'~')
t 2
MR. STEINDLER:
Why is the NRC involved at all?
I 3
thought the issue was manifest numbers and local 4
regulations, state regulations.
5 MS. FAIROBENT:
John, do you want to answer that?
"I 6
MR. GREEVES:
We are involved in the sense that we 7
have a responsibility for the manifest process, so that's 8
one reason we arc involved.
A second reason we need to be 9
involved is, somewhere down the road there are some non-i 10 agreement states with host-states, and the NRC is going to 11 be right in the middle of this issue.
.12 We have an interest obviously in what is going on
.,q
' \\s,/
13 here, and we need to for our own part sort out how-we are i
14 going to do with this issue.
In addition to that, we have a 15 responsibility to advise the regulators in the agreement 16 states on difficult issues, and they are bringing this issue
-17 to us.
So, we have about three different angles on why 18 we need to be involved in this particular issue.
!19 MR. STEINDLER:
Can you help me out, what is the 20.
issue on the manifest -- where do you folks get involved in 21-the manifest?
22 MR. GREEVES:
We wrote the regulation that sets in r
23 place what is required to be reported on the manifest.
To
(}
the extent that there are problems associated with that, we 24 25 need to be aware of those.
If there is a need, for example,
=
62 1
on any piece of the regulation for revision, we are
,e 3 i).
2 naturally a source to-take a look at that issue.
3' MS. FAIROBENT:
I think the other thing, Dr.
1 4
Steindler, is that one reason I personally would like NRC 5
involvement on behalf of the utility industry is so that we i
6 don't get into an inconsistent interpretation of what might.
7 be acceptable state to state.
We are not going to totally 8
eliminate that, but the agreement states do look to NRC for 9
overall guidance in how they should go about doing their 10 regulatory programs.
11 I think the first place we have to start is in 12 dealing with NRC and then dealing with anything additional -
/"N
-f
(_,)
13
- the utilities with any specified agreement state would 14 have to work out with the agreement state regulators 15 ultimately.
16 MR. STEINDLER:
Another question is indicated in 17 one of your slides that the current method.of compliance is 18 using scaling factors, is apparently satisfactory for all 19 radionuclides with the exception of technetium and Iodine.
20 If Iodine was as apparently far off as you indicate, whence 21 do you get this faith that the rest of the scaling factors 1
22 are in pretty good shape?
l 23 MS. ROBINSON:
Primarily we know that the rest of Ii 24 the radionuclides that we have the analytical sensitivity in V
l L
25 the commercial laboratories.
We are not running into the 1
1
63 1
situation of conservatisms introduced by the use of that LLD 7
s-x
/
2 factor that we had used.
However, we do believe that we can 3
improve those numbers once this model is expanded.
I think 4
this is a powerful tool, as more-of the validation data will 5
confirm.
6 I think that this tool will at least provide us 7
the basis to at least judge from a fundamental base how 8
accurate or what is the over conservatism in the other 9
radionuclides.
10 MS. ROBINSON:
Cesium 137 and 129 -- Iodine 129 11 and your predicted versus measured comparison, did you do 12 any analysis on technetium?
f3(,)
13 MS, ROBINSON:
We are doing the analysis,.but.we 14' do not have any data to bring forth at this moment.
15 MR. HINZE:
If I may, I understand correctly that 16 the input parameters for the model are readily available for 17 past performance.
Is it therefore the intention to.use this 18-to correct, modify the manifest documents that have come in-e 19 the past?
20 MS. FAIROBENT:
I think that is something that we 21 would have-to have discussions with NRC to take a look at.-
L l
.l 22 It may be'something that, under the site closure performance 23 assessment requirements for the three facilities that are l
/\\
24 going to close down -- the two that are going to close in
%A 25 1993 -- that is something that I thin'k those operators are
64
- r 3 going to have to look at.
1
()
4 2
John, maybe you could provide some clarification.
3 I also don't know that they track-Iodine 129.though since 4
day one of'their operation, so it may be that they will be 5
okay under the current procedures.
6 MR. GREEVES:
I think anybody that is following
' l 7
this process knows there's a great deal of sensitivity about 8
that issue, so I don't want to go too far into it.
The 9
states that are involved in this process now -- we have 10 three states out there -- what is clear is that everyone is 11 agreeing that you have a lower limits of detection problem.
12 We know it's no higher than that, we don't know how low it
, m,
(
)
13 is.
14 There is currently some debate about questions of 15 correcting manifests.
You can conjure up a lot of concerns 16 on people's part about the industry going back and cooking 17 the data.
I don't think it serves any purpose to jump into 18 this discussion here.
What is clear is what is on there at 19 the present time, is a lower limit of detection and that the 20 number is something below that.
Whether any particular i
21 state will elect to go back and try to define what that 22 number is, I think is pretty much up to them.
If they want 23 to consult with NRC, we would be happy to consult with them, j
l
[O~T.
24 but I don't want to predict or project what I think they 25 might do.
)
l l
E a.
s 65 e,
1 MR. HINZE:
I understand that.
If there are any j-ms.
(
l'
--' ~ ~ ' '
2 technical problems that anyone can envision that might h ar9 3-in the way of that prediction of past events --
4 MS, ROBINSORl The code itself has the capability 5
to go back historically from day one to determine-the fadine 6
release rate from every reactor.
7 MR. HINZE:
And, the input parameters have all 8
been measured in the same way for the historical. time over 9
which we might be interested.
10 MS. ROBINSON:
Pretty much.
11 MR.-HINZE:
There is nothing that would invalidate 12 it?
C\\
(_)
13 MS. ROBINSON:
Not that I can recall right off L
14 hand. Iodine detection is pretty straightforward.
I think 1
L 15 you could use it to give a number. One point that I do want 16 to make is, we are using the best available technology that 17 was available to us up until even now, we are using the best l
18 available technology.
The Battelle methodology really is a V
j 19 research tool in terms of the analytical method that they l
20 are using.
1 21 There is no commercial laboratory that provides
'22 this service.
In fact, I would strongly advise that the L
23 exact wrong thing to do is to run out and do more waste L.
['I 24 stream analysis with the current methodology or even with
%_/
25 the analytical capability that Battelle has on a waste l
i r
66 1
stream basis.
2 You can't overcome the practical limitations ~of i
3 sampling high dose rate waste st'.eams.
We have the tool and i
4 we are validating a tool that is going to-give us the best, 5
most accurate and I believe very defensible determination of-6 the Iodine 129 release rate.
7 MR. GREEVES:
Just to clarify, the NRC staff has 8
met with them several times.
We really don't have a report 9
to review.
Effectively what needs.to happen here is,_they 10 need to finish their work, submit this work to interested
\\
11 parties and then it needs to ne critically reviewed by the 12-NRC staff. None of that has happened to date.
That needs to,
(~
\\,)/ '
13.
play out in the future.
I 14 MS. ROBINSON:
And, of course, that is the 15 industry plan.
You will first I believe, see the j
16 methodology and the application from the license disposal L
l 17 sites will be used for derivation of the source term, but a L
18 separate NRC submittal will be proposed as an alternative 19 methodology to sampling analysis for compliance.
20 MR. GREEVES:
Personally, the NRC can't wait to l
21 get it in an application.
We need to see an approach to 1,
22 this methodology early on, and we need to decide how we are 23 going to deal with that way in advance of an application f
24 coming to us.
We also, on a consulting basis to the states, 25 need to give them our best advice on this topic.
l
67 1
We are not looking to wait until we get an O
2 application.
We are neeting with these folks and are 3
looking forward to receiving their report, and than 4
critical 3y reviewing it and see where we agree and whera we 5
don't agree, and where we go from here.
6 MS. FAIROBENT:
pat, you might just want to touch 7
upon when the results are going to be done.
8 MS. ROBINSON:
Well, I believe at this point in 9
the progress of the analytical -- there are a lot of power 10 stations that we have done the test column program at, 11 Because of the limited availability of this equipment, it is 22 DoD equipment.
We line up behind DoD priorities and in some
)
13 cases DOE priorities for the use of the instrument.
14 We are a bit delayed in getting analytical results 15 for the validation for those very reasons.
I certainly hope 16 that at the absolute latest we have a validation report that 17 could be submitted to the NRC for review -- technical 18 review, as an alternative methodology by the end of this 19 year.
We are absolutely moving as fast as our little feet 20 can go, 21 We realize ; ere all the compacts are.
Certainly, 22 as the industry knows, the interest as quickly as poscible 2?
incorporating this methodology into our current reporting
()
24 and compliance programs.
25 MR. MOELLER:
You mentioned going back to the
i 68 1
methodology would enable you to go back and calculate from 2
day one for various nuclear power plants.
Does the nuclear 3
industry keep the records of the coolant --
4 MS. ROBINSON:
Yes, they do.
They are required 5
to.
6 MR. MOELLER:
From day one?
7 MS. ROBINSON:
Yes.
8 MR. MOELLER:
That's correct?
9 MS. ROBINSON:
Yes.
10 MR. STEINDLER:
How big is this code thst we are 11 talking about?
12 MS. ROBINSON:
You mean lines of code?
13 MR. STEINDLER:
However units you want to give it.
14 Hours of running on a Craig?
15 MS. ROBINSON:
I can give you an example of how 16 long it takes to run to batch one job of coolant data for 17 one cycle, it will take on the order of about seven hours 18 probably to run on the 38625 megahertz machine on a PC.
W3 19 could run one data point real fast, but it's doing a lot of 20 math.
21 MR. MOELLER:
Are there other questions?
- Yes, 22 Gene.
23 MR. VOILAND:
Would you remind me what the
()
24 inventory limits are, please, on the --
25 MR. MOELLER:
For tne disposal site?
i 69 1
MR. VOILAND:
For the disposal site.
2 MS. ROBINSON:
I realized that after I sat down I 3
never really did give you a very good picture of what 4
exactly do those curies mean in a preliminary performance 5
assessment standpoint.
There is no curie point per se on a 6
disposal site.
The source term number is input into l
r 7
performance assessment codes.
i 8
MR. VOILAND:
I thought you said there was an 9
inventory limit?
10 MS. ROBINSON:
Not on the disposal sites.
There 11 could be one imposed if a disposal site -- of course, they 12 have to-meet the performance requirements of 10 CTR Part 61.
13 They could put a cap for the Iodine 129 inventory on that i
14 disposal site.
15 John, you were going to comment?
16 MR. GREEVES:
Basically what happens is, you have 17 the performance requirements of 25 millirem per year.
What 18' you would do is go into a code and back out how much Iodine 19 it would take to bump up against that, and then a compact 20 would say my code tells me I can't have any more than this 21 amount of Iodine in my site so that's my inventory limit.
22 There is none explicitly required.
You would back the 23 number out.
c.
24 MS. ROBINSON:
To give you just a point of 25 reference where we are with the numbers, there was one t
.-_y-y y
.c.,
70 1
compact that had done some preliminary performance s
(
)
2 assessment with three curies of Iodine, and they found that 3
they exceeded the 10 CFR 61 requirement, I think it was by a 1
4 factor of 100 in which case they would need one-one 1
5 hundredth of the number of reactors or 100 more disposal 6
sites.
That is how they would have had to limit it.
7 MR. VOILAND:
When you talk about three curies of 8
Iodine 129, that represents a whale of a lot of --
l l
9 MS. ROBINSON:
That's over --
10 MR. VOILAND:
I have been doing some scribbling 11 here, and it seems to me that a ton of spent-fuel contains 12 about 15 millicuries of Iodine 129.
That makes about 60 i O
(,/
13 tons per curie or 180 tons.
I cannot conceive of having 14 that much fuel fail in a reactor.
15 MS. ROBINSON:
You are absolutely right.
16 MR. VOILAND:
If you go back into the reactor and 17 look at their experience -- because they have conditions 19 that they have to meet -- their technical requirements on 19 release of gases, and those gases can be related back to the l
20 amount of failed fuel.
It just seems to me with that 21 incredibly long half-life that there is no way that you can i
22 get that amount of -- the source term is of interest to me l
23 is, what comes from the fuel into the waste.
('~N 24 MS. ROBINSON:
You are absolutely right.
We just
\\b 25 don't make that much.
That is exactly right.
l
71 1
MR. VOILAND:
Why are we concerned with it if we
)
2 don't make that much?
~
3 MS. ROBINSON:
Because what our current 4
methodology and reporting on shipping manifests says that S
because of all these problems we described we have a 6
fictitious -- a number that certainly doesn't represent 7
reality or even practicality.
8 MR. VOILAND:
If the regulation allows you to 9
compute that if you have a good basis for it -- the facility 10 that I was involved with, we used to calculate what our 11 wastes were.
If you can demonstrate by other kinds of 12 measurements that the amount of Iodine which can got into
)
13 the waste from the reactor operation is of little 14 significance, why isn't that an adequate calculation?
15 MS. ROBINSON:
We have done some preliminary 16 calculations assuming a fuel performance failure rate and 17 certain burn on that fuel.
18' MR. VOILAND:
How does that work out?
19 MS, ROBINSON:
When you run it through the 20 preliminary performance assessment and you assume that all 21 reactors for 30 years operate at those conditions, you would 22 exceed the performance objectives of 10 CFR 61, which is why 23 we have gone to a more rigorous resolution of what is being C\\
24 released from defective fuel.
\\_)
25 MR. VOILAND:
Is that all reactors?
You make some
i I
72 I
assumption then about how many disposal sites you have?
I
! \\
2 MS. ROBINSON:
Yes.
Right.
Disposal sites --
3 MR. VOILAND:
I am astonished that you have that 4
much because the fuel does perform exceeding --
5 MS. ROBINSON:
What we really have is one-half a 6
millicurie for, in this case, a compact that had seven i
7 operating nuclear power plants versus the three curies that 8
was actually derived from sampling analysis data.
9 MR. VOILAND:
One-half a millicurie, again, is 10 equivalent to 30 tons of fuel.
I have a hard time seeing 11 it.
12 MS. ROBINSON:
Over 30 years, did you calculate 30 (m) 13 years.
14 MS. FAIROBENT:
You're hitting upon the 15 conservatism that was left in, which is the assumption that 16 everything released ends up in the waste.
We know that that 17 is not true.
We know that some Iodine is going to get up 18 through the gaseous effluents as well as a little bit out 19 through the liquid, but one of the conservatisms that we 20 have left in was the assumption that everything released 21 ends up in the waste.
22 That may be where you are getting a little bit of 23 a discrepancy --
24 MR. MOELLER:
You are saying that 30 tons or 25 whatever of fuel just totally got out of the cladding and
73
~
1 shous up or the Iodine from that much.
,s'
)
2 MR. VOILAND:
The Iodine from it.
3 MR. MOELLER:
Are there other questions?
4 MR. STEINDLER:
You indicated in this table where 5
fou had validation testing of five, six or seven plants and 6
in parenthesis you had the estimated number of defective 7
fuel pins.
Was that estimate obtained by methods 8
independent of the Iodine --
9 MS. ROBINSON:
That was really estimates provided 10 to us from the reactor fuel performance individuals at the 11 power reactors.
j 12 MR. STEINDLER:
They measure off gas rates or
/^'s V
13 whatever have you.
14 MS. ROBINSON:
You know, these truly are 1
15 estimates.
I think if they are plus or minus two they are 16 barely in the ballpark.
17 MR. STEINDLER:
What I am getting at is the same 1
18 kind of analysis that Gene just went through, roughly what 19 the fuel inventory and burn up and Iodine content is.
In l
20 tact fuel pin, and if you failed two, you know how much 21 Iodine you could possibly get out, you know what the volume 22 is and would be able to calculate again on the same kind of 23 assumptions.
How does that match up with your actual
(T 24 experience?
V 25 MS. ROBINSON:
We don't know yet.
We haven't i
r 74 1
correlated it.
But I know these actual pin predictions are
<3 I
)
2 very difficulty -- very questionable.
Let me give you a 3
real life example.
4 EPRI produced a code out of their fuel performance 5
group called CIRON.
CIRON is essentially only that 6
empirical diffusion model from trying to determine defect --
7 number of defects only, not even defect size.
That code is 8
predicting within a fact of two with the exception of a case 9
in point was Yankee Row recently had exercised the code and 10 predicted they had 20 pin defects.
When they shut down and 11 went in they had over 400 pin hole defects.
It was from 12 some previous work that was done.
,sk,)
13 Part of what happens with fuel in zirc cladding 14 is, one they have a defect they have a secondary effect of 15 secondary effect of hydrating from water coming in, so you 16 see diffusion.
Maybe you don't have it at the beginning of 17 the cycle but toward the end of the cycle you do.
Yankee 18 Row happens to use the old stainless clad which doesn't 19 secondary hydride, and what they had were a lot of pin hole 20 defects.
They weren't able to resolve the actual number of 21 defects because of the limitations of the code primari; 22 geared toward diffusion controlled plants.
5 23 That is the strongest tool that the industry has 4
(~T 24 available today, is the CIRON code.
We are going to do some
-)
25 very interesting things in the future.
Part of what we will
+
i 75 1
be doing, what NUMARC will be doing is sponsoring some work
,s
)
2 to correlate fuel performance using EPRI's CIRON code, using 3
Vance and Associates 3R-STAT code to predict the occurrence 4
of hot particles, in fact, fuel hot particles.
If you think 5
about it, if the code can tell you when you are in a 6
diffusion controlled mechanism reliably, those are the same 7
conditions that one would think that okay, fuel particles 8
are being released into the coolant and those in fact are i
9 the fuel particles that give us problems with skin exposures j
10 in the future.
11 So, we hope that this tool finds many applications l
12 in nuclear power plant operations.
I think it will be very l (3 i,/
13 interesting to see how these things do correlate out.
t 14 MR. STEINDLER:
One other quick comment.
I think 15 there may be more than one outfit that can do your mass spec 16 analysis to the kind of level that you --
17 MS. ROBINSON:
I would love to know about them 18 right now, because I am backlogged to the teeth with --
19 MR. VOILAND:
Is that Battelle PNL?
20 MS. ROBINSON:
Yes, it is Battelle Northwest.
i 21 MR. VOILAND:
I used to manage that laboratory.
22 MS. ROBINSON:
Oh, you did.
Well, I used to work 23
(~'
24 MR. VOILAND:
Indirectly.
(
25 MS. ROBINSON:
I used to work there, five or six I
76 1
years ago.
I would be very interested in finding that out.
2 MR. MOELLER:
Does that about wrap it up?
3 MS. FAIROBENT:
Dr. Moeller' I would just like to 4
say that when the results are in and the report is prepared, 5
after you have had a time to look at the report, we would 6
certainly come back and talk to you about where did we 7
really end up versus where we think we are going to end up 8
as we are today.
9 MR. MOELLER:
Let me, on behalf of the Committee, 10 thank you for coming.
I, personally, found it to be a very 11 exciting and interesting report.
It is something that I 12 wasn't aware was going on, and it has the potential for
(~N
's,)
13 making a major contribution to a real problem -- to solving 14 a real problem.
15 So, thank you again.
16 MS. ROBINSON:
You are welcome.
Thank you.
17 MR. MOELLER:
The Committee will take a 15 minute 18 recess.
19 (Brief recess.)
20 MR. MOELLER:
The meeting will resume.
The next 21 item on our agenda is a briefing on the BEIR V report, the i
22 National Research Council's Committee on the Biological j
23 Effects of Ionizing Radiation.
The report is entitled:
()i 24 "The Health Effects of Exposure to Low-Levels of Ionizing
\\.
25 Radiation."
We have with us Dr. Arthur Upton', who chaired
77 1
the BEIR V Committee, who also is well known for laany other 7s
\\
2 activities including his representation as one of the U.S.
3 Members on the International Commission on Radiological 4
Protection and his years of contribution to the comparable 5
organization here in the United States, the National Council 6
on Radiation Protection and Measurements.
t 7
He serves when he isn't busy with all of these 8
other things, as Chairman of the unit at the Institute of
(
9 Environmental Medicine at New York University.
Arthur, it i
10 is a distinct pleasure to welcome you.
You will notice 11 that by the size of the audience that there are many people l
l 12 who share with no the excitement at hearing from you the
(,)
13 latest developments as reported in BEIR V.
14 MR. UPTON:
Thank you very much, Dave. I 15 appreciate those kind words of introduction.
I am never 16 sure that when I speak about a report like this what the 17 reception will be.
I have a friend at NYU who has a poster 18 on his wall that says, when you think they are going to run 19 your out of town get out in front and make it look like a l
l 20 parade.
21
( Laughter. )
22 I can remember the way Joe Califano opened a talk 23 that he gave at NIH when I was there about a decade ago.
At
/~'\\
24 the time, he was the Secretary of Health, Education and
. V l
25 Welfare with his background as an attorney.
He stood in the
78 1
auditorium looking out on the many faces of the scientists
,s
(
)
I 2
from NIH and said he wasn't really sure how to address a j
l 3
group of scientists as an attorney.
4 He recalled an experience he had when he went to 5
speak to his daughter's third grado class.
He pondered how 6
to do that, and finally decided how he would tell them what 7
it was like to be the Secretary of Health, Education and 8
Welfare and so on.
At the end of his talk a little girl 9
came running over to him and he said, what did you think of 10 that?
She said she thought it was very dull.
11 (Laughter.)
i 12 Joe was taken aback and crest fallen and began to l\\
l (,/
13 lick his wounds, and his daughter came running over and she 14 said, Daddy, don't pay any attention to her.
She is a smart 15 allec.
She just repeats what everybody else says.
16 (Imughter.)
17 I did bring along some slides for cues.
If I 18 could ask someone to turn on that machine, please.
19 (Slide.)
20 Dave did indicate the subject of my remarks, the 21 summary of the so-called BEIR V report, the Committee on the 22 Biological Effects of Ionizing Radiation, the fifth l
23 incarnation of this Committee, if you will.
It represents j
the fifth study under the auspices of the National Academy 24 25 to look into what we know about the risks of ionizing
+-
79 1
radiation.
The charge to this Committee was really to y
I
)
2 update the BEIR III Committee which appeared in 1980.
3 The BEIR III report looked broadly at ionizing 4
radiation from all sources, with particular reference to 5
effects on the population from exposure at low-levels.
The 6
sponsor of this BEIR V study is Chirpic.
The Committee was 7
a committee of many disciplines.
The V.'.ce Chairman, Dan 8
Hartl, is a geneticist in Washington Uriversity, St. Louis.
9 Bruce Becker is from Albuquerque, a r.pecialist in internal 10 emitters.
Kelly Clifton, a radiobiologist from the 11 University of Wisconsin.
Carter Denniston, the geneticist 12 from Madison, University of Wisconsin.
Ed Epp from Mass
()
(jl 13 General Radiological Physics.
Jack Fabrikant from Berkaley, 14 physician, biologist, radiobiologist and so on.
I won't go 15 through the entire committee.
16 It represented, I think, an excellent assortment 17 of specialists from different disciplines, each a noted 18 authority in his or her field.
19 One of the main purposes in reassessing what we 20 know about radiation effects at low-levels today, what we 23 can predict in the way of risks, came out of the changes in 22 dosimetry of the atomic bombs.
The old dosimetry, the so-23 called T-65 has changed.
Hiroshima is substantially lower.
1()
24 The contribution of gamma rays is somewhat increased as one
'd 25 moves out from ground zero.
80
,73 1
Again, this very much smaller neutron component at 2
Nagasaki -- traditionally there was a substantial difference 3
between the two cities and the magnitude of the cancer 4
excess at a given distance.
This had always been attributed 5
to what was thought to be a very much larger neutron 6
component at Hiroshima, neutrons being known from many i
7 experiments and clinical observations to be more effective 8
radiobiologically than gamma rays.
9 If we look at the situation in Hiroshima comparing 10 the new dosimetry, the DS-86 with the T-65, we see here 11 looking at the free air, the much smaller neutron component, 12 a larger gamma ray component without going into the details
()
(_)
13 of the effects of shielding, structures and the shielding 14 within the body.
One sees, allowing for tnese factors, 15 about one-half the effective dose equivr. lent with the new 16 dosimetry assuming an RBE of 20 for neutrons.
17 Similarly in Nagasaki, a very much smaller dose 18 equivalent -- average dose equivalent.
That means, of 19 course, that at a given distance there is a greater effect 20 per unit dose if the dose equivalent is smaller.
Everyone 21 foresaw that risks were likely to go up as a result of these i
22 changes in dosimetry.
It wasn't clear how much they would 23 go up.
[/)
24 Another factor is the fact that in the various s_
25 cohorts of A-Bomb survivors, those irradiated less than ten
81 1
years of age, those irradiated between 20 and 29 -- that 2
should be ten to 19, 20 to 29, 30 to 39 and so on.
In each 3
cohort with increasing age, the excess plotted here on tho 4
left has gone up.
Here is the youngest cohort.
There was 5
an excess of cancer during the time the cohort was in 6
adolescence, the excess was larger during the time the 7
cohort was in its 20's, still larger when the cohort was in 8
its 30's, still larger now, some 45 plus years later, not 9
the cohort is in its 40's.
10 So, the excess in each cohort has grown with 11 attained age.
The excess plotted on log scale -- so these 12 are really substantial increments -- plotted in relationship l n(,)
13 to the natural age dependent incidence of cancer, which also i
14 goes up exponentially with age.
This is plotted in a i
15 straight line against the scale on the right.
So, one is 16 seeing this radiation induced excess growing at pace with t
{
L 17 the natural increase in risk.
18 MR. MOELLER:
Art, to help me, are the excesses at 1
19 between age 20 and 30, is that a cumulative total including 20 the excess between zero and 20?
l 21 MR. UPTON:
That is age-specific.
I 22 MR. MOELLER:
Age-specific, okay.
23 MR. L' a'ON :
This is age-specific death rate over
/^)
24 here.
l
\\_)
25 MR. MOELLER:
Okay.
F i
82 1
MR. UPTON:
Actually, it's deaths per 10,000 i
.~
7
'{
)
2 person year Gray.
It is an annual excess during that time.
i 3
(Slide.)
4 At the time of the BEIR III report which came out 5
in 1980, it wasn't clear how to model the excess with the 6
exception of leukemia.
I'm sorry that these are not clear 7
for you.
8 MR. MOELLER:
I think it's the slido projector or 9
something.
10 MR. UPTON:
I can't seem to get the whole thing 11 clear.
One knew at that time that the excess of leukemia 12 was beginning to go down, so the leukemia appeared.
The
/^N
(,)
13 curves showed irradiation at this point in time, this is a 14 time scale, this is incidence on the vertical radiation at 15 times zero.
Following the elapse of a latent period, take 16 some time for a cancer to grow into clinical size, clinical 17 expression.
After the latent period, the irradiated 18 population would show an increase over the background rate.
19 This is the spontaneous or baseline rate, which is going up 20 with age.
21 It was already clear in 1980 that leukemia made 22 their appearance and risk went down. One really didn't know 23 how to model the solid cancers.
One knew that one had an l
I 24 excess at this po!'it in time, but one didn't'know whether
(~'}
\\/
l 25 the excess would simply continue essentially the same number
83 1
of extra cancers each year for the rest of the lifetime of zs
(
l 2
the population or whether the excess, once it appeared, 3
rather than be a constant number of extra cases would be a 4
constant percentage of the baseline incidence.
So, it would 5
grow in time and the curves would spread.
6 As we have just seen, since these curves are on a 7
log scale, what one appears to be seeing with the solid 8
cancers is an increase in relative riskt that is, the 9
relative risk is staying about the same.
As the population i
10 grows the extra numbers of cancers are growing with time.
11 The relative risk model, the multiplicative risk 12 model seems to be a model that fits the data better today.
(~N
(_)
13 At the time of BEIR III both models were used to make 14 projections.
This model projected substantially moro, l
15 several times more cancers than did this one.
The fact that i
16 the data are more compatible with the risk model is another 17 basis on which to explain the larger excess today, larger l
18 risk estimate today.
l 19 (Slide.)
20 Traditionally, the leukemia excess that I should 21 emphasize at this point that the BEIR V Committee tried to 22 review all of the available data that would enable it to i
23 assess what we know today about irradiation effects, what we
(~)T 24 can predict to be the risk at low-levels.
It was only for q
25 the A-Bomb survivors that one had an essentially random
84 1
exposure of both sexes and virtually all ages, the
,3
/
i
\\I 2
population then representing both sexes and all ages, a big 3
population.
4 In many respects, it is the experience of the A-5 Bomb survivors that provides the largest source of 6
information and major source of information, but not the 7
only source insofar as possible.
Other radiation 8
populations were examined and, to the extent that there were 9
discrepancies between results in different populations, 10 these were considered in arriving at the risk estimates.
11 I should mention that the Committee, particularly l
12 the epidemiologists, biostatisticians, mathematical modelers
(\\
(,/
13 wanted insofar as possible not to have to rely on published 14 data.
All of the publications from the big studies have 15 presented the data, usually grouped in some way, grouped by 1
16 dose range, grouped by age and so on.
The Committee thought 17 that these groupings of data, pooling of data that are 18 customary in publishing epidemiological or experimental data 19 might stand in the way of fitting the most reliable models 3
20 to the data.
So, insofar as possible the Committee I
21 requested from investigators original data.
22 We were fortunate in being given by the Radiation 23 Effects Research Foundation data tapes from Japan which
(N 24 enabled the Committee to model the data in its own way.
i
(/
25 Traditionally, as I mentioned earlier, Hiroshima showed a 4
5
85 1
leukemia curve more or less linear and much steeper than the
!, ~ ')
s 2
curve for Nagasaki.
Several years ago Strami and Dobson at 3
Livermore, looking at Loy and Mendelson's reanalysis of the 4
Hiroshima / Nagasaki A-Bomb dosimetry fitted the leukemia data 5
to the Loy Mendelson dose estimates, and suggested that the 6
two cities were really no longer different if one used new 7
dosimetry.
8 They suggested that the curve for Hiroshima was 9
not linear but more likely to be a linear quadratic kind of 10 curve which was characteristic of the animal leukemia data.
11 The new data out of Japan for leukemia, both cities 12 combined, show a curve that looks curve linear.
It fits a j
(,)
13 linear quadratic model best.
It saturates at somewhere 14 between three and four seivert and then bonds over.
15 The Committee just didn't think it likely that 16 survivors had experienced a dose of this magnitude.
We 17 didn't trust these data, so we essentially threw the data 18 away, truncated the data at four Gray and fit the data 19 within that dose range.
20 (Slide.)
21 This is the model, a linear dose term, quadratic 22 dose term, and then terms for time following irradiation, 1
23 time since exposure and age at the time of exposure.
Each 24 of those factors makes a difference.
['
}
25 (Slide.)
I
86 73 1
It would appear from the data, not just the A-Bomb
]
2 survivor data but other published data, that susceptibility 3
to leukemia is substantially higher early in life than 4
later.
This is relative risk plotted against attained age, 5
risk drops down, and as had been published earlier the 6
excess rises to a maximum and falls.
With chronic 7
granulocytic leukemia, the excess in any age group appears 8
to peak within about a decade and then almost disappear.
9 Now, the important thing here is not just that we 10 are dealing with a wave phenomenon but that we are decling i
11 with what would appear to be a family of diseases, not one 12 specific disease.
The latent period, as you con see, is l
(m.,
(_)
13 very different for the acute leukemia among those irradiated 14 in middle age than in the case of chronic granulocytic 15 leukemia.
Here, the peak is very much later, i
16 In those irradiated in childhood, on the other 17 hand, the acute leukemia still reaches its peak in the first 18 ten years.
So, I question whether it is really valid to 19 lump these together as we did in the previous figure in 20 plotting dose effect relationships.
It may turn out that 21 they are really quite different dose effect relationships, 22 irrespective of the time distribution.
23 When one begins to fracti6nate the data into age 7.~s}
24 and disease-specif*7 intervals, the numbers get so small
(
25 that one really can't model.
The data imply to me that in 4
1
r 87 1
childhood irradiation increases the incidence of this
\\'
2 disease, which Court, Brown and Doll showed some years ago j
3 for Great Britain and Wales.
These are the natural 1
4 incidents in the population of England and Wales.
5 This is childhood leukemia.
Children are 6
substantially higher risk of this disease than are adults, 7
the acute lymphatic leukemia.
The acute myeloid is almost 8
flat with age.
Remember, this is a log scale.
Chronic 9
granulocytic or chronic myeloid leukemia is predominantly a l
10 disease of the elderly, relatively infrequent in childhood, 11 increasing systematically with age.
12 None of the irradiated human populations have
(N
(_,)
13 shown a significant believable excess of this disease, l
14 chronic lymphocytic leukemia, classically the disease of the l
15 elderly individual.
l 16 (Slide.)
17 We are dealing with a number of cancers.
We have l
18 spoken about leukemia and, because of the early peak we just l
l 19 mentioned, has been one of the major diseases to be studied 20 in irradiated populations.
These are data for Hiroshima and l
31 Nagasaki published by Shimizu, et al., the attributable risk l
22 of leukemia is so high.
It is estimated that about 40 23 percent of the leukemia in the population, survivors, are
)
attributable to radiation, 81 of the 202 leukemia.
24 l
25 Notice if you will, there are now some -- were l
t
88 a
1 some 5,900 cancers between 1950 and 1985 in the survivor i'
2 population, only about 350 can be plausibly attributed to 3
irradiation, less than six percent.
If we didn't know that 4
the populations had been irradiated and we simply looked at 5
the cancer rates in different cities in Japan, we would not 6
realize that there was an excess. Epidemiological methods 7
are not sufficiently sensitive to detect a six percent 8
excess.
We certainly wouldn't have attributed this to 9
radiation.
There is that much variation from city to city 10 from other causes.
i 11 The excess of leukemia probably would have stood 12 out.
The excesses of the other cancers are nowhere near as
,,(,)
13 large.
Only because we an stratify the population on the 14 basis of dose or distance from ground zero, are we sure that 15 there is an excess, at least at the high doses.
16 As I emphasized, for leukemia, the data are most
)
17 consistent with the quadratic function or linear quadratic 18 that saturates.
The Committee really didn't believe the 19 reliability of the doses above four Gray but fits almost i
20 equally well using this model.
For the solid tumors, if you 21 take the solid tumors as a class, the data are more 22 consistent with the linear regression than with any of the 23 other.
So, the linear model was chosen by the Committee,
(~')i 24 not because we know it's right but because it gave a better
(
25 fit to the data.
l
89 1
(Slide.]
3 2
For the breast there are now a number of sizeable 3
population groups. This is a figure published some years ago 4
by John Boyce at the cancer Institute, Charles Land and 5
their co-workers.
They emphasized at the time that there is 6
a striking similarity between these different groups of 7
women, women exposed to the Atomic Bomb, women treated to x-8 ray of the breast for post-partum mastitis, inflammatory 9
condition that interferes with the nursing of a newborn 10 infant.
Radiation used to be the treatment of choice, a few I
11 rads and the inflammation subsided, and the mother could 12 nurse her infant and everything was fine, except the g) i 13 carcinogenic aftermath.
14
[ Slide.)
15 These curves represent breast cancer in women 16 whose breasts were irradiated not instantaneously as in the 17 case of the A-Bomb survivors, not in one or several brief 18 exposures in radiation therapy, but rather in multiple 19 fluoroscopic examinations of the chest in the treatment of 20 pulmonary tuberculosis.
21 Phen I was an intern, I can remember sitting 22 patients up in bed every couple of weeks and fluroscoping 23 them to determine whether or not we needed to put more air
(')
24 into the pleural space.
In the pre-antibiotic era, we V
25 treated tuberculosis by putting people to bed.
Putting air
F 90 1
in the pleural space to collapse the lung, rest the lung if
,s i
\\'
2 you will, we had to refill the air from time to time because 3
it was absorbed.
L 4
The refills of the pneumothorax required periodic 5
fluoroscopic examination, many small increments of dose.
6 The surprising thing is that when you normalize for age at 7
irradiation, duration of follow up, there seems to be 8
virtually the same excess irrespective of the duration of I
9 expocure.
That implied that if a small dose were somehow 10 were reparable or partially reparable so that by the time a Il second small dose came along it wouldn't be fully additive 12 with the first, then there ought to be a less steep curve, a
,rm()
13 shallower curve here than here; that that isn't clearly l
14 evidenced suggests that a small dose may not be well 15 repaired and there may be no threshold for this cancer.
16 So, for breast cancer based on the A-Bomb data and l
17 based on the other data, the model involved -- the 1
1 18 assumption of linearity between risk and dose -- the 19 decrease with time after irradiation and the correction for 20 time at irradiation, the young being very much more 1
21 susceptible than the elderly.
If you look at relative risk l
22 in relationship to attained age, the curves drop off.
We 23 know that after the menopause the risk of breast cancer in
(~]
24 the general population goes down dramatically.
This now is
%_)
l 25 the numerical excess plotted against attained age for I
91 1
different ages at the time of irradiation, five year olds, 7s
(
)
2 15 year olds and so on.
3 Notice again that the area under the curve is very 4
much larger for these younger groups than for the groups in 5
middle age or beyond.
That is comforting to the radiologist 6
because it argues that mammography involves relatively 7
little radiogenic risk of breast cancer, but we know a 8
substantial benefit in the form of early detection.
9 For lung cancer too, the data would argue that 10 relative risk is dropping off very fast with time after 11 exposure, so that the curves are not strictly parallel as we j
i 12 suggested they might be at the very beginning.
Similarly,
(
)
13 the relative risk in relation to age at exposure is dropping i
14 down.
l 15 So, one would predict from this that those exposed 16 very young would have a much smaller risk by the time they 1
17 reach the age at which lung cancer is normally expressed in 18 the population.
Again, a linear model, terms for time after 19 irradiation, decay with time once the latent period elapses.
20 Then, a less susceptible response in the female.
21 For digestive tract cancer, again, linear.
- Again, 22 terms for age at the time of irradiation and sex, female up 23 here greater susceptibility.
All other tumors, again,
()
24 linearity.
Not because we know it's right, it looks as
\\_/
25 though it fits better at this time at least and age at the
92 1
time of exposure.
-~
3
]
8
~
2 Taking the different models, one then derived 3
estimates for one-tenth of a seivert to 100,000 people, a 4
given sex but all ages at the time of irradiation.
For 5
males one would project 770 extra cases, 10 leukemia, 660 6
solid tumors.
Somewhat larger excess for women or females -
7
- women and girls -- fewer leukemia but more solid cancers, 8
in part because of the inclusion of breast cancer in the 9
female which does not appear in the male.
10 (slide.)
11 The Committee wanted to look at specific organs, 12 that is, the stomach, the large bowel, liver, et cetera.
(h
( ;)
13 Again, when the numbers of cases in a given age group for a 14 given dose were broken down that way, the numbers became so 15 small that our model said it was hopeless.
We better take 16 all digestive tract organs, all respiratory tract organs.
17 We have enough numbers there to develop some plausible 18 models.
19 This is the breakdown then.
In males you will
(
20 recall that there were 660 extra tumors and they partitioned
[
1' 21 themselves as shown.
In females, this is the breakdown, 22 about 70 breast cancers.
A word about age.
Roughly one-l l
23 half of the A-Bomb survivors who were alive in 1950 are
[ )T 24 still alive.
There is a lot of uncertainty about the
\\-
l j
25 ultimate cancer risks in those who were irradiated fairly
93 1
young.
Those irradiat9d at age 25 are now 50.
Bear in mind 7s' 'i 2
that nearly half of all the cancers in the population in 3
this county appear in people over 65.
That is true in Japan 4
as well.
5 Those irradiated at age 25 in Japart aren't yet at 6
the -- have not yet reached the age when most cancers would 1
7 appear anyway.
If one uses the models that I have described 8
to you, one would project that the lifetime excess is going 9
to be very much larger in those who were irradiated at early 10 ages than those who were irradiated let's say at middle age 11 or beyond.
Clearly, at age 85, life expectancy is no longer 12 sufficient to allow a full expression of risk.
' (
(_,
13 If the risks decay with time faster than we predictedthemto,thenthesenumberswillprovetobeverg 14 l
15 much too large, and only time will tell.
l 16 (Slide.)
l 17 We know from experiments in the laboratory that if 18 we spread the dose out in time we get fewer tumors in 19 animals.
Emphasize that for the breast cancer data in l
20 humans the information from patients whose breasts were 21 radiated fluoroscopic examinations don't show clearly a less l
22 carcinogenic impact than in women whose breasts were L
23 radiated in a single sitting or, at most, several daily
-()
24 exposures.
N/
25 Whether the breast is an exception or whether
94 1
that's the rule for humans, only time will tell.
We need 2
more data.
These are data from the Argon Laboratory, John 3
Thompson, Doug Gron and their co-workers, life shortening 4
plotted against dose to the whole body in mice.
Mice 5
exposed to gamma rays in a single sitting snow a curve that 6
rises fairly steeply with dose.
Most of the life shortening 7
is the result of the accelerated development of tumors, 8
tumor induction.
9 If, instead of irradiating the mice in a single 10 exposure the gamma rays are given in 24 daily fractions, the 11 curve is shallower.
If the radiation is delivered in 60 12 daily fractions, the curve is even less steep.
These daily 13 fractions are delivered in ten minutes or so each.
If the 14 animals live in the radiation fields so they are exposed 23 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> a day -- the source is turned off only long enough to 16 let the caretaker go in and fecd and water and clean the 17 animals -- the curve becomes almost flat.
There is still 18 some effect, but very little.
s 19 With gamma rays, as we spread the dose out in 20 time, we greatly reduce the tumorgenic offectiveness at 21 least experimentally for tumor induction and many other 22 kinds of effects, mutation induction and other kinds of 23 biological effects.
24 We know that there is DNA repair, so it is 25 tempting to attribute much of this dose rate dependency to
m 95 1
repair of DNA damage and other homeostatic mechanisms.
That 4
2
.is not'just the fact of age that the animals, simply because 3
they are getting much of their dose at later ages are less 4
sasceptible, it is indicated by the fact that with fast 5
neutrons if we fractionate the dose rather than giving it in 6
a single exposure, if anything, it is more effective.
7 We know that the kinds of lesions that protons --
f i
8 we call them protons -- produce in DNA tend to be very much-9 less reparable than the lesions that are produced with gamma 10 rays.
11
[ Slide.)
12 The Committee agonized over this dose rate 13 business, what allowance should be made for protracted 3
14 irradiation.
Most of our epidemiological information, the 15 anchor points that we use f en making risk estimates of low 16 doses -- human anchor points -- are from relatively high-17 dose rate, high dose exposure.
There were those who said 18 well, we ought to put in a factor of four or five.
The i
19 epidemiologists, the biostatisticians looking at the high
~
20 dose data -- particularly the Japanese data -- say we don't 21 see any evidence for this.'
22 So, the Committee really couldn't arrive at a 23 confident num'oer. The committee did conclude that it was 24 reasonable to suppose that protracted irradiation, low dose,
- 25 low dose rate, would be less effective -- less tumorgenic l
1 96 js 1
than high dose rate exposure by a factor of two or more.
-f Y
\\w' 2
The Committee couldn't chose a number.
There'was 3
no scientific basis on which confidently to arrive at a 4
single number.
When we projected the cancer mortality from 5
one-tenth of a rem per year over a lifetime without any dose 6
rate effectiveness factor, these are the numbers that 7
emerged.
Smaller numbers than if one gave the dose in a 8
single exposure, but still substantial.
i 9
There is an implied DREF for leukemia because you 10 will remember that the model was a linear quadratic model, 11 which means that as one goes down to lower doses -- lower 12 dose rates one loses the quadratic component.
Since the A
i
)
13 model for solid cancers was linear, no such implied DREF was 14 included in the model.
If one puts in a factor here then t
15 one gets a very much smaller total risk estimate.
In my 16 view, that would be a reasonable allowance to make.
17 (Slide.]
18 How about those who are exposed occupationally in 19 this age range?
Again, the Committee model for one rad per.
20 year gives us the substantial excess population of 100,000
- 1 and increase of about 14 percent over normal.
The leukemia 22 excess is close to 50 percent at a rad per year.
If one 23 puts in a DREF of two for solid tumors, then the total comes
(~T 24 down to sond hing like six or seven percent of normal.
\\_)
25 How do these numbers compare with the numbers we
l
/
97 l
would have dealt with in the past?
If we take the same 7~ F 2'
model, linear relative risk model, lifetime expression of 3
risk, there really isn't that much difference over a 20 year 4
period.
If one too the BEIR I report and said this gives us 5
an envelope of estimates, a range of credible estimates, we 6
are'still really within the same range or' barely different 7
from the top of the old range, even BEIR III.
8 The problem is that there was, in the past, more 9
inclination to believe the so-called additive risk model 10 which, as I mentioned earlier, was substantially smaller at' 11 the time of BEIR I than the relative risk model.
UNSCEAR 12 didn't make any relative risk projection in its 1977 report.
/~
!q,T/
13 This range really reflects differences in dose rate, high 14 dose rate, high dose versus low dose rate, low dose.
15 BEIR III tended to prefer this model over that 16 one.
The NUREG gave us both projection.
UNSCEAR, in its 17 latest report, used the additive as well as the 18 multiplicative.
Notice if you will that with time, as the 19 excess in the A-Bomb survivors and other irradiated 20 populations has grown with time, with attained age, the 21, additive risk model has grown correspondingly.
22
[ Slide.)
23 If you look now at the range, BEIR I, BEIR III,
Y 24 BEIR V, you are seeing -- if you look at the 90 percent (G
25 confidence intervals -- substantially higher numbers.
98 n\\
(Slide.)
1
/(g 2
Cancer isn't the only effect of concern, of 3
course.
Back in the 1950's when fallout first began to 4
contaminate the globe, it was the geneticists who argued 5
that we had to be careful; that if we didn't watch our step 6
we would follow the human genome and perhaps even threaten 7
the survival of the species.
These dire predictions, I 8
think, contributed to the impetus for a nuclear test ban 9
treaty.
10-Well, it is reassuring I think that there have 11 been no demonstrable effects of radiation in the A-Bomb 12 survivors, despite -- in their children -- despite heroic
- O
(_,).
13 attempts to detect them.
If one looks at the BEIR V 14 estimates which were predicated on the assumption that it 15 would take at least one Gray, 100 rem to double the mutation 16 rate in the human gecm cells, the numbers here are really no 17 different from the numbers in the BEIR III report.
18 (Slide.)
19 The difference really is in this category, In l
20 BEIR III, an estimate was made for irregularly inherited 21 traits.
There is reason to suppose that most of the 22 diseases that we all see in our population, particularly as 23 we age, have some genetic component; heart disease,
[
24 hypertension, cancer, arthritis.
N 25 The problem is, we really don't know how much of l
99 2
1 this load of genetically related diseases can be attributed p
2 to new mutations.
So, UNSCEAR in its latest report and ti.9
]
'~'
3 BEIR V Committee threw up their hands and said we can't 4
really make confident estimates of the extent to which 5
irradiation may affect the irregularly inherited diseases so 6_
we won't try to do that.
But if we ignore those, the 7
numbers really haven't changed substantially over the last 8
20 years.
9 So, there is no big surprise there.
New 10 development reflected in the BEIR V report relates to the I
11 re-examination of the frequency of severe mental retardation 12 in the A-Bomb survivors who were irradiated before birth.
k,)
13 It was observed back in the 1960's that there was an 14 increase in the proportion of children with small heads, l
15 dose dependent increase, children with small heads who were 16 irradiated before birth.
17 It was known that somelof the children were feeble 18 minded.
But only recently, the last decade or so, has there l
19 been a systematic effort -- part of Otake and Shaw to re-20 examine these data quantitatively.
And now, with the new 21 dosimetry, it would appear that there is a very steep rise 22 in the frequency of severe mental retardation among those j
23 children who were irradiated in the last part of the first
[)
24 trimester.
\\_J 1
25 If you will look at all of the children at all l
100 1
prenatal ages, there is an observable effect but'less ry 2
pronounced.
If you look at those who were irradiated at the 3
last trimester -- in the second trimester -- there is a 4
smaller effect beyond 25 weeks; that is, the last trimester 5
there really is no demonstrable increase with dose or before 6
eight weeks there is no demonstrable increase with dose.
7 You might be inclined to say this is some sort of 8
a fluke, it isn't consistent and-let's dismiss it.
The 9
problem is that this time during interuterine development is 10 just when we would expect the brain to be most sensitive.
11 It is at that time in the development of the brain that the 12 primitive neuroblasts -- the-cells that are giving rise to (D
(_,/
13-the cerebral cortex -- are going through their most rapid-14 phase of proliferation.
We know that dividing cells are 15 more sensitive than non-dividing cells.
These cells are 16 dividing and they are migrating.
One would expect these two 17 ages to be sensitive.
18 What we don't know clearly is what the shape of 19 this curve is down here in the range below about 30 rads.
20 Is there a threshold or is there no threshold.
The data are 21 consistent with a threshold in the range of 20 rads, two-22 tenths of a Gray.
The data don't exclude the possibility of 23 no threshold.
One can suppose that the loss of a single I h 24 neuroblast or a small number of neuroblasts might, at a
%-)
25 critical stage, effect the development of the entire cortex,
101 1
that part of the cortex that those cells are going to give g\\
o4 l-2 rise to.
3 Before that sensitive time, the killing of cells L
4 may find the brain still so primitive that other cells that b
5 survive are able to move in and take the places of those 6
that are injured or destroyed.
We just don't know.
7-Certainly, this is a sensitive effect and it is mirrored by l
8 changes in the I.Q. score.
L 9
The Japanese examined the children and once again, 1
l 10 in those within the critical age range there l's a dose 11 dependent diminution in I.Q. score, somewhat less pronounced 12 than those of this other age but really not significant
- 1. !3 l X,,)
13 before or after the sensitive stages.
The data would amount 14 to something like 30 I.Q. points per Gray.
This doesn't say 15 that there is an effect down at the level of a rad or two l
l 16 rads, but it is a substantial level of a Gray.
We don't Ic 17 know whether there is a threshold, 18 School records, the Japanese are very meticulous g
1 19 in keeping school records, and scholastic achievement scores 20 are similarly depressed in relation to dose in those 21 individuals who were irradiated at the sensitive stages in 22 development.
23 (Slide.)
V)
I 24 I think if I may close -- that's the end of the --
25 I have one more.
The big different I think from the 1
+
102 1
standpoint of risk estimates, we really can't conclude that 7--
)
2 the effects on the developing brain are non-threshold
}
3 phenomena but we have long been concerned about the L
4 sensitivity of the embryo.
I think the big thing to be l
5 concerned about here is the difference between the ICRP 1
6 estimate in 1987 predicated on an additive or absolute risk l
l 7
model, old dosimetry if you will, and the new numbers, 8
whether they be BEIR V or UNSCEAR.
9 I guess the challenge to us now is how to use the l
10 new information in a way that is consistent with the 11 uncertainties but also consistent with public expectation.
12 I think I will open up now for comments and questions, if I O) i 13 may.
14 MR. MOELLER:
Thank you, Arthur.
Do we have 15 questions?
I guess one question many of us would ask is as L
16 sort of a bottom line, what is the number -- what is a 17 number that you would suggest be used for the types of 18 exposures that we deal with most commonly in every day 19-living where we have low doses and low dose rates?
20 MR. UPTON:
The projection we made for one-tenth 21 of a rem per year is the relevant projection, I think.
The-22 Committee made no projection with the DREF for solid tumors.
23 There is an implicit DREF for leukemia.
The Committee
(~)h 24 concluded that it would be appropriate to assume that
\\_
25 irradiation at low doses and low dose rates would be less
103
/~
1 carcinogenic, perhaps by a factor of two or more.
'~')
2 It would be my inclination to adjust that 3
projection with a DREF of two or more for solid tumors.
4 MR. MOELLER:
Go ahead, Bill.
5 MR. HINZE:
I am curious here, if in terms of 6
reducing the uncertainty in the risk estimation looking to 7
the future, do we have any expectation that we can receive 8
any information from the Russians from the Chernobyl 9
incident and also from the disaster that they had -- I can't 10 name the place.
11 MR. UPTON:
In the Urals?
12 MR. HINZE:
In the Urals, right.
I am wondering
.O
\\)
13 with glasnost, whether that kind of information is becoming 14 more available to us.
Certainly one of the problems in all 15 of this is the statistical samples.
Can we look forward to 16 any enhancing of our estimation of risk?
17 MR. UPTON:
Let me say first of all a simple 18 answer, yes.
I think we can by continued follow up of the 19 A-Bomb survivors, by systematic follow up of other 20 irradiated populations where we can get data. I think we can 21' narrow the uncertainties substantially and I think we must 22 do that.
23 Now, what we can learn from the accident in the
[
24 Urals, what we can learn from Chernobyl, I must confess that
-(
25 I don't know.
I have spoken recently with Gilbe ' Bebee at
]
l 104 1
NCI with Warren Sinclair at the NCRP, with Richard Wilson at
<v s
Y,'
2' Harvard. Each of them has been in touch with colleagues in 4
.3
.the Soviet Union.
It is my understanding that the Russians c
4 now want very much to learn what can be learned from those f
5 catastrophes.
6 At the same time, there appears to be turmoil in i
7 the Soviet society.
The authorities who, heretofore-largely 8
from the Central Government, have called the-shots and.have 9
been responsible for overseeing activities and managing 10 investigations, research and so on.
They are now no longer 11 held in high repute it would appear.
Their capacity to 12 mobilize support and mobilize the expertise that may be A(-)
13 needed is questionable I understand.
14 On the other hand, local populations are anxious 15 for help.
It is my hope that IAEA and their organizations 16 involving the states and perhaps private relationships 17 between investigators and the states, colleagues in the 18 Soviet Union, can look into the accidents and learn 19 something from them.
I am not sure at the moment as to how; 20 successful those efforts will be.
I think there has been a 21 fair amount of turmoil.
22 On the other hand, I understand there have been 23 many measurements.
I just don't know, rs
- t j
'24 MR. HINZE:
There is a potential for a quantum
\\_
25 jump.
105 gm(.
1 MR. UPTON:
There is a potential.
There is a i
2 potential.
I think the accident at Chernobyl gave rise to
~
3 massive contamination with radio iodine I understand, some 4
of'the projections of thyroid doses are alarmingly high in 5
children especially.
This is protracted radiation.
We 6
don't have very much good information about protracted 7
exposure.
8 So, to the extent that studies of that nature can 9
be carried out and definitive information obtained, the 10 information could be very valuable to us.
11 MR. HINZE:
Following on that same line in terms 12 of decreasing the uncertainty, certainly as your. report
,_ n k_,)
13 indicates that one of the most solid ways to get at this is 14 through looking at the mechanisms for carcingenesis.
That 15 would be approached presumably largely through animal 16 related studies.
17 I am wondering how -- you have also shown us some 18 animal situations here, some statistics from animals -- how 19 well can we expect the extrapolation from these animal l
l 20 related studies to man?
l 21 MR. UPTON:
There is always a very large 22 uncertainty in extrapolating from one species to another.
23 In chemical toxicology we are confounded by differences in 1
-l 24 metabolic capacities, toxification, detoxification of 25 substances.
These kinds of problems don't enter in so much
106 w(
with radiation, although they affect us in our treatment of 1
L i ')
2 radionuclides, radionuclide metabolism comparable in man as 3
compared with a dog and compared with a rat and so on.
4 Beyond that level of complexity, I think that the 5-repair systems -- the homeostatic mechanisms that operate 6
between cells and between tissues are very important and we 7
don't really know enough now to extrapolate confidently 8
across species.
What has been done in animals to date I 1
9 think has been very largely descriptive and not analytical.
10 I think the stage is set for very much more analytical 11 cellular biology.
12 We have begun to identify in the clinic, activated
("')
e
(_/
13 oncogene associated with certain tumors, deleted or 14 inactivated tumor suppressor genes.
We are getting down to 15 pay dirt.
We are beginning to understand what makes cells 16 to behave the way they do.
I think with the new techniques 17 of molecular biology, recombinant DNA methodology, very much 18 more analytical research can be done in experimental model 19 systems, and give us insights into the mechanisms in 20 carcingenesis.
21 I say mechanisms because I think we will have to 22 acknowledge that cancer is a huge family of diseases that 23 don't all behave the same way.
One of the flaws in our
N
~
present approach to the problem is simplistic models.
They (b
24 25 are the best that we can do now.
The models that the BEIR V
1
~107 j
1 Committee came up with are not biological models, they are f q; t
+\\_f 2
descriptive--
3-MR. HINZE: ' Mathematical.
4 MR. UPTON:
Mathematical models.
The biologists 5
of the Committee pleaded for better models but were unable 6
to say this is how to make them.
I think the next ten years 7
will get us closer, but I think we are going to have to work 8
at it from every vantage point where we can make progress.
'l 9
Not to do the epidemiology I think would be very wrong.
I-10 know that there are people, and Ralph Lapp is one of them -
I know Ralph and like him and admire him very much.
11 12 1 think that he has argued that the study of D
Yx_)
13 nuclear plant personnel is likely to be a mistake because 14 not enough person rem, the likelihood of false various 15 positives, you won't get good information.
We can get 16 information that can be misleading and alarming.
17 On the other hand, I think if we don't get the 18 information and try to pool the data across the world 19 through meta analyses, we'really will not be in a position 20 to determine the upper limits of risk for the human species.
'l 21 MR. HINZE:
On the opposite side of the fence from 1
22 the genetic aspects, I am sure we have all heard about the 23 leukemia that-has shown up in the children of some of the
(
24 workers in the nuclear power plants in Britain.
What is the 25 story on that?
l 108-i 1
FGt. UPTON:
I think the whole radiobiological'
.r.
2 community and'public health comnunity is agog over the 3
Gardner report.
You are referring ta) the study by Gardner r
4 of the cluster of leukemia around Sullfield.
-5 MR. HINZE:
Right.
6 MR. UPTON:
Where it would appear that the excess 7
is most pronounced in the children of workers at the 8
Sullfield plant who accumulated doses on th order of ten 9
milliselverts within the first six preceding the lifetime of 10 the child or a lifetime occupational dose in excass of 100 11 milliselverts, i
12 Those are interesting data.
They are provocative.
(~
(
13 They generate hypotheses.
There are some data in'the 14 experimental literature that lends. support to these, but I 15 don't think we are at all close to concluding that the 16 association that Gardner has demonstrated is a cause-effect 17 relationship.
It may be, and I can't say it isn't.
It-does 18 not look plausible genetically from what we know about 19 mutation rates.
20 That doesn't mean that it is wrong.
Again and l
21 again in science we have stumbled on new phenomena that are 22 totally unexpected.
I think it would be a great mistake to 23 say that Gardner is right and we have to believe it and we
(~')
24 have to act accordingly.
I think on the other hand, to say
\\J 25 that it is wrong and we can ignore him is equally mistaken.
?
109
,-~(
I understand that the Richland group is going-to be looking
'l
'~'
2 at the Richland population to see whether there is evidence 3
for a Gardner-like relationship in that population.
4 We don't have in this country, by in large, the 5
capacity to trace childhood cancers is effectively as they 6
do in Great Britain and some of the Scandinavian countries.
7 That kind of study is more difficult here.
I think it's 8
another reason why those of us concerned with nuclear energy 9
-- and I am and I wotild hate to see us forego the nuclear 10 energy -- must clamor to provide the resources not just 11 financial but in vita) statistics and other kinds of 12 capabilities that vill enable us to get some answers.
,O
(_)
13 MR. !i1NZE:
And, international cooperation.
14-MR. UPTON:
International cooperation, exactly.
15 MR. HINZE:
Thank you very much.
16 MR. ORTH:
You referred to the study on the 17 nuclear plant workers.
18 MR. UPTON:
Yes.
19 MR. ORTH:
Of course, the Department of Energy is j
20 in the process of turning over 200,000 records to the Three 21 Mile Island Health Fund which is certainly going to draw 22 conclusions.
So, whether or not we are going to get very 23 good data out of it or not, it certainly looks like somebody
'~h 24 in what I will call the establishment also ought to be b
25 looking at the data.
i-
110 1
MR. UPTON:
You are absolutely right.
I serve on l
_,x
'-)
l 2
the National Academy Community that is advising the I
3 Secretary on the treatment of the DOE epidemiology program.
4 The Committee became aware that the Gebbe panel had been 5
formed about the same tiue the National Academy Committee i-l-
6 had been formed, and the Gebbe panel has come out with its V
7 recommendation.
8 As you point out, the data are to be released 9
apparently, and I am concerned as I trust you are, about 10 possible misinterpretation, misrepresentation.
I think that 1
11 it is very important that there be the opportunity for other 12 groups of scientists to look at those data.
A
%,)
13 MR. ORTH:
The point is, it has to be done 14-relatively rapidly --
l 15 MR. UPTON:
It has to be done rapidly.
16 MR. ORTH:
They have announced that they will have 17 preliminary results before the end of the year.
18 MR. UPTON:
I am aware of this, and I am concerned i
19 about this.
One of the things that our National Academy 20 Committee has discussed is the changing ethics surrounding l
21 science.
When I graduated from medical school and decided I 22 wanted to go into science, it really meant doing science on 23 the weekends, something that I did at night after the J
'( )h 24 hospital work was finished, something that I managed to do 25 on Saturday and Sunday.
111 1
only in my lifetime really has there developed a
~s
)o 2
fulltime employment for scientists on any large: scale.
It 3
isn't a rich man's hobby today.
By the same token, most of 4
us in science get our support from the public'in one form or 5
another, and are accountable to the public.
That means that 6
the data that we produce really belong to the public.
My 7
notebooks, I really had no right to lock in my office or in 8
my files and say that you really can't see them.
9 The committee is aware of the report by another 10.
National Committee that takes the position that there needs i
11 to be developed an etiquette, a philosophy accepted in 12 society, that after a certain period of time elapses during Al,)
13 which the investigator has an opportunity to harvest his or y
14 her information and publish it, the data then ought to be 15 available.
They ought to be part of the public domain so 16 that anybody can have that information, independently go 17 through the data and see whether the interpretation that was 18 published makes sense.
19 How to do this, given the sensitivity of all of 20 the issues, the privacy concerns that have to be protected 21 for the workers themselves, concerns about responsible 22 interpretation and avoidance of mischief, these are thorny 23 issues and I don't pretend that I know the answers.
But I L [~'j 24 see us going down that road very fast and having to work
.a 25 hard to find answers to avoid heartache.
l l
- )
112' i
1 MR. STEINDLER:
Let me ask a different kind of lc') -
2 question.
A number of your data are couched in. conclusions 3
dealing with death from cancer.
4 MR. UPTON:
Yes.
5 MR. STEINDLER:
In the future world of 200 years 6
from now there will be no deaths from cancer, let me assume 7
that.
What do you do at that point in the interpretation of 8
your results as they relate to the issue at hand?
j i
9 MR. UPTON:
The report does emphasize that at this l
10 point in time the most reliable information one has 11 concerning effects on cancer in irradiated populations come 12 out of mortality data.
For the United States there is now l
l
,CN l
(,)
13 at the' National Cancer Institute a so-called SEAER system, l
)
14 surveillance, epidemiology and end result system, that j
-l 15 covers about ten percent of the population for incidence.
I 16 Hospital records are scanned and the incidence of
{
i 17 cancer is determined.
One would prefer I Think if possible, 18 to deal with incidence, and the report indicates how one can 19 crudely translate mortality into incidence, not that 20 mortality itself is very good either because death 21 certificates are notoriously inaccurate.
22 One can give, using the information that one has, I
23 convert the mortality estimates into incidence estimates
]
t k
24 crudely.
I would hope that the time will come when your l'
25 prediction is borne out, that nobody dies of cancer anymore.
e l
113 L
1 Cancer is a common disease.
About one of every three to 7S
'~>
2-four people in this room will develop cancer at some time 3
during his or her life and one out of five us will die of 4
the disease.
l 5
I think any of the estimates cannot really be 6
taken a face value.
One has to look at them carefully, look L
7 at the uncertainties and look at the impacts.
Cancer in a 8
90 year old person is not as much of an impact on that 9
person or that person's family or co-workers or community 10 than a cancer in a 15 year old or a five year old.
So, just 11 counting cancers doesn't begin, I think, to convey the l
l 12 nature of the impact.
<s k_,)
13 We did try to express the loss of life associated 14 with cancer and get a handle on it.
I think any group that i
15 wants to translate these estimates, crude as they are and as 16 uncertain as they are, into social policy, must make the 17 effort to read into the data their broad significance.
18 MR. STEINDLER:
How does a change in ability to --
19 let me use the word cure, because it's shorter -- influence 20 the comparison between let's say BEIR I and BEIR V or BEIR V 1
21 and BEIR XII, what will we be able to do with the data on a 22 comparison basis?
23 MR. UPTON:
I don't recall whether BEIR I had any 1
\\
[)
24 discussion of conversion of incidence to mortality or vice w
25 versa, BEIR III did and BEIR V does.
So that, I would hope
114 s-1 in future editions of the BEIR report -- I am sure you are
- r 1
L.)
i 2
aware that each time the BEIR report is undertaken a new 3
BEIR Committee is formed.
It's not the same people again 4
and again, although there may be some continuity of 5
membership.
6 I would hope that the basis would be there for the 7
kinds of inter-comparisons that you indicate will need to be 1
8 made.
Simply counting tumors isn't enough.
You may have 9
noticed in one of the tables that I showed, there was an i
10 entry-for earth deaths as well as cancers themselves.
11 Cancer is such a common disease that the number of early 12 deaths-will exceed the number of cancers.
Some of the
,f
(
13 people will develop radiation cancers and will develop them
'14 earlier than they would have developed them had there been j
15 no radiation.
l 16 They don't represent a new cancer, another cancer, 17 but an earlier cancer.
It is an earlier death in that 18 sense.
19 MR. STEINDLER:
Let me ask one other question.
20 MR. UPTON:
Sure.
~
21 MR. STEINDLER:
As you review the world of 22 research and the focus and level of effort or support that 23 is attributed to it, where do you see inadequacies?
T 24 MR. UPTON:
That's a huge question, and I am sure (d
25 that I-am myopic.
I ask the question like that and look in
r-115
-1 the mirror-and see myself.
7ff 2
MR. STEINDLER:
As you wish.
3 MR. UPTON:
In looking at radiation -- wo are all 4-talking about risks of radiation -- I am really deeply 5
troubled that so many of the people that I see as eminently 6
experienced, knowledgeable, credible scientists are l
7 approaching retirement age.
We came into this business ~of 8
nuclear energy after World War II.
The AC had a mission, it 9
organized itself to conduct that mission effectively and the l
10 nation stood behind it.
Whether it was because of national l
l_
11 security interest is beside the point, the job was done.
1 12 People were recruited into the field, good people.
(
Q, 13 That generation is dying out now, and I don't see another 14-younger cohort coming along to take its place, hamnot 15 aware that AC, URDA, DOE ever assumed that it had the 16 responsibility to sustain the national. scientific capability 17 in this subject area.
I am not sure it was ever given that 18 responsibility by the nation.
19 As a result, I don't think that we are today.
20 producing people nor have we produced in the last 20 years, 21 the people who we will need ten years hence or 20 years 22 hence to solve some of the problems that we ata trying to 23 solve.
That, to me, is the most serious problem.
(
I don't think it's beyond solution.
I think that 24 25 if positions were created, if training monies were made
116 7-q 1
available, people would come into the business as they did 2
in the 1950's.
People came into the field from neighboring 3
fields, and they would come back. There needs to be an 4
incentive for them to do so. There need to be career 5
opportunities.
I don't see us providing for those in our 6
society.
If we don't have them, I think the future of 7
radiation research is in jeopardy.
8 MR. MOELLER:
Very interesting and thought 9
provoking comments.
Are there any other questions by 10 members of the Committee or its consultants?
We were 11 scheduled to go another few minutes.
Let me just ask if 12 there are members of the NRC Staff here that have any O
As-)
13 questions, we would certainly invite you to ask them.
Go to
[
14 a microphone Hal, and just ask your-question.
15 MR. PETERSON:
I am Harold Peterson, Etnior Health 16 Physicist, Office of Research.
There are a number of -- it t
17 is not so much of a question of.Dr. Upton as much as 18 additional information for the ACRS staff.
19 There is a cooperative agreement between the U.S.
20-and the USSR on reactor safety, civilian reactor safety I-21 should point out, that has within it ten subdivisions, one 22 of which, number seven, deals with environmental transport 23 and health effects.
This group is, at the moment, actively (nu) 24 planning a cooperative program of both environmental 25 transport research, mainly model comparisons with Chernobyl
l' 117--
U gg' 1
data which the Soviets will furnish and, in addition, some i
V 2
uptake studies.
3 Also, the health effects studies, both the early 4
and late health effects of Chernobyl, so that this area is 5
being explored actively by the NRC.
Those working groups 6
are NRC, DOE and NCI.
Dr. Bebee, as Dr. Upton mentioned, is 7
on that group.
8 MR. HINZE:
Is there any chance that you will 9
receive some information about the Urals disaster?
10 MR. PETERSON:
Yes.
In fact, when we were over in 11 the Soviet Union last fall we received about an hour and 12 one-half briefing on that activity.
j^s
' (,,)
13 ~
MR. HINZE:
Are their statistics qualified?
14 MR. PETERSON:
I haven't looked exactly as to 15-whether you could get meaningful epidemiological study.
The 16 data look a little lower on average dose than Chernobyl if I 17 recall.
The total size of the population was several tens 18 of thousands, so it may be feasible.
That's one thing I 19
'think they are going to look at with regard to the health
- M) effects study, is whether that population can also be
- ~ 21 examined.
'22 MR. HINZE:
That's excellent.
23 MR. PETERSON:
Secondly, we also have a limited l
. [~)
24 study looking at molecular and cellular biology to determine s_/
[
25 what the mechanistic studies could be used for in further l
t t
118 u.
1 shedding light on low-dose, high-dose extrapolation and the L]
2 risk estimates themselves.
Lastly to point out that as far 3
as epidemiological studies, the new Part 20 -- the revised 4
Part 20 does now require reporting for the seven categories 5
that we required detail information on including power 6
reactive, will require reporting by named individual and 7
social security number.
8 NCI has been. heavily consulted in the formulation i
9 of the reporting requirements and the data forms that will l
10 be used for that to make sure that they will be useful for.
11 epidemiology. studies.
12 MR. STEINDLER:
What is the size of_your research
/N
' _,/
13 budget in this area at this point?
(
14 MR. PETERSON:
I don't know how much of the total 15 is going to be split -- for the total NRC research budget?
16 MR. STEINDLER:
Right.
i 17 MR. PETERSON:
It's less than $2 million.
18 MR. STEINDLER:
Thank you.
That's the point that 19 I was getting to.
20 MR. MOELLER:
Thank you, Hal.
Are there other 21 questions?
Gene Voiland.
22 MR. VOILAND:
You mentioned something abotit the 23 identification of genes that were associated with repair.
f~
24 What did the Committee feel the future of that kind of C
25 research might lead to?
119 l
j~
1 MR. UPTON:
The Committee looked at this l
~N~,
2 fundamental research in DNA repair as vital, very important.
3 I think it is the basis for much of the dose rate effect L
4 that I referred to.
It has been studied systematf.cally in-l.
'3 very few model systems today.
i 6
MR. VOILAND:
Does that relate back to the high 7
frequency or the high incidence in that early age group, the f
l 8
.zero to ten or 15 years or something like that?
L 9
MR. UPTON:
It may.
I suspect that the remarkable 10 age difference may be attributable in large measure to the 11 so-called Mugar Knudson hypothesis.
If you damage a cell 12 with radiation, damage the genetic material in that cell,
.f3
(_,).
13 the data would suggest that -- first of all, it isn't' going 14 to effect the cell until the cell divides, because you 15 damage one gene on one chromosome and there's a normal gene 16 on the other chromosome that maintains the necessary 17 function and then the cell divides and it may lose the 18 capacity all together.
19 It takes more than a single faulty gene -- and 20 much of the data would suggest that that is the case-- then 21 you have to effect two.
Ten to the minus six, one in a 22 million chance of affecting one, then it's ten to the minus 23 twelve chance of affecting two.
If the cell with one
()
24 altered gene multiplies and grows out into a population of a 25 million then the chance of having that second damage affect m
120
,cs one of its daughter cells becomes very much larger.
'I
'2 When we injure a child, a child continues _to grow.
3 The stem cells that have to continue to multiply in the bone 4
marrow, the skin and tha lining of the bowel and other 5
organs to maintain those organs, they expand.
One then has 6
an expanded population of cells already injured, just 7
waiting for that second injury to occur.
That is the 8
essence of the theory.
9 We know that there are other factors involved.
10 The capacity to recognize foreign substance, immunological 11 reactions, many age-dependent things are happening.
We 12 really haven't studied them in great depth.
/~N
(,sl 13 MR. VOILAND:
Thank you.
I might just comment 14 that one of my favorite technical detective stories was in 15 the BEIR report relating to the analysis of the mine 16 workers, which I thought was just a splendid effort.
17 MR. UPTON:
Thank you.
18 MR. MOELLER:
With that, I think we will bring the 19 meeting to a conclusion.
Let me thank Dr. Upton once again 20 for coming.
You have honored us by being here today.
In 21 mentioning the National Cancer Institute, I failed to 22 raention in my introduction that Dr. Upton was formerly the 23 Director of the NCI and also an Assistant Surgeon General in
[')
24 the.U.S. Public Health Service.
v l
25 With that, we bring the 21st meeting of the i
I
6f3
-g.'
./'
121
..s.
c.9 q) q l l' Advisory Committee on Nuclear Waste to a conclusion.
Let me v
jg. -
2 thank everyone who has appeared, and our three ladies that 3
appeared earlier this morning.
That presentation was also-4 very helpful.to us, and I am sure the members of the-public 5
who have joined us today have enjoyed both of these 6
presentations.
i 14 <
7.
Thank you very much.
8
[Whereupon, at 11:58 a.m., the Committee 9
concluded.]
10 11 L
L' 12 (D
\\m,)
13 L_
14
\\c 15 16 ll 17 18 19 20 21 22 p
i'f 23 2'4 25
-r,
)
4
E
+
l REPORTER'S CBRTIFICATE I
This is to certify that the-attached proceed-ings before-the Unitec'. Staties Nuclear Regulatory Commission j
in the matter of:
NAME OF PROCEEDING:_
21st ACNW Meeting DOCKET NUMBER:
PLACE OF PROCEEDING: Bethesda, Maryland were held as herein appears, and that this is i
the original transcript thereof for the file of the United States Nuclear Regulatory Commission taken by me and thereafter reduced to typewriting by me or under the direction of the court report-1 ing company, and that the transcript is a true and accurate record of the foregoing proceedings.
j
[
Mary C.
Larkin Official Reporter =
i Ann Riley & Associates, Ltd.
dl
l e
o og I
REVIEW OF TECHNICAL BASES FOR 10CFR61 WASTE FORM STABILITY REQUIREMENTS l
CAROL HORNIBROOK JUNE 29,1990
O O
O WASTE FORM STABILITY RELATIVE TO GROUNDWATER IMPACTS FINDINGS:
MOBILE ANIONS OF C-14 AND l-129 CONTRIBUTE ESSENTIALLY ALL OF THE GROUNDWATER DOSES DOSES ARE CONTROLLED BY RELEASE RATE OF NUCLIDES i
INFILTRATION WATER VOLUME CONTROLS RELEASE RATE SITE STABILITY CONTROLS INFILTRATION WATER VOLUME
O O
O
'i 1
GROUNDWATER IMPACTS (CONT'D)
CONCLUSIONS:
1 SITE STABILITY FEATURES ESSENTIALLY MUST BE PERMANENT WASTE FORM STABILITY MAY ENHANCE SITE STABILITY BY DECREASING THE ORGANIC CONTENT PER UNIT VOLUME OF TRENCH l
LIKELY, DOMINATING FEATURES CONTRIBUTING TO SITE STABILITY:
PACKAGE VOID VOLUME l
EMPLACEMENT BACKFILLING COMPACTION j
COVER DESIGN BECAUSE OF l-129 AND C-14's LONG HALF-LIVES, WASTE FORM l
STABILITY WILL HAVE NO IMPACT ON RETARDING MIGRATION l
i f
1
O O
O 9
l WASTE FORM STABILITY RELATIVE TO INTRUDER IMPACTS CONCLUSIONS:
l l
WASTE FORM STABILITY PROVIDES INTRUDER PROTECTION ONLY FOR SHORT-LIVED NUCLIDES l
l PASSIVE WARNING DEVICES COULD PROVIDE THE SAME OR GREATER l
LEVEL OF PROTECTION l
WASTE FORM STABILITY PROVIDES ESSENTIALLY NO INTRUDER PROTECTION FOR LONG-LIVED TRANSURANICS.
j i
SIMILARLY, ENGINEERED BARRIERS WILL NOT PROVIDE INTRUDER I
PROTECTION BECAUSE OF FINITE LIFE OF THE BARRIER r
i i
l r
o
O O
O
(
INTRUDER IMPACTS (CONT'D)
CONCLUSIONS:
DEEPER BURIAL IS SIGNIFICANTLY MORE EFFECTIVE IN REDUCING INTRUDER DOSE THAN WASTE FORM STABILITY l
DEEPER BURIAL OF ALL WASTES GREATER THAN CLASS A ELIMINATES THE CLASS B AND C WASTE FORM STABILITY l
REQUIREMENT WHILE PROVIDING MORE EFFECTIVE INTRUDER l
PROTECTION.
l
0 O
O t(
WASTE FORM STABILITY TESTING CRITERIA i
l WASTE FORM STABILITY TESTING CRITERIA DO NOT:
i l
t RELATE ON A ONE-TO-ONE BASIS WITH 10CFR61 I
PERFORMANCE ASSESSMENT I
REPRESENT A SIMULATION NOR ACCELERATION TESTING OF ACTUAL l
DISPOSAL ENVIRONMENT
{
-.-...n
~
O O
O
(
SUMMARY
WASTE FORM STABILITY:
PROVIDES NO PROTECTION FOR GROUNDWATER DOSES PROVIDES INTRUDER PROTECTION FOR SHORT LIVED GAMMA NUCLIDES PROVIDES NO PROTECTION FOR LONG-LIVED TRANSURANIC NUCLIDES CAN BE EFFECTIVELY REPLACED BY DEEPER BURIAL AND PASSIVE WARNING DEVICES s
O O
O
~
~
LLW Disposal Facility 1-129 & Tc-99 inventory Development Advisory Committee on Nuclear Waste (ACNW)
June 1990 Patricia J. Robinson (415) 467-0719
~'
O O
O l
LLW Disposal Site Development
- Recall from 10CFR61, ground water doses are inventory i
controlled not concentration controlled i
l i
i l
I i
- Groundwater doses are controlled by mobile anion isotopes of:
i l
- Inventories of Tc-99 & I-129 have the potential to limit the i
siting of new compact LLW facilities i
Performance Assessments indicate that I-129 contributes >
90% of the total dose i
).
i 4
i l
l l
t
2 O
O.
O
~~
1 r
i 4
i I
r
~
Ned I
L Development of a realistic l-129 & Tc-99 disposal site inventory I
I
[
I 4 Accurate i
4 Defensible i
i l
l l
s l
l a
i i
i i
l i
i Scaling Factors
}
r I
i Research data shows that I-129/Cs-137 scaling factors are conservative by factors of 100-10,000 depending on waste i
stream j
I I
Consequently, inventories derived from 10CFR61 sampling l
programs & shipping records over predicts the dose impact l
by a significant margin 1
I i
r l
i 1
l e
4 O
O O
~
1 l
1 i
i Industry Scaling Factors i
i Industry Average I-129/Cs-137 Ratios from Waste Sampling & Commercial Lab Analysis l
i l
Primary Resins 3.0E-06 2.6E-06 SECONDARY RESINS 2.7E-05 l
Evap Bottoms 3.5E-05 Condensate Resins 1.9E-04 DAW 6.3E-03 9.9E-05 Primary Coolant 3.5E-02 2.5E-04 All Waste Streams 9.2E-05 4.4E-05 I
i Expected ratio ~1.0 E-09 to 1.0 E-07 l
1
O O
O
~
PWR Typical LLW Volumes Primary Resin 2
-4 Radwaste Resin -
EEEE Evap. Boltoms s
i-i-i-i i
o 1000 2000 acco moo 5000 e000 no Annual Volume (cu ft) m:
j r
i e
I i
t k
a Distribution of I-129 (by waste stream) i i
i I
i h
d
. a..
~-w, ere
,+
y~
'l
.ar E
4
' ['E l.m^f
([~ ~
E j
g.;
j.
,: s. - ' ~
._e-inlgog j
.- w-er p.,,
t
'r..
1 r.;,
..g.
~
+
sy,
,., ~
- s. :ik[ j.IfV. 7.
i 0.3%
i
'1 l
El Primary Resin (20-200 R/hr)
I Radwaste Resin l
DAW i
I
)
i i
1 1
I l
~
O O
O q
l Causes of Overestimates Analytical Limitations Sampling Limitations
- Isotope Chemical Behavior ALL CONTRIBUTE TO OVERESTIMATES
i
~
O O
O Causes of Over Estimates Analytical Limitation Best available analytical measurement techniques are not sensitive enough to measure I-129 or Tc-99 in these dilute wastes Industry database has only 70 measured values of I-129 out of 3000 sampies 2930 samples are below detection limits Reported at the minimum detection limit (MDA) and. many are used in scaling factors 4
l 0
o o
i 1
l 1
1 i
i Causes of Overestimates l
i i
i Operational Limitations 1
Representative sampling is impractical for waste streams of interest because of:
l r
Extremely High Radiation Dose Properties of waste i
i Limited sampling provisions Limited mixing provisions 1
i i
I
- Sample does not reflect changing plant and fuel conditions i
?
i i
i i
O O-O o
i i
'l I
i A More Accurate Way...
l Development of a Predictive I-129 & Tc-99 i
Inventory Methodology Based On:
l
\\
l Modeling ofI-129 & Tc-99 Release from Fuel
- Release rates based on fuel conditions
- Fuel conditions assessed based on 5 short-lived gamma emitting iodines and cesium routinely measured in RCS Validated Against Controlled at Reactor Primary Resin l
Testing and Measurement I
t l
l
&--4s,.4,4._4 jus s.a J h4 4 4
..Aa-malmJed--+4 44d', we e.t 4
- A g e.em at h+ mm4deam_so-+.-dem-gaee.-
Asw i,
O:
I 1
s l
l l
1 I
I l
4 e
7 1
Oa 9,
o
~
i O
C<
l l-l-
l 9
i f
l
_..._,__..._...-__..______._.___.._.___.__e_
O O
O l
~
What is the Source of I-129 & Tc-99?
1 i
i l
l l
- Fission of Power Reactor Fuel 1
i Reactor Fuel May Be Inside or Outside Fuel Pins
- Fuel Contained in Fuel Pins Will Release Fission l
Products from Defect in Cladding 1
I-129 is released at different rates from fuel depending on the release mechanism:
1 Recoil DifTusion Knockout l
l l
l i
i t
i l
1 i
1 l
Fuel Sources for Release of I-129 a
i l
l 1
2 l
l i
l Rod Defect Exposed Fuel (Trainp) l L
- 'U(.:.-t
- -e
t.
~
t }, $ ~l '~ '~,
..8.4 i.
' L.'.
48 2
,; ;J..
' f,;$[p,
l Faset Cap imama
+:];
i.'.'
^
.. uv-..a
/
. - e e
w 4
[$'[*
". },...y ;,
% [y;. -v..cs.M-i dh
].} $
.. +
s..
m=: n 3 La n ?
2.., :' 17
^~3, I
l.131[.f VNQ I
t c.
- -4 6443
]. 'j)<*-
l 4
"^
QQ%;_
i Q., ' ;
Cs-137 $$ y,1
. [..~
k^;
ww
?'gTQ '.*gf M. +
..,~j ', -
... g,- __;..
s t w m~.
b
[
7*
p
,.g
~
- ,- T>.5 '.) ~.. -' ' $
[.',
- 4. g. ;;.. ; : g, g ~ '
- ?
L-
%g;
- l;"'t
- * '..fW_. ',
?
g
....e i.
l I
V n
i d
i I'
l i
L 1
i i-I I
l i
L i
O O.
O I
~
1 i
Core & Fuel Release Variables i
r l-129 Release Rate is controlled i
by the following parameters:
Age (burn-up) of Defective Rod i
Defect Size i
Contribution from Exposed Fuel Contribution from Defective Fuel i
i Contribution from Recoil i
Contribution from Diffusion l
Contribution from Knockout i
'l i
Y
~
.. ~ - -
Fundamental Model Concepts The 5 short-lived Iodine Isotopes are released from exposed fuel and defective fuel rods They are released from both sources by three fundamentally different release mechanisms:
RECOIL DIFFUSION KNOCKOUT
~
O O
O t
?
[Cs-134/Cs-137] vs Irradiation Time 1000 P
+
)
800 r
2.
600 l
O Days j
400 y
i l
200
[
o
.2
.4
.6
.8 1.0 1.2 1.4 i
[Cs-134/Cs-137]
1 a
i i
i.
1 i
?
I l
O O
O I
l-129/1-131 Ratio From Diffusion l
s 5
i
,p>
I
~y-.
4 3
- ~,.v I-129/l-131 f
amio
/
(x E-09) 2 E
2 l3" i
O 1
o 3
/
T 0
200 400 600 800 1000 Irradiation Age of Fuel, Days i
1 1
~
_O 0
t 0
u 0
1 o
kc 0
o 5
s n
8 ya K
D l
m e
0 u
F o
b f
r 6
o f
e
/
g o
A O
i n
t 0
o a
i d
t R
a 4
ida 1
rr 3
I/
1 0
5 l
2
/
92 1
~
l 6
5 4
3 0
3
)s 1-o l/
92 x
~
1
(
l O
l
=
i
O O
O i
t 1
l i
Ex: 1-129 / l-131 Release Ratios l
i l
l Release Mechanism I-129 / l-131 Ratio i
Recoil 4.55 E-10 1
Diffusion
- 3.43 E-09 3
I Knockout
- 1.70 E-08 i
i Dependent on fuel burn-up, shown here at 500 EFPD i
4 i
1
)
t 6=ws w
t-----
= - - - -
-w m
=
w r
. =-
1-129 Dominant Release Mechanism Varies ANO-2 Surry-2
,gr Likss;~ %~~
r-
~
a
~
2.8%.
3.6%
a %xnockout 89.4 %
% Diffusion E:'e % Recoil 1.4%
9.2%
SONGS-3 C
O O
O Methodology Develop concentration and release rate profiles for I-131 Determine core and fuel release parameters using:
I-131,I-132,1-133,I-134,I-135 (all measured RCS)
Once we know core and fuel release parameters, then we know release rate relationship as a ratio of I-129/ I-131 l
Multiply release ratio times actual I-131 release rate to produce the I-129 release rate Develop I-129 release rate prordes on an annual basis over the 3 fuel cycles (5 yrs) and integrate to yield total release
.~
s
~ ~.
l Ea
[l-131] RCS During Fuel Cycle 4.00E -
A [i-131] pCUcc (b
~
3.00E-02 -f---
- (c-~ '
~
L "a
!!)
EME -
-~ #
j 2,
,. /
ae
-i u
i b*
l"' ~ '
~
= - ~ -, - -
1mE42-
%agj$
u u u a u s s u a a ETFB 5Tu a u ETRrW a a u s a ETRu a u BTUTETs u a RETETs a u urs s s ETFW 12/29/86 2/1/86
+
0 O
O PWR-3, Cycle 4 Release Mechanisms s
m
-A-g-
m,,we
. 7%-
i w+
M a^4 N #M A
4,-A.
A_
y 2
{
.. M. N -
+
+-a,,.en.
~ ~ - '
~ '~' '
I gg_
42 Q1 (Recoil) 3 2
B O2(Diffusion) 4 03(Knockout 0.0 % M ? i Y f 7 f h ! F ! Y k W ! W $ 9 ! h k ~ h f W Y f 7 7 ! V ! Y !'
6/3/85 1/14/85 W+
x t
~.
a PWR-3, Cycle 6 L
100.0% -
~
I l
80.0% -
i 1
i M
4
=.%w, ea a.
. c.va... - -
- i w
- a, w
m e_ w.+w v.
s..e A
A A
A Q1(Recoil) i E Q2(Diffusion) t l
l
@ Q3(Knockout 5
t
. M.0% *.
l I
enn
~.
~n.m s.
,,m...
s-w.n w m.s.+ :
.+.n.-
,.nn.
.i r
T h!
g
~1 t%
.,.,. p, r;,.,.,.- p; 4_,_g a. s4g-0.0 %,.,.,., %,.=E - t A,
,v, 10/31/87 l
I l
t 1
e--
-'w 6
+ - =
v
,1 O
O O
1 4
l-s i
l-129 Release Rate i
i 4 Fuel Cycles i
i 1/14/85 12/28/89 J
.4
-in a
w 2
w w
w -,
CYCLE 4 CYCLE 5 CYCLE 6 CYCLE 7 3 5
. ~~~"
i,.m g 4+
y,.jo42% f% k' '
m,.c<n~ _
ggge wa,.*:* *##
h,
}
s
\\
%'dF
g 3
g
- %#s#V#
9 i
4,,,
_~
~. -,,
-. ~ ~.
l l
r
+
- )-
9 L
i i
1 '
E.
i
t.
~
~
o O.
O 3
Compact 3 Preliminary Source Term j
Using Validated Model Method l
PWR-1 PWR-2 I-129 l-129 l
C ycle #
fuCi/ MWD)
Cycle #
fuCi/ MWD) 16 1.30E-05 8
1.40E-05 17 3.32E-06 9
6.34E-07 18 6.13E-07 10 1.06E-06 i
19 2.13E-05 Average = 9.56E-06 Ci/ MWD Average = 1.84E-06 Ci/ MWD l
i PWR-3
~
BWR-1 F129 b129 C ycle #.
(pCi/ MWD)
Cycle #
fnCi/ MWD) 4 1.51E-05 7
6.50E-07 i
5 6.91E-05 8
2.20E-07 6
3.70E-07 9
4.80E l 7
6.00E-06 i
i Average = 2.35E-05 pCi/ MWD Average = 1.11E-06 Ci/ MWD b
llllllll!1l1 7654 C
P v
W c
R ie
-3 P
W 636 1 (1p1 R
.0795 C2 0011 i/9 EEEE-M 3
0000 W
1 6755 D
1
)
29
,1,1 5323 F
1168
,4,9,7,7 MyC u
3898 Wc e
1087 2910 Rl l
e Re le R
a e
8 c-s I
3 075 a1 e
s2 0468 e9 9926 d
b y
(uC C
i) yc le
m m_
O 100 w;T 0
/
A
- s. S T
5 0
0
.-)
- 00. -
)
oRiC y
C3(
s o
r r
tc o
a
+
f t
g n
n i
y 0
a l
e r
c o*
6 s
v t'
820
/
ni s
n eC 713- -
s i
i *y v
y l
n a
i n
9r a
2a
,g m
n 1
O ilp rr0 m
m 1
e yy6 ya y
/
r s t
00000 a(
ru S
33333 na i
imt aS a
id let n
r s i
pe yf m
i r n ea r
v m i
Y 64 d
l
/
201 eg e
r n iC 000 - -
ei r
dp ip P
nh si o s cm s
12345 io r t
ttttt af c
ccccc ad t
a aaaaa de p
ppppp v
si r m
mmmmm ie h
o ooooo TD C
CCCCC O
~
~
l Advantages of Approach 4
Conservative - assumes total inventory released from fuel i
ends up in the disposal facility
\\
More accurate than sampling and analysis limitations l
l Utilizes >100 RCS measurements / year l
Use of more accurate inventory reduces the dose impact Low cost l
l l
Method can be independently validated using in plant testing l
i and measurement a
I 4
A-
w a
4.._awa
_w a#
.m.
- .adeium
+_h+da3.r+m._
e.h sas, wama au m,m-4 m a.
m amma.ma44a_,
m_g_4 m. sa2w_4 4ew.;.+
3 e.h e,4 m.s 4
,*Ess,.m.,a,._maa.4.,
I i
O
. 1 i
O.
\\
)
V.
m 5
u ra O
e
.,,.m s
O O
O i
Validation Testing e
i.
i t
I Plant Testing Complete Planned Oconee-2 B&W(2-3)
Calloway-W(2-5)
Palo Verde-2 CE(2-5)
CrystalRiver-3 B&W(4-6)
Palo Verde-3 CE (1-2)
Indian Point-2 W(0)
Ginna W (<10) l Indian Point-3 W(1-2) i Susquehanna-1 (1) i Fitzpatrick (0) j.
WNP-2 (3-4)
Vermont Yankee (10-12)
Sponsoredby EPBI, Sponsored by ESEERCO/NYSERDA i
\\
}.
- ( ) denotes estanneed number of desective sues pins 1
i 1
i o.
1
O O
O l
i I
Validation Effort i
Test equipment installed on reactor sample line to collect released 1-129 and Cs-137 i
- flow rate controlled
- duration of test accurately recorded
- 2 day minimum test period 1
- 120 liter test
.l I
i
. i.
_ _ _ __ _ _j
.- _=
._=
. _ _ =
_ _ _ = - -
O O
O PWR Validation Test Cooling Water from a ciar vessen 2-30 day Test
__gw.p4_,',
Dose Rates Contact = 4-400
/L_e'"
R/hr Dose Rate Shield = 20-50 mr/hr 10 GPM/sq.ft.
8.8*
Volumetric Flow Rate = 81 ml/ min g
l T
i i
ll I
~
~
O O.
O l
BWR Vali# tion Test i
lq Cooling Water i
2-4 day Test
., rom Reactor vesssi Dose Rates
=
'Precoated Contact = 1.5 - 5 R/hr Powdex Resin pjgger Volume Flow Rate = 41 ml/ min 100 Liters T
t t
{
O O
O Validation Effort All test resin sent to BNW for analysis and determination of total quantity of I-129, Tc-99 and Os-137 loaded onto test resin 3R-STAT
- used to predict total quantity loaded on resin for the same time period of test column Results of 3R-STAT & BNW analyses compared
- 3R-STAT ( Reactor Radenuchde Release Status) proprietary Vance & Associates code
y y
q Results of 1st Validation Testing, PWR-5 Cs-137 Test Column inventory:
Predicted
= 317 Ci Measured by PWR-5 = 293 Ci 1-129 Test Column inventory:
Predicted =
2.4 E-05 Ci BNW Measured =
1.32 E-05 Ci
1-129 & Tc-99 Special Case Compliance Viable Ootions for waste streams other than DAW:
(1) Not list on manifest, < 1% of the tabie values-And... use semi-annual solid eff!uent report for total 1-129 & Tc-99 released from reactor (2) List something on manifest, an LLD, or an estimate And... use semi-annual solid effluent report for total 1-129 & Tc-99 released from reactor i
l.
O O-O i
Compliance & Reporting l
i Te
_i, s
l
. Current method hf compliance using SF ok for all l
l other radionuclides ^except Tc-99 & 1-129 l
s.
N j
. Generic scaling factors for DAW l
I j
4 I
t l
[
- ~,
+~
O O
O i
i' 1-129 & Tc-99 l
Special Case Compliance L
l-Periodic re-validation of method with RCS test l
columns 1 validation check every 3-5 yrs.
l l
l t
s
>a DRAFT AN EVALUATION OF THE TECHNICAL BASES FOR THE WASTE FORM STABILITY REQUIREMENTS IN 10CFR61 NP............
Research Project 2724-4 1
Prepared by VANCE & ASSOCIATES P.O. Box 997 Ruidoso, New Mexico 88345 Principal Investigator j
L J. N. Vance l'
r, l
l' l
Prepared for Electric Power Research Institute 3412 Hillview Avenue i
Palo Alto, California 94304
-EPRI Project Manager C. Hornibrook Radioactive Waste Management Program l
Nuclear Power Division
(
f 4
f~'N Q
1 Section 2.0
SUMMARY
AND CONCLUSIONS-1 For the original.10CFR61 Impacts Methodology, both the 10CFR61 Draft and the Final Environmental Impact Statements (DEIS and FEIS) were reviewed to establish the technical bases for the waste form stability requirements contained in 10CFR61.
The Update 10CFR61 Impacts Methodology was also reviewed to determine if any significant changes had taken place since the original impacts methodology.
The original' and update 10CFR61 Impacts computer programs were also exercised as an element in the investigation into the technical bases.
- However, coding errors in several of the logic statements produced results that' (3
r
)
were counter to expectations based on the model logic and descriptions contained in the supporting documentation. -
Therefore,. greater-
- eliance was placed on the written descriptions of the 10CFR61 impacts models.
The following.summartzes the findings and conclusions regarding-the L
technical bases for the 10CFR61 waste form stability requirements.
L Groundwater Doses l
o Groundwater doses are dominated by the three mobile anions of Tc-99, C-14 and I-129.
o Because of the relatively long half-lives of these nuclides (213,000 years, 5,730 years and 15.7 million years for Tc-99, C-14 and I-129 respectively), the durability of l
engineered barriers and waste form stability are not likely l
/^T l
2-1 g
O to be long enough to provide any protection for groundwater doses from these nuclides.
o Groundwater doses are controlled by the release rates of these nuclides into the groundwater pathways.
o The release rates are governed by the infiltration water volume into the site, which is an assumed direct function.
of the site stability, o
Because of the half-lives of these nuclides, site stability must be essentially permanent to reduce the groundwater doses from these nuclides, o
It is possible that waste form stability may enhance site stability by decreasing the organic contcnt per unit volume.
of trench.
However, it is likely that other site design and operating features such as waste emplacement, backfilling, compaction and minimizing package voids-are the most important features for achieving long-term site stability, o
Using the reference site waste volumes and I-129 curies, the calculated thyroid doses for the intruder well are below the 10CFR61 dose limits for sites with improved. site stability-provided by trench cover, backfilling and compaction,. even without waste form stabilization.
Intruder Doses o
Waste form stability provides intruder protection only for short-lived nuclides, by reducing the number of exposure hours assumed for the direct gamma exposure pathway.
9 2-2
1
~-
()
LJ o
Passive warning devices' constructed of materials with-greater durability than packaged waste and placed in the trenches.would provide the same or greater levels of protection.
o Deeper burial is significantly more effective in reducing the intruder doses than is waste form stability, o
Requiring all wastes greater than Class A to be buried at deeper depths would in effect eliminate Class B wastes and -
eliminate the need for waste form stability for intruder protection purposes.
o Waste form stability will provide essentially no intruder L
protection for the long-lived transuranic nuclides.
This E
,f s was recognised by the NRC in 10CFR61 by not establishing
(_s)
-Class B limits for the long-lived nuclides, o
Intruder protection for the long-lived nuclides is provided by limiting their concentrations in waste and by intruder j
barriers, such as deeper burial.
L o
Because of some of the'long-half-lives of the transuranic
- nuclides, the intruder barriers must-be essentially I
permanent.
Engineered barriers will not provide intruder l-protection because of the assumed _ finite life of. the barrier.
' Waste Form Stability Testino Criteria o
The waste form stability testing criteria provided in NRC guidance documents do not relate on a one-to-one basis with L
1 1 Ov 2-3
Q the 10CFR61 performance assessment, and they are ' not intended to represent a simulation nor acceleration testing of the actual disposal environment.
Rather, they are assumed by the NRC to provide a level of assurance that a waste form that meets the test criteria will provide adequate stabilization in the disposal site.
's l
1 n
4