ML20140J536
| ML20140J536 | |
| Person / Time | |
|---|---|
| Issue date: | 05/29/1985 |
| From: | Meyer R NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Ross D NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| Shared Package | |
| ML20140J438 | List: |
| References | |
| FOIA-85-772 NUDOCS 8604040352 | |
| Download: ML20140J536 (9) | |
Text
.
Distrib'ution:
M Subj Circ ,
MAY 2 91985 Chron Br en r/f (RMeer RMeyer r/f MSilberberg MEMORANDUM FOR: D. F. Ross , Deputy Director WMorrison Office of Nuclear Regulatory Research RMinogue THRU: M. Silberberg, Assistant Director Accident Source Term Program Office, RES _
FROM: R. O. Meyer Accident Source Term Program Office, RES
SUBJECT:
SOURCE TERM CODE PACKAGE, ,
At your SARP review group meeting of May 22, it became evident that 'information about the Source Term Code Package had not been distributed widely enough.
I hope to remedy that situation with this memo and its enclosures.
Two things ought to be emphasized. (1) The Source Term Code Package remains basically the reviewed BMI-2104 suite of codes. The changes that are now
- b:ing made involve developments that were made (and discussed) during the course of the review. (2) The quality assurance work at BNL addresses the code .
package itself and was not intended to cover QA for a new round of calculations i at BCL. j The enclosures describe the Source Term Code Package now being assembled at b BCL and the related quality assurance at BNL.
Original Sigaed By R. Meyer Accident Source Term Program Office Office of Nuclear Regulatory Research
Enclosures:
i
- 1. A new section (Sect. 3.5.3) in l NUREG-0956. '
- 2. 4/16/85 memo describing the i code package.
- 3. Part of the BNL 189 on A-3284 (FY 85). ,
I l
RES:ASTP0 RES:ASTP0 M2yer:md Silb.erberg 8604040352 860218 5/18/85 5/7 /85 $$$I 72 7 PDR
- !' 0;j - l M
-e - - - _
l
, ., , , . .e Distribution for memo dated gay 2 91985 ,
G. Arlotto R. Denning, BCL M. Ernst J. Gieseke, BCL R. Bernero R. Bari, BNL
- 0. Bassett- T. Pratt, BNL -
F. Gillespie T. Spets ._
E. Jordan M. Silberberg Z. Rosztoczy M. Cunningham D. Pyatt ,
G. Marino
, J. Mitchell 4 J. Rosenthal C. Ryder L. Soffer K. Goller J. Hulman D. Muller
- C. Allen G. Bagchi .,
- e- i a
< , - _. , , - . ._. , . _ . , . , . . - . . . . . ~ . , - - ,
\
. ! I l as largd as iodina within the fu21 p211ets. The cesium not reacted with iodine is believed to react with steam to form cesium hydroxide. However, the presence-of some elemental iodine is possible for certain sequences. Elemental iodine )
has been observed experimentally, as a consequence of hydrogen combustion in the presence of aerosols containing CsI (Ref. 3.22). Elemental iodine and I hydrogen iodide are also believed to be formed in-vessel in BWR reactors because
~
of reactions with 8 4C control materials (Ref. 3.23). Further, in-early s eparate effects measurements at Oak Ridge National Laboratory, a very small amount of ___
elemental iodine was observed during experiments on releases from fuel in steam (Ref. 3.24). Therefore, the effect of some elemental iodine in the containment -
atmosphere is being investigated parametrically with the iodine model in use in the Federal Republic of Germany (Ref. 3.25).
A further consideration about the chemistry of iodine is its, tendency to react with organic materials to form volatile organic iodine compounds. Organic compounds from many sources are available in the containment atmosphere to form these reaction products. Since removal processes that would lower airborne concentrations of cesium iodide and other aerosols are not effective for organic ,,
forms of iodine, the process of formation of organic iodine will likely form a limit for iodine release following containment failure below which removal processes are not likely. The range of this release is uncertain, but is likely to be on the order of one percent of the core inventory.
Tellurium (Te) is an extremely reactive metal, forming compounds with unoxidized Zircaloy and remaining with the core material as long as most of the cladding remains unoxidized. This process is modeled in CORSOR, where the release of Te is dependent on the portion of Zircoloy remaining unoxidized. Following cladding oxidization, the release rate of Te is increased sharply. This modeling of the retention of Te with the fuel until ex-vessel oxidization of the cladding for certain sequences accounts for a calculated Te leakage release fraction larger than cesium and iodine . leakage release fractions.
3.5.3 Source Term Code Package .
In the previous sections, remedies have already been described for some of the problems affecting the BMI-2104 calculations. For example, an improved model 05/22/85 3-48 NUO956 CH 3
. -- -- - - - - -~
for tha core-concrate interaction, CORCON Mod 2, became available shortly after publication of BMI-2104 This newer version does not suffer from the viscosity-related problems of CORCON Mod 1 (see Section 3.1.3). In addition, I certain inconsistencies, like the use of INTER instead of CORCON within the MARCH code (see Section 3.1.1), could be rectified with some reprogramming.
Furthermore, most of the data handling problems (see introductory paragraphs ;
in Chapter 3) could be alleviated with automated coupling of the codes. ~
1 Work was thus begun in August 1984 to produce an improved version of the BMI-2104 suite of codes that is now referred to as the Source Term Code Package.
Major features of this code package are described in Table 3.5. The Source .
Term Code Package is seen to retain all of the basic features of the BMI-2104 analytical procedure, which was extensively reviewed, and the modifications l
that have been made resulted largely from the peer review process. The code package should be operational in June 1985, and code release and documentation are scheduled for September 1985.
i i 3.5.4 Summary
)
In summary, the BMI-2104 suite of codes reflects the state of the art as it 4 **
existed in the 1983-84 time period and represents a major advancement in the source, term analytical procedure since WASH-1400. While the NRC recognizes the desirability of further review and documentation, it can be noted that the codes are fully operational, that an extensive peer review has been conducted, l and that a large amount of' documentation has already been published on the codes. A slightly improved version of the codes, referred to as the Source Term Code Package, will be available soon to facilitate the analysis of
, additional accident sequences.
The BMI-2104 codes provide best estimate analyses. They address natural j phenomena that were either omitted or treated simplistically in previous analyses, such as WASH-1400, in a manner that produced large source terms. Large uncertainties exist in many of these codes and numerous areas have been identi-fied for improvement, but no egregious errors have been uncovered in the analyt-ical procedure. Research programs that will lead to code improvement and reduced uncertainties are discussed in Chapter 8.
05/23/85 3-49 NUO956 CH 3
-.._e
ro -
Table 3.5 Major features of tha nIw 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 equiation has been used along with vaporiza-tion properties, when appropriate.
CORCON Mod 2 has been used with its imporved treatment of viscosity.
NAUA and A more realistic treatment of water droplets has been MARCH used at the interface between codes. -
All Codes Codeinterfaceshavebeenchangedtoutilizetape-read output and input.
O t
- l l
05/23/85 3-50 NUO956 CH 3 -
~~ _ _ _ _ _ _ _ ___ - ---- --
UNITEDST TES Q(,~~'d,,g
- NUCLEAR REGULATORY COMMISSION
, .; , g uswiscrow.o. c. aosss
' '.. . . .
- APR 161985 MEMORANDUM FOR: M. Silberberg, Assistant Director Accident Source Term Program Office -
Office of Nuclear Regulatory Research FROM: Ralph 0. Meyer _
Accident Source Term Program Office Office of Nuclear Regulatory Research
SUBJECT:
NRC SOURCE TERM CODE PACKAGE On April 10, 1985, Hans Ludewig (BNL) and I met with J. Gieseke, P.' Cybulskis, H. Jordan, and K. Lee at BCL to discuss final plans for packa.ging the BMI-2104 Battelle suite of codes. This packaging will be done in several steps as follows. t In the next one to two months (i.e., by June 15,1985), additions (asopposed to subtractions to be mentioned below) will be made to produce a good working code package. The constituents of this package will also be discussed below. ,
i This work will proceed with a high priority, and the BCL staff believes that this is the quickest route to new sequence calculations -- quicker than starting
- immediately with the uncoupled codes. BNL will assist BCL in this early phase, i particularly in the area of core-concrete interactions. ,
i l
In the following four to six months, subtractions from the code will be made.
That is, unused options (like 4 unused in-vessel fission product release options, i 2 unused melt'Jown models, etc.) will be removed from the package to make it more efficient and less user-dependent. A code manual will also be prepared during this time. Since we will still be utilizing the basic codes, reviewed in the BMI-2104 study, the manual for the code package will rely on existing manuals for the individual codes. The new manual will merely discuss any code i
changes that have been made and describe how to use the package. During this I time, BNL will be actively ennaged in quality assurance verification, including some detailed comparisons with BMI-2104 cases. i I
Following verification fnd documentation of this HRC Source Term Code Package I work will continue at BCL and BNL to provide additional validation (or bench-
, l marking) with new data and mechanistic codes leading to improved versions of th2 code package (e.g., Mod 1).
i ge kg,_ . g, -
N.:5ilbirberg 2 APR 16 B85 Tha code package will consist of the following:
l MARCH-3 l MARCH-3 will be MARCH-2 with CORSOR-M, CORCON2 and probably VANESA (a recent version) built right into MARCH. CORSOR has already been put into MARCH at l BCL. CORCON has been put into PARCH at SNL. And CORCON has been coupled ;
directly with VANESA at BNL. Therefore, no development work should be required _
to assemble MARCH-3.
TRAP-MELT-3 TRAP-MELT-3 is a combination of TRAP-MELT-2 and MERGE. This work has been -
completed at BCL, and a draft report desciibing some of the revaporization calculations with this code has been completed. .
NAUA/SPARC/ICEDF .
l Near-term changes will concentrate on better interfaces with MARCH and TRAP-MELT. An improvement is being made in the treatment of water droplets in the MARCH /NAUA interface. A small program has also been written to un-bin the ,
VANESA output so that NAUA will get individual species. A few other modest i changes are also planned. .
in summary, the code package as it will exist this sumer will not require I any.new developmental work. It will consist of the basic BMI-2104 methodology, -
which was reviewed, along with some improvements that were made during the b course of that review (e.g., CORCON 2).
Procedurally, BCL was to give us a letter report proposing the specifications for the code package, and they did that on November 9,1984. We were then to reply to BCL with further instructions to proceed. Our meeting of April 10, 1985 and this memorandum are intended to provide that feedback.
M Ralph 0. Meye Accident Source Term Program Office Office of Nuclear Regulatory Research 3 i ,
cc: D. Ross, RES '
O. Bassett, RES )
J. Gieseke, BCL l P. Cybulskis, BCL 1 H. Jordan, BCL )
K. Lee, BCL 1 H. Ludewig, BNL FSRB !
ASTP0 l
5 m, . - , . - - - . _
P. 3 l
Prajaqt
Title:
Sourca Toro Code Packcge Verification ;
PROJECT DESCRIPTION i
- 1. OBJECTIVE OF PROPOSED WORK
- s. Background During accidents in Light Water Reactors (LWRs) the reactor core could be damaged cnd fission products may be released to the primary system. If the primary system ,
is breached fission products could in turn be released to the containment building. l In containment there are a number of systems available to help prev'ent the fission products from being released to the environment. If these systems fail or are com- ~-
promised, a fraction of the radionuclides may be released to the atmosphere with corresponding adverse effects on the surrounding environment. There are potentially a large number of different accident sequences that could lead to core damage and ultimately to core meltdown. Each individual accident sequence could result in sev-Each path will oral possible paths for fission products ,to reach the environment.
have a unique fission product release characteristic or " source ters."
In order to define a " source term", information is needed on the am'ount and chemical form of the fission product species released and also on the characteristics of the release. The release characteristics are the timing of release, release duration, release height and release energy. In the Reactor Safety Study (RSS) models of the physical processes associated with particular accident sequences were developed to assess the magnitudes and timings associated with the release, transport, and depo-cition of the radioactive materials from the core through the primary system and centainment and into the environment. However, it has been suggested that the meth- ,
odology used to generate source terms in the RSS may contain simplifications, which would tend to overpredict the release of fission products and hence result in overly '
conservative estimates of off-site consequences.
Significant research activity in this area has been undertaken following the publi-
~
cation of the RSS in 1975. An updated basis for estimating fission product behavior w s published in NUREG-0772 by RES/NRC. In addition, updated fission product source to'rm methods were developed under the direction of the Accident Source Term Program Office (ASTPO) and published in BMI-2104. Several computer codes (refer to j BMI-2104) were developed to calculate source terms. Consequently, calculating indi- ;
vidual source terms is now a highly complex processes involving significant data transfer between all of the ASTP0 codes. In addition, as these codes are not '
coupled, a number of coupled phenomena cannot be readily addressed, e.g. , local haating effects due to primary system retention of fission products. It was there- l fore decided by the NRC staff to fund Battelle Columbus Laboratories (BCL) to inte- l grate these codes into one self-consistent code package. This package would elimi-i nate the need for assuring correct data transfer and compatibility between codes and also allow the user to assess the influence of the coupled phenomena.
- b. Obj ective The objective of the activities described in this project is to provide quality as-BNL staff will ob-curance of the BCL code package described above in Section la.
tain the code package from BCL and install it on the BNL computing system. The code package will be reviewed specifically to ensure that models and options have (See Continuation Sheet) i l
- 1 Prajset Tit 19: Scurca Tarn Cada Package Verifientien p. 4 l
~ ' 1. OBJECTIVE OF PROPOSED WORK (Csnt.)
basn correctly implemented. In addition, the coupling of the various codes in the pnckage will be carefully checked. Finally, the portability of the code package and ito ease of use will be assessed.
- 2.
SUMMARY
OF PRIOR EFFORTS Not Applicable. _
i
- 3. WORK TO BE PERFORMED AND EXPECTED RESULTS _
- a. Work Requirements As directed by the NRC Project Manager, R. O. Meyer, BNL staf f will perform the fol-loving tasks:
TS,sk 1 BN1 staff will obtain the Source Term Code Package from BCL and install it on the BKL computer. ,
Tack 2
, Ths code package will be reviewed at BNL for (a) correct implementation of BMI-2104 '
modnis and options, (b) adequacy of code couplings, and (c) portability and ease of .
uso.
Tnrk 3 Tha code package will be exercised for selected plants and accident sequences. The !
opacific plants and accident sequences to be analyzed will be selected by the NRC Project Manager in consultation with BNL staff.
Tack 4 It will be necessary for BNL staff to iterate with BCL staff to obtain corrections fer any problems discovered during the review.
i Task 5 l A rsport will be provided to the NRC Project Manager briefly summarizing the BNL ef-fort and describing the state of readiness of the Source Term Code Package.
- b. Meetings and Travel Tha BNL staff will participate'in meetings at the NRC Headquarters in Silver !
Springs, Maryland. In addition, BNL staff may visit other laboratories or institu-tiens and participate at professional meetings. ,
(See Continuation Sheet) l 1
- l
e~ .
MONTHLY HIGHLIGHTS FOR MAY 1985
" Source Term Code Package Verification" (FIN A-3284)
BNL Principal Investigator: W. T. Pratt (FTS 666-2630)
NRC Project Manager: R. O. Meyer (FTS 427-4461),ASTP0
- 1. Scope /
Purpose:
Updated fission product source term methods were developed under the di-rection of the Accident Source Term Program Office (ASTP0) and published in BMI-2104. The several computer codes developed as part of this effort are being integrated at BCL into one self-consistent code package. The objective of this BNL project is to provide quality assurance of the BCL code package. BNL will obtain the code package from BCL and install it l on the BNL computing system. The code package will be reviewed to ensure .
correct implementation of the models and options. In addition, the cou-pling of the various codes in the package will be carefully checked.
- 2. Work Performed During Period:, ,
Work on this project was initiated and disci sions were held with the NRC Project Manager related to the scope of work. A meeting was held at BCL to discuss final plans for packaging the suite of BCL codes.
- 3. ~ Problems / Delays:
W ~0 The level of effort on this project will be incr4ased after the working code packa'ge is received at BNL.
- 4. Summary of Progress to Date/ Milestones:
See item 2.
- 5. Next Reporting Period:
A working version of the code package is expected by June 15, 1985.
-772 gf l