NUREG/CR-2142, Research Info Ltr 121:submits Results of Research Re CORCON-MOD1,improved Computer Model for Interaction Behavior of Molten Core Matls W/Concrete.Title Page & Refs for NUREG/CR-2142 Encl
| ML20004D102 | |
| Person / Time | |
|---|---|
| Issue date: | 05/27/1981 |
| From: | Minogue R NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| CON-FIN-A-1019, RTR-NUREG-CR-2142 RIL-121, NUDOCS 8106080375 | |
| Download: ML20004D102 (9) | |
Text
- _.
8(
4,
'o UNITED STATES
'g NUCLEAR REGULATORY COMMISSION o
nE WASHINGT ON, D. C. 20555
}
bck MAY '2 7 1981 ffh h.
{
t JUfl 9 4 1981 u
)
y 7.
E&,%r MEMORANDlM FOR: Harold R. Denton, Director
~
Office of Nuclear Reactor Regulation M y g.?
FROM:
Robert B. Minogue, Director Office of Nuclear Regulatory Research
SUBJECT:
RESEARCH INFOR1ATION LETTER 121, CORCON-MOD 1 1
AN IMPROVED COMPUTER MODEL FOR THE INTERACTION BEHAVIOR' j
OF MOLTEN CORE MATERIALS WITH CONCRETE Introduction i
This memorandum transmits the results of completed research on the interaction of molten core materials with reactor cavity concrete.
Specifically, v51s letter describes the advanced melt / concrete interactions i
code - CORCON-MODl.
The primary objective of the molten fuel anteractions program, which was j
initiated in 1975 at Sandia Laboratories, is to investigate the physical phenomena associated with interactions of a molten core with the contain-ment concrete basemat followins reactor vessel melt-through during postula'.ed core melt accidents.
Specific objectives are to detemine the ra ce of penetration of the melt through the concrete and to characterize the rate of production of steam and noncondensible gases generated durirg the interaction. These phenomena are important to the assessment i
of core melt accident risks since they impact directly on the timing and 1
ndt of containment failure. Melt penetration rates through the concrete detennine the timing of containment fatiure via basemat melt-through.
The steam and noncondensible gases produced during the interaction are significant contributors to containment overpressure failures and to containment failure modes resulting from the combustion of flammable gases (H, CC, hydrocarbons).
2 l
The CCRCON code embodies.he state-of-technology of our understanding of melt / concrete interacyops.
The CORCON code improves upon a preliminary conputer model INTER geveloped previously at Sandia and has similar capabilities to the WECHSL code being developed at the KfK Laboratory in the Federal Republic of Gennany.
The CORCON code provides the Nuclear Regulatory Commission (NRC) with the best available tool for evaluating 1
the melt / concrete interaction phase of postulated ~ core melt accidents.
1 i
810 6 08 0 W
I u,
W 2 71981 Harold R. Denton The code can be used in the following applications:
(1) Sever.! accident risk and consequence studies, (2) Evaluation of plant specific severe accident risks (e.g. Zion /
Indian Point Study),
(3) Evaluation of the performance of ESFs and proposed mitigation features during severe accidents, and (4) Develorment of rules for Degraded Core Cooling, Minimum Engineered Safeguards, etc.
This research program is being conducted oy Sandia National Laboratories (Dr. M. Bennan, Program Manager) under the ciirection of the Severe Accident Assessment Branch, Office of Nuclear Regulatory Research (RES). Enclosure 1 is a draft user's manual for the CORCON-MOD 1 code. The code has been l
set up and is operational on the Brookhaven National Laboratory (BNL) computer system and can be remotely accessed from NRC headquarters. contains the necessary information for accessing the code.
I l
Code Description The CORCON code describes the physical phenomena associated with the interaction of a molten core with the reactor cavity concrete basemat under pcstulated core meltdown accident conditions. The major physical t
systems which the CORCON code models are the concrete, the molten pool, and the gas atmosphere and surroundings directly above the pool.
Analyti-l cal models are provided for the important physical and chemical interaction phenomena including: melt / concrete heat transfer, concrete decomposition and ablation, growth of the cavity, heat transfer within the pool and from l
the pool surface to the atmosphere and surroundings, gas evolution from the decomposing concrete and melt / gas chemical reactions, and decay heat-9t: aeration within the melt. The code can be applied to the ar,alysis of both reactor accident phenomena and prototyp1c interaction experiments.
l An indepth discussion of the physical models in the CORCON-M001 code is contained in Enclosure 1.
The following section briefly describes the principal code features.
The code allows the molten pool to exist as layers of immiscible phases.
Currently, the code treats up to three layers; a heavy oxide layer, a r
metallic layer, and a light oxide (slag) layer. The coding structure i
c l
MAY 2 71981 Harold R. Denton will allow for up to a total of six layers including a coolant layer above the melt (not currently modeled in CORCON-MOD 1).
Relative layer densities detennine their vertical orientation (i.e. layers can flip during a problem calculation).
Time-dependent concrete ablation and growth of the molten pool cavity is modeled by a routige developed by ACUREX/Aerothenn Corpore tion under contract to Sandia.
The ablation and cavity shape change models assume steady-state, one-dimensional concrete ablation coupled to a two-dimensional axisymetric shape change procedure. The shape change procedure employed in the CORCON code is based on a method that was first developed for modeling the shape change of ablating reentry vehicle nosetips.
The thenaal dehydration and decomposition of the concrete produces copious quantities of gases (principally water vapor and carbon dioxide).
The large amounts of gas released by the decomposing concrete exert a controlling effect on many physical processes governing the interaction behavior. For example, heat transfer between the molten pool and the concrete, and between layers within the pool, is modeled by applicable gas film models and natural and gas-driven convection models respectively.
The concrete decomposition gases are assumed either to escape from the l
melt cavity region along the gas film between the melt and the concrete, or to enter the melt and to rise as bubbles through the melt.
In passing through the various molten layers the gases are assumed to come 1nto
~
thennal equilibrium with each layer.
In the metallic layer the team and C0 are reduced by the metallic elements present to the combastible 7
species H,3 and CO.
These rise through the remainder of the pool and are released to the atmosphere.
Heat is transported away from the melt at the upper surface of the melt by both convection and radihtion to surrounding structures. The current j
radiation model in CORCON-M001 assumes unifonn temperature gray body radiation with a shape factor for two infinite parallel plates.
Energy sources ( ad sinks) within the molten pool itself include the heats of chemical reactions and radionuclide decay heat.
The CORCON-MODI code is written in FORTRAN-IV and requires approximately 105 K of core to run on the CDC 7600.
l Code Application and Limitations i
The CORCON-MODI code represents the state-of-technology of our understanding of the melt / concrete interaction phenomena.
However, the following limitations should be understood and accounted for by users of the code.
The CORCON-MODI code is currently applicable only to the high temperature phase of the interaction phenomena when the melt is sufficiently hot
n MAY 2 71931 Harold R. Denton '
that it is entirely liquid and is eroding the congrete at a rapid rate.
This limitation is critical. Recent calculations indicate that during some core melt accident sequences the melt may begin to freeze prior to complete penetration of the basemat.
Since the solidification temperature of the molten core materials is above the melting temperature of most concretes, penetration of the concrete by the hot solid core debris will cantinue. CORCON-MOD 1 does not model this latter phase of the interaction and should not be applied to this phase.
The CORCON-MODI code cannot be applied to the interaction of molten core.
debris with most core retention materials. This restriction results from the many code models that assume copious gas flow through the mol ten pool.
Experimental evidence indicates that most candidate core retention materials (e. g. magnesium oxide, high alumina cement, firebrick, etc.) produce much less gas than concrete during interaction with molten core materials.
l The current version of th' ^0RCON code does not include models for an I
overlying liquid coolant layer.
The thenno-chemical model for gas phase reactions (oxidation,.hydrogena-l tion, etc.) in the atmosphere directly above the melt has not been l
incorporated into the current code version.
The heat transfer models from the pool surface to the atmosphere are approximate (e.g. constant convection heat transfer co-efficient, incorrect fonn factor for radiation heat transfer).
The mass and energy additions to the pool caused by melting and ablation of materials above the pool which may fall into the pool are not modeled.
~
Oxidation of the metallic phase of the melt by gases which escape from the cavity through the gas film between the melt and the concrete is not i
modeled. These and other limitations and uncertainties are discussed in more detail in Enclosure 1.
4 Recommendations (1) NRC staff and NRC consultants use the CORCON-MOD 1 code for future analysis of the interaction of molten core debris with concrete and that the use of the INTER code be discontinued.
l Future and Related Research 4
l Current plans are to issue an improved version of the CORCON code (MOD 2) near the middle of calendar year 1982.
In this version of the code every effort will be made to include at least rudimentary models for l
..--._-.~,. --__-_ _
MAY 2 71981 Harold R. Denton 5-concrete attack by hot solidified debris.
Efforts to model the initial refreezing and resolidification phase of the interaction will be delayed until 1982. The code development effort in these areas is chiefly limited by the lack of an experimental basis for the models.
Additionally, in MOD 2 of the code we plan to add or improve models which will eliminate or relax some of the more important limitations identified in the preceding section.
Planned sensitivity studies with the code will aid in identifying which models require immediate improvement and which model improvements can be delayed.
Melt / concrete experimental programs are planned both at Sandia (Large Melt Facility) and at the German KfK Laboratory (Beta Facility) to provide data needed for code development and to aid in validating the code.
Efforts are currently underway to evaluate the best method for interfacing the CORCON-MOD 1 code with the large meltdown accident systems code:
(e.g. CONTAIN, MARCH).
For further information on the CORCON-MOD 1 code, on its use, or on future CORCON code development plans, please contact Richard Sherry of my staff (427-4329).
b Robert B. Minogue, Director Office of Nuclear Regulatory Research
Enclosures:
(1) Draft Sandia NUREG Report, "CORCON-MODl: An Improved Model for Molten-Core / Concrete Interactions,: SAND 80-2415, by J. Muir, R. Cole, M. Corradini, and M. Ellis (2)
Information fnr CORCON-MOD 1 use on the BNL computer system.
cc: See next page e
E
/
MAY 2 71931 Harold R. Denton cc w/ enclosure:
G. Knighton, NRR A. Marchese, NRR J. Carter, NRR R. Denning, SCL cc w/o enclosure:
T. E. Murley, NRR D. Ross, RES M. Silberberg, RES C. Kelber, RES J. Meyer, NRR M. Berman. Sandia J. Muir, Sandia T. Speis, NRR M. Cunningham, RES G. Arlotto, RES
.i e
4 e
ROUGH DR!E NUREG/CR i /'ffs SAND 80-2415 R3 e
b CORCON-MODl:
AN IMPROVED MODEL FOR MOLTEN-CORE / CONCRETE INTERACTIONS J.
F. Muir R.
K. Cole M.
L. Corradini M. A.
Ellis Reactor Safety Studies Division Sandia National Laboratories Albuquerque, New Mexico 87185 operated by Sandia Corporation for the U.
S.
Department of Energy Prepared for Division of Reactor Safety Research Office of Nuclear Regulatory Research U. S. Nuclear Regulatory Commission Washington, DC 20555 Under Iiteragency Agreement DOE 40-550-75 NRC FIN No. A1019 R O U G H D R L F 1-1
k
'%/
$h
$+f)+%
i
/4 #
4
...e... <e 1,.
TEST TARGET (MT-3) l.0
!!2 EA a
gl=u i
m as l,l '., h,
bb 7
./
1.8 II.25 1.4 1.6 i
/
jc 6"
l
+, 4
+44
'$i V h,'
N)g4,}
g,ppp
A R$.;)>
k.f e
TEST TARGET (MT-3) 1.0
'e m Br2 IS IM R1112.2 u
Will=s=a=
t s
g
- bb l,l
-]
^
i.e l.25 1.4 1.6 4
6" l
%?
'O(h
- %f Mr&,,
sp%%;<
t g,,
MA,Y 27 1?M References (1) Murfin, W.
B., "A Preliminary Model for Core-Concrete Interactions,"
SAND 77-0370, Sandia Laboratories, October 1977 (2) Research Infomation Letter No. 28. " Melt / concrete Interactions,"
Memorandum from S. Levine, RES to E. G. Case, NRR, and R. B. Minogue, SD, May 9,1978 (3) Memorandum from J. F. Muir, Sandia National Laboratories, to Distribution, " Acquisition of German Fuel Melt / Structural Materials Interaction Computer Codes," June 19, 1980.
(4) Kwong, K. C., Beck, R. A. S., and Derbidge, *;. C., "CORCON Program Assistance" ACUREX Final Report FR-79-10/AS, ACUREX Corporation /
Aerothem Aerospace Systems Division, Mt. View, Calif., July 1979 (5) Murfin, W.
B., " Report of the Zion / Indian Point Study: Volume 1,"
NUREG/CR-1410, August 1980 9
9 e
<~
we,
,,~,,-,v
,,.,r-----e,-r v-.
r-,-
,,--,--e-,e n,.. - -,---,
p-,--
-,-e--
,----,e w-,
.-n4.e,w
-r,
MAY 2 71981 Infonnation for Running the CORCON-MOD 1 Code on the BNL Computer The CORCON-i1001 code has been set up and is operational on the BNL CDC 7600/6600 system.
The code is set up on a disc file under the filename "SHERRYCORC0fN0Dl".
ID = ZZRNRC.
The file contains three partitions:
the first partition is in CDC UPDATE fonnat, the second is a relocatable binary file, and the third is an absolute binary file.
Input data for a sample problem is " currently" set up on an INTERCOM file "CORC0flDATA", ID = SHERRY.
For infonnation on how to set up a CORCON run ci the BNL computer refer to the NRC's " Computer User's Guide" available from the NRC Office of Administration, Division of Automatic Data Processing Support.
e e
,,,-.w,,w
. -,,