ML20195G584
| ML20195G584 | |
| Person / Time | |
|---|---|
| Issue date: | 04/18/1988 |
| From: | Parry S Advisory Committee on Reactor Safeguards |
| To: | Savio R NRC |
| Shared Package | |
| ML20151C617 | List:
|
| References | |
| FOIA-88-470 NUDOCS 8811280167 | |
| Download: ML20195G584 (10) | |
Text
. . . .. s. .....m.
t v
- #g UNITED 8 TATE 8
! NUCLEAR REGULATORY COMMISSION -
{ ADVISORY COMMITTEE ON REACTOR SAFEGUARDS WASHINGTON, D. C. 30688 e,
- ...*. April 18, 1988 -
PEMORANDUM FOR:. R. P. Savio, Chief Project Review Branch #1 0 /
FROM: S. J. S. Parry, Senior ACRS Fellow d*
$UBJECT: RECOMMENDATIONS PELATIVE TO "B" AND "C" LLW FORMS AND PACKAGING FOR DISPOSAL Un January 28, 1988, a survey report on LLW solidification processes and high integrity containers (HICs) was submitted. In the cover letter to that report, a comitment was made to provide the subcomittee members
( with recomendations relating to that report. On February 11, 1988, the recorrendation report was deferred until the staff had ccepleted their report to NMSS management. Given HMSS's staff statement that their report will be delayed to the June 1 time frame, it seems appropriate to provide my draft views on these topics at this time.
Attachment:
As stated cc: D. W. Moeller, ACRS Member H. Steindler, ACRS Fember R. F. Fraley M. h. Libarkin ,
T. McCreless
- 0. S. Merrill j f,
,ne 0011200167 00111D bN1 .470 PDl ;
\"
e
-. , , -- v 3 e MAFI DRAFT St
. PARRY /bys 4/18/88
. PARRY:LLWASTE April 18, 1928 l
~
MEMORANCUM FOR: D.W. Met 11er, Chaiman, Waste Management Subcomittee FROM: S.J.S. Parry, ACRS Senior Fellow
(
I
SUBJECT:
Recomendations regarding LLW Fom and Containers A survey report on the solidification and packaging of C1sss B and C LLW has beenpreparedanddistributed(January 28,1988) Pe the subcomittee's use ,
and infomation. While Dr. Mapp, Mr. Sumier and Dr. Tokar and their associates provided active assistance with the research of data for the survey report and this documento the author assures full and complete responsibility for the ccerpleteness of the data and conclusions drawn in the sumary report and the recoerendations presented herein.
The process of licensing a waste fem or solidification process and the - ,
several responsioilities associated with it will to briefly outlined. NRR is responsible for what equipment and/or processes are utilized in the reactor planttoNolidifyI.LW. It autho,izes the use of whatevEr system is chosen to
. DDAET
,1 . .
be used. On the other hand NMSS reviews, and grants (if warranted), an
~
approval, of the waste fem that is produced in the plant and shipped to the disposal site. The latter review and approval procedure consists of the vendor of the process submitting a topical report to the HMSS Staff which describes the process, and presents experimental evidence that the product will meet the perfomance criteria placed on the waste fem or the package, .
by 10 CFR 61 or applicable branch technical positions.
In actuality, the -
process is more comp 1'ex than that, since vendors usually submit an experi-mental program plan which outlines how they plan to oitain data to be pre-sented in the topical report prior to perfoming the actual tests. While not belaboring the complexity of these efforts it should be noted that only one solidificationprocessandswohighintegritycontainer(h!C)designshave
( been approved by the hMSS Staff (up to the date of this draft. February 3, 1986). This is cut of perhaps 3D topical reports that have been submitted over the past 5 years or so.
The statistks given above should not be construed as a criticism of the NMSS Staff. It is, in fact, an illustration of the problems facing hM55. If one were to literally apply the current requirements on each source of 8 and C low-level waste, it would be necessary to ha',e each plant submit a topical report on each batch of B and C wks'te as it was processed. Given the time ;
and financial expenditure that each report represents that would be an impossible and impractical situation. In matter of fact what has happened is - l that she three states containing the currently active disposal sites have .
granted authorization for the disposal of the waste foms on an interim
[
basis while the tcpical reports are being reviewed. But as one fomulation ,
is being reviewed, a slight variation in the composition of the waste stream DDAET
' "~
DMFI may occur, requiring the submission of additional test data. For example,
~
sodium sulfate solutions of various compositions and with the adr' ion of small amount.'s (1%) cf a chelating agent may each require an additional review and testing program.
The key point, regardless of the review performed, is that there is ng ,
guarantee that the waste actually produced at the plant, whether it be by .'
utility cr contract personnel, has any similarity to the waste originally tested and approved and thus, the performance of the solid on its actual condition at the disposal sites may not be assured. Since detailed perfor-mance assessment analyses of the sites, which take into consideration the variable nature of some of the solid waste streams, have not been perfonned
( it cannet be said that this situation will not represent a present or future risk to the public.
Reference is made to Attachment I which is drawn from the survey report on LLW solidification processes and HICs. The sections cited describe the nature cf the two classes of solidification processes, i.e., chemical bonding and encapsulation. A conclusion that may be drawn from these citations is that those processes that rely on bonding between the waste particles and the matrix will likely produce variable products of uncertain properties. There is, of course, both experimental evidence (prelimintry BNL data) and a practical illustration (failure of the TMi-2 liners) of this view (a second plantproblemwasrecentlydisclosedbythestaff). Again, however, it has yet to be demonstrated that such a situation represents s'present or future threat to public he.ilth and safety.
.- an a w
f5IN[
On the other. hand it can be stated that if the solidified waste is disposed of in a container which ensures the retention of the waste 'until it decays radioactiveIytoinnocuouslevels,.henthepublichealthandsafetycanbe assured. Such a position, however, may not be supportable on cost-benefit i grounds.
As a result of my studies and analysis, I suggest that the ACNW might explore i the following topics:
- 1. Should risk assessments prepared to support 10 CFR 61 be updated to include the possibility of an inadeouste waste form being emplaced in the burial site?
(
- 2. If chemically bended waste foms are not as satisfactory as encapsulated materials from a standpoint of unifemity and reproducibility, should these materials be banned until a demonstration of acceptable risk to the public is ccepleted?
- 3. Would the use of metallic or structurally stable HICs essentially ;
eliminate any risk to the public? Should a demonstration of acceptable risk to the public be rcquired?
l
- 4. Does the current system of approving topical reports after the solidt-fication process is authorized for use at a reactor give assurance that. .
the wastas entering the disposal site will meet the criteria set for the
( ,
waste? Is the arount of resources expended by the present process justified by the results.
e
m--__ __
p p ,r., g7
, . U11:;l B
(
Attachment:
As Stated .
cc: WM Subcomittee Members l R.F. Fraley l M. W. Libarkin '
T.G. McCreless R.P. Savio -
- 0. Merrill ;
, l . ,
i l ,
' ( .
'l i f 1 !
4 f
j i
, 1 t
i i .
t l
i i
i l
/h# .
i 1 . .. i J, l i \
I e ;.
I !
l R FlA P*P
. ~ ,. .
,r . !!. ' solidification Processes ,
A. Background Discussion Approximately 15 topical reports on various solidification processes and the resulting waste forms have been submitted to the NRC Staff for review and approval over the past 5 years. Of thesa, only one, the AZTECH process, ,
developed by General , Electric, has received official NRC approval. Note:
Nuclear Packaging bcught out all of G.E.'s activities in the low-level waste area in mid 1987, including the AITECH process.
The various solidification processes may be divided into two general categories or classes. They are: 1) Solidification processes in which the
(
waste itself is chemically involved in the process, that is to say that the waste itself forms a chemical bond with the solidification agent; and 2)
Encepsulation techniques in which the waste itself is not chemically active in the process, but is, in fact, physically surrounded by the matrix material. The Portland and gypsum cement based processes are representative of the fomer group, while the bitumen and polymerization processes would be included in the latter.
It may be seen that in the first case the waste materials themselves help On detennine the physica1 and chemical properties of the final waste form.
the other hand, in the lattar group, the waste materials prcbably do not ,.
- materially affect the properties of the waste form,' until very high waste loadings are reached. It is also worthy of note that some of the
( . .
99 9
. ~ .
]
s encapsulation processes either completely or.largely drive off any water i contained in the wasta stream and may physically fractura the individual waste parti.'cles. Alternatively, the cement based processe.s must take into account the presence of variable fractions of water in the waste streams. -
Realization of these intrinsic differences between these two types of pro-cesses leads to an understanding of some of the reasons for the variable .
performance of the waste forms and processes themselves. It should be noted that in either case the resultant solid, the taste fom, is not homogenous.
The matrix or bending phase surrounds discrete particles of waste which may or may not be chemically beund to the matrix.. Yhus in the case of the cementaceous processes the constitution and properties of the matrix may be quite variable. On the other hand in the encepsulation processes the matrix,
( and consequently the solidified mass, will tend to have reproducible and predictable preperties, largely independent of the initial waste material.
l' \
I I
I 1
l l -
)
?:l. ,-
- (
i l
s e T
.. ; . c i
.
- l B. Chemical Gonding Processes .
As noted previous'y these processes are largely based upon the use of Portland cement. One process uses a gypsum cement, but the fundamental l approaches are similar for all of the processes. The following. sections will !
present significant aspects of several processes. All processes 7equire small scale testing of the specific batch of waste to be solidi'ied, prior'to ,
the start of the processing campaign. This procedurt. generally involves a -
through mixing of the waste, to enhance uniformity, followed by sampling and i i
the preparation of laboratory scale samples of the waste form. Based upon this laboratory scale testing, additives, such as 11me for pH control, or emulsifiers to break up oil contamination, may be added and re-testing i perforead. Upen detemination of a "suitable" femulation the process
(
campaign would then be initiated. It should be noted that the F,rincipal criteria for "suitability" is whether or not the fomulation sets within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> or so, withcut free standing liquid. Thus, it may be seen that each waste stream, during eaa solidificatien campaign, may produce a product that has no guaranteed similarity in either chemistry or physical properties to that prepared during another campaign or in the original testing program for i i
the topical report data.
~
'I
( *. ^
. I g .
r
[ C. Encapsulation Processes The encapsulation processes use organic binders rather than inorganic materials such is Portland cennt. At the present time there are four such processes in use or proposed for use in the reactors. Only one of these, the AZTECH process, has been approved by the NRC Staff. There are two general groupings of binders, the as@nitic or bitumenous mat. trials and the '
~
monomer /polperic substances. In several of the processes the mixtures are heated and a vacuum is applied thus eith'er wholly or largely removing the water centained in the waste. In all cases the resulting solidified naterial centaining the waste consists of ficely dispersed waste particles surrounded by an inert binder which forms a continuous matrix. Thus, the binder or matrix detemines the general physical preperties until very high loadings of waste are reached. As a consequence the tc. reducibility and unifomity o','
, physical properties might be expected to be better than that of the
, chemically bonded materials. Further, sint.e the aquecus portion of the waste streams is eliminated in several of these processes the volume of material l
placed in the disposal site is generally less than the initial volume of the waste.
I **
l I .
- 11-7
. .