ML20141A102
| ML20141A102 | |
| Person / Time | |
|---|---|
| Issue date: | 08/29/1985 |
| From: | Silberberg M NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Denning R Battelle Memorial Institute, COLUMBUS LABORATORIES |
| Shared Package | |
| ML20140J438 | List: |
| References | |
| FOIA-85-772 NUDOCS 8604040386 | |
| Download: ML20141A102 (9) | |
Text
. .
~
Distribution:
j'{
Subj TMargulies, RES QUG 2 91985 Circ. HRichings, NRR Chron MErnst Branch r/f JMitchell Mi hell r/f RMeyer Dr. Richard Denning i erberg-Battelle Columbus Laboratory OEBassett 505 King Avenue WMorrison Columbus, OH 43201 RMinogue D?oss D2ar Dr. Denning:
After many discussions, the following describes our understanding of the decay h:at situation in the MARCH-3 code (MARCH 2-CORCON) and our request for a modifi-cation to alter the decay heat calculation.
The decay heat calculations in MARCH 2 and in the stand-alone version of CORCON are based on different types of calculational models. MARCH 2 uses the simplified formula of the 1979 ANS 5.1 decay heat standard, which assumes only U-235 fission at a constant power level (except for a constant multiplicative factor of 1.02 to account for U-238 decay heat at extremely short times after shutdown).
CORCON uses the decay heats of inventories of individual radionuclides based on an ORIGIN calculation for a PWR or a BWR plant. These different models result in about a 20% difference in decay heat at the times of interest for core concrete interactions. Therefore, in the Source Term Code Package, when control is switched from the MARCH 2 model to the CORCON model, there would be a drop in the decay heat were no action taken to resolve the difference. . Following investigation of the sources of the difference, it was concluded that for reasons outlined below the MARCH 2 model should be adjusted downward by about 8% and the CORCON model normalized to agree with the MARCH 2 model, producing roughly a 12% upward adjustment. An input quantity, Q MeV/ fission, is available for producing about the right adjust-ment in MARCH, but for the long term, the model in MARCH should be changed to accommodate fissions in Pu-239 and U-238 as well as in U-235.
The default value in MARCH 2 for Q, the total recoverable energy associated with one fission, is 200 MeV/ fission. This value has been used in all the cal-culations to date. The value of Q to use as input to the standard for any given fissioning species is, actually, dependent on several parameters including the neutron flux spectrum and the time in the cycle. There fere. , it is necessary to provide a value that is appropriate for the particular Wd lem. At the end of an equilibrium cycle in an LWR, about equal numbers M Th ions are taking place in U-235 and in Pu-239 (about 40% in each) with atot. 93 numbers of fissions also taking place in 81-238 and in Pu-241 (about IM in a t). Based on evaluations by R. Sher,* values of Q for each of the fissioning species have been determined by H. Richings of the Core Performance Branch in NRR. His values are:
202 MeV/ fission for U-235, 211 MeV/ fission for Pu-239, 206 MeV/ fission for U-238 and 213 MeV/ fission for Pu-241. .Since the decay heat for core melt accident
- R. Sher, " Fission Energy Release for Sixteen Fissioning Nuclides,"
EPRI report NP-1771 prepared by Stanford Univ., March 1981.
go SHOLL.g86860218 g YB5-772 PM
- Vd
~
.Dr. Richard Denning AUG ' . F.
considerations is dependent mainly on the operating history in the few days just before the accident, to provide an average Q value consistent with an inventory at the end of a cycle, a value of 207 MeV/ fission should be used.
Further, in the ANS standard the constant multiplicative factor of 1.02, which provides a bounding decay heat in the first 10 seconds following shutdown, is not appropriate for the long times from shutdown of core melt accidents. Since the factor appears in the numerator of the decay heat equation and the Q factor appears in the denominator, raising the Q value by 1.02 will effectively eliminate this factor.
In addition, the nuclides born of Pu-239 fission exhibit less decay heat than those born of U-235 fission independent of the effects of differences in Q. .
In the time period between 10,000 and 100,000 seconds, the differences between l the decay heats is changing from about 10% to about 3% with Pu-239 fissions l producing lower decay heat. An overall 4% reduction would account for the equal fission fractions and be closer to the value at 10,000 seconds, where the decay hsat is a larger fraction of the total available heat. This correction can also be made by adjustment of the value of Q input.
To accommodate all the above effects, a final effective value of 0 to input is !
thin:
207 x 1.02 x 1.04 = 220 MeV/ fission. :
The decay heat equation in MARCH 2 (i.e., in the simplified version of the l j standard) also includes the heat produced by the decay of U-239 and subsequently -
by Np-239. There is a normalizing factor R in the equation to give the number of atoms of U-239 produced by one fission in U-235, which is also dependent on reactor parameters. The default value of 0.8 has been used and continues to be appropriate for reactor conditions at the end of an equilibrium cycle. Both T. England (LASL) and H. Richings agree that the magnitude of this contribution to the toal heat is less than 20%. Since this second term in the ANS standard is not affected by the Q value, the total correction to MARCH 2 due to a 10%
increase in Q value would be a net reduction in decay heat of approximately 8%.
The ANS 5.1 decay heat standard also includes a time dependent factor, GMAX (t),
which accounts for the decay produced by neutron absorption in fission products.
The value in the standard is conservative, but based on conversations with T. England it is appropriate to use this factor in the time period of interest, 10,000 to 100,000 seconds. It is his opinion that neither ORIGIN nor CINDER '
would correctly account for this process, given the libraries available through the ENDF system. Since in the time period of interest GMAX is about 1.1, this -
effect accounts for the underprediction of-the decay heat by CORCON, which is based on ORIGIN results.
In summary, the difference between the decay heat calculations of MARCH and 1 CORCON has been resolved. The present MARCH model is too high because it does not account for fissions in Pu-239 and it includes a. factor to bound the 1
. . ___ -, . . _ ._ - . - . _ . _ - ~ . _ .
.- . AtJG 2 91985
.Dr.MichardDenning d: cay heat in the time period between shutdown and 10 seconds. On the other hand, CORCON is too low because it fails to account correctly for the heat produced from neutron absorption in fission products. The adjustment to the Q value described above has been discussed with Howard'Richings (NRR), Tal England (LASL), and David Bennet (SNL). All agree that this is the appro-priate adjustment to make at this time.
Sincerely, Vilgliuss $4 M 07 M. Silberberg, Branch Chief Fuel Systems Research Branch Division of Accident Evaluation Office of Nuclear Regulatory Research cc: T. England, LANL D. Bennet, SNL D. Powers, SNL K. Perkins, BNL ,
T. Pratt, BNL J. Gieseke, BCL j T. Kress, ORNL
________ _ _ _ ./_ _ _ ___ _ ______________________________________________
DAE:FSRB , DAE:FSRB DAE:FSRB Mitchell:md Meyer Silberberg 8/14/85 8/,; /85 85
- ; ': . IT USNRC SOURCE TERM CODE PACKAGE AND ITS APPLICATION R. O. MEYER
' ~
USNRC f* ***oq,
! i i 'E
\...../ !
STH ANNUAL INFORMATION EXCHANGE MEETING ON CORE MELTDOWN RESEARCH Shelter Island, NY October 17, 1985
l l i
l l
Fission Product Thermal Hydraulic l Transport Behavior ;
g _ _ _ _ _ _ _ _ _ _ _ _ _ _ ,l MARCH 1 ORIGEN
! Fission Product
, Overall Inventory in Fuel Behavior of l
El
- I h= Reactor Coolant System, Motten Core and
=1 a flgl CORSOR I Containment l
g
- u l l vl 8 Release l Retained l ls from Fuel ; in Fuel l l l
- l
=, I I I e 8 a l
TRAPMELT g i MERGE I '
l o l '
Reactor Coolant System m m, Detailed Temperature, I i l Transport and e Pressure, and Flow in g l B ! Reactor Coolant System I l l Retention u I l l !
l_ _ _ _ _ _ _ _ _ _ _
_ _ _ _ al 1 1 ' '
l l l l CORCON l VANESA
-l r I a E S Release from '===='s Detailed Core- -m J Concrete Temperature
$ Core Concrete Melt i l and Interactions
- i; l E I al l
!l E
, r N AU A, SPARC, ICEDF j
g l
Containment Transport l l l and Retention g l '
I L_______ _ _ _ _ _ _ __ ;l 1 r Release of fission products to the environment: Source Term Figure ES.1 BMI-2104 suite of codes as used in the source term reassessment.
1 Table 3.6 Major features of the new Source Term Code Package Code Improvement MARCH CORSOR and CORCON have replaced FPLOSS and_ INTER within the MARCH code, thus eliminating related inconsistencies.
MERGE and These codes have been combined into a single code to TRAP-MELT treat fission product reheating.
CORSOR An Arrhenius equation has been used along with vaporiza-tion properties.
CORCON Mod 2 has been used with its improved treatment of viscosity. >
NAUA and A more realistic treatment of water droplets has been MARCH used at the interface between codes.
All Codes Code interfaces have been changed to use tape-read output and input.
! I
- 1
~
1 I
i l
Source Terms in the Regulatory Preccao .',
1960 1970 1980 1990 S.ndi.
Siting Study NUREG/CR 2239 s
/
Source Term Reassessment NUREG 0772 Severe Accident Research Program BNL Accident Consequence Study WASH 740 K
Reactor Safety SourceTerm Study Reassessment Study WASH 1400 NUREG 09rg Tf D-14844 in Plant Release Assumptions j
~
Containment WASH 1400 Source Term Performance Credit Release Tables Analytical Procedure Siting Regulations Emergency Planning 10CFH100 Environmental Impact Planned Regulatory 1, Regulatory Guides Statements implementation Design Basis Accident PHA Analyses Assumptions Figure 2.1 History of source term assessment and relationship to regulatory process.
a-- -- ___ _ _ ___ _ _m-_-_-m- __ _ -__ _ _ _ _ _ . _ _ _ - _ _ _ - _ _ _ _
- _ _ _ __ _
t REGULATORY AREAS TARGETED FOR SOURCE TERM RELATED CHANGES +
- 1. RESOLUTION OF SAFETY ISSUES RELATED TO SEVERE ACCIDENTS -
PREVENTION AND MITIGATION OF SEVERE ACCIDENTS, MAINTAINING CONTAINMENT INTEGRITY IN CASE OF SEVERE ACCIDENTS
- 2. CONTAINMENT PERFORMANCE REQUIREMENTS - LEAK RATE TEST, UNDETECTED BREACH OF CONTAINMENT INTEGRITY
- 3. ENVIRONMENTAL QUALIFICATION.0F EQUIPMENT IMPORTANT TO SAFETY - QUALIFICATION REQUIREMENTS FOR DESIGN BASE ACCIDENTS
- 4. EMERGENCY PLANNING - ONSITE PLANNING, OFFSITE EMERGENCY PLANNING ZONES, GRADED RESPONSE
- 5. ACCIDENT CONSEQUENCES AND INDEMNIFICATION - DEFINITION OF AN EXTRAORDINARY NUCLEAR OCCURRENCE, EXTENSION OF THE i PRICE-ANDERSON ACT
- 6. AIR FILTRATION AND OTHER FISSION PRODUCT ATTENUATION SYSTEMS - CONTAINMENT SPRAY SYSTEMS, RECIRCULATING AIR FILTERS, CONTROL ROOM AIR FILTERS AND FILTERED BUILDING EXHAUSTS WERE DESIGNED FOR ELEMENTAL IODINE AND FOR INSTANTANEOUS RELEASE - AUTOMATIC ACTUATION OF PWR SPRAY SYSTEMS
- 7. ACCIDENT MONITORING AND MANAGEMENT, ONSITE AND OFFSITE -
I INSTRUMENTATION - DIAGNOSTIC CAPABILITY OF 0FFSITE MONITORS
..I
- 8. OFFSITE CONTAMINATION, AND RECOVERY - PRESENT BASIS IS i WASH-1400, PREDISTRIBUTION OF POTASSIUM IODIDE '
l
- 9. SAFETY ISSUE EVALUATION - PRIORITIZATION OF GENERIC ISSUES IS PRESENTLY BASED ON WASH-1400.
- 10. SITING - EXPLICIT CONSIDERATION OF SEVERE ACCIDENTS IN -
SITING, USE OF SOURCE TERM IN SITING I
d'e 1
October 23, 1985 Status of the Source Term Code Package (STCP) at BNL Code Status Comments -
MARCH 3 Operational . Sample Problem runs to completion (MARCH 2/CORSOR/CORCON)
. Several inconsistencies have been uncovered, which are being inves-tigated and resolved.
TRAP-MERGE Operational . Sample Problem runs to completion (TRAP-MELT / MERGE) ~
. Use of 1971 Standards instead of 1979 Standard Decay Heat in MARCH
. Many errors have been found and '
corrected at both BNL and BCL l
. BCL asserts that the impact of these errors on Peach Botton calculations are not signifi-cant. This must be verified through calculations. ,
VANESA Loaded . Sample Problem is being exercised SPARC Loaded . Sample Problem is being exercised NAVA Loaded . Sample Problem is being exercised b