ML20213F357
| ML20213F357 | |
| Person / Time | |
|---|---|
| Issue date: | 06/21/1985 |
| From: | Rosztoczy Z Office of Nuclear Reactor Regulation |
| To: | Speis T Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20213E209 | List:
|
| References | |
| FOIA-87-113, FOIA-87-60 NUDOCS 8507020137 | |
| Download: ML20213F357 (11) | |
Text
~.
- a.
ma asyg j%
UNITED STATES
..[
NUCLEAR REGULATORY COMMISSION
~
.'70 a
o WASHINGTON, D. C. 20S$5
[ j' June 21, 1985
/'
MEMORANDUM FOR:
Themis P. Speis, Director Division of Safety Technology FROM:
Zoltan R. Rosztoczy, Chief Research and Standards Coordination Branch Division of Safety Technology
SUBJECT:
MINUTES OF THE NRC/IDCOR MEETING ON APRIL 30, 1985
Reference:
" Minutes of the NRC/IDCOR Meeting on March 26, 1985", memo from Z. Rosztoczy to T.'Speis, dated May 9, 1985 t
Meeting Sumary T. Speis opened the meeting sumarizing the status of the outstanding i
technical issues.
Seven out of the eighteen issues were discussed at the March 26 meeting. Agreement was reached on the approach to resolution for six issues.
One issue, hydrogen generation, will need further discussion.
The purpose of the meeting is to discuss the remaining eleven issues plus j
one new issue:
retention of fission products in ice beds.
T. 3uhl in his opening remarks acknowledged receipt of Mr. Dirck's response to C. Reed's March 21, 1985, letter.
He expressed concern about the schedule change indicated in Mr. Dirck's letter, and urged NRC to accelerate NRC analysis'of the reference plants, if possible, in order to reduce the delays. Copies of the April 15, 1985, ACRS letter on " Severe Accident Policy - Systematic Review of Nuclear Power Plants' were provided for the IDCOR participants.
T. Speis requested that IDCOR study the ACRS letter and use it in the development of the methodology for generic applicability.
Discussion of the twelve remaining technical issues followed, chaired by Z. Rosztoczy.
R. Henry lead the IDCOR presentations. An approach to resolution was agreed on for each of these issues.
The agreements reached are summarized below under " Resolution of Technical Issues".
Resolution of technical issues is the first step toward the systematic review of operating plants.
Close cooperation between IDCOR and NRC will continue in the coming months.
The tentative schedule is as follows:
IDCOR will complete development of the methodology to be used for the systematic review of operating plants by the end of June 1985.
IDCOR will probably present_the_results to NRC in July, 1985.
f ~f h
- (4 I\\ w v
~%.__..
JUN 211985 The new methodology will be applied to the four reference plants to accomplish two goals.
It will demonstrate the feasibility of the new methodology, and at the same time it will update the reference plant analyses to the level agreed on in the resolution of the technical issues.
Expected completion date: September 1985.
IDCOR will also apply the methodology to a few selected plants, which do not have PRAs.
These plants will include a CE and a B&W design, since the four reference plants represent only GE and W designs.
Expected completion date:
end of 1985.
NRC's Severe Accident Research Program will complete, as soon as possible, the analyses of the four reference plants. The schedule of these analyses is presently under review.
NRC will evaluate both the IDCOR analyses and the SARP analyses for each reference plant and will develop guidelines for the systematic review of operating plants.
The evaluation will start with the Peach Bottom plant, a BWR with Mark I containment.
NRC will review the IDCOR' proposed methodology for the systematic review of operating plants as soon as it is available.
Similarly, r
' application of the methodology to selected plants will be reviewed when available.
The effort is expected to culminate in the issuance of guidelines for the systematic review of operating plants together with an assessment of the IDCOR methodology as an acceptable tool to perform this review.
IDCOR and NRC will be ready to meet with ACRS in August 1985.
The purpose of the meeting will be to provide a status report for the j
Committee, outline future plans and solicit the Committee's comments, i
A list of attendees is enclosed. Copies of handouts are available from the NRR contact, James Watt, RSCB (X28279).
Resolution of Technical Issues Summaries of the agreed-on approach to resolution of issues 8 through 18 i
addressed at the meeting are provided below.
- Issue #8 Direct Heating of Containment by Ejected Core Material Corium dispersal in high pressure PWR sequences provides a mechanism for i
direct heat transfer of molten core material, latent heat, and exothermic
. chemical reaction energy to the containment atmosphere.
Sufficient energy i-transfer provides a mechanism for early containment failure; however s
qualitative and semi-quantitative assessments show that sufficient energy l'
transfer to fail containment is unlikely to occur.
/
JUN 211985 1
IDCOR is assessing the likelihood of corium dispersal based on detailed cavity and containment designs.
The staff concurs in this approach.
Corium dispersal is unlikely in many cavity designs based on geometric
-considerations.
These efforts provide a valuable screening process.
IDCOR has assumed that in the Zion-like cavity 50% of the corium will be dispersed from the cavity onto the reactor floor where it will ultimately be s
cooled.
None of this material was assumed to participate in direct heating phenomena, due to the structures above the exit of the instrument tunnels.
IDCOR cited tests at ANL to support their view.
The staff believes that these tests were not prototypical and do not serve as an adequate analytical benchmark.
Testing of a high pressure blowdown of a melted core into a tenth scale model of a Zion-like cavity performed at Sandia demonstrate the direct heating phenomena, and provide information on particle size distribution versus mass of the corium.
These tests vented from the cavity to an unobstructed geometry.
Future testing will incorporate prototypical geometric obstruction.
Since obstructions in the lower containment and steam generator compartments act as separators, it is important to know the behavior of corium dispersal as a function of particle size as well as total mass. Credible quantitative models of these phenomena have not been developed to date by either party.
IDCOR has agreed to benchmark its analyses against prototypical data as available.
The staff agreed with IDCOR's screening approach. The staff reconsnended that IDCOR-developed screening criteria be reviewed by the staff as soon as feasible.
By July, IDCOR will categorize all plants according to cavity geometry and expected debris dispersal.
IDCOR will provide estimates of predicted debris oispersal for each plant.
However, IDCOR is not planning to provide an uncertainty band with the predicted dispersal.
The uncertainty associated with dispersal should be evaluated and included in the overall uncertainty analysis.
NRC:
J. Rosenthal IDCOR:
R. Henry
- Issue'#9 Ex-Vessel Fission Product Release NRC's model tracks more species and allow chemical reactions that produce volatile oxides and hydroxides of refractory materials and generally calculate a higher release of fission products.
IDCOR, while agreeing more chemical species should be considered in their analyses, contends that the following considerations included in IDCOR's analyses should make their results more realistic as compared to NRC's:
1.
, Core melt is discharged from the vessel over several hours as compared to immediate discharge of all core melt in NRC's model.
~
'JUN 211985
. 2.
Core debris is assumed to be dispersed over a larger area in more 4
readily coolable geometry.
i 3.
IDCOR's model calculates the debris heat-loss more mechanistically.
IDCOR also contends that ex-vessel fission product releases need not be considered for cases involving suppression pool scrubbing or without containment failure. NRC agrees with IDCOR that suppression pool scrubbing
, can substantially reduce the amount of ex-vessel fission product releases, but not completely eliminate them.
NRC believes that even this reduced rate cannot be neglected in accident sequences that involve containment bypass (such as equipment hatch left open).
1 While NRC agrees that instant discharge of 100% core materials at the vessel failure is not realistic, we need further evidence to support IDCOR's t
contention that core melt will take place over a period of a few hours.
k Core. debris dispersal ~was verified by SNL's high pressure melt ejection tests.
For the melt discharged under gravity, dispersal of core debris out of the reactor cavity is highly unlikely. Moreover, in SNL's high pressure
. melt ejection tests, a huge cloud of aerosols was observed.
It is the NRC's position that source terms associated with the process of pressurized melt ejection should be considered if one is to take credit of the core debris "di'spersal.
The effect of' different heat transfer models was discussed in Issue #10.
NRC believes that the divergence between the NRC and the IDCOR methodology i
is caused primarily by the first two assumptions.
Resolution for Issues #9 and #10:
IDCORwillimprovethefihsionproductmodel(morechemicalspecieswillbe considered, chemical and' thermal equilibrium of important materials) and treatment of the bottom ci'ust.
NRC will evaluate the effect of realistic, timr 6 pendent discharge of core materials instead of instantaneous discharge.
l NRC will provide IDCOR with test resulti 93 te a concrete interaction available by the end of this summer from the Beta facility and from the
-Sandia program.
Both IDCOR and NRC will compare their respective analytical models to available-test renits and will make iraprovements as needed.
Both IDCOR and NRC will update the reference plant analyses to reflect changes made in the calculational models and insight gained from the
. experiments..
f NRC:
R. Curtis IDCOR:
M. Plys
,o
~
. ~... - - - - - - - - - ~ - - - - - - - - - - - - - - - - - - - - - - -
,,,e
JUN 211985
- Issue #10 Ex-Vessel Heat Transfer Models From Molten Core to Concrete /
Containment NRC analyses predict more heat going into concrete attack.
IDCOR cited heat transfer models that are included in IDCOR's analyses, but absent in NRC, and maintained that their modeling is more realistic because they include:
1.
Natural circulation between containment compartments.
2.
Hydrogen recombination.
NRC questions that natural circulation could have an appreciable effect on upward heat transfer from the melt into containment.
Both NRC and IDCOR agree that analytical models should be benchmarked against experimental data.
SNL's large melt tests and BETA tests in FRG should provide an improved data basis for such benchmarking.
For agreed-on resolution see issue #9.
~
NRR:
R. Curtis IDCOR:
M. Plys
- Issue #11 Revaporization of Fission Products in the Upper Plenum IDCOR agrees with the NRC recommendation to consider the revaporization of chemically reacted fission products in their model although they disagree with NRC on the focus of the issue.
- NRC believes that revaporization is a timing problem and that if revaporization occurs late into the accident where the containment source tenn is low, then this source term could be altered, even with relatively i
small amounts of revaporization.
IDCOR believes that the assignment of high vapor pressure to the volatile fission products necessarily implies i
conservatism in the resultant calculations and that heat loss from a PWR reactor vessel after vessel failure may be too great t'o allow surface heating and fission product revaporization.
l While NRC disagrees with IDCOR that the use of the conservative vapor,
pressures would resolve the issue -- since in this case revaporization would occur earlier than realistic, we agree with IDCOR that thermal hydraulics might play an important role in late revaporization.
IDCOR agrees to carry out.. integrated accident analyses using lower vapor pressures for the volatile fission products to demonstrate the influence of such uncertainties.
t' Both parties agree that an integrated calculation is needed to assess the full.effect of this issue.
In order to resolve this issue, parametric studies on time delays and release rates will be performed by IDCOR and also by NRC.
NRC:
M. Silberberg IDCOR:
M. Kenton
JUN 211985
- Issue #12 Fission Product and Aerosol Deposition in the Containment This issue is closely related to Issue 4, " Fission Product and. Aerosol Deposition in the Primary System", and to Issue 138, " Retention of Fission Products in Ice Beds", for cases when the. ice has melted, since the same empirical aerosol model is used by IDCOR in the primary system and the containment. The NRC continues to have concerns about the validity of this model for the full range of conditions and phenomena applicable to severe accident conditions.
IDCOR and NRC agreed to the follo~ing path w
to resolution.
1.
IDCOR will continue to compare the empirical model against appropriate experimental data, and will improve the model as needed for the range of applications involved in the reference plants including retention in empty ice baskets.
2.
IDCOR will assess the empirical model with a detailed (mechanistic) aerosol model.
3.
NRC and IDCOR will select an appropriate severe accident sequence to make a clear comparison between IDCOR analysis and CONTAIN results.
NRC:
R. Curtis IDCOR:
M. Epstein
- Issue #13A Amount and Timing of Suppression Pool Bypass IDCOR proposes to use the Vaughan plugging model in IDCOR calculations without limitations. According to this model, aerosol will plug any leakage path, even large openings (i.e.,1 ft. wide) as long as a sufficiently large amount of aerosol passes through the path. The geometry and orientation of the opening as well as the rate of discharge have no effect on the. predicted plugging. NRC questioned the use of this correlation.
Agreement was reached on the following approach to resplution:
IDCOR will analyze the stuck open vacuum breaker case both ways:
(1) using the plugging correlation; and (2) assuming no plugging.
IDCOR will review the basis of the plugging correlation including potential limitations on the correlation and the applicability of the correlation for containment leak paths.
IDCOR will consider in the analysis of the reference plants all possible bypass paths'and the fluid conditions pertaining during potential accidents.
- =..
~
NRC will review any further evidence for aerosol plugging of leakage paths that IDCOR considers relevant to the suppression pool bypass issue.
JUN 21 NES 7-NRC:~ J. Read IDCOR:
J. Gabor
- Issue #13E Retention'of Fission Products in Ice Beds NRC and IDCOR agreed that retention of fission products in ice beds is important only for large releases to containment before the ice melts combined with early containment failure.
The NRC has a subroutine in NAUA based on the ICEDF code, that explicitly models the aerosol processes in the ice bed and associated structure.
IDCOR models the ice compartment as a specific volume in their containment description, thereby invoking the standard deposition models of steam condensation and aerosol deposition.
To resolve this issue, IDCOR will:
1.
Provide information from the Sequoyah analyses on the fraction of material reaching the containment that is deposited in the ice compartment.
2.
As an explicit part of the verification process, provide a discussion of the applicability of the aerosol deposition model to the ice compartment.
Successful resolution depends on agreement on the ability of the aerosol deposition model to predict adequately fission product retention in ice beds.
Long term verification should be accomplished by comparing the steam condensation / aerosol deposition model with ice compartment experiments.
NRC:
M. Silberberg IDCOR:
.M. Plys m..
--Issue #14 Modeling of Emergency Response s_
- s IDCOR stated that they would continue to assume that 5% of the population would neither participate in evacuation, but would be relocated 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after plume passage.
This is acceptable to NRC, and this issue is closed.
NRC:
J. Read IDCOR:
Rv Henry
- Issue #15 Containment Performance
- Model experiments. on steel containments at Sandia have indicated that (1) containment function is expected to be available for membrane strains between 2 percent and 3 percent; and (2) the total observed leakage up to these strain levels remained below 2 percent of volume per day.
The large scale model experienced a bursting type of failure at approximately five times design pressure.
JUN~ 21 ESS In contrast, the IDCOR analyses assume that the dominant containment failure mode is " leak before break" which occurs as a result of large strain of the containment boundary.
The leakage in some cases in the IDCOR analyses prevented containment failure.
NRC commented that, if a relief path is needed for safety, a valve should be opened rather than relying on equipment failure.
In order to resolve this issue:
1.
IDCOR will develop containment strain methodology for failure based on experimental data and analyses performed to date.
2.
IDCOR will incorporate the strain-induced failure methodology in the integrated plant analyses such that individual penetration behavior and likely failure sizes can be estimated.
3.
IDCOR will consider uncertainties in failure mode in source term calculations.
4.
A small IDCOR/NRR subgroup will be set up to evaluate proposed containment failure methodology.
NRQ:
G. Bagchi IDCOR:
R. Henry
- Issue #16 Secondary Containment Performance Although NRC analyses of severe accident radionuclide releases do not always account for fission product retention in the secondary containment buildings (auxiliary and safeguards buildings), NRC concurs with IDCOR that inclusion of these effects is necessary for an accurate risk perspective.
Furthermore, given the resolution of questions related to fission product deposition mechanisms (Issue 12), NRC does not anticipate any disagreement with IDCOR over their physical models for fission product retention.
However, the NRC staff questions the high levels of fission product retention predicted by IDCOR.
For instance, in the V sequence Sequoyah, IDCOR predicted a decontamigation fgetor of about 10 for Zion an
(
For Peach Bottom, the DFs ranged from 10~ to 10 NRC emphasized that the N
i calculation of fission product reduction factors must account for situations in which the secon'dary building is unisolated at the start of the accident, or is compromised as a result of the accident sequence.
Other factors, such j
as ventilation systems which could transport fission products out of the building, must also be considered.
In contrast, NRC analyses show typically a decontamination factor of about two for an impaired structure surrounding the containment b'uilding.
These analyses modelled the structure volume but not the detailed geometry.
Residence times of radionuclides in this additional volume are based on calculated mass flow rates of steam, non-condensible gasses and aerosols from tne fault containment through the additional structure (typically the auxiliary building) to the environment.
-a-
.m-
.-e.-,
.,--c,
.---rc.m..
-,c,.,--.-----.-e...-,.,---.e
,,..-,a
-.+m-
I JUN 211985 IDCOR recognizes the NRC concerns.
In future analyses for reference plants and operating reactors, IDCOR will carefully consider the effect of these mechanisms on secondary building retention of fission products.. NRC will include secondary building retention in future calculations of fission product releases.
NRC:
R. Barrett IDCOR:
J. Gabor.
- Issue #I7 Hydrogen Ignition and Burning Both the NRC and IDCOR have developed computer codes to analyze nuclear reactor accidents involving the transport and combustion of hydrogen.
However, these codes use different assumptions and models for ignition, burning and flame propagation. Modeling differences result in HECTR predictions of higher temperatures and pressures for BWR Mark III, ice condenser, and large dry PWR containment types.
It was agreed that the following steps would be taken to resolve the remaining differences.
NRC and IDCOR would define and calculate a standard problem which would.
1.
include but not be limited to natural circulation and recombination models. One purpose of the standard problem would be to assess the importance of the differences which result from the IDCOR assumption of critical flame temperature criteria when compared with the corresponding modeling assumptions used in HECTR.
2.
The NRC will assess the effects of the IDCOR compartmentalization and report back to IDCOR the results of this analysis.
It is concluded that successful completion of these steps would provide the basis for final resolution.
NRC:
R. Curtis IDCOR:
M. Plys
- Issue #I8 Essential Equipment Performance IDCOR has not yet identified for the reference plants all essential equipment needed in case of severe accidents.
These equipment should be identified together with the time interval during which equipment need to function.
For essential equipment the predicted severe accident conditions should be compared against the equ.ipment qualification test conditions.
It is important that the comparison is done on a consistent basis.
Comparing the expected internal temperature of the equipment against the maximum l-environmental. temperature of the test is not meaningful.
Time at v.
temperature is one of the important parameters as is temperature, pressure, radiation, and submergence.
I n.--..,.-.,,
,n,.
nn.,_._,,,_-.-,,.,.,--,---_,.
,, _, -, -, - - -. - - - - -. -,,. -,.,, - - -. -, -,,,...,,. - ~. _,. -. - -
JUN 211985 Aerosol and particle deposition on equipment can increase radiation dose to the equipment. Similarly, radioactivity of water pools can have an effect on submerged and nearby equipment.
These conditions must be ta. ken into account.
IDCOR will identify essential equipment for the reference plants, will specify the time interval for which the equipment is needed, and will check whether
~
the qualification of equipment for design basis events is sufficient to support the assumed performance of these equipment during severe accidents.
NRC:
Z. Rosztoczy IDCOR:
J. Carter Z
a.n ca c.
Zoltan R. Rosztoczy, Chief Research and Standards Coord nation Branch Divisfor,of Safety Technology
Enclosure:
-List of Attendees
_ _ _ _. _ _ _ _ _ _. - _. - _,..__._,_.~--_.__ -.
JUN 2 1 fo85
~
List of Attendees NRC/IDCOR Meeting 4/30/85 J. Rosenthal NRC 492-9447 R. Barrett NRC 492-7591 M. Silberberg NRC 427-4737 G. Marino NRC 427-4270 M. Fontana ENERGEX (615)481-3300 R. Henry FAI (312)323-8750 M. Leverett EPRI (415)855-2936 T. Buhl ENERGEX (615)481-3300 Z. Rosztoczy NRC 492-4221 T. Speis NRC 492-7517 R. Bernero NRC 492-7373 P. Morris WESTINGHOUSE (412)374-5490 M. Kenton FAI (312)323-8750 M. Plys FAI 312)323-8750 J. Carter ENERGEX 615)481-3300 M. Epstein FAI 312)323-8750 J. Mitchell NRC -
427-4614 J. Read NRC 492-8301 M. Cunningham NRC 443-7984 P. Cybulskis BATTELLE (614)424-7509 J. Watt NRC 492-8279 P. Ellison FAI (312)323-8750 A. Sharon FAI (312)323-8750 H. Fauske FAI (312)323-8750 C. Nilsen NRC 443-7662 K. Winegardner PNL (509)375-3839 D. Yue NRC 492-8129 J. Burns NRC 443-7860 R. Curtis NRC 427-4262 K. Parczewski NRC 492-8593 T. Lee NRC 427-4568 C. Allen NRC 492-7347 D. Moorehead WESTINGHOUSE (412)374-5991 A. Wooten WESTINGHOUSE (412)374-7128
- 8. Mims TVA (615)632-7263
- 0. Bassett NRC 427-4281 O
-