ML20136G101
| ML20136G101 | |
| Person / Time | |
|---|---|
| Issue date: | 06/18/1985 |
| From: | Dircks W NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | Asselstine NRC COMMISSION (OCM) |
| Shared Package | |
| ML20136F845 | List: |
| References | |
| FOIA-85-448 NUDOCS 8506250497 | |
| Download: ML20136G101 (8) | |
Text
CY:RU55 KELBER MARCUS BASSETT 6
a
' SILBERBERG JUN 18 N MARINO 1
ARLOTT0/SHA0 GILLESPIE/ERNST FILE MEMORANDUM FOR:
Commissioner Asselstine FROM:
William J. Dircks Executive Director for Operations
SUBJECT:
STEAM EX?t0SIONS The staff has prepared the enclosed response to your questions on steam explosions.
(3iped) WLiliam !. Dircks i
William J. Dircks Executive Director for Operations
Enclosures:
1.
Response to Questions Regarding Steam Explosions.
2.
List of Stamm Explosion Review Group Members.
3.
Executive Lanmary for NUREG-1116.
- 4. -List of Some Published Material on Steam Explosion Phenomena.
5.
Statement of Work Describing Program on Analysis of Steam Explosion Uncertainties, d-Topical Note on Reassessment of TMI-2 Core Damage.
- Chaiman Palladinc
}C{en!it ribution l
Commissioner Rooerts tral File Commissioner Bernthal Contnissioner Zech W0ircks OIT JRoe OGC TRehm SECY VStello RMinogue contact:
GCunningham C. Allen, NRR HDenton 492-7347 DEisenhut pI TSpeis CAllen N",cnemenc
{ Reading 4
0 RR EDO-000576
,h5dMg f
" Smith EDO-000576
$**da.
DS N
..p....f.....f I.
D.j.{RR E
,, ;,Q,,,,,,,,
on.c,
m>
=,,
m esoam ais ooooi w cw on.o OFFICIAL RECORD COPY usom mi-m
)
0
.., Enclosure 1 Response to April 25, 1985 Question From Commissioner Asselstine Regardina Steam Explosions QUESTION: Commissioner Asselstine asked for information regardi g the potential for a steam explosion within the reactor pressure vessel to generate missiles that might breach the containment and allow early release of fission products.
In particu1ar he noted that he
~
has been told that there is a growing consensus that such a con-tainment failure mode (denoted as a-failure) has a very low probability (Pa).
In that regard he asked three questions:
(1) Does staff now believe there are scientifically accepted principles and methodologies for calculating that probability?
(2) If so, does staff intend to publish the bases for calculating the probability in an archival, peer-reviewed journal before making use of that probability?
(3) If not, when does staff believe a justifiable basis for calculating the probability will be available?
Response: The staff agrees with the observation that there is a growing i
l consensus that the probability of a-mode failure is very low. The basis for the staff's current position is documented in NUREG-1116, A Review of the Current Understanding of the Potential for Containment Failure Arisina from In-Vessel Sigam Explosions. A draft version of this report has been provided to the Commission (memorandum Dircks to Palladino, March 7,1985).
The report presents the results from a recent review of the potential for a-mode failure l
'm u
. sponsored jointly by the Office of Nuclear Regulatory._Research (RES) and the Office of Nuclear Reactor Regulation (NRR).
The document
~
has come to be known as the SERG (Steam Explosion Review Group) report.
The members.of the SERG consist of individuals that have considerable expertise in some area of steam explosion phenome-nology and have made significant contributions to studies involving the potential for steam explosion induced containment failure. A list of the members is provided in Enclosure 2.
The final version of the executive summary of the report is included here as.
One of the objectives of the review was to obtain from the members of the SERG their independent estimates of the conditional prob-ability (i.e., given a core melt) of a-mode failure. The members
- were also asked to review a report by Dr. Marshal Berman, et. al.,
issued by SNL (NUREG/CR-3369) describing studies of uncertainties in the likelihood of steam explosion' induced containment failure which concluded that the probability of containment failure must be assumed to " span the range from 0 to 1." As noted in the executive summary, all members of the SERG provided a best estimate value which is less than the best estimate value of 10 estimated in WASH-1400 for this probability. The members also disagreed with the methodology as used in NUREG/CR-3369 to reach the conclusions with regard to uncertainties in estimating Pa reported in that document.
The members of the SERG were not constrained to follow any particular prescription in making their independent estimates.
A review of NUREG-1116 reveals, in fact, considerable variation in the approaches selected by the various participants.
As the report notes, however, there was agreement regarding several
(
j scientifically accepte.d concepts involving key aspects of steam explosion phenomena.
These include agreement on the validity of transient jet breakup and flooding and/or fluidization limits on the extent of fuel-coolant mixing.
Most members of the SERG agree that limitations on the mass of molten fuel that can participate in the steam explosion is an important factor in reducing their' estimates for Pa.
Some members believe careful evaluation of the distribution of energy (including energy losses) and the mechanical response of the system also provides strong bases for estimating low values of Pa.
Still others believe that realistic assessment of the initial conditions (e.g., the melt progression leading to core melt-water contact) will also result in low estimates of Pa.
The fact that the various analysts reach low estimates for Pa by emphasizing different phenomena provides additional confidence that, overall, the probability is indeed low.
It should be noted here that the authors of the SNL report, NUREG/CR-3369, disagree with many of the st.atements of the SERG experts as set forth in NUREG-1116.
Their objections can be found in letters wtiich are included as Appendix E to NUREG-1116.
There is no single currently accepted methodology that embodies all of the important principles in an integrated analysis.
How-ever, as noted above, there are scientifically acceptable principles and concepts that can be used in estimating such probabilities. Applications of these physical concepts in evaluating steam explosions have been reported by members of the SERG (and others) in archival peer-reviewed journals. A partial list of several currelt publications is attached to this memorandum (Enclosure 4).
The 1ist is not a complete one.
Ir. the two decades or more that this typu of phenomena has been studied hundreds of l
l i
., articles have been published.
The phenomenon of molten material-coolant. interactions has been a subject of international study both independently and in joint cooperative efforts.
The history includes extensive studies in the LMFBR program and in non-nuclear activities (paper, aluminum, steel and LNG) as well as in the LWR industry.
The attached list represents some of the more recent publications concerning key aspects of steam explosions and a-mode failure phenomena in LWR.
A few examples from this list can
'be used to illustrate the point that important concepts used in the evaluation of Pa have been published in peer reviewed archival journals.
In particular the articles by M. Corradini (items 10 and
- 19) and T. Theofanous (item 26) illustrate the use of an integrated approach (item 10) and the application of physical modeling of flooding /fluidization (item 19) and transient. jet breakup (item 26) respectively.
I
' The staff believes that the application of physically reasonable models using concepts such as those described above together with reasonable engineering judgement provides a satisfactory basis.for reaching the conclusion it has adopted regarding the low value for l
Pa.
The staff currently has work in progress at the Franklin Research Center that we believe will provide a more coherent methodological approach for evaluating Pa.
The statement of work describing this program is includea as Enclosure 5.
This methodology will provide i
a structure by which the significance of the modeling assumptions and phenomenological uncertainties involved in estimating Pa can be t
examined.
Further, the methodology will also allow us to make l
reasonable judgements regarding the range of uncertainty that l
sh'ould be assigned to estimates of Pa.
We currently. expect results
b
. from this task to be available in six months. A review article describing this methodo-logy (and other approaches) will be submitted to a peer-reviewed journal for publication within the next few months.
It is worth noting that, based on its review, the APS Study Group on Radionuclide Release from Severe Accidents at Nuclear Power Plants concluded that the probabil~ity of containment failure from steam explosions is very unlikely.
Finally, it is relevant to nota here that current evaluations of the TMI accident tend to support the concept that limited energy release is obtained from molten material interactions with water
- in a severe accident.
(See the note by George Kalman on Reassessment of TMI-2 Core Damage in the May 2,1985 NRR Technical Report - a copy is included here as Enclosure 6).
Data from TMI is, of course, the most prototypic experimental data currently available. Other experimental results and ' programs have been designed to obtain specific information in limited areas such as conversion efficiency and generally involve non-prototypic conditions. Although the TMI results are preliminary they do tend to provide additional confirmatory support for the conclusion that l
contact between large molten debris masses and water, under l
prototypic conditions, do not produce an energetic steam explosion.
The staff is continuing to evaluate the information available from the various TMI studies (INEL and others) in this area to determine i
the specific applicability of these studies to improving our l
understanding of steam explosion phenomenology.
In particular we will pursue the relevance of questions of prototypicality, including the time scale of the sequence at events.
/
l
. In summary, the responses to Commissioner Asselstines's questions are:
(1) There is no current accepted. methodology that embodies all of the important principles in an integrated analysis for -
calculating Pa.
(2) The staff has work in progress to develop such a methodology.
A review describing that methodology (and other approaches) will be submitted to a peer-reviewed journal within the next few months.
(3) The staff believes the review presented in NUREG-1116 provides an adequate basis for its current position with regard to Pa.
l t
i
r
., Enclosure 2 Steam Explosion Review Group (SERG) Members Dr. S. G. Bankoff Mr. P. Cybulskis W. P. Murphy Professor of Chemical Nuclear Systems Section and Mechanical Engineering Battelle Columbus Laboratory Northwestern University 505 King Avenue 2145 Sheridan Road Columbus, OH 43201 Evanston, IL 60201 Dr. H. Fauske Dr. William Bohl FAI 16 W 070 Los Alamos National Laboratory West 83rd PO Box ~1663 Burr Ridge IL 60521 Los Alamos, NM 87545 Dr. Anthony Briggs Dr. T. Ginsberg Safety and Engineering Science Division Brookhaven National Laboratory UKAEA Atomic Energy Establishment Building 820M.
Winfrith Dorchester Upton, NY 11973 Dorset DT2-80H England Dr. F. Mayinger Mr. T. Butler Institute for Thermodynamics Los Alamos National Laboratory University of Munich PO Box 1663 Federal Republic of Germany Los Alamos, NM 87545 Dr. D. Squarer Dr. I. Catton Nuclear Safety'and Analysis Department of Nuclear and Electric Power Research Inst.
Mechanical Engineering 3412 Hillview Avenue University of California Palo Alto, CA 94304 2567 Boelter Hall Los Angeles, CA 90024 Dr. T. Theofanous 132 Pathway Lane Dr. D. Cho West Lafayette, IL 47906 Argonne National Laboratory RAS Division 9700 South Cass Avenue Argonne IL 60439 Dr. H. Corradini University of Wisconsin 1500 Johnson Drive Madison, WI 53706
_