ML20217G990

LWR Severe Accident Research Accomplishments & Future Plans, Presented at 970601-04 Second Intl Conference on Advanced Reactor Safety in Orlando,Florida

ML20217G990
Person / Time
Issue date:	06/01/1997
From:	Ader C, Rubin A, Tinkler C NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:
References
NUDOCS 9804290251
Download: ML20217G990 (8)

Text

,

The Second International Conference on Advanced Reactor Safety Orlando, Florida, June 1-4,1997.

This is a preprint (draft) of a paper intended for publication in a journal or proceedings. Since changes may be made before publication, this preprint (draft) is made available with the understanding that it willnot be cited or reproduced without the permission of the author.

LWR SEVERE ACCIDENT RESEARCH ACCOMPLISHMENTS AND FUTURE PLANS Charles E. Ader, Alan M. Rubin, Charles G. Tinkler Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Co==61an Washington, D.C. 20555 ABSTRACT After more than 10 years of rather broadly directed research, the USNRC issued in May 1989, its ne U.S. Nuclear Regulatory Commission has revised Severe Accident Research Program (SARP) been conducting Ij ht Water Reactor severe accident Plan. The SARP revision (NUREG 1365)' was g

research for over 17 years. his research has provided motivated by the need to focus efforts and crystallize a large body of technical data, analytical methods, and positions of value to the regulatory mission of the l

the espertise necessary to provide for an understanding agency. Early containment failure, or avoidance of it, of a range of severe accident phenomena. This became the focal point, and issues associated with it understanding of the ways severe accidents can constituted the key elements of that plan. The SARP progress and challenge containments has been was revised again in December 1992 (NUREG 1365, sufficient to support plant risk assessments, Rev.1), to provide an update on the progress made in development of accident management strategies and severe accident research and to refocus on resolve certain severe accident issues for operating and phenomenological issues (e.g., fuel-coolant interactions future plants. Much has been accomplished, however and debris coolability), direct containment heating and certain issues remain. His paper discusses recent on severe accident code development. De December accomphshments arising from the Severe Accident 1992 revision also outlined the plans for hydrogen Research Program and what is currently being done to combustion and source term research and the plan for address the remaining issues in order to further completion of remaining research on the BWR Mark I tahance our understanding and ability to predict Containment Liner Failure.

certain severe accident phenomena.

Since the SARP was revised in December 1992, I. INTRODUCTION significant progress has been made on many of the technical issues identified in NUREG-1365, Rev.1.

He U.S. Nuclear Regulatory Commission his paper provides an update on the more recent (USNRC) began the Severe Accident Research accomplishments in severe accident research in a Programs aher the TMI-2 accident in March 1979, to number of areas and provides an overview of the provide the t'a==61a= and USNRC staff with the current status and future plans of the severe accident l

technical data, analytical methods, and expertise research program to further enhance our l

necessary for assessing plant response to a range of understanding and reduce residual uncertainties in a severe accident scenarios, assessing the efficacy of number of these areas.

various strategies to prevent or mitigate the consequences of severe accidents, including improved II. DISCUSSION design features or meeldleat management strategies, and assessing the consequences of severe accidents. This The Severe Accident Research Plan issued in understanchag of the ways severe accidents could December 1992 (NUREG 1365, Rev.1) provided an progress and challenge containments was crucial to the update on the progress made in severe accident USNRC's moving forward in addressing the risks from research and outlined research plans to address a severe accidents, number of the remaining severe accident issues.

\\

()

9804290251 970601 PDR ORG NREB y

PDR

,m Ader 1

Recent accomplishme ts arising from the SARP since Examinations, steps to ensure water addition to the the publishing of NUREG D65, Rev.1, and future drywellin the event of a severe accident.) Research on plans are presented for the following major program this issue is now complete.

areas: closure of the Mark I containment liner failure issue; closure of the dired containment heating issue; B. High Pressure Melt Ejection (HPME)/ Direct fuel-coolant interaction and debris coolability; reactor Containment Heating (DCH) vessel integrity; bydrogen combustion; source term; and severe accident codes.

In certain reactor accidents, degradation of the reactor core can take place while the reador coolant A. BWR Mark 1 Containment Shell (Liner) Failure system remains pres 6urized. A mohen core left uncooled will slump and relocate to the bottom of the An accident sequence leading to early containment reactor vessel. If the reactor vessel fails, the core melt failure had been postulated for BWR Mark I will be ejected into the containment cavity under I

containments. This sequence involves the direct attack pressure. If the material subsequently should be of the containment steel liner by molten core material ejected from the reactor cavity into the surrounding following vessel failure. At the time of the publishing containment volumes in the form of fine particles, of NUREG 1365, Rev.1, considerable research had thermal energy can be quickly transferred to the been pcrformed by the USNRC to addressed key containment atmosphere, pressurizing it. The metallic phenomena associated with the liner failure issue, such components of the ejected core debris could further as melt conditions at the time of vessel failure; melt oxidize in air or in steam and can generate hydrogen spreading characteristics; thermal hydraulic and chemical energy that would further pressurize the characteristics of molten core-concrete interactions containment. This process is called direct containment both with and without an overlying water pool; heat heating (DCH).

transfer characteristics at the interface of the molten core, overlying water pool, and liner; and fission Direct Containment Heating was identified as one i

product attenuation in the presence of an overlying of the major areas of emphasis in NUREG 1365, j

water pool.

Rev.1, because of its potential for early containment failure, and a significant experimental and analytical Integration of the research information derived program had been undenaken in support of resolution from these programs into an assessment of the of the DCH issue. Considerable progress has been conditional probability of liner failure both with and made toward the resolution of the DCH issue for without an overlying water pool in the drywell, given a PWRs. The planned experimental programs have now core melt accident that proceeds to vessel failure, was been completed, along with the resolution of DCH for l

2 completed and provided in NUREG/CR 5423. By a number of PWR plants. The resolution for the developing probability distributions for important remammg plants should be completed in early 1997.

parameters that factor into the analysis from data The USNRC research progrars consisted of (1)

(where available), computer analyses, or other insights, integral testing at different scales, (2) separate effects the authors of NUREG/CR 5423 were able to obtain testing, and (3) analytical model development and estimates of the likelihood of liner failure both with validation. In terms of scaling, the severe accident and without a water pool overlying the molten corium scahng methodology that was published in in the drywell. As a result of peer review of NUREG/CR 5809' guided the formulation and NUREG/CR 5423, additional work was completed and execution of the experimental programs and analytical reported in NUREG/CR-6025' invohing: (1) the methods.

initial conditions of melt release, (2) the possible range of melt spreading and depths against the liner, (3) the The first terolution report, an assessment of the duration of melt superheat, and (4) the liner failure DCH load for the Zion type configuration, was criteria. The conclusions in NUREG/CR-6025 completed in December 1994, and was issued as confirmed the conclusions in NUREG/CR 5423; that NUREG/CR 6075'. The DCH loads for Zion were is, with wamr available to the drywell, such as via estimated using models that have been benchmarked sprays, the ; 'obability of liner failure is extremely low, against the experimental results for the Zion la the absent of water, however, the same approach configuration, and uncertainties were included in the lead to the conclusion that failure would be certain. (In evaluation using the same probabilistic approach that parallel with this research effort, licensees were was used in the Mark I issue resolution. The pressure requested to evaluate, as part of their Individual Plant capacity of the Zion containment was specified in Ader 2

terms of failure frequency as a function of containment dispersal at lower pressures than those considered in pressure. Dere was essentially no overlap between the the existing test series.

load and strength; therefore, k was concluded that the probabihty of failure of the Zion containment due to C. Fuel-Coolant lateractions and Debris loads===aci*d with a DCH event is ceremely small.

Coolability A screening criterion of candmianni containment fauure

<0.01 given a HPME was used as a success criterion.

Fuel-coolant interaction (FCT) is a process by which moken fuel transfers coergy to the surrounding la May 1995. NUREG/CR4109' was issued which coolant, leading to breakup and quenching of melt with

===>==>d the DCH issue for the Surry plant using the possible formation of a coolabic debris bed or, methodology and assumptions similar to those used in akernately, the fuel-coolant interaction could take the NUREG/CR4075. Like Zion, for Surry, there was no form of ao coergetic steam explosions that could intersection of the load distributions with the challenge reactor vessel and containment integrity. In containment strength distnbution, and thus, DCH was those scenarios in which the core debris falls into the considered resolved based on consideration of reactor cavity, the areas of concern are associated with containacet loadag from DCH alone. In addition to meh concrete interactions.

evaluating the likely failure of the containment from DCH, the likelihood of HPME was also evaluated for Since the quantification of the containment failure Surry and abown to be low. This information could mode induced by in vessel steam explosion-generated have been combined with the likelihood of containment missues in the Reactor Safety Study, WASH-1400' failure froni DCH if there had been some likelihood (identified in the study as the alpha. mode failure),

that the containment loads could have challenged signi6 cant progress has been made in understanding containment integrity.

the processes and parameters that effectively limit the potential of missue-induced failure by an in vessel The most recent milestone in the resolution of steam explosion. Most recently, in June 1995, a second DCH for PWRs was the issuance, in February 1996, of Steam Explosion Review Group (SERG 2) workshop NUREG/CR4338'. Like Su.ry, the methodology was convened by USNRC to review the current developed in NUREG/CR-6075 was used to perform a understanding of the complete spectrum of FCIissues load versus strength evaluation for each of the by a panel of international experts. The first Steam Westinghouse plants with largc dry (34 plants) and Explosion Review Group (SERG 1) workshop took subatmospheric (7 plants) containments. Tbc results of place in 1985. The SERG-2 experts generally this evaluation showed that likelihood of conditional concluded that the alpha-mode failure issue was containment failure from DCH is less than 0.01 for resolved or *cssentially' resolved meaning that this each plant analyzed. Agam,like Surry, there was no mode of fauure is of very low probability and of little need to integrate the likelihood of HPME with or no sipiificance to the overall risk in a nuclear power i

containment fauure probabilities to resolve DCH for plant. NUREG 1524' was issued in August 1996 the plants analyzed, and DCH can be considered summarizing the deliberations of the experts, resolved for these plants based on containment loading only.

The SERG 2 experts noted that with the essential resolution of the alpha mode failure issue, the Resolution of the DCH issue for Westinghouse emphasis of FCI research shifted to other issues such j

plants with ice condenser containments, and all as mild quenching of core melt during non-explosive Combustion Engineering plants and all Babcock and FCI, and shock loading of lower head and ex vessel Wilcox plants is nearing completion. At this time, it structures arising from explosive localned FCI. These appears that resolution of DCH for these plants will issues are relevant with regard to assessing certain not be able to based only on a consideration of accident management strategies for operating reactors containment loads, but will have to also consider the and the adequacy of certain passive system design likelihood of HPME. Accordingly, akernate success features of advanced light water reactors.

criteria which consider both the likelihood of a HPME and candmianal containment failure are under As part of its ongoing research effort to address consideration. Finally, a limited number of additional these issues, the USNRC is participating in the melt i

experiments at SNL are being considered, in quenching experiments at the FARO facility and the conjunction with FZK in Germany, to explore debris steam explosion experiments at the KROTOS facility", tsoth at the Safety Technology Institute of Ader 3

the Joint Research Center at Ispra, Italy. In the When this molten core material is relocated into the PARO facibty, large masses (typicaDy, up to 250 kg) of lower head of the reactor pressure vessel, a molten prototypic reador mek are generated and poured into pool forms and can impose a significant heat load on a water pool of varying depths at a range of system the reador vessellower head. Post acadent pressures. De FARO tests that have been carried out examinations of the TMI-2 reactor core and lower show generaDy cuasistent mek quenching with no plenum found that approximately 19,000 kg of molten steam explosion. KROTOS esperiments conducted at material had relocated onto the lower head of the smaller scale (up to 4 kg) with prototypic melt (UO -

reactor vessel. Resuks of the OECD TMI 2 Vessel 3

ZrO ) produced so steam emplosion, even under Investigation Projed (VIP) concluded that a locahzed 3

condaions of high water subcoohag, high mek hot spot of approximately 1 meter diameter had existed superheat, or presence of a trigger.

on the lower head. De maximum temperature on the inner surface of the reactor pressure vesselin this Other aspeds of PCI are being addressed through region reached 1100'C and remained at that caperiscatal programs at the University of Wisconsin temperature for approximately 30 minutes before and at Argonne Natianallaboratory. The cooling occurred. A final report'2 on the margin-to-esperimental program at the University of Wisconsin is failure analysis was issued in March 1994.

examining the effects of melt superheat, water subcoohng, water viscosity, system pressure, Even under the combined loads of high fuel / coolant mass ratio (alternatively, volume ratio),

temperature and high primary system pressure, the and the presence of trigger on energetics, with more TMI-2 vessel did not fail. Analyses of vessel failure recent experiments concentrating on the fuel / coolant under these conditions, using state-of-the-art computer mass ratio (volume dio) effect on energetics. The codes, predicted that the vessel should have failed via experimental progru ! the Argonne National local or global creep rupture. Knowledge of in vessel Laboratory has low i a whether chemical and ex-vessel heat transfer phenomena to the lower cugmentation of d urgetics can occur in Zr-water head is needed to assess the ability of the reactor and Zr ZrO watti 4:am explosions. Results to date, pressure vessel to maintain its integrity during a severe 2

with quantities up to 1 kg of both Zr water and Zr-accident. Since completion of the TMI 2 VIP, research ZrO -water have shown limited chemical augmentation.

has been ongoing in three major areas on RPV lower 2

head integrity to understand the important in vessel De currently active experimental research on and ex-vessel heat transfer phenomena. He first area debris coolability, called the Melt Attack and involves experiments with prototypic materials and Coolability Experiments (MACE) program", was analytical investigations'under the OECD RASPLAV developed as an extension of the Advanced project" on melt pool natural convection, crust raatat== cat Experiments (ACE) program under the formation and growth, and beat flux distribution on the sponsorship of USNRC, EPRI, and other, mostly RPV lower head.

governacotal, agencies in several countries. The MACE program is intended to determine the ability of The overall objective of the RASPLAY program is water to cool prototypic ex-vessel core debris of urania-to provide analytical and experimental information that zirconia composition. Five tests, including a scoping can be used to assess whether, and under what test, were conducted under the MACE program during conditions, molten core materials can be 1992 through 1995. A simb test, M3b, was conducted cooled / retained inside a reactor pressure vessel. The in January 1997. His test was performed at a scale experimental program includes several integral more than two times larger than earlier tests. This test experiments utilizing ceramic UO /ZrO and metallic 2

2 was designed to provide information on the effect of Zr melt of varying compositions in a slice geometry scale on crust formation, stability, and debris representing the lower head of the RPV, and a number raalahiley. The need for add tional tests is an open of separate effects experiments. The first RASPIAV issue which will need to be addressed following the integral experiment was performed successfully in analysis of resuks from the M3b test.

October 1996, at the Russian Research Center, Kurchatov Institute. This experiment utilized 200 kg of D. Reactor Vessel Integrity ceramic UO /ZrO, and metallic Zr He side walls of 2

the experimental facility were heated inductively.

During the late phase of a severe accident, a Preliminary data from the experiment show that the significa=e amount of core material may relocate temperature in the melt reached 2970K and 60% to downward into the lower head of the reactor vessel.

80% of the ceramic material was meked. In addition, Ader 4

4

=

the temperstwe distribution along the vessel wall the RPV lower head, the vessel failed early as a result revealed that natural convection in the mek was of weld penetration failure following global

• dFM this being an important technical objective deformation of the vessel. Additional tests are planned of the experissent. Postaest analyses and examination with penetrations and different pressures and heat of this experiment are presently underway.

fluxes. De result of these experiments will be used to support the development and assessment of analytical ne second caperimental project is an models of RPV failure.

internationaDy funded project being conducted at FAI, Inc,. to obtain data on two==*eh==ia-s that could E. Hydrogen Combustion protect the vessel waB from overheating from relocation of molten core debris with water present in ne major concerns regarding hydrogen in LWRs the lower plenum, ne first involves the lack of are that the static or dynamic pressure loads for adhesian between mek material and the vessel wall that hydrogen combustion and detonation may pose a can create==h=n==eist ea-sad resistance between the challenge to containment integrity or to the survival or

]

melt and the wall. He second merh==ta= involves functioning of essential safety equipment. When creep of the RPV wall due to both internal pressure hydrogen combustion alone is insufficient to threaten and elevated wall temperatures. He relative growth of containment integrity, combustion may still represent a the vessel wall can form a gap between the wall and significant contribution to containment loadings when the crust, thereby allowing water in the gap to enhance considered conjunctively with other phenomena.

the cooling of the wall and the debris.

Research conducted world-wide over the past 17 ne third experimental program is under way at years has extensively investigated a number of issues the Pennsylv2=h State Univerdty to address ex vessel related to hydrogen combustion and transport during floodmg of the reactor cavity to prevent vessel failure.

severe reactor accidents. Much of the work, performed his program investigates boiling heat transfer on to experimentally investigate the design and evaluation downward facing curved surfaces. The results of this basis for reactor containment performance, focused on study include data on the critical heat flux (CHF) and global deflagrations of premixed volumes of hydrogen, the development of an analytical model that can be air and steam. Significant information exists on used to predict the spatial variation of the CHF on the hydrogen combustion to assess the possible threat to external bottom surface of a RPV during saturated and containment and safety related equipment. Some subcooled boihng conditions. Additional experiments ancillary issues remain related to a better will be conducted during 1997 to expand the data base understanding of the likelihood of various modes of and model to take into account the effect of insulation combustion at high temperature and in the presence of around the outside surface of the RPV.

large quantities of steam.

Each of the areas discussed above addresses some One area of remauung uncertainty regards high aspect of whether or not molten core material can be temperature hydrogen combustion. A joint USNRC retained in the RPV, either from in-vessel cooling with and Japanese Nuclear Power Engineering Corporation water in the lower plenum, or ex vessel flooding program is addressing this high-temperature hydrogen Another research project being conducted at Sandia combustion phenomena. The program is being carried National Laboratories involves scaled experiments on out at BNL and will build upon prior combustion RPV lower head failure. Dese experiments are research to develop a database to predict the hydrogen providag data on the strain behavior prior to creep combustion modes for the early phase of a severe ruptwe, ruptwe time, and the resulting rupture size accident in a nuclear power plant when the from creep rupture of the lower head under the containment atmosphere may contain a hydrogen combined efects of thermal and pressure loads. For mixture with large steam concentrations at elevated all the caperiments, initiation of appreciable vessel temperatures. A large number of experiments have deformation m M at vessel wall temperatures above been performed to date, and the results indicate that:

900K, and the vessel typically failed at approximately (1) the effect of elevated temperature oc hydrogen-air 1000K. He size of failure was always observed to be mixtures is to increase the sensitivity of the mixture to smaller than the heated region. For experiments with detonate, (2) increasing the steam content decreases non-uniform heat flux distributions, failure typically the sensitivity of the mixture to detonate, and (3) the occurs in the region of peak temperature and areas of Zeldovich-von Neumann Doering (ZND) model gives a slightly reduced wall thickness. With penetrations in reasonable prediction of the dominant trends. In Ader 5

~

addstion, experiments on deGagration-to-detonation and the resulting source term. USNRC research in transition (DDT) at high temperature and the effect of this area has been reDected in the MELCOR and venting on DDT also have been completed. These VICTORIA severe accident codes, that are discussed results confirm that, while elevated temperatures later in this paper, and has been reflected in the increase the likelihaad for DDT, venting decreases the update of the TID-14844" source term, which has likelihood for DDT.

been in use for three decades in connection with plant siting assessments. His update of TID-14844 was la another rescanh effort, a number of published as NUREG 1465" in February 1995. The experiments have been performed at the Russian revised source term is currently being used in the Research Center, Kurchatov Institute to investigate the assessment of the AP600 design and is being evaluated scahag phenomena in DDT and the separation criteria by both the USNRC and utilities for use for current for placing igniters. Dese experiments, at large scale reactor licensing applications.

(480 m'), have served to confirm our understanding of the DDT phenomena. In addition, the experiments on The USNRC's involvement in research in this area i

igniter placement, again, at large scale, have indicated is primarily through the participation in the PHEBUS-that for typical igniter placements (i.e., igniter located FP (fission product) program. This program is

)

above the injection source), the igniter produced a very sponsored by the Cnemissariat & L'Energie Atomique slow deGagration with a very small pressure spike, of France (CEA) and the Commission of the European Communities and is aimed at studying, under In support of the certification review of the ABB sufficiently prototypical conditions in an in pile facility, Combustion Engineering System 80+ design, those phenomena governing the transport, retention, experiments were performed at the SURTSEY facility and chemistry of fission products under severe accident ct SNL to determine hydrogen combustion behavior conditions in LWRs. Phenomena to be studied are under condations of rapidly condensing stream. In those occurrmg in the core, in the primary reactor these experiments, sprays were used to de-inert initially coolant circuit, and in the containment. The USNRC steam inerted hydrogen air mixtures. The hydrogen will be able to obtain integral experimental data to was ignited by thermal glow plugs as the mixtures further validate its analytical models for fission product became flammable. The experiments successfully transport in the reactor coolant system and showed that the thermal glow plugs were effective in containment, and for iodine chemistry in the burning the hydrogen by multiple deflagrations, with no containment. The experimental data from PHEBUS-detonations.

FP is confirsnatory in nature and will be used to confirm the conclusions reached from the USNRC's Finally, with the proposal by Westinghouse to use completed fission product research program.

passive autocatalytic recombiners (PARS) as part of the AP600 design to control combustible gases in the G. Severe Accident Codes containment followmg a design basis loss-of-coolant accident, experiments are also being performed at the Because of the difficulty in performing prototypic SURTSEY facility to verify the performare af PARS.

experiments for a variety of severe accident scenarios, These experiments are examining the startup substantial reliance must be placed on the characteristics, the hydrogen depletion rate and the development, verification, and validation of computer performance of PARS in the presence of steam. He codes for analyzing severe accident phenomena. The results will be used to develop an independent audit objective of these efforts to develop and improve the capability to evaluate the performance of PARS.

severe accident codes is to capture the understanding of the phenomena related to severe accidents to F. Source Terms provide the necessary analytical tools for experienced staff to evaluate issues related to severe accidents.

" Source Terms" refer to the magnitudes of the There is currently a two-tiered approach to radia=+ive materials released from a nuclear reactor development of severe accident codes. An integrated core to the containment atmosphere, taking into code provides the capability to evaluate the progression account the timing of the postulated releases and other of severe accident sequences from initiation through information needed to calculate off site consequences containment failure. Other codes are developed to of a hypothetical severe accident. Over the past 30 allow a more detailed anclysis of in-vessel severe years, substantsal information has been developed accident progression or containment response, as well updating our knowledge about severe LWR accidents as detailed modeling on specific phenomena such as Ader 6

nas' = product transport. The following is a brief SCDAP/RELAPS (SCDAP/REIAPS MOD 3.2) is u

overview of the major severe acadent codes, recent also planned for the Spring of 1997. In addition, an accomphshments and future plans.

international forum, similar to MCAP, is being planned to begin in May 1997 to enhance the exchange of I. MELCOR. MELCOR" has been information on the applicability, limitations and developed as an integrated computer code that can be operational experiences with SCDAP/REIAPS.

used to evaluate the progression of severe accidents from initiation through containment failure and to

3. CONTAIN. CONTAIN" is a detailed estimate severe accident source terms as well as their code for the integrated analysis of containment

, sensitivities and uncertaimeira in a variety of phenomena that provides the capability to predict the applications. Recent insprovements in the code include physical, chemical, and radiological conditions inside a radial relarme*w= of mek in the reactor core, reactor ~=ealament in the event of a severe accident.

incorporation of the larson-Miller vessel failure Recent emphasis on the development and application criterion, improvements in the modeling of the of the CONTAIN code has been to develop and scrubbing of fission product vapors through a vahdate models related to the ALWR containment suppression pool by incorporation of the latest versio's performance for both design basis and severe accident of the SPARC code into MELCOR and incorporation calculations. The CONTAIN code was modified to of models for fission product chemical reactions with model these unique ALWR safety features and was surfaces and hygroscopic aerosols. In addition, a assessed against selected tests from Westinghouse's significant effort has been made to perform calculations AP600 Passive Cooling System (PCS) Large Test cf a potential severe accident for the AP600 plant Facility. Future plans for CONTAIN will qualify the design. The AP600 plant has several design features code to replace the current design basis accident codes not found in current operating nuclear power plants that have been used by the USNRC for a number of which required new or revised models. Another years for plant licensing calculations.

byproauct of this effort was the incorporation of code enhancements which resulted in substantial

4. VICTORIA. VICTORIA"is a computer improvements in code running time.

code designed to provide detailed predictions of the fission product release from the fuel and the transport The===*anaent of MELCOR continued both by in the RCS of radionuclides and non-radioactive the USNRC and through the MELCOR Cooperative materials during core degradation. Recently the code Assessment Program. In the later case, an has been used to perform pretest analyses for the international forum has been created with membership Phebus FIT 3 and FIT 4 fission product release and from nineteen countries who are successfully transport experiments and has been used for full plant exchanging information on the applicability, limitations calculations in support of the steam generator and operational experiences of MELCOR. The next rulemaking effort. A peer review of VICTORIA will major release of the MELCOR code (MELCOR 1.8.4) be complete in early 1997 and future development is planned for the Spring of 1997.

activities will be guided by the recommendations of this peer re5iew. With completion of the VICTORIA peer

2. SCDAP/RELAPS. SCDAP/RELAP5" is review, all of the major severe accident codes will have a computer code that has the capability to perform been peer reviewed.

detailed analyses of the in-vessel progression of LWR severe accidents as well as detailed experiment CONCLUSIONS analyses. Recent accomphshments for SCDAP/REIAPS development include: (1) completion The USNRC's severe accident research program of SCDAP/REIAP5/ MOD 3.1 full plant calculations has provided and is continuing to provide valuable for support of the steam generator rulemaking effort; information for closure of severe accident issues.

J (2) conspletion of SCDAP/REIAPS/ MOD 3.1E Because of the extensive severe accident research j

rpdates and systematic assessments (the key elements efforts over the past 17 years, there is now a large body of these updates are (a) debris oxidation model of technical data and well developed analytical tools to improvements, (b) Ag-In-Cd control rod material allow for an understanding of severe accident interaction model, and (c) BWR control blade / channel phenomena and to address the severe accident issues box model improvements); and (3) completion of on existing and advanced reactors. A number of the general PWR/BWR upper plenum component model major severe accident issues have been resolved and development. The next major release of accident management strategies have been devised to Ader 7

deal with the potential severe accident scenerios in light 9.

USNRC,"A Reassessment of the Potential for an water readors. In addition, the understanding of Alpha-Mode Containment Failure and a Review of severe accident phenomena and the progression and the Current Understanding of Broader Fuel-consequences of severe accidents gained from the Coolant Interaction Issues - Second Steam severe accident research program is also fundamental Explosion Review Group Workshop,"

to the ability to move forward with risk-informed NUREG 1524, (1996),

regulation.

10. D. Magallon,1. Huhtiniemi, A. Annunziato, A.

Even though a number of the major severe Yerkess, and H. Hohmann, " Status of the accident issues have now been resolved, additional FARO /KROTOS Melt-Coolant Interactions severe accident research is expected to continue, Tests," 23rd.WRSM, October 23-25,1995, pgs.157-although at a lower level than in the past, to improve 171, NUREG/CP-0149, Vol. 2, (1996).

the technical understanding of the residual issues. In addition, results of severe accident research that is

11. B. Sehgal and B. Spencer,

• Melt Attack and being conducted world-wide will be used to continue to Coolability Experiments Program," 2nd CSNI update and validate the severe accident codes. It is SneA=L' Meetino on Molten Core D_ekig-also possible that additional severe accident issues may Concrete Interactions. April 13,1992, pgs. 345-arise in the future.

356, Kernforschungszentrum Karlsruhe, (1992).

REFERENCES

12. USNRC, TMI-2 Vessel Investigation Project Integration Report," NUREG/CR-6197, (1994).

1.

U.S. Nuclear Regulatory Commission (USNRC),

" Severe Accident Research Program Plan,"

13. T. Speis and A. Behbahani, "RASPLAV: A Unique NUREG-1365, August 1989; NUREG 1365 Rev.1, OECD/ Russian Experimental / Analytical Program (1992).

in Severe Accident / Analytical Program in Severe Accident Management / Mitigation," TOPSAFE '95.

2.

USNRC,"The Probability of Liner Failure in Mark pgs.180-187, European Nuclear Society, (1995).

I Containment," NUREG/CR-5423, (1991).

14. U.S. Atomic Energy Commission, " Calculation of 3.

USNRC,"The Probability of Mark I Containment Distance Factors for Power and Test Reactor Failure by Melt-Attack of the Liner,"

Sites," Technical Information Document (TID)-

NUREG/CR-6025, (1993).

14844, (1 % 2).

4.

USNRC, "An Integrated Structure and Scaling

15. USNRC," Accident Source Terms for Light Water methodology for Severe Accident TechnicalIssue Nuclear Power Plants," NUREG-1465, (1992).

Resolution," NUREG/CR-5809, (1991).

16. USNRC, *MELCOR Computer Code Manual,"

5.

USNRC,"The Probability of Containment Failure NUREG/CR-6119, (1995).

by Direct Containment Heating in Zion,"

NUREG/CR-6075, (1993).

17. USNRC, "SCDAP/RELAP5/ MOD 3.1,"

NURER/CR-6150, (1995).

6.

USNRC, The Probability of Containment Failure by Direct Containment Heating in Surry,"

18. USNRC," Reference Manual for the CONTAIN NUREG/CR-6109, (1995).

1.1 Code for Containment Severe Accident 7.

USNRC," Resolution of the Direct Containment Heating Issue for All Westinghouse Plants with

19. USNRC," VICTORIA: A Mechanistic Model of Large Dry Containments or Subatmospheric Radionuclide Behavior in the Reactor Coolant Containments," NUREG/CR-6338, (1996).

System Under Severe Accident Conditions,"

NUREG/CR-5545, (1992).

8.

USNRC," Reactor Safety Study; An Assessment of Accident Risks in U.S. Commercial Nuclear Power Plants," WASH-1400 (NUREG/75-014), (1975).

l 1

Ader 8