ML20029C134
| ML20029C134 | |
| Person / Time | |
|---|---|
| Site: | Zion File:ZionSolutions icon.png |
| Issue date: | 03/31/1991 |
| From: | Ginsberg T, Grimshaw C, Park C, Tutu N BROOKHAVEN NATIONAL LABORATORY, KOREA ATOMIC ENERGY RESEARCH INSTITUTE, MARGROVE CONSULTING, LTD. |
| To: | NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| References | |
| CON-FIN-A-3293 BNL-NUREG-52181, NUREG-CR-5282, NUDOCS 9103260234 | |
| Download: ML20029C134 (57) | |
Text
-
BNL-NUREG-52181 Estimation of Containmerr:
Pressure Loading Due ':0 3
Direct Containment Heating for the Zion Plant
. ETut C. K. Park, C. A. Grimshaw, T. Ginsberg i
Brookhaven National Laboratory Prepared for U.S. Nuclear Regulatory Commission i
laR 28#!; 92=32 CR-52B2 R PDR
AVAILADILITY NOTICE Avartabmty of Reference Matenals Cned en NRC Pubkataons Most documents cited in NRC pubbcations will be avaltable from one of the fonowing r.ources' 1
The NPC Puboc Document Room 2120 L Street, ffW Lower Level, Washingten. DC 20$55 2.
The Superintendent of Documents, U.S. Governmont Printing Off+co, P.O. Box 37062, Wash 6ngton, DC 20013 7082 3.
The National Technicalinformation Service, Springfield, VA 22161 Although the bsting that follows represents the majority of documents cited in NRC pub 0 cations, it is not Intended to be exhaustive.
Referenced documents avanable for inspection and copying for a fee from thr, NRC Pubuc Document Room include NRC correspondence and Internal NRC memoranda; NRC Office of Inspection and Enforcement buhethe, circulars, information notices, inspection and investigation notices; licensee Event Reports; ven, dor reports and correspondence; Comrnl'ston papers; and applicant and %censee documents and corre.
spondence.
The following documents in the te. MEG series are avaAable for purchase from the GPO Sales Program; formal NRC staff and contractor reports NRC. sponsored conference proceedings, and NRC booklets and brochures. Also available are Regulatory Guides, NRC regulations in the Code of Federal Regulations, and Nuclear Regulatory Commission Issuances.
Documents available from the Natk>nal Technical information Service include NUREG series reports and technical reports prepared by other federal agencies and reports prepared by the Atomic Energy Commis, sion, forerunner agency to the Nuclear Regulatory Commission.
Documents available from pubhc and special techn! cal libraries hclude all open literature items, such as books, Journal and periodical articles, and transactions. Federal Register notices, federal and state legista-tion, and congressional reports can usualty be obtained from these libraries.
Documents such as theses, dssertations, foreign reports and translat6ons, and non.NRC conference pro.
ceedings are available for purchase from the organization sponscring the publicatlon cited.
Shse cop!es of NRC draft reports are avaltable free, to the ecent of supply, upon written request to the Office of information Resources Management. Distribution Section, U.E Nuclear Regulatory Commiss6on.
Washington DC 20555.
Copies of industry codes and standards used in a tubstantive manner in the NRC regulatory process are mahtained at the NRC Llorary,7920 Norfolk Avenue, Bethesda, Maryland, and are available there for refer.
ence use by the pubt:c, Codes and standards are usua&y copyrighted and may be purchased from the originating organization or, if they are American Natbnal Standards, from the American National Standards institute,1430 Broadway, New York, NY 10018.
DISCLAIMER NOTICE This report was prepared as an account of work sponsorud by an agency of the United States Govemment.
Neitner the United States Govemment nor any agency ttwreof, or any of their employees, makes any warranty, expresed or implied, or assumes any legal liability of resportibility for any third party's uso, or the results of such use, of any infor mation, apparatus, product or process disclosed in this teport, or represents that its use by such third party would not intnnge privately owned rights.
i
_ s.
NUREO/CR-5282 BNLcNUREG-52181 Estimation of Containment Pressure Loading Due to Direct Containment Heating for the Zion Plant l
1 Manuscript Completed: September 1989 Date Published: March 1991 Prepared by N. K. Tutu, C. K. Parki, C. A. Grimshaw2, T. Ginsberg Brookhaven National Laboratory Upton, NY 11973 Prepared for Division of Systems Research Omce of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555 NRC FIN A3293 iPresently with Korea Atomic Energy Research Institute, Daejon, Korea.
2 Presently with Margrove Consulting Ltd., London ECIR 3 AL).
l
k ABSTRACT For some: core meltdown accident sequences in light _ water reactors-(LWRs)
- it-is possibleifor the primary system.to remain at high pressure.
Under these circumstances,Tas the molten core. debris penetrates the reactor vessel, the core-idebris would be ejected under high pressure and, subsequently, dispersed into the containment 1 atmosphere. During the process, thermal:and chemical energies are 1directly-transferred from the core debris to the containment atmosphere. This
' phenomenon has become known as direct containment heating-(DCH).
1This, report. presents the results of a series of calculations at Brookhaven NationalLLaboratory (BNL) to provide estimates of the DCH containment pressure loading;in Lthe Zion plant subject to a wide range of initial conditions and phenomenological assumptions.
The containment loading. calculations were aer-formed using< a version of the CONTAIN code 'with update modifications, watch parametrically characterize DCH phenomena (CONTAIN DCH, Version 1.10). The range-of: calculation parameters was selected to represent many of the current-uncertainties = in DCH -initial-conditions -and uncertainties in. modeling _ DCH
' phenomena.
The parameters varied in the sensitivity. study included:
primary-system pressure at vessel failure, core melt inventory,L melt and steam flow rates-through-the reactor' cavity, melt droplet size, melt trapping rate, extent of hydrogen' combustion, quenching of trapped debris, and co-dispersal of water from the reactor cavity.-
- The ch' ice of CONTAIN calculation input parameters is discussed and results o
are ' presented' for both a seven-cell and a single-cell nodalization of 'the Zion
-containment building._ The seven-cell _ calculations incorporate all the features of-the= CONTAIN-DCH model.
The -single-cell calculations are included only for comparison purposes and. represent upper-bound estimates of.DCH loads,- assuming complete mixing, adiabatic conditionsLand thermal equilibrium between the gas and-core debris.
Calculation results are_ presented and discussed, s
)
F
-[
9 i
111
1 EXECUTIVE
SUMMARY
For some core meltdown accident sequences in light water reactors (LWRs) it is possible for the primary system to remain at high pressure.
Under these circumstances, as the molten core debris penetrates the reactor vessel, the core debris would be ejected under high pressure and, subsequently, dispersed into the containment atmosphere. During the process, thermal and chemical energies are
'l directly-transferred from the core debris to the containment atmosphere.
This 1
-phenomenon _has become-known as direct containment heating (DCH).
I
-This report presents the results of series of calcu'ations-performed at Brookhaven National Laboratory (BNL) to provide estimates of the DCH containment
. pressure loading in the Zion plant subject to a wide range of initial conditions and phenomenological assumptions.
The containment loading calculations were performed using a version of the CONTAIN code with update modifications, which parametrically characterize DCH phenomena (CONTAIN DCH, Version 1.10). The range of. calculation parameters was selected to represent many of. the current uncertainties-in DCH -initial r.onditions and uncertainties in modeling DCH phenomena.
The parameters varied in the sensitivity study included:
primary system pressure at vessel failure, core melt inventory, melt and steam flow rates through the reactor cavity, melt droplet size, melt trapping rate, extent of hydrogen combustico, quenching of trapped debris,-and co-dispersal of water from the reactor cavity.
j The CONTAIN DCH: calculation. results presented in this report should be i
viewed as preliminary. _Much basic experimental data are required in. order to provide the basis for rational: selection of the basic parameters required by the CONTAIN code., Furthermore, integral experiments have not yet been performed in a multi-compartment model of the Zion containmmt using high-temperature melt simulants and using steam as the blowdown gas.
A suitable database is not yet available, therefore, for assessment of the ability of the CONTAIN DCH models to accurately. predict the DCH accident scenario.
Experiments based upon scaling analyses are planned at Sandia National Laboratory and at Argonne National 1
Laboratories using 1:10 and 1:30 linear scale models of the Zion reactor cavity and containment vessel. :These experiments will provide data to. assess the suitability of' the ' CONT 61N DCH models.
j The choice of CONTAIN calculation input parameters is discussed and results y
are presented.for both a seven-cell and a single-cell nodalization of the Zion containment -building. The seven-cell calculations incorporate all the features of the CONTAIN-DCH model and are therefore a more mechanistic representation of
.DCH loads; The single-cell-calculations were designed to provide an upper-bound estimate of DCH containment pressurization and involved assumptions of adiabatic containment, complete mixing -of chemically reactive-constituents and thermal-equilibrium'of gas and core melt components. Calculation results are presented and discussed.
')
J The calculational results (seven-cell nodalization) indicate that the DCH containment loadings predicted by the CONTAIN-code, using the input-parameters selected by BNL, are close to the estimated Zion containment capacity for initial conditions involving high primary system pressure (>7 MPa) and large partici-pating melt mass (>70% of core _me_lt inventory). These results are driven, to a L
large extent, by DCH parametric model assumptions which lead to extensive q
hydrogen generation due to metal-steam reaction in the reactor cavity and y
intermediate subcompartments. The results are, in addition, strongly inflaenced by assumptions leading to extensive burning of high-temperature hydrogen, which is predicted to enter the oxygen-rich containment dome.
The seven-cell sensitivity calculations indicate that large uncertainties exist with respect to the predicted containment pressure rise.
DCH parameters which have the strongest influence on the uncertainty in DCH pressure rise are those involving hydrogen burn conditions and those involving the influence of water on DCH interactions. Within the range of selected parameters, the seven-cell calculations lead to the prediction of increasing containment pressure rise due to the presence of water in the reactor cavity and water on the containment floor (melt quench).
A comparison of the seven-cell calculational rer.ults with the upper-bound single-cell results indicates that the mechanistic CONTAIN treatment of DCH (albeit parametric) leads to calculated DCH pressure rise magnitudes which are significantly lower than (by about 50%) those predicted by the single-cell, upper-bound calculation.
This difference in the predicted results is attrib-utable to mechanistic treatment of various (mitigating) rate-dependent heat and mass transfer processes.
The CONTAIN calculation results indicate a trend of lower DCH pressure rise with decreasing initial reactor cooling system (RCS) pressure.
This result is driven, in part, by the assumed increase of melt droplet diameter with decreasing RCS pressure and the consequent reduction of surface area available for heat and mass transfer.
The trend, however, is also believed to be the result of the influence of RCS steam inventory and the effect of the resulting steam flows on the processes of hydrogen production and convection. These calculational results are rather speculative because of the lack of definitive experimental data to support assumptions regarding the behavior of basic DCH parameters with RCS pressure.
Experience gained from the study reported here suggests that more realistic estimates can be made if the Zion containment is divided into a larger number of cells than used in the present study and if a separate momentum equation for droplets is implemented in the CONTAIN code. The importance of melt entrainment into the gas flowing in the reactor cavity suggests that the model of reactor cavity phenomena which is incorporated in the CONTAIN code include a melt entrainment rate description which is based upon appropriate experimental data.
Experimental data are needed to support development of the methods required to extrapolate the data for such quantities as fraction of melt dispersed from the reactor cavity and droplet diameter from small-scale laboratory conditions to full-scale accident conoitions.
Droplet diameter data are needed, not only to characterize the droplet size exiting the reactor cavity, but also to characterize the droplet size which results from impaction of the melt stream with the first structure downstream from the cavity exit. These data are needed as a function of driving vessel pressure and vessel breach diameter.
I vi l
i TABLE.0F-CONTENTS Page-ABSTRACT tiii EXECUTIVE-SUMARY-.. =--......................
v LIST OF FIGURES...-..............,......-..-.
viii-
- LIST OF TABLES ix-ACKNOWLEDGEMENTS -.-...........-,.......... =..
xi
-1.
INTRODUCTION,.....................-....
l=
1.1: Background........................
-1 1.2-The Direct Containment Heating Scenario 1-1.3 Outline-of Report 2
- 2. CONTAIN_ CODE MODEL OF DCH FOR THE ZION PLANT 3
4 1
- 3 2.1= Containment Model' 2.2 Initial Conditions:
Primary Coolant System and Containment 3
l 2.3-DCH Assumptionscand Parameters 4
2.3.1 Primary System Pressure -
4 i
2.3.2 Melt and Steam Blowdown Parameters 4
2.3.3 Reactor Cavity MeltL Dispersal 4
'2.3.4 Melt, Droplet Diameter 4
2.3.51 Melt Trapping :in Intermediate Subcompartments and Dome 5
l
. 2.3.6 ---Hydrogen Burn Assumptions 5
. 2.3.7 Quenching of Trapped Debris
-5 j
- 3. CALCULATION RESULTS...... -.................-
12
=4 e
3.1 ' I ntroduct i on..... -.. -.................. -
-12
~12 3 2L Results for Case A-1
.,..,..=............
- 3.3 -Sensitivity Studies 13 j
3.3. llI ntroducti on -....... -.... -...... -... :. -
13.
3.3.2 Elnitial Vessel-Pressure -
14
...- -...--,....=..--
- 3. 3. 3 Mel t I nv entory,.. -.... -....... :.......,
14 3'.3.4 Flow Rate-Through the Cavity 15 15 3.3.5' Particle Size
......_.t.
3.3.6-Trapping Rate '-
16 1
3.3.7-_- Hydrogen Burn Conditions... -.. --..... -. -...
16 3.3.8LCo-Dispersed Water and Debris Quenching-17
- 3.4f Single-Cell-Adiabatic Equilibrium Results 17-
- i 3.5 = Summary of. CONTAIN Resul ts -...... --.., -......,
18 A.'
SUMMARY
AND-RECOMMENDATIONS-.................-.
.'39 1
5.
REFERENCES 42 vii l
-l
w LIST-0F FIGURES t
Figure Page
- 2.1:
Schematic'of-Zion plant showing nodalization for CONTAIN Calculations.-....:...................
6:
3.1-Containment atmosphere pressure during DCH calculation for
' Case A-1 20
-3.2 Gas temperatures in various cells for Case A-1 21 3.3-Temperature ~ of suspended debris in various : ells for Case'A-1 22
- 3.4
- Cumulative mass of debris trapped for Case A-1 23 3.5
- Airborne debris mass in various cells for Case A-1 24 3.6 Hydrogen suspended mass by cell for Case A-1 25
- 3.7:
0xygen suspended mass by: cell for Case-A-1 26 13. 8 Effect of initial vessel pressure on peak pressure rise and melt-oxidation.
Cases A-1,-8-1, and C-1_........
27 3.9-Influence ofL hydrogen burn criteria on DCH. loading 28 3.10 Zion DCH calculation results for high RCS pressure (16.4 MPa)-initial conditions 29-
1-y l
\\;
i
-viii
LIST OF TABLES
-Table Page h
L 2.1
-Outline of Five Series of Zion Calculations 8
2.2 Molten Corium Characteristics
.....i....-...-.
8
'2.3_
Input Parameters for CONTAIN Calculations of Zion DCH Containment Loadings -...................
9 2.4 Input Parameters for CONTAIN Calculations with Quenching and 11 Co-dispersed Water 3.1-CONTAIN Results of Zion DCH Loading Calculations 31-3.2 Variation of DCH Pressure Rise with Melt Inventory....
-37 3.3 Effect of Melt Flow Rate Through the Cavity-37 3.4
--Variation of DCH Pressure Rise With the Trapping Rate 38 Parameter q
u
)
i ix
4 E
. ACKNOWLEDGEMENTS MThe authors.are grateful to Mr. D. Pyatt and Dr. P. K. Niyogi, the NRC; pr_oject manager _s.for support of:-this project,
.. The authors are especially grateful _to Mr. K. Washington, Dr. K. Bergeron, and 'Dr. D. Williams of Sandia National ~ Laboratory ~ SNL) -for their help :in Limplementing the CONTAIN DCH code _ at ' BNL and for___ th(eir many other va
. suggestions and insights'.
1TheJ staff. of. Jack Tills and: Associates, Inc. (Albuquerque,, NM), :under
-subcontract to SNL, provided the CONTAIN input deck for the Zion plant, including _
the containment-nodalization.- Mr. A. Pasculi (ENEA, Italy) was responsible for:
Ldevelopment:-of the-Zion ~ plant-description.-
.The report has benefitted from a review by Dr. W.-T. Pratt.
The authors h
- are also grateful :to-M. Raia for her excellent typing: skills,
'S e was ably
= assisted by S. FlippenLand CL Conrad.
j 4
3 r
r h
y t
i k
1 l
- xi
.,--..vv.7--.y,..L,46,.,,-_1 m r,'7,,...y
,,.-.,.m.
,,-7.---L..-,-y--
m,
,,,_o.4
.,l
l
- 1.-
INTRODUCTION 1.1 Backoround (ZRRS)jn February 1987, a preliminey version of thq Zion Risk Rebaselining Study was published for comment.
This report which provided one of the The approach adopted for the prelim-technical bases for the Draft NUREG 1150., t-elicited information generated for inary Zion study was one in which the exper the analysis of Surry (another reference plant studied for NUREG-ll50) was adapted, through scaling considerations, to provide input to the Zion study.
Since the publication of the draft Zion study, extensive modifications have been undertaken, among which was the generation of a plant-specific phenomenological data base the application of mechanistic severe accident analysis computer codes.
a As identified in the draft Zion study, a dominant contributor to the public risk was the issue of pressure loads to the containment building at the time of vessel breach, especially due to the direct containment heating phenomenon.
Direct containment heating (DCH) refers to a series of complex physical and chemical phenomena occurring inside the containment building subsequent to the ejection of molten core debris from the reactor vessel under high pressure conditions.
The objective of this report is to present and discuss the results of a series -of mechanistic DCH containment loading calculations for the Zion plant 3
performed using CONTAIN-DCH, Version 1.10 of the CONTAIN code with updates which include the description of DCH phonomenology.' The calculations were performed at Brookhaven National Laboratory (BNL) to provide support to the containment loads expert review panel established for the NUREG ll50 effort, The series of calculations reported here was designed to provide estimates of the Zion DCH loading response for a range of assumptions which are representative of the current uncertainties in DCH initial conditions and DCH phenomenology. CONTAIN input parameters were identified which were felt to have the greatest impact on the magnitude of predicted DCH p-essure ~ rise.
These parameters, which covern both DCH initial conditions and phenomenology, were varied over ranges which are consistent with major in vessel accident progression analysis uncertainties, with experimental observations of DCH phenomenology and with uncertainties in the
- phenomenology 1.2
-The-Direct Containment Heatina Scenario The following description of a postulated DCH scenario is based upon a conceptual assessment of the processes which are hypothesized to occur in the various regions of a containment building subsequent to reactor vessel failure:
Molten corium, consisting of oxides and unreacted zirconium and stainless steel, is assumed to have -accumulated within the lower -plenum of the pressure vessel. The reactor coolant system is assumed at high" pressure. The contain-ment' building atmosphere initially contains a mixture of air, steam and hydrogen released from the primary system during the in-vessel-core disruption phase of the severe accident. Water may be present in the reactor cavity and in pools or sumps within the containment building.
When the vessel fails the corium is ejected as a jet through a breach in the vessel into the reactor cavity.
In the-absence of water in the cavity, the molten cortum spreads under its own momentum across the cavity and accumulates there. Steam, shich follows the melt from the vessel, fragments the melt and carries some fraction of the melt as droplets into 1
-the subcompartments above'the cavity. Thermal and-chemicallinteractions occur between the steam and droplets within the cavity and.also within the su'acompert-
)
-ments between the cavity and thelupper containment dome. The hyPogen initially--
presentiin containment; and the: hydrogen produced ac a result of stea:a-metal-reaction is.. transported to the containment dome, where it encounters oxygen and
- may burn, depending on local-- flammability conditions.
The presence of wat_er in the reactor cavity may lead to steam explosions andLco-dispersal of water, steam, and melt from the cavity -to-the downstream -
-regions of the containment building. The presence of water pools in containment
-would lead to the possible generation _of_ large quantities of-steam if melt were.
deposited and quenched-in the-pools.
The combination of thermal and exothermic chemical interactions described above would lead to pressurization of the containment 'atmosphers.- A more de-tailed description-of the direct containment-heating phenomena in presst rized water reactors (PWRs) ccn be found in Ref. 5.
j V
- 1. 3 --
Outline of Report The remainder of. thkreport _ describes the analysis performed oy Bhl of the-
~
DCH scenario in the; Zion nuclear plant, subject to a range of postulated accident
. initial conditions.
The analysis was performed using the CONTAIN containment
- analysis-computer code, Version ~1.10.
The CONTAIN ccde ir.put parametnrs which
, control the DCH containment pressurization calculation were selected by BNL brsed -
-upon interpretation of the available experimental data base nd were supplemented by stand-alone calculations and enaineering judgements wheie necessary.
4 Section-2 describes the-various elements of the model of DCH for the lion plant ' including the nodalization of the-containrxat -building, the accident
-initial Econditions and the specific assumptions related to the-' thermal and chemical-interaction characteristics of direct containment heating. The computa-tional results-
'DCH analysis /.are described in Section 3.References are also made to the Surry wherever appropriate; Conclusions are w esented in Section-4.
i
-. i ik j
I 2.
2.
CONTAIN CODE MODEl'0F DCH FOR THE ZION PLANT 2.1 Containment'Model-Figure 2.1-is a schematic diagran' of the Zion containment building showing the nodalization of containment into six internal cells and one additional cell representing the environment.
The seven-cell nodalization was used in most calculations in the current report, The structural input data was provided by J. Tills crd Associates, it is noted that the nodalization divides the Zion plant,into a reactor vessel (or primary cooling system denoted as cell 2 in t
Figure 2.1), reactor cavity (1), steam generator room (3), outer annular region (4), lower containment doma region (5) and upper dome region (6). The environ-ment is represented as Cell 7.
In addition to the seven-cell calculations, two series -of adiabatic single cell calculations were performed which embodier conservative assumptions of thermal equilibrium and coa.plete mixin -of cher scally reactive constituents.
a Thase assumptions-provided upper-bound conuin..ent pressure rise calculation results which could be compared with results based upon the more mechanistic treatment-using the CONTAIN-DCH models.
2.2 initial Conditions:
Frimary-Cgolant System and Containment Two accident sequences, TMLB' and S20, were selected as accident scenarios which serve to define the direct containment heating initial conditlons for the primary coolant system and for the containment. These tecidents are described in detail in Reference 1 but are basically a station blackout accidegt (TMLB')
and a small break LOCA with loss of all coolant _ injection. MARCH code calcula-tions were performed to simulate the initial phase of the accident scenario. The l-TMLB0 calculation-simulates a " typical" high primary system pressure initial L
condition, while the :S2D calculations simulate low and intermediate ' initial l
primary system conditions. -The output from MARCH TMLB' calculations were pre-
. processed by J. ' Tills and Associates, Inc. and used in the CONTAIN calculations -
to initialih the primary system and containment conditions at the time of vessel breach.
The S2D calculations were treated in-a'similar manner at BNL. -These
_ pre-DCH calculations are significant only in establishing the primary system and containinent pressure, temperature and composition on the one hand, and the -
distribution of water in containment on-the_other hand, it is noted that the initial conditions compited in the above: fashion 'are.quite. siqilar. to.those specified in the context of the Containment loads' Working-Group
- Table 2.1 presents an overview description of the five series of Zicn DCH
(
calculations. A qualitative case description is given along with the proc 1rsor transientt(TMLB' or S20) from which the DCH initial conditions ~ are derived' It includes' the reactor coolant system (RCS) and containment pressure at vnssel.
breach and' the number of cells in the containment model.
Note that W,tial-primary system pressure. ranges from 3.4 toil 6.4 MPa.
Cases D and E were per-formed using a single containment cell with an adiabatic boundary (no heat lasses to structures) -in order to provide a-basis _ for: assessment of the degrt e-of conservatism in the seven-cell CONTAIN calculations. A more detailed descrig tion of case input parameters is presented in Section.2.3.
specification of the Containment Loads Working Group Standard Problem 1., rom the The mass and temperature of the molten core material are taken f Table 3
2.2 presents the total corium mass (assuming 100% of core molten) along with its assumed temperature and assumed distribution of significant species from the standpoint of containment loading.
2.3 DCH Assumptions and Parameters-CONTAIN Version 1.10 was used for these calculations.
It is a luned parameter code and therefore a complete mechanistic analysis of all DCH phenomeno could not be performed.
It was necessary, therefore, to perform calculations with a range of input parameters to study the effects of various processes and variables on the predicted containment response.
A detailed discussion of the DCH data base which forms the basis for many of the judgements regarding the choice of CONTAIN DCH parameters is found in Ref. 5.
Tables 2.3 and 2.4 present a sunnary of major initial conditions and parameter values used in the calcule-tions.
A brief discussion of the rationale behind the choice of the major CONTAIN input' parameters is presented below:
2.3.1 Primary System Pressure The A, B, and C sets of cases represent the variation in the pritrary system pressure at the onset of vessel breach. They are high, intermediate, and low RCS pressure conditiors, respectively.
2.3.2 Helt and Steam Blowdown Parameters it is assumed that the reactor vessel fails in the region containing tae-molten pool of corium at the bottom of the reactor vessel and that the failure occurs at an instrument tube penetration. A penetration is assumed to fail and to ablate as a result of heat transfer er in the vessel is taken as 0.55 m.,from the melt. The resulting hole di: met-A major fraction of the taelt is likely to be delivered to the reactor cavity prior to the beginning of the steam bic+
down from the reactor pressure vessel.
The fraction of the available mit inventory that is delivered to the reactor cavity upon failure of the reactor vessel was varied between 0.33 and 1.0 to reflect the wide range of uncertainties -
in -in-vessel melt progression and vessel failure phenomena.
The primary system gas blowdown was modeled assuming either a constant or a linearly-varying gas flowrate froa the vessel over a time period which was chosen to correspond to the calculated primary system blowdown time based upon a vessel hole size of 0.55 m, assuming isentropic choked flow through the hole, 2.3.3 Reactor Cavity Melt Olspersal Experiments conducted thus farma suggest that under high-pressure melt ejection conditions nearly all of the melt discharged into the Zion reactor caviO' will be dispersed from the cavity by the reactor vessel blowdown gases.
Thus, it was assumed that all of the melt which enters the reactor cavity from.
the pressure vessel is dispersed from the cavity.
This assumption was imple-mented in the CONTAIN code by specification of a " trapping rate" of zero in the reactor cavity (Cell 1). The assumption remains to be verified in experiments designed to assess the effect of scale on the extent of melt dispersal.
4
2.3.4 Helt Droplet Diameter The CONTAIN code assumes all the suspended melt to consist of droplets of a single size. This droplet size is then used in calculations in all regions of co'tainment:
cavity, subcompartments, and containment dome.
'lo methods are currently availabit to enable one to predict the characteristic droplet size resulting from the reactor cavity debris dispersal process. It is expected that the droplet size should depend upon the reactor vessel pressure, the vessel breech area and melt mass. No systematic data base is available for development of models or correlations of droplet size resulting from the cavity debris dispersal process.
The experiments of Ref. 10 (SPIT-18 and SPIT 19) were designed to study debris dispersal ftom the Zion reactor cavity under driving vessel pressure conditions neaa the " upper end" of the reactor operating rsnge.
The vessel pressure in these experiments at melt ejection time was 12.6 MPa. These experi-monts were carried out using 1/20 th scale models of the Zion reactor cavity.
I The mass-mean droplet diameter measured in these two experiments were 0.75 mm and y
0.43 mm, respectively. The experiment of Ref.13 (DCH 1) was carried out using an initial vessel pressure of 2.55 MPa and using a 1/10-th scale apparatus. The l
mass mean droplet diameter measured in this experiment was 0.55 mm.
Methods for extrapolatinn of the measured mass maan droplet diameters from the above experiments to D0tl accident conditions have not been developed, for the purposes of the parametric calculations presented below, it is assumed that the " base case droplet diameter for the Series B (vessel pressure 6.5 MPa and vessel breach diameter-0.55 n) calculations presented below is 0.5 mm.
it is expected that as the primary system pr2ssure at vessel failure increases, the gas velocities within the cavity increase and the size of melt droplets would de-crease. As a consequence, a narticle diameter of 0.D mm was selected for Series A as the base case astumption.
By a similar argument, for the cases in Series C, a particle diameter of 1.0 mm was selected for the base case since the primary system pressure is lower.
Additional sensitivity calculations were performed within each series using different droplet diameters increased by a factor of two.
2.3.5 L.t Trapping in Intermediate Subcompartments and Dome Although CONTAIN does not mechanistically model the trapping of melt droplets by containment structures, it uses a simple model to simulate trapping.
The suspended debris in a given cell is removed at a rate equal to the product of an input parameter A (trapping rate) and the mass of the suspended debris in that cell.
The trapped debris is assumed to be deposited on the floor of that cell. Table 2.3 shows the Wapping rates selected for various cases. The value of the base case trapping r a for each of the cells is indicated in footnote 4.
These were chosen so as to ensure that a maximum of approximately 20% of melt particles would enter the containment dome region for the base cases A 1, B 1 and C-1.
For the sensitivity calculations, trappMg fractions were varied by three orders of-magnitude.
2.3.6 Hydrogen Burn Assumptions There are two options in the CONTAIN code to burn the hydrogen within a given cell. The " Default" option burns the existing hydrogen only if the male fractions of hydrogen and oxygen exceed a certain threshold while the mole b
fraction of steam stays below a certain limit. This option is designed to take into account the steam inerting effect at low yas temperatures. H wever, since at high gas temperatures (say > 1000 K) s)ontaneous reaction between hydrogen and oxygen may take place irressective of tie steam concentration, the code allows another option to burn the 1ydrogen.
This is designated as the " Unconditional Hydrogen Burn" (UCHB) option. With this cption, hydrogen is allowed to burn as long as there is oxygen available in the cell.
However, the user must specify a burn time, t, during which the hydrogen is to be burned.
Once the 1,ydrogen b
burn occurs in a given cell for the time t, the code then stops the combustion of hydrogen for ; period equal to t beIore reinitiating the combustion of 3
hydrogen. The Wowdor. time parameter 1 was chosen to ha 1 second for the base 3
case value sr.
to approximate a continuous burn within the time frame of the s
vessel blotdown and debris dispersal as long as cxygon is available within a subcomparthent.
The sensitivity of the calculation to t was assessed with a b
case execu"ed for t, equal to 10 seconds.
2.3.7 Quenching of Trapped Debris To simulate the quenching of trapped debris that has fallen on the fic)r of the containment building (in Cells 3 and 4) in the pool of water that is likely to be present there for some accident sequences, a few more calculations were performed.
Th' input parameters for these are given in fable 2.4. This table also shows two runs for which 20,000 kg of w:ter was added to the blowdown source of melt and steam.
These :ases were run
- *imulate those accident sequences for which water is already present in tho
- actor cavity prior to vessel failure.
i 6
i
/
+
_A
1
/
a l
)@
POLAR CRANE-STEAM
\\ l OENERATORS-3 M
,e i
,,NY,.
~
t rl~f I I
4 SEAL TABLE 8
i 3
- d -
)
J gt.
IN+ CORE INSTRUMENT
..J OUiDE TUDES -
REACTOR YESSEL REACTOR CAVTTY.
Pressurizer nei,e Tank.
-=
concrete (typ) f hy tj
^#g,e,' i._
k O
-shield Watt
$h) g O
% @ }_
Sump-c-
hf, j-ymPeina Vent Duct-
/
Door J SECTION A - A Figure 2.1 Schematic of Zion plant rhowing nodalization for-CONTAIN calculations.
7 l
Table 2.1 Outline of Five Series of Zion Calculations Containment Volume:
75,850 m3 Model RCS-Containment Qualitative Accident Pressure No. of Pressure Case Series Sequence (MPa)
Cells (MPa)
Description A
TMLB' 16.4 7
0.3 High-Pressure B
S2D 6.5 7
0.33 Interm-Pressure 3
C 3.4 7
0.33 Low-Pressure D,E TMLB' 16.4 1
0.31 High-Pressure Adiabatic
- Similar initial conditions to Series B except P.CS pressure.
Table 2.2 Molten Corium Characteristics Temperature:
2533 'K 4
Total Corium Mass (100%):
1.38x105 kg 4
Mass of V0 :
9x10 kg 2
4 Mass of Zirconium:
1.1x10 kg 4
Mass of Fe:
1.87x10 kg 4
Mass of Zr0 :
1.49x10 kg 2
4 Mass of Cr and Ni: 0.33x10 kg 8
Table 2.3 input Parameters for CONTAIN Calcatations cf 21en DCH Containmant Loadings initial
! Primary H In H In Zr Fe Contaireant PSystes Particle Burn Blow g n y
2 Pressure Melt Diameter Trapping Hydrogen Time Time Containment vessel Injected injected Pressure g
Case *
(MPa)
Fraction (mm)
Rate Burn (sec)
Fle= type (sec)
(kg)
(kgt (kgt ikg)
(Wa!
3 5
A-1 in.4 1.000 0.25 4"
UCHB 1
FT, 6, 15 190.8 320.7 It000 18700 0.3 l
(CHB 1
FT t5. 15 190.8 320.7 11000 18700 0.3 A-2 16.4 1.000 0.25 At b
A-3 16.4 1.000 0.25 At UCHB 1
FT 25, 25 190.8 320.7 11000 18700 0.3 e
Default FT, 6, 15 190.8 320.7 11000 18700 0.s A-4 16.4 1.000 0.25 At A-5 16.4 1.000 0.5 A
Default F T, 6, 15 190.8 320.7 1t000 18700 0.3 6
A-6 16.4 1.000 0.5 At UCHB t
FT, 6, 15 190.8 320.7 11000 18700 0.3 UCHB 1
FT, 6, 15 190.8 320.7 11000 18700 0.3 I
A-7 16.4 1.000 0.25 10 At UCHB 1
FT, 6, 15 190.0 320.7 11000 18700 0.3
[
A-8 16.4 1.000 0.25 t/10 At tCHB 1
FT, 6, 15 190.8 213.8 7333.3 12466.7 0.3 l
l A-9 16.4 0.667 0.25 At A-10 16.4 0.333 0.25 At UCHB 1
FT,
- 6. 15 190.8 107.0 3666.7 6233.3 0.3
{
A-11 16.4 1.000 0.25 1,
UCHB 10 FT, 6, 15 190.8 320.7 11000 15700 0.3
{
A-12 16.4 1.000 0.25 At UCHB 1
FT, 6, t 5 t90.8 320.7 11000 18700 0.3 A-16 16.4 1.000 0.25 100 At UCHB 1
FT, 6, 15 190.8 320.7 11000 18700 C.3 r
UCHS I
FT,
- 6. 15 190.8 107.0 3666.7 6233.3 0.3 A-17 16.4 0.333 0.25 to At A-18 16.4 0.333 0.25 100 A UCHB t
FT, 6, 15 190.8 107.0 3666.7 6233.3 0.3 t
A-t9 16.4 0.333 0.25 1/10 At UCHB t
F 7, 6, 15 190.8 107.0 3666.7 6233.3 0.3 A-20 16.4 1.000 0.25 At
'JCHB 1
FT, 6, 15 190.8 320.7 1100('
18700 0.3 7
B-i 6.5 1.000 0.5 At UCHB 1
F T,
- 8. 11 418.8 125.0 t1000 18700 0.33 UCHB 1
FT 18, 11 418.8 125.0 11000 18700 0.33 B-2 6.5 1.000 0.5 At b
UCHB 1
FT
- 17. 17 418.8 125.0 11000 18700 0.33 B-3 6.5 1.000 0.5 At c
B-4 6.5 1.000 0.5 At Default FT,
- 8. 11 488.8 125.0 11000 18700 0.33 0
Default FT, 8, 13 418.8 125.0 11000 18700 0.33 6-5 6.5 1.000 1.0 At UCHB 1
FT, 8, i1 418.8 125.0 11000 18700 0.33 B-6 6.5 1.000 1.0 At UCHB t
FT
- 8. 11 418.8 125.0 11000 18700 C.33 B-7 6.5 1.000 0.5 10 At a
B-8 6.5 1.000 0.5 1/10 At UCHB 1
FT,
- 8. 11 418.8 125.0 11000 18700 0.33 B-9 6.5 0.6(7 0.5 At UCHB 1
FT, 8, 11 418.8 83.3 7333.3 12466.7 0.33 UCHB 1
FT, 8, 11 418.8 41.7 3666.7 6Z33.3 0.33 B-10 6.5 0.333 0.5 At UCHB 10 F T,
- 8. 11 418.8 125.0 11000 18700 0.33 B-11 6.5 1.000 0.5 At 6
Table 2.3 Input Parameters for CONTAIN Calculations of Zion DCH Containment Loadlags (Cont'd)
Primary Inittaf Systen Particle Burn Blow D H
3" "2 In Zr Fe Conteinment 7n 2
Pressure Malt Diameter Trapping Hydrogen Time Time Containment vessel Injected injected Fressere g
Case *
(Wa)
Fraction (mm)
Rate Burn (sec) Flowtype (sec)
(kg) 8kg)
(kg)
(kg)
(Wal S
8 A"
UCHB 1
FT 12.2. 12.2 418.8 40.7
!!000 18700 0.33 C-1 3.4 1.000 1.0 t
c 9
C-4 3.4 1.000 2.0 at UCH3 1
FT 12.2. 12.2 418.8 40.7 11000 18700 0.33 e
C 3.4 0.667 1.0 It UCHB 1
FT 12.2, 12.2 4!8.8 27.1 7333.3 12466.7 0.33 c
C-6 3.4 0.333 1.0 At UCHB 1
FT 12.2, 12.2 418.8 13.6 3666.7 6233.3 0.33 c
10 D-2 16.4 1.000 0.25 0
UCHB 1
FT,
- 6. 15 143.25 606.6 11000 18700 0.3t D-3 16.4 0.667 0.25 0
UCHB I
FT, 6, 15 95.5 606.6 7333.3 12466.7 0.3 t D-4 16.4 0.333 0.25 0
UCHB 1
FT 6, 15 47.75 606.6 3666.7 6233.3 0.31 a
D-5 16.4 0.006 0.25 0
UCHB t
FT, 6, 15 47.75 606.6 73.3 124.7 0.31 1I D-6 16.4 1.000 0.25 0
UCHB 1
FT, 6, 15 143.25 606.6 11000 18700 0.31 12 0-7 16.4 1.000 0.25 0
UCHB 1
FT, 6, 15 143.25 606.6 11000 18700 0.31 5
E-1 16.4 1.000 0.25 0
Default FT, 6, 15 143.25 606.6 11000 18700 0.31
- Groups A,B. C: 7-cell cases; ' D E: 1-cell adlebatic cases.
Notes:
Oebris
+ Steem + H 3
2 i
Steam + H 0'D"'S Si'** * "2 2
'FTy=
FT FT
=
=
DoorID N
?
e 2 Melt ejection tiene, gas (steers and hydrogen) blowdown time.
Particte size = 0.25 mm, terminal velocity = 3.0 m/s.
Trapping rate, 4 = 0, 0, 2.0, 2.0, 0.2, 0.2, 10, for cet ts 1 to 7. respectively (t/sec).
6 UCHB = Unconditional Hydrogen Burn.
Particle size = 0.5 mm. terminal velocity = 5.1 m/s.
7 Adiabatic (all cells).
g gear,iclesize=1.0 men, Part terminal velocity = 9.1 ets.
,c t. s i,e = 2.0 mm,,ermi na,.e, c i t, =,., m/s.
No trapping.
gg 5,000 kg = ster added to blowdown source.
12 20,000 kg water added to blowdown source.
t Table 2.4 Input Parameters for CONTAIN Calculations with Quenching and Co-Dispersed Water i
Amount of Water!
Temp. of Heat Trans.
l Case
- Input Parameters of Floor (kg)
Water (K)
Coef.
2 (W/m,g)
A-13 As for Case A-12 127061 375.33 IE6 A-13a As for Case A-1 127061 375.33 IES A-13b As for Case A-1 127061 403.0 1ES i
A-14 As for Case A-1 2 127061 375.33 1E3 A-21 As for Case A-1 127061 375.33 IE6 B-12 As for Case B-1 127061 375.33 1E3 C-2 As for Case C-1 127061 375.33 1E3 E-2 As for Case E-1 3 100000 407.0 1ES
- Groups A, B, C:
7-cell cases, Group E:
1-cell case, lin 7-cell cases, total f rom Cell 3 and Cell 4.
220,000 kg water also added to blowdown source.
Except for:
UCHB, Burn time = 1 sec, Particle size = 1.0 m.
3 Velocity = 3.0 m/s Trapping Rate = 1.
11
3.
CALCULATION RESULTS 3.1 Introduction The results of all the sensitivity calculations are presented in this section.
The parameters that were varied include:
reactor vessel pressure, droplet diametar, melt and steam flow rate through the reactor cavity, the fraction of melt inventory participating in DCH, hydrogen burn criteria, co-dispersed water and the quenching of trapped debris, in order to better appreciate the physics of the DCH phenomena, and also to understand the limitations of the CCNTAIN code, we first discuss the results for Case A-1 in some detail (Section 3.2). The sensitivity of results to various parameters is the subject of Section 3.3.
Finally, single cell CONTAIN calcula-tions are discussed in Section 3.4.
3.2 Results for Case A 1 Figure 3.1 shows the computed gas pressure in various cells of the contain-ment as a function of time for the Case A 1.
As expected, the pressure dif-ference between Cells 3, 4, 5 and 6 is very small. However, the computed pres-sure in Cell 1 reactor _. cavity)isdeemedtoohigh. The calculated pressure is based upon a si(mple model for pressure drop between adjacent cells rather than momentum equations for gas and molten core. However, this unusually hi pressure does not seem to influence the other results of calculations. gh cavity This was verified by another sensitivity calculation with different cross sectional flow The cross-sectional flow area between Cells 1 and 3 (an input parameter) areas.
was increased until the cavity pressure showed a large reduction. A comparison of the other results of computations showed virtually the same results for the two sots of calculations.
The temperatures of gas and suspended debris in various cells of the containment are shown in Figures 3.2 and 3.3, respectively.
If we look at the Cell 3 (steam gencrator room) temperatures, we note that the gas and debris reach thermal equilibrium (gas temperature = debris temperature) r,ather quickly.
It is only af ter all the debris has been ejected from the cavity that the tempera-ture of the suspended debris decreases in this cell due to mixing with the incoming cold steam from the cavity. However, as is clear from Figures 3.1 and-3.5, the-peak containment-pressure occurs just before-the completion of-debris ejection from the cavity.
Since the debris temperatures in Cell 3 are rather high ( 2000 K), and because the bulk of the debris remains within this cell (Figure __3.4), the fraction of melt thermal energy contributing to the peak containment pressure rise <is small.
Figure-3.4 shows the cumulative debris mass trapped in various cells of the containment as a function of time. The following two observations are made.
First, most of the debris is trapped in Cell 3 (steam generator room). Secondly, with the base case trap)ing rate (4 =tantaneous masses of airborne debris in 2.0) selected for this case, the debris is~ trapped fairly quic(ly.
The ins various cells are shown in Figure 3.5.
A comparison with Figure 3.4 shows that
' Note that (see Table 2.3) for this case while the debris blowdown lasts only 6 seconds, steam blowdown. lasts 15 seconds.-
12
the instantaneous airborne debris mA is only a small fraction of the total melt inventory.
The masses of hydrogen and oxygen suspended in various cells as a function of time are plotted in figures 3.6 and 3.7, respectively. As Tabic 2.3 shows, for this calculation,190.8 kg of hydrogen is initially present in the contain-ment and 320.7 kg of hyd. gen exhts in the reactor pressure vessel at the instant of vessel failure.
Since the in vessel hydrogen is released into the cavity (Cell 1) over a period of 15 seconds, the rapid rise in hydrogen mass in Cell 3 shows that a large amount of hydrogen is being produced due to metal-steam reactions within the cavity and Cell 3.
However, because the volume of Cell 3 is small, the oxygen in this cell is quickly depleted (see Figure 3.7) and not much hydrogen combustion takes place in this cell. The major reason for oxygen depletion is its transport to upper containment cells (5 and 6). In other words, as the hydrogen-steam mixture enters the lower cells u the containment, the existing atmosphere of these cells (therefore, the oxygen present) is pushed (or convected) to the upper cells. This can be seen from Figure 3.7 which shows a large inmng in the mass of oxygen suspended in Cell 6.
The periodic oscil-lations in hydrogen mass in Cells 4, 5 and 6 are due to the peculiar nature of the hydrogen combustion model in the CONTAIN code.
This model requires the specification of a " burn time" as an input parameter.
For this calculation, a burn time of one second was chosen. Therefore, as seen from the plot for Cell 5 in figure 3.6 hydrogen combustion is turned on for one second and turned of f for another second.
This explains the two second periodic oscillations (A B C in Figure 3.6) in the mass of hydrogen. At point C, as is evident from F'gure 3.7 the oxygen in Cell 5 is completely depleted, and from this point on the mass of hydrogen in Cell 5 continues to increase until the end of steam hydrogen blowdown from the reactor pressure vessel. Note that a large amount of hydrogen remains trapped and unburned in Cell 5.
Considering the very artificial division of the containment dome into Cells 5 and 6 (see figure 2.1), and the fact that Cell 6 is oxygen rich, this phenomenon of unburned hydrogen trapped in Cell 5 likely to be unrealistic because hot hydrogen plumes coming out of small{s very open-ings in the boundary between Cell 3 and Cell 5 should rise quickly into the oxygen rich upper dome region (Cell 6),
3.3 Sensitivtty Studies 3.3.1 Introduction Table 3.1 shows the results of all the CONTAIN calculations in a tabulated form. The input parameters for these calculations are given in Tables 2.3 and 2.4 (Section 2.3). The results are divided into four major groups: high initial vessel pressure cat.es (A-1 through A 21), intermediate initial vessel pressure cases (B 1 through B 12), low initial vessel pressure cases (C-1 through C 6),
and single cell adiabatic calculations (D 2 through 0 7, E-1, E-2).
For each case the table of results provides information on the extent of melt oxidation, hydrogen production and combustion, and extent of melt trapping on the contain-ment structures, in reviewing these results one should bear in mind that the estimated failure pressure for the Zion containment is obout 1.0 HPa (14). These results will be discussed in greater detail in this section.
2Small relative to the cross shtional area of Cell 5.
13
m The initial conditions at the instant of vessel 'ailure during a high pressure melt ejection accident are highly uncertain.
This does not merely reflect our inability to predict these accurately; t,ut it also reflects the fact that different accident sequences can lead to dirferent initial conditions at the instant of vessel failure. Therefore, the parameters that were varied to study the influence of initial conditions were: initial reactor vessel pressure, mass of molten core debris, and the presence or absente of co-dispersed water. Since CONTAIN is not a mechanistic three dimensional fluid dynamics code, it cannot simulate the physics of melt entrainment, flow around objects, melt droplet-structure interaction, and details of hydrogen combustion. Therefore, the code provides the user with a set of input parameters that can be changed to reflect the user's knowledge of the physics of these phenomena.
This, of course, is a crude way of simulating the physics of these phenomena.
Therefore, these parameters were varied to study the sensitivity of the peak containment pressure, to the magnitude of the parameters.
3.3.2 Initial Vessel Pressure The primary system pressure at the instant of vessel failure, Po, can influence the DCH loading in several ways. A large value of Po implies a large steam inventory in the primary system, and high steam velocities in the reactor cavity. Both of these effects can lead to extensive hydrogen generation (due to metal-steam reactions) in the reactor cavity.
Higher steam velocities in the cavity would also lead to the formation of smaller size melt droplets.
As Po increases, we also expect the rate of entrainment of melt from the reactor cavity to increi.se.
Therefore, for very large values of Po the melt is likely to be dispersed out of the cavi+, long before the blowdown of steam from the pressure ve:sel is complete.
Thase observations led to the use of different sets of particle size and the melt ejection time for different initial vessel pressures (Table 2.3).
Figure 3.8 shows the effect of initial vessel pressure on the peak pressure rise due to DCH (AP) and the fraction of Zr and fe that remains unoxidized until the instant of peak pressure.
As can be seen, the effect of initial vessel pressure is nonlinear.
The computed loading due to DCH, for example, is much more sensitive to Po at low values of Po than it is for very large values of Po.
The observed behavior is qualitatively consistent with the assumptions discussed in the previous paragraph.
3.3.3 Melt Inventory Table 3.2 shows the effect of melt inventory that is released from the reactor pressure vessel on the peak containment pressure rise.
Although, as expected, the pressure rise is indeed a monotonically increasing function of the melt bventory, the containment pressure rise is not directly proportional to the melt inventory. There are two reasons for this. First, a portion of the comput-ed pressure rise is due to the combustion of hydrogen aiready present in the containment prior to vessel f ailure, and we have assumed the mass of this hydro-gen to be independent of the melt inventory.
Secondly, the efficiency for thermal and chemical energy exchange between the melt and containment atmosphere is believed to be high when the mass of melt involved is small compared with the mass of the gas atmosphere with which the melt is interacting.
Under these conditions, thermal and chemical interaction rates would be only minimally limited by the available thermal capacity of the gas or by the available nass of gas oxidant species.
14 J
l 3.3.4 flow Rate Through the Cavity The stesm mass flow rate through the reactor cavity is a function of the hole size in the reactor pressure vessel bottom and the instantaneous pressure in the primary system. Given the hole size, the time dependent mass flow rate of steam lehving the vessel can be easily computed by assuming isentropic choked flow through the vessel hole.
Because of time constraints for these calcula-tions, we could only simulate the steam t, lowdown time for a single hole size (0.55 m). The flowrate was assumed to either vary linearly with time or to be constant with time as indicated in Table 2.3.
The rate at which the melt is entrained (by steam within the cavity) and ejected from the cavity is not easily calculated because no verified models in the required flow regimes exist.
To stt, y the influence of the melt ejection rate on the DCH loading, two runs with different values of
- melt ejection time" but with otherwise identical conditions were made.
Table 3.3 gives one seh example.
Notice that a change of melt ejection time from six seconds to fif teen seconds leads to an increase of 20'4 in the DCH pressure rice.
The difference in the maximum pressure loadings may be explained as follows: A combination of small particle size (therefore efficient molt-atmosphere energy exchange) and unconditional nydrogen burn makes the Dressure peak to occur soon af ter all the debris is ejected from the cavity. As a result, for Case A 2 the pressure peak occurs 9 seconds later than it does for Cate A 1.
Therefore, as Table 3.3 sk ws, more hydrogen can burn during this time and contribute to the DCH per From the earlier discussion in Section 3.2 (see also figure 3.4) r-
. ~st of the debris is trapped in Cell 3.
Therqfore, most of the mel ssphere energy exchange takes place in the reactor cau ty and Cell 3.
For Case A-1 the melt is ejected at a high rate and the melt-gat equilibrium temperature is relatively high (2000 K). As a result, the debris
- 1. unable to lose all of its thermal energy.
Since the debris is ejected at a much lower rate in Case A 2, the temperature of suspended debris in Cell 3 for this case drops to a lower value (1500 K). Thus more energy is released into the atmosphere and, hence, a larger peak pressure rise is computed. Similarly, for Case A 2, since steam blowdown and melt ejection times are the same, more energy exchange can take place between the melt and steam within the reactor cavity.
Hence, a larger peak pressure rise is computed.
3.3.5 Particle Size Since che time constant for thermal and chemical energy exchange between a melt droplet (particle) and the containment atmosphere is expected to decrease with decreasing particle size, the DCH pressure rise should increase with de-creasing particle size. The sensitivity of the pressure rise to changes in the particle size, on the other hand, is a function of the particle size:
For very small particle size d when the chemical reaction and thermal exchange time constants are much smal,ler than the droplet flight time within the containment, increasing the particle size to d ' would not have a noticeable effect on DCH AP because even though do' > d d'ois still small enough to allow complete energy o
exchange between the particle and the atmosphere (subject to thermodynamic limits).
However, when the particle size becomes large enough to barely allow complete energy exchange between the particle and atmosphere, increasing the particle size should have a much stronger influence on the DCH AP. The CONTAIN calculations show that when the assumed particle size is small (0.25 mm or 0.5 mm), doubling the particle size has negligible effect on the DCH pressure rise.
This can be seen by comparing the results in Table 3.1 for Cases A-1 and A-6, A-4 and A-5, B 1 and B-6, and B-4 and B-5.
The results for cases C-1 and C-4 show 15 s
I that increasing the particle site from 1.0 mm to 2.0 mm reduces the DCH pressure rise by 15.9%.
3.3.6 Trapping Rate Table 3.4 shows the variation of the calculated peak pressure rise with the assumed trapping rate.
The reactor cavity was, in all cases, assumed to be completely dispersive and no trapping was, therefore, assumed for Cell 1. Table 3.4 presents the values of trapping rate parameter assumed for the reraining cells along with the corresponding predicted containment pressure rises for the various cases. The calculation results indicate that the CONTAIN code predicts littic sensitivity of the containment pressure rise to the assumed trapping rate.
For example, reducing the trapping rate by three orders of-magnitude (Cases A-16 and A 8) changes the DCH pressure rise by only 11 percent.
There are several possible reasons for this observed computational result. One, ssibility is that 4
extensive thermal and chemical interactions between melt and steam takes place in the reactor cavity (Cell 1). Little additional interactions may be possible in the downstream subcompartment denoted here as Cell 3.
For example, if nearly l
all the unreacted metallic is oxidized in Cell 1 to form hydrogen, then the trapping rate within Cell 3 would have little influence on hydrogen generation in Cell 3, and hence, on the containment pressure rise (provided the contribution of melt thermal energy transfer to the DCH pressure rise is small (see Section 3.3.4.).
Another possibility is that the combination of the lumped parameter l
approach, which assumes complete instantaneous mixing of melt and gas, and the assumption of small particle size (more rapid molt-gas heat transfer) makes DCH loading somewhat less sensitive to the trapping rate. A more detailed cell-by-cell accounting of energy transfer (hydrogen produced, melt oxygen reaction, melt gas thermal energy transfer, melt-structure thermal energy transfer, gas-structure thermal energy transfer) is needed to better understand and to assess the importance of melt trapping in the containment building.
3.3.7 Hydrogen Burn Conditions There are two options for hydrogen combustion in the CONTAIN code.
The first, termed " Default Burn," initiates the combustion of hydrogen in a given cell if the mole fractions of hydrogen and oxygen exceed a certain threshold while the steam mole fraction stays below a certain limit. Since at sufficiently high gas temperatures spontaneous oxidation of hydrogen can be expected to take place irrespective of steam concentration, the second option, termed "Uncondi-there is oxygen avail (able in the given cell.UCHB)" allows burning of hydrogen to tional Hydrogen Burn Figura 3.9 shows the effect of hydrogen burn criteria on the DCH peak pressure r. <. This figure also presents the cumulative mass of hydrogen burned this figure the amount of hydrogen burne0)with the default option is very small.
in~the containment until the instant (t of peak pressure.
As can be seen in The UCHB option leads to combustion of a significant portion of the hydrogen inventory, and consequently to a much larger pressure rise.
However, it is important to note that not all of the available hydrogen is burned despite the-use of the UCHB option.
For example, Table 3.1 shows that for Case A-1 while B80.6 kg of hydrogen is burned by tp, 507.8 kg of hydrogen is left unburned. As i
discussed in Section 3.2 and shown in Figure 3.6,--the reason for the large amount of unburned hydrogen is the fact that it remains-trapped in oxygen-depleted Cells 3'and 5.
The mass flow rate of gases between Cells 3 and 5 and Cells 5 and 6 is 16
too low to move all the hydrogen to the oxygen-rich upper dome of the contain-ment.
3.3.8 Co dispersed Water and Debris Quenching Since some high pressure melt ejection accident sequences are likely to result in the accumulation of water on the containment floor and in the reactor cavity, a few calculations to study its effect on the containment peak pressure rise were performed. Because any water present in the reactor cavity is likely to be ejected from the cavity along with melt following the vessel failure, the presence of water in the cavity is simulated by adding water to the blowdown source of melt and steam in the CONTAIN code.
When the initial primary system pressure is high (cases in Group A), the effect of including the quenching of trapped debris on the floor of Cells 3 and 4 is to increase the peak pressure rise slightly.
The pressure rise for Case A13a(sameas4-1butwithquench)is13%higherthanitisforCaseA-1. As Williams et al.
have pointed out, there are two reasons for this increase.
First, the quenching of trapped debris in the water pool on the floor releases additional steam into the containment atmosphere.
This, of course, directly contributes to the incremental pressure rise.
Secondly, the addition of this steam into the lower regions of containment causes additional trapped hydrogen in these cells to move upwards into the oxygen rich upper dome where it can burn.
When quenching was included for cases at intermediate and low initial primary system pressure (Groups B and C), no change in the peak pressure rise was found. Thu can be seen from Table 3.1 by comparing the results for Cases B 1 and B-12 (same as B-1, but with quench), and C-1 and C 2 (same as C-1, but with quench).
The addition of co-dispersed water has a strong influence on the DCH pres-sure rise.
As can be seen from Table 3.1, the peak pressure rise for the Case A-13 (same as A-1, but with 20,000 kg of co-dispersed watr.r and quenching on floor) is about 64% higher than that for the Case A 1.
Qualitatively similar results have been reported in Ref. 6.
3.4 Sinale-Cell Adiabatic Eauilibrium Results The calculational results presented in Table 3.1 and discussed above reveal many cases in which the containment peak pressure during the DCH accident sequence approach or exceed the estimated Zion containment failure pressure of 1.0 MPa.
An effort was made, therefore, to assess the degree of conserntism implicit in the CONTAIN seven-cell calculations by providing the basis for comparison of the calculation results with those of a clearly conservative, upper bound calculation. The approach taken was to perform single cell CONTAIN calculations which assume:
(i) adiabatic containment boundary and adiabatic internal structures no gas or melt heat transfer to either containment walls or internal structures)(, (ii) complete mixing and thermal equilibration of melt and containment atmosphere, and (iii) complete mixing of all reactive constituents.
Various assumptions were made regarding the quantity of co-dispersed water and water on the containment floor and regarding the combustion criterion of avail-able hydrogen. All melt and containment atmosphere gases were allowed to come to thermal equilibrium. The results of these calculations are presented in Table 3.1 (Series D and E) and are also presented in Figures 3.10 and 3.11 along with 17
results of seven-cell calculations.
Figures 3.10 and 3.11 also present an estimate of the failure pressure for the Zion containment.
Figure 3.10 presents the DCH pressure rise as a function of participating melt inventory fraction. Cases 0 2 through D 5 represent the single-cell, high-pressure results assuming complete combustion of available hydrogen (using the unconditional hydrogen burn option and assuming no water present in containment).
These results are in the sgme range of pressure rise as predicted by Bergeron et al., using the DHEAT code.
This set of conservative single-cell calculations demonstrate that for a participating melt inventory fraction greater than 0.2, the computed peak pressure exceeds the estimated failure limit of the Zion containment. Such results point to the need for detailed, realistic, mechanistic modeling of DCH phenomena since conservative assumptions lead to predictions of containment loads which are unacceptably large.
It should be pointed out that Case D-5, where the melt fraction is extremely small (0.0067), results in an unrealistically high predicted pressure rise of 0.46 MPa. This results from the ccmbustion of hydrogen present in containment prior to pressure vessel failure and of hydrogen released from the vessel.
Without substantial melt release, however, the temperatures in containment are unlikely to support hydrogen com-bustion in the presence of inerting steam.
For Case E-1, which assumes 100% of melt inventory participating in direct containment heating, the " default hydrogen burn" option is used. As Table 3.1 shows no hydrogen is burned in this case.
This taale also shows that for this case virtually all the Zr and Fe remains unoxidized even though enough oxygen is available.
This happens because in a single cell calculation all the melt entering the containment is instantaneously mixed with all the atmosphere.
As a result, the debris temperature very quickly drops below the threshold for melt-oxygen chemical reaction.
For cases run with the option of " Unconditional Hydrogen Burn," the combustion of hydrogen releases enough energy to keep the debris temperature above the oxidation threshold. Since a virtually negligible amount-of the melt chemical energy is released to the atmosphere for Case E 1, we conclude that for this case the melt thermal energy alone is responsible for a pressure rise of 6.7 Bars.
While the Series D assumptions are arguably conservative, they are not necessarily the most conservative that can be defined. This is demonstrated with Case E-2 in which the melt debrir is allowed to fall into a pool of water after 4
equilibration with the containment atmosphere. This results in a pressure rise which is 1.3 bars greater than that of Case D-2.
3.5 Summary of CONTAIN Resu111 Figure 3.10 compares the high-pressure, single-cell, adiabatic Series D calculation results with the Series A results for the seven-cell model.
The seven cell DCH pressure rises (open squares) lie significantly below the single-cell (open circles) predictions.
It is clear that mitigating mechanisms are being cont'iered in the multi cell CONTAIN modeling of DCH phenomena.
These mitigating mechanisms include, for example, incomplete mixing, rate dependent processes, and energy transfer to structures. It is these mitigating mechanisms that lead to a considerable reduction in the computed DCH pressure rise.
l Figure 3.10 demonstrates that relatively large pressure loads are being predicted for high-pressure scenarios desnite the incorporation of mitigating I
physical mechanisms into CONTAIN DCH calculations. For large participating melt i
18
masses the computed pressure rises lead to pressures which are near the estimated failure pressure of the Zion containment building.
The uncertainties are, however, quite large, as represented (as an example) by the line between the points R and S in Figure 3.10.
Case A 13, which includes the effect of both debris quench and co dispersed water, leads to a prediction of a pressure rise above failure.
In the absence of hydrogen burn, Case A 4 leads to a pressure rise below failure.
Figure 3.11 summarizes the computed pressure rise results for the low,
intermediate-and high primary system pressure DCH initial conditions.
The results indicate a substantial influence of initial primary system pressure. At low RCS pressure (3.4 HPa or about 500 ptia) containment pressure rise is pre-dicted to be considerably below the estimated failure pressure for Zion.
As discussed in Section 3.3.2, there are several reasons for this. Since the mass of steam in the primary system is directly proportional to the initial RCS pressure, P, this influences the DCH loading in several ways.
first, as P is increased, The mass of steam available for chemical reaction within the cav,ity and downstream sub-compartment (Cell 3) increases, and therefore causes an increase in the )roduction of hydrogen.
Secondly, an increase in P causes an increase in the ) lowdown time and mass flow rate of steam flowing fr,om the RPV into the containment. Therefore, this increase in the convective flow of steam causes more hydrogen to be pushed into the oxygen rich upper donie of the contain-ment where it can burn.
Another reason for the sensitivity of the DCH loading to the initial RCS pressure is due to the fact that we have assumed the " effec-tive' melt particle diameter to be dependent upon the value of P. The physical basis for this assumption is the fact that when the initial kCS pressure is increased, the resulting steam velocity in the cavity will also increase. Since it is the high speed flow of steam that causes the breakup of melt into droplets, t
it is reasonable to assume that as the velocity of steam in the cavity is in-creased, the diameter of droplets created will be reduced. As an example, recall that the diameter of a stable droplet in a uniform flow of gas is given by the condition that Weber number equals a constant (Ref.16). As mentioned above, it is this dependence of melt particle size on P, that is partly responsible for the sensitivity of DCH loading to P,.
However, it is important to bear in mind that the choice of particle size (as a function of Po) is not the dominant reason for l
the sensitivity of DCH pressure rise (AP) to P. To see this, consider Cases B6 (P - 6.5 MPa) and C1 (P = 3.4 HPa), both of which assume the me particle di,ameter (1 mm). As Table 3.1 shows, the computed difference in DCH AP for these two cases is 0.1 MPa. First, we note that this difference in DCH AP (0.1 MPa) is caused by something other than the particle size.
Next, let us now compare this pair of cases to the base case pair B1 (P = 6.5 MPa) and C1 (P, 3.4 MPa) that are plotted in Figure 3.11 as data poiEts for melt fraction = 1.
The difference in DCH AP for these cases is 0.15 F.Pa. Therefore, only 0.06 MPa (0.16
- 0.1) of the 0.16 MPa change in the DCH AP due to change in P, could be at-tributed to the effect of particle size.
As a final remark, we note that no relevant and reliable data for the particle size as a function of initial RCS pressure is presently available, i
19
"o 9 ea O
LEGEND P-CELLI
_ _ P_ CEI,L3 _ _
o_
o P-CELL 4 ~
~~
CELL 5 p
P-CELL 6 9$-
i 9
I b9 to 8 -
e Drn o
~
(n d _
rd -.
CL 9
L-2-
x
/
9.a -
9 a,
10710 10720 10730 10740 10750 10760 TIME (SEC)
Figure 3.1 Containment atmosphere pressure during DCH calculation for Case A-1.
.c-e :
U b bb O1 4'2 113 CdC4lI4 Cd Cd CU U.U.O O Ed i I'Iii i*
4b Hlb.F H i.
c 8
C6O m
e We S
O D
0
- t~
O e
C 7
w e
. e v
c
.: a 8
t c
I se 4
O
.g7 3
OmU
's M
(4 h i
I l
u 6
C1 g.
w c a gpe
._ b a
e N
d E
i 8,
,y*
i w
O m.
N
- f3 l
s O
~
m.
- s.
~.g C
.-e bQ i
6 i
i i
i 0'00980'09880'00080'094I0'00910'09810'0001 0'094 0'009 0'098
(>0 annivusanal 21
Ii Q-3* l..v S.c Za 2
W a M 31 3:
(.D (4U;U U U Cd'C4 Es] Cd U
NI lil:1 l
AC Q40 0
l i o
O o
w 8
~~
e e
m
%s
- tw w
C o3
\\ 'e u
I..:'l; 3
e c
O o
.t2 e
i SCE U.1
[L) ?.K 2g I
a m Mb@
- t~
v o
u
,o G
m h
i 9
M
.. i**
A e
n w
O S
__D D
,)
o.
bd i
i i
i I
0'0000 0'0098 0'0008 0'0091 0'0001 0'009 0'0
(>0 sanivuaansi 22
w l
'o. c39 LEGEND T-MASSI
[
_ _ T_-M ASS _:} _,
..T-M ASS 4_.
o 6 -.
i T-MASS 5 f. MASS. 6 o.
co -
^O M
l "o
Cn d -
CO g.
n o
4-o
.g.
l
/-
q
, f i 10b0 10b0 10h40 10h50 10760 l
10710 l
TIME (SEC) l Figure 3.4 Cumulative mass of debris trapped for Case A-1.
i t
...._z_--.
I l
"o o
--< ol LEGEND D-MASSI
}
D-MASS 3 o
-]ljfhiAggf}
6-
,.} i
, D-MASS 5
,/';i D-MASS 6 l
l
~ ;
o d-l l
l*
m O
Z o
v Cn d -
- 4 2
(f)
- 1 1
2 o
+
l g
.I e
V
/
- s o
i
/
- \\
ci -
s
/
r U
I \\
/
N
[ -
~ (,,
s
~
N_
N a
d.
10710 10720 10730 10740 10750 10760 TIME (SEC) l Figure 3.5 Airborne debris mass in various cells for Case A-1.
~ M.at d:03
- C <**C < *C E 2'.E E 2 O NIN.N N N-Ed = =:= = =
A e
g C6 O
t
- 4 aC W
m O
...m u
3 o
o i
g U
w e
4 1
m
,e m
N a
O O
f l
-5 2
i l
GE NE m
3 0
M f
CA 8 l
1
~ w 4 O O
l MbM e
-6 e
C3 O
m o
s
,o' x-3
~.
LJ o
D-co C
t Q
~
(~; m(.
w-m2 O
.-e bO s
6 e
i 0*009 0*001' 0'000 0'002-0'00i 0'0 (o>0ssva 25
l nece g< <<<<
2 2l2 2 2 ON N'N N N
Wo o;o c o A
l :
ooNO m
a 3
c
-A h
O w
e W
V
,i a
i e
2
-g v
f G$
m 8.
I
% ae w
W-4 l
8E5 1
-No N,
M b'g-
- i':.
i
-g h
(
y r/
w
~-
. sq; o
5'
-O m
l l
4-I k
0*21 0'01-0'O 0'9 0*r
- O*8 0'0-01.
(D}I) SSVN C
26-
-o
1 i
l i
l I
i 1.0 10 g
i i
i i
i i
a 02-i a
DCH 6P k
I m
i gS
- 1 a
0.. g g.
- g-g e.
g
- n.
. 4 y
l o,4 I l 19 4
5.
m go Fo H
0.2 2
k E
a i
3-
't 0i 0
~ '
0-
-4 8
12 16 --
20 WITIAL PRES 8URE N VE88El Po (MPa)?
Figure 3,8 Effect of initial vessel pressure on peak pressure-rise and melt -oxidation;. - Cases A-1, B-1, and C-1.,
-11 A
J 27 1
y y
v-
,cwm, w,m.--
,-.vE+,ww%--->,e.-m.,~,,.+r-5+.
-r,e,-,-
,.w--
v v
p,We+
~
~
e.%,*-
I
l 1000 10
- r e N*
E 8
800 g
On O w 1
O H, Burn g
m 2
6 600 m
O O
n o~s c
)
I
'O 400 m
O 4
N U
's
$i N,
O m
fAW~__O z
C Default
-i o: DCH ar,with UCHB 2
H Burn 200 F
DCH Ap with Default H Burn o:
2 burned with UCHB l
=
e: Mass of H2
~
burned with Default Barn 3
p a: Mass of H2 c
N
%-4_ _ _ _( _ a. ______f g
s i
g A-1 A-6 B-1 B-6 j
A-4 A-5 B-4 B-5 l
CASE l
Ficure 3.9 Influer.ce of hydrogen burn criteria on DCH loading.
l 1
i l
I 16 O
P SINGLE CELL ADIADATIC 7
HIGH PRESSURE (16.4 MPa) 14 O D-5.D-4,D-3.D-2 e E-1 (Default H Burn) 12 0 E-2 (Quench) 2 RS a n
r y
10 S
CONTAINMENT FAILURE 0
(ESTIMATE)
(
4 4_
O 6
o g
S lLE3 c
j t
7 CELL HICH O A-10.A-9.A-1 PRESSURE A-( uench l
2 co-dispersed water) l 16.4 MPa 3 A-4 (Default H Burn) 2 t
i e
i a
e e
a e
f l
O 0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 MELT INVENTORY FRACTION Zion DCH calculation results for high RCS pressure (16.4 MPa) initial conditions.
Figure 3.10
16 14 SINGLE CELL ADIABATIC
)
HIGH PRESSURE (16.4 MPa) 12
.O D-5,D-4,D-3 D-2'
.n
_y 10 m
v CONTANMENT FALURE o.
g (ESTIMATE) 4'
-. _ - -. _. g oO' 6
o i
{f t
4
' 7 CELL J l
v c--
2 O A-10 A-9 A-1: High Pressure (16.4 MPs) a:B-10, B-9, B-1:Irrtermediate Pressure (6.5 MPa)
O
- v:C-6, C-5, C-1; Low Pressure (3.4 MPa) i i
0 0.1 0.2 0.3 0,4 0.5 0.6 0.7 0.8 0.9 1.0 i
MELT INVENTORY FRACTION -
Figure 3.11 Zion DCH calculation results for various RCS pressures.
- w __ _ -
t
o l
Table 3.1 CON 111N Results of Zion OCif Leadin9 Calculations
(
j Time,t,.
Irapped' Suspende<!
Tr6pped Suspended Tra!Ted Suspended Trapped SusreMed Frection Fractic*
,_~_
S 2
y,3, ?
Zr 2r F?
Fe 2r0 2r0 Fe0 Fe0 2 r-Fe Frax 3 to Pmai Case Wa)
(sec)
(K)
(kg)
(k g)
(ki)
(kg)
(kgf (kjf (kg)
(k g)
Unesidized Unawidfrrf l
A-1 0.99 6.10 1714 12 0
) 313 78 23554 6153 M 10 4867 0.01 0.04 l
A-2 1.13 15.32 1510 12 0
1 700 0
27668 2040 21 'M 16G 0.01 0.04 l
r
- A-3 1.04 15.82 1472 10 4
594 59 24307 1452 18S54 10 'z 0.27 0.17 l
A-4 0.78 15.00 1111 26 1
961 31 29030 654 22216 490 0.01 0.06 A-5 0.74 15.00 1084 294 9
5983 149 28726 647 15779 341 0.03 0.33 A-6 0.97 10.32 1639 196 0
4540 0
28129 1331 17078 1075 0.03 0.25 A-7 0.95 6.10 1657 83 0
2S81 59 28116 1497 19077 1131 0.02 0.16 A-8 1.01 6.35 1728 1
109 69 118 8163 21550 6496 17386 0.01 0.01 74 0.92 14.32 1403-8 0
431 0
19420 383 1512 307 0.01 0.04
[
' - 10 0.77 14.82 1038 4
0 238 0
973i 200 7551 167
'L OO 0.04 i
a-Il 1.01 11.32 1615 12 0
619 0
28489 1221 22194 955 0.01 0.04 I
w A-12 1.12 6.30 1603 11 0
621 3
23867 Sa46 18622 4711 0.01 0.03 A-13 1.43 C.IO 1930 19 0
1150 67 23042 5729 17131 5287 0.01 0.07 A-13a 1.08 6.10 1744 20 0
1028 74 24015 5685 12068 4491 0.01 0.06 A-135 1.03 6.10 1739 20 0
1024 74 23997 5670 18061 4502 0.01 0.06 A-14 1.37 A.25 2028 11 0
728 6
23209 6501 17800 5236 0.01 0.04 A-16 0.94 14.82 1563 232 0
4499 0
29409 0
18209 0
0.03 0.24 A-17 0.76 14.82 1014 32 0
1125 0
9864 0
6550 0
0.02 0.19 i
A-18 0.74 14.82 976 92 0
1755 0
9783 0
5750 0
0.03 0.29
- -19 0.79 14.8?
1057 0
0 27 0
5634 4223 4551 3407 0.01 0.01
'O 1.29 23.40 2055 11 0
605 0
29614 96 23126
'T 0.01 0.03 4s.
1.07 6.10 1723 20 0
1025 0
24124 5577 18166 4404 0.01 0.06 l
l
.)
Table 3.1 CONT 4fM Results of Zion DCH Leading Calctriatlons (Cont *d)
Trapped" Suspended Trapped Suspended Trapped Suspended Trapped suspended Fraction Fraction 5
Time, t toPmah, Tman3 Zr -
2r Fe lFe Ir0 2r0 Feo Fe0 Zr -
Fe 3
Case (MPa)
(sec)
(K)
(kg)
(kg)
(kg)
- (k g)
_{kgf (kgf (kg)
(kg)
Pmat ynor'dized Unoxidized i
B-1 0.91 8.52 1337 216
.0 4499 109 26453 2728 15785 1319 0.05 0.29 B.2,
0.94 12.02 1420 201 0-4377 21 27731 1727 16716 1380 0.03 0.25 B-3 0.90
.14.50' 1377
.18
-11 4274 150 27434 1387 16847 937 0.07 0.26 l
B-4 0.60 11.0!
862 316 22 5487 279 26069 1198 15947 633 0.04 0.31 i
8-5.
0.59 11.02 822 1534 72 14017
> 595 26430 1125 5016 223 0.16 0.78 l
B-6..
0.85 10.02 1205 1377 0
13249 48 26654 1222 6006 925 0.13 0.71 I
Be7 0.97 12.02 1330 877 0
10060 0
28539 3
11075 3
0.09 0.54 B-8 0.%
10.02-1388 30 0
745 3
14704 1074 10942 12077 0.01 0.04 B-9 0.82 12.02 1165 130 0
3057 0
19176 461 11691 372 0.03 0.25 G-10 0.72 12.02~
945 73 0
1678 0
9588 222 5663 179 0.03 0.27 I
I (a
8 11 0.93 11.02 1435 238 0
4637 0
28454 953 17269 769 0.03 0.25 l
3-12 0.91 8.35 1343 216-1-
4461 244 25918 3515 15422 2527 0.03 0.25 C-1 0.75 10.15 1062 1203 102 11940 i762 24975 2147 6152 867 0.18 0.71 C-2 0.75 10.15 1062 1203-102' 11940
- 762 pa975 2147 6152 867 0.18 0.71 C+4 0.68 10.15 888 5036 303 1647#s 1496 19823 1851 346 294 0.55 0.97 C-5 0.69 10.15' 932
' 844 549 8162 514 16625-1411 3869 586 0.19 0.72 270 4203 442 8251 713 1772 298 0.20 0.74 C-6 0.63
'10.15 780 463 I
1
_________m ea.
+.._r---
m
- - - - - ^-
~-"#-Z-*-'*-^'-
- ^ - -
'F
^-
-a
-#-.2-..
Y 8
~Me--MM
.6888888 8F um-e e e e e
e e g
gC00000 00 6
5 4
w m O, m m 00000 t f ed
& LM C0000 4A O uNe e e e e e e e e g
g000000 00 g.
m I
^O thOOO OO O th m c5 @ m th m
@m th Os (1 th m to Ca 4 w M@N MM
^
N-NN T
3 M
^
U gDPOOOOOO Og
.m-e e
6 M
g N
v
(
w o
e vC %
0 0 0) t0 0 0 OO
-p
& CD NNOmNN O
U C6 Le A to th - A A N
eNw mmm th m to 4
3 N-NN L3 M
t C'
@[k h OOOOOO OO
- V u O LP N
9 9 Lm N
O L Nw m
ad
>=
N C
h Gp 00
^
000000 a)
O t% w.as M
w so w
k N
3 e=e
%O V
M C.
000000 O to
('
- a ga, e o c3 nw w a N
"3 L
w e
>=
tK v
E
~
y e
4 C la F C00000 CO P-
@ Nm 4*
CL w
'O O
+
Q U
S e
M er*
T M
6 m
LU Pwm G
=
j 000000 00 o
u e.,
M^
N N tO O M e=e M@
gW m to e
- w N w Q@
lil w N w em Ps N to ON s=e sw,=e
- og e=4 ese e tt et w 64 9^
LA LA O O O LA O in Y U NNOQON ON
- ts. @
e e e e e e o e g., My
., LD =O N W to N to W -M~--
~-
m e
emM O
MM
'4 0. ' N w e
- A f% N
@ CO g
7 j
pr e e e e e e e e t - em e.e e e O e o -
O eo
@so N M w tn 40 h
- e N e t e e e e e e U QOOOOO bi be
-Table 3.1 CONTAlfi Results of. If on DCH toadin9 Calculations (Cont'd) -
Cue. F-p. Hy Cum. H p Cum. Hp Cem. Energy Cwn. Energy *Fractice
- i Unburned Unburned Burned l Burned Produced Produced Fros H Burn From M Burn of Melt-2 2
p (3 8J) at End (J) in Cells 612 at End at t at.End at t,'
6 at End at t H,at t, Case (kg)
(tq)
(kg),
(tc)'
(kg)
-(kg)
' x 10 7
x 10 1 5 and 6 A-1 507,8 542.3 880.6-1038.8
'7931.9 9356.9
~1.259 1.486 0.18 A-2 653.9-513.0 :
942.4 ' 1083.3 8489.8 9758.8 1.348 1.549 0.14 A-3 464.6 587.1 933.9 1009.9
-8410.7 9096.7 1.335 1.444 0.12 A-4 1603.8 1293.8 106.8 980.9 961.5 2854.3 0.135 0.453
.0.22 0.0 0.0 0.000 0.000 0.22 A-5 1406.8 1406.7 0.0 0.0
.8077.4 8818.3 1.282 1.400 ~
0.18 A-6 383.8
'401.4^
896.7 978.9 A-7 472.5 454.1 850.8 1061.3 7664.0 9560.1 1.217 1.517 0.05 A-E '
528.7 578 4 883.9 1019.0 7961.5 9179.5 1.264 1.457 0.26 A-9, 289.2
?i0.3 849.2 898.0 7649.4 8088.7 1.215 1.284 0.18 A-10 122.6 72.9 517.0 563.0 4657.6 5116.7 0.739 0.812 0.17.
A-11 554.8 542.7 943.3 1034.1.
849).3 9315.5 1.349 1.479 0.f '.
A-12 477.3 477.3.
916.6 1008.1 8258.0 9082.0 1.311 1.441 u
A-13 539.1 603.6 E?0.5 938.1 7301.0 8451.0 1.159 1.342.
l.8 A-13a '447.4 533.0.
914.9 1018.0
'8205.1 9169.1 1.303 1.456 0.20 A-13b 455.1 537.4 904.?
1012.5 8144.9 9138.9 1.293 1.451 0.20 A-14 539.5 604.0.
848.1 970.8 7638.4 8745.4 1.213 1.338 0.28 A-16 494.1 389.1 974.8 1C93.6 8781.0 9760.0 1.394 1.550 0.03 A-17 115.3 70.7 502.6
.549.5 4527.6 4940.5 0.721 0.784 0.05 A-18 110.9 67.1 490.8 5 36.3 4420.7 4827.4
-0.702 0.766 0.03-A-19 118.3 69.7 526.5
.576.4 4743.2 5192.3 0.753 0.324 0.47 A-20 665.4 650.8 917.1 931.7 6259.7 8392.7 1.311 1.332 0.20 A-21 470.4 546.6 888.4 1004.T 002.3 9050.3 1.270 e
1.437 0.20 -
l 1
l l
E n.
_y-l i
Table 3.1 CONTAIN Results of I?on CCH Loading Calculations (Cont'd) t 1
Cun. H Cum. Hg Cum. Mp Cum. H p Cum.' Energy Com. Energy Fractiod Burn of Melt i
y From H2 2(Burnat End (J) in Cells Unburned Unburned Burned Burned Produced Produced From H J) at t,'
at End' at t at End H at End at t, H2 at t, z$088 x 10 :1 5 and 6 7 2
Case (kg)
(kg)
(kg)
(kg)
(kg)
(kg) _
1.631 0.10 B-1 500.4 291.0 S23.0 1140.9 7414.3 10277.6 1.177 B-2 508.9 798.0 930.6 1141.4 8383.1 10282.7 1.331 1.632 0.10 B-3 462.1 301.2 934.1 1123.1 8413.8 10116.6 1.336 1.605 0.09 B-4 1436.6 1~32.5 8.6 212.5 77.1 1915.4 0.012 0.316 0.15 B-5 1067.1-1057.7 0.0 9.5 0.0 85.8
'O.000 0.014 0.15 B-6 347.4 158.6 692.4 892.4 6237.2 8038.4 0.990 1.276 0.11 8-7 334.2 140.7 911.8 1105.3 8213.2 9956.6 1.304 1.580 0.04 B-S 641.4 518.8 904.6 1038.4 8147.4 9352.8 1.293 1.485 0.30 B-9 329.7 152.9 754.7 931.6 6799.2 8391.9 1.079 1.332 0.11 E-10 135.3 10.9' 592.0 716.4 5333.8 6452.9 0.847 1.024 0.10 B-11 533.1
'285.2 884.1 1132.0 7963.5 10197.0 1.264 1.619 0.12 1
<sa B-12 575.3 292.0 822.9 1140.9 7413.1 10277.5 1.177 1.631 0.10 C-1 356.1 100.3 600.4 890.4 5408.4 8020.3 0.859 1.273 0.08 C-2 356.0 '
100.3 600.5 890.3 5408.6 8020.4 0.859 1.27' O.08 C-4
.130.2 26.9 489.7 626.0 4411.5 5639.5 0.700 0.8' S 0.08 C-5 209.6-23.5.
544.4 755.9 4903.1 6869.5 0.778 IJ 81 0.08 C-6 70.5.-
13.4 476.5 547.0 4292.2 4926.5 0.681 0.722 0.08 l
i
_.....y____
[
3
& " " ll qq
[
\\' C Lik l
~
+
~ '
...v.
t <..
_ ;: ~
.~
..q- } %~
~
~
Et j jM y
.js
+
3 t
~
7 1.
R
'I g
+1'
- x hTablei.lf CONTAlkResults' of Ifon'DCH Loading Calculations (Cont'd) y.
' &f){
f
..:i '
~
Cum. H Cum. H Cum. H p Cum. Mp Cum. Energ Cum. Enery 4
Unburned Burned Burned Produced Produced From H g
. 2 I'
Unburned 2(Burn From H Burn 2
H at t H at End at tg at End.
. at t,'
at End s at t J) at End (J)-
Jt Case
'2 2
(kg),
(kg)
.(kg) '
'(kg):
- (kg)
"(kg).
x$0 33' x 1033 i
D 0.003 10.003 749.8 749.8 6748.2 6748.2
- 1.073 1.073' D 0.0031
' O.0631 702.1:
702.1l 6318.9 6318.9
- 1.005 1.005 t
D 0.003' "0.003 654.3
'654.3
.5888.7:
.5883.7' O.936':
.0.936
~
D-5 0.003.
0.003 654.3 654.3 5888.7
.5888.7
. 0.936 D-6 5.975 1 0.003
.743.9 749.8 6701.0 6755.0 1.064-
'l:0.936 1.072 f
4.
i D-7 0.003-0.003 749.8 749.8-6755.0 6755.0
'I.072 :
.1.072 i
g E-1 821.1-821.1
^ 0.0 '
O.0
~O' 0 0.0 0'.000 ti.000 E-2 0.003L
-0.003 749.8 c749.8 6748.2 6748.2 1.073
- 1.073 e
a 1 Maximum pressure of gas in Cell 6.
,2 Time elapsed;L Tp, (secs) to peak" pressure. -
8
. Maximum tenperature'of. gas in' Call'5..
] Trapped mess 'offmaterials)(Ir, Zro. Fe, Fe0) is total from sti cells at' time T.
s p
'5 Suspended mass 'of materials. (Zr, Ir0, Fe; Fe0) is total from all cells at time T.
2 p
I
- Total amount'.'of-H 0, produced during H burn.
2 2
~
7 Total (tsapped and suspended) fraction of melt at end. -
For input parameters, see Table 2.3 (Section 2).
+
.=
r f
ya'.
+
9>.
.y
}
e'=[
4 y s y" m'
,*gi
'g' y % - '[
7"
"""f~4f I' 5
't=',+d y
'u
il$'y9"'Jp'dJJ'"1 p
4'i V:
gg yu'3-g g c c.
L
- - p-.[-'M
'T'i5
.{
88y'('
't-pg4 is' S q y%
y'-p
'-g$gge>'
4-g 1yqv +$
t.
9 a-=.-rw,-per'w4
,, +
'i
- =
-.s I
b
.(_
.' '- p
)
i Table-3.2 Variation'of DCH Pressure Rise With
- Melt --Inventory Case Melt Fraction AP (MPa)
E A-1 1.0 0.69=
A-9.
0.666 0.62 A-10 0.333 0.47-B 1.0-0.58 B-9 0.666 0.49 B-10 0.333 0.39 x.,
-)
LTable 3.3 - Effect of--tielt Flow Rate Through the Cavity Case A Case A-2 4
0.69 0.83-Steam Blowdown Time,-(s) 15 15-
[
~!
Melt l Ejection Time Through Cavity, (s) 6
.15-
' Time to Peak Pressure Rise,:-t, (s).
6.1:
15.32
-l p
881-942-Hydrogen _:Burnedtill;tp,'(kg)
~
- Debris Temperature in Cell' 3.During 0 t t:1 t,-'(K)
- = 2000 :
- 1500-
- P o
T 37 t
.m
9
' Table-3.4' Variation of-DCH Pressure Rise With the Trapping Rate Parameter Trapping Cum. H 2 Trapping Time Constant Burned Fraction of Fe DCH:
Particle Size Rate-in Cells 3 & 4 Unoxidized AP at tp)(kg.
by t (MPa)
Case (mm).
'(s-1)
(s) p A 0.25 A /10 5.0 883.9
-0.01 0.71 t
A-1 0.25' At 0.5 880.6 0.04 0.69 A-7 0.25 10A _
0.05 850.8 0.16 0.66 t
A-16 0.25 100At 0.005
-974.8 0.24 0.64 6-8 0.5 A /10 5.0 904.6 0.04 0.63 t
B-1.
-- 0. 5 At 0.5..
823.0 0.29 0.58 B-7 0.5
-_10 At 0.05 911,8 0.54 0.54 i
L t
38
]
. ~
m 4.
iSU'MMARY AND RECOMMENDATIONS This report presents the results of a study performed at Brookhaven National Laboratory (BNL) designed to provide estimates of the Direct Containment
' Heating-(DCH).contai_nment pressure loading in the Zion plant subject to a wide
- range of initial-conditions and phenomenological assumptions.
CONTAIN DCH,
- Version 21.10 of the CONTAIN code with update modifications to characterize DCH phenomena - was used to perform the containment loading-calculations. The range of calculation parameters-was selected-to represent many of the current uncertainties in DCH initial conditions and uncertainties in modeling DCH phenomena.
The parameters varied in the sensitivity study included:
primary system pressure at vessel failure, core melt inventory, melt and steam flow ratee through the reactor ' cavity, melt droplet size, melt trapping rate, extent of hydrogen combustion, quenching of trapped debris and co-dispersal of water from
- the reactor cavity. - The magnitude of the parameters were based upon experimental data, where available, supplemented by engineering judgement.
It is noted that the available experimental data derived from small-scale experiments and that methods for extrapolation of such data to full-scale accident conditions yet
- remain to be developed.
The CONTAIN-DCH calculation results presented in this report should-be viewed as preliminary.
Much basic experimental data are required in order to provide the basis _ for rational selection of the basic parameters required by the CONTAIN code. Furthermore, -integral experiments have not yet been performed in a multi-compartment _ model of the Zion containment using high-temperature melt simulants and using steam as the blowdown gas. A suitable database is not yet
- available, therefore, for assessment of the ability of _the CONTAIN DCH models to accurately predict the DCH accident scenario.
Experiments based upon scaling analyses are planned' at-Sandia National Laboratory and at Argonne National-Laboratories using 1:10 and 1:30 linear scale models of the Zion reactor cavity
_and containment. ' vessel. -
These-experiments will provide data to assess the suitability of _the CONTAIN DCH models.
The choice of CONTAIN calculation input parameters is discussed and results are presented for both a seven-cell and a single-cell nodalization of the Zion containment building. The seven-cell calculations incorporate all the features of the CONTAIN DCH model. The single-cell calculations were designed to provide an upper-bound estimate of DCH containment pressurization and involved assump-tions _off adiabatic containment, complete mixing of chemically reactive con-stituents and-thermal-equilibrium of gas and core melt components. Calculation results are presented and discussed.
~
The calculational results (seven-cell nodalization) indicate that the DCH containment loadings. predicted by the CONTAIN code, using the input parameters
' selected by BNL, are close to.the estimated Zion containment _ capacity-for initial conditions involving high primary. system pressure (>7 MPa) and large partici-pating melt _ mass (>70%.of core melt inventory). - These results are driven, to_ a large extent, :by1DCH. parametric model assumptions which lead to extensive hydrogen ? generation due to L metal-steam reaction-in the reactor cavity and
- intermediate subcompartments. The results are, in addition, strongly influenced
. by assumptions _ leading to extensive burning of high-temperature hydrogen,_which is predicted-to enter the oxygen-rich. containment dome.'
The seven-cell sensitivity calculations indicate that large uncertainties exist with respect to the predicted containment pressure rise.
DCH parameters 39
l which have the strongest influence on the uncertainty in DCH pressure rise are those involving hydrogen burn conditions and those involving the influence of water on DCH interactions. Within the range of selected parameters, the seven-cell calculations lead to the prediction of increasing containment pressure rise due to the presence of water in the reactor cavity and water on the containment floor (melt quench).
A comparison of the seven cell calculationai results with the upper bound single-cell results indicates that the mechanistic CONTAIN treatment of DCH (albeit parametric) leads to calculated DCH pre'sure rise magnitudes which are significantly lower than (by about 50%) those predicted by the single-cell, upper-bound calculation.
This differenca in 'he predicted results is attrib-utable to mechanistic treatment of various (mi.1 gating) rate-dependent heat and mass transfer processes.
The CONTAIN calculation ;esults indicate a trend of lower DCH pressure rise with decreasing initial reactor cooling system (RCS) pressure.
This result is driven, in part, by the assumed increase of melt droplet diameter with decreasing RCS pressure and the consequent reduction of surface area available for heat and mass transfer.
The trend, however, is also believed to be the result of the influence of RCS steam inventory and the effect of the resulting steam flows on the processes of hydrogen production and convection. These calculational results are rather speculative because of the lack of definitive experimental data to support assumptions regarding the behavior of basic DCH parameters with RCS pressure.
Experience gained from the study reported here suggests that more realistic estimates can be made if the Zion containment is divided into a larger number of cells than used in the present study and if a separate momentum equation for droplets is implemented in the CONTAIN code. The importance of melt entrainment into the gas flowing in the reactor cavity suggests that the model of reactor cavity phenomena which is incorporated in the CONTAIN code include a melt entrainment rate description which is based upon appropriate experimental data.
The CONTAIN code provides the user with an " Unconditional Hydrogen Burn" (UCHB) model-as an option. This ontion allows the hydrogen in a cell to burn as long as oxygen is available, irrespective of the steam concentration. The reason for including this option is the expectation that at high gas temperatures (say
> 1000 K) spontaneous reaction between hydrogen and oxygen would take place N spite of the presence of inerting steam. However, the user is aqi provided wlon a choice of an input " trigger taperature" which must be exceeded before the combustion of hydrogen is allowed to take place. As a result, as our calcula-tions have demonstrated, the UCHB option results in the physically unrealistic combustion of hydrogen even when the gas temperatures are too low to overcome the steam inerting effect.
Therefore, until a better hydrogen combustion model becomes available, the UCHB option.chould be modified to include a user-specified
" trigger temperature" for hydrogen combustion.
In other words, the hydrogen would be allowed to burn only if the gas temperature exceeds the " trigger temperature."
Experimental data are needed to support development of the methods required to extrapolate the data for such quantities as fraction of melt dispersed from the reactor cavity and droplet diameter from small-scale laboratory conditions to full-scale accident conditions. Droplet diameter data are needed, not only to characterize the droplet size exiting the reactor cavity, but also to charac-40 I
l s
terize, droplet; size which results from impaction of the melt stream with the first' structure'~ downstream from the cavity exit.
These data are needed as a'
' function:of driving vessel pressure and-vessel breach diameter.
+
f-h 3
r 41
u 5.
REFERENCES I
1 Khatib-Rahbar, M.,_ et al.,-" Evaluation of Severe Accident Risks and Poten-tial for Risk Reduction:
Zion Power Plant," Brookhaven National.-Labora-tory, Draft NUREG/CR-4551, Vol. 6 (February 1987).
- 2. - U.S. _ Nuclear Regulatory Commission, " Reactor Risk Reference-Document,"
DraftNUREG-1150-(February 1987)._
3.
Bergeron,- K.D., et al., " User's Manual for CONTAIN 1.0, A Computer Code for Severe ' Nuclear Reactor-Accident Containment Analysis," -NUREG/CR 4085,
-SAND 84-1204, Sandia-National-Laboratories: (May 1985);
4.-
Williams,'D.C.. et al., " Impact of Chemical Phenomena in Direct Containment I
Heating," Presented _at American Chemical Society Severe Accident Chemistry l
Symposium,--Anaheim, CA (September 1986).
5.
Ginsberg, T. and N.K. Tutu, " Progress in Understanding of Direct Contain-ment Heating Phenomena in Pressurized Light Water Reactors," Proceedings of Third-International Topical Meeting on Nuclear Power Plant Thermal i
Hydraulics and Operations, Seoul, Korea (November 1988).
6.
Williams, -D.C.,
et al., " Containment Loads Due to Direct Containment-Heating and Associated Hydrogen Behavior:- Analysis and Calculations with4 the CONTAIN Code," NUREG/CR-4896, SAND 87-0633, Sandia National Laboratories -
1
-(May1987);
7.
Wooton,- R.O.,- et ' al., ~ " MARCH 2 Code Description - and User's Manual," - Bat-telle Columbus Laboratories-NUREG/CR-3988, BMI-2115 (August 1984).
-8.
,USNRC, " Estimates of Early Containment _ Loads from Core Melt - Accidents,"
NUREG-1079 (January 1986).
-9.-
.Pilch, M. and W.W. Tarbell, "High-Pressure Ejection of_ Melt from a Reactor Pressurevessel-TheDischargePhase,"NUREG/CR-4384, SAND 85-0012, _Sandia National. Laboratories-(September 1985).
10.-
- Tarbell, W.W.,
et al., " Pressurized Melt Ejection Into - Scaled - Reactor
]
. Cavities," NUREG/CR-4512,--SAND 86-0153, Sandia National Laboratories (August 1986).
- 11. - Spencer, B.W.,
et al., "Sweepout-Thresholds nine P.eactor Cavity-Interac--
tions," ANL/ LWR /SAF 82-1, Argonne National: Laboratory (April 1982).-
H
- 12.. Tutu,.N.K., et al., " Debris Dispersal from Reactor Cavities Ouring High-i h
Pressure Melt Ejection Accident Scenarios," NUREG/CR-5146, BNL-NUREG-52147,-
Brookhaven National Laboratory (July-1988).
- 13. /Tarbell, W.-.W; et al., "Results from the DCH-1 Experiment," Sandia National Laboratories.NUREG/CR-4871z(June-1987).
- 14. : Containment; Loads Working Group, " Estimates of Containment Loacs from Core Melt Accidents," U.S. Nuclear Regulatory Commission, NUREC-1079 (December 1985).
42 1
__a
' 15. = Bergeron, K.D. t and D.C. - Williams, "CONTAIN Calculations of Containment
-Loading ^of Dry PWRs," Nucl. Ena; Des.,-1Q,.153-159 (1985).
16.
Pilch, M., C.A.= Erdman, and A.B. Reynolds, " Acceleration Induced Fragmenta-tion.of L.iquid. Drops," NUREG/CR-2247 (August 1981).
a
-l j
- l l
1 4
'i 4
43
v.s.NuctsAR R ovoAtoav eweissioN
- i. g ag g a g g,,o,u us
~
E" 'E BIBLIOGRAPHIC' DATA SHEET NUREG/CR-5282
.3,,,,,,w,.,.a,,,,,,,,,,,
BNL-NUREG-52181
- 2. ritLa ANo sustiits Estimation of Containment Pressure Loading Due to Direct s
oar: Rt* ort rueuswt o Containn.ent Heating for the Zion Plant
[
e naa March 1991
- s. AUTHORt5; 6.1YPE of REPORT N. K. Tutu, C. K. Park, C. A. Grimsha/, and T. Ginsberg Technical i
L P E R100 CoV E R E D,Jaesw.-. 0 cess 1Presendy with Korea Atomic Energy Research Institute. Dae;on, Korea.
- Pretently with Margrove Consulting. Ltd., lendon ECIR 3AD.
12I87 ~ 9/89 gny go,A,Niz At ion - N Au a Aho ao o a tss,,' aac.,- a o- -. ca= - aw. n aan a-= c.~~. - a'-'e -e==~m ***
- s. P Brookhaven National Laboratory Upton,- NY 11973
- s. gogngoyc ANiz AiioN - Nasa ANo AcoR ess,,, ac.. 3
,n~~~m,+
aae o-.- o". - a.,-a. n a-- a-- c.--
- Division of Systems Research Office of Nuclear Regulatory Research U. S. Nuclear Regulatory Commission Washinoton. -DC - 20555 10, sVPPLEM4NTARY NOTis ll. A8siRACI < Joe.we w mus This' report presents the results of a series of calculations at Brookhaven LHational Laboratory (BNL) to provide estimates of the DCH containment pressure
-loading in the Zion plant subject to a wide range of initial conditions and phenomenological assumptions. 'The containment loading calculations were-performed using a version of the-CONTAIN code with update modifications,-which parametricallv characterize DCH phenomena (CONTAIN-DCH, Version 1.10). The range of calculation parameters was selected to represent many of the current uncertainties.ii nCH initial condjtions and uncertainties in modeling DCH phenomena. The pt 6. ars i
varied in the sensitivity study included: primary system pressure at vessel failure,
' core melt inventory, melt and steam flow rates through the reactor cavity, melt droplet size, melt trapping rate, extent of hydrogen combustion, quenching of trapped debris, and co-dispersal of water from reactor cavity. The choice of CONTAIN calculation. input parameters is discussed and results are presented for both a seven-cell and a single-cell nodalization of the Zion containment building, a iu v wonoSotsca:erons,t...,,.
,,., ~.,.m 3
...u,,,i.., %,,,
Zion-l' Reactor--containment systems; Zion-2 Reactor--containment Unlimited systems; meltdown -C Codes; containment--pressuriz 'ng; reactor
.. ucw.us.c u...,*
,v~
accidents; reactor core. disruption; pressure ves!,ls--simalation; failures; chemical reactions; hydrogen--combustion; mathemacical Unclassi fied models; heat transfer; hydraulics; computer codes; corium; radioactive aerosols; fission product release; pressure measurementi Unclassi fied PWR type reactors; US NRC; BNL; dispersions i.,. Nuyisin ue,At.e s 14 PMICt h1C eo.eu Jn tup
l 1
i 1
J i
t' l.. '
I t-l l
i l
l l
i l-I.
I THIS DOCUMEPT WAS PRINTED USING RECYCLED PAPER.
UNITED STATES-isnc,u.ouan. ctass mi L
. NUCLEAR REGULATORY COMMISSION Po5$3fe'"'*
f 6
WASHINGTONi D.C. 20555
,g o..,
g e
OFFICIAL BUSINESS 7
PENALTY FOR PHIVATE USE,4300 f
. t-P 120555139n31 i
1".o., ' A ' V l * *, '
US Nsc-oAjy
'IY FCIA yJt$t7_;7 ION 5 SVCS
'T g, a P j " - y ij a. g o.';3 db H}NOION
- r-d'
, 3 ;) m s
C:
4 5:
c t
na 9C.
g
> fi I5' W *i C
ti E
Ql d;
e, O[
m t' c r'
. F.i !
d; bl 9 o:
~4 :
- v I 4
i I
,4 E
..c E'
l l
8 e
G
_ _ _ _ _ _ _. _ _ _ _ _. _ _ _ _ _ _ _ _. _ _ _