ML17298B370
ML17298B370 | |
Person / Time | |
---|---|
Site: | Palo Verde ![]() |
Issue date: | 07/31/1983 |
From: | ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY |
To: | |
Shared Package | |
ML17298B367 | List: |
References | |
RTR-NUREG-0737, RTR-NUREG-737, TASK-2.B.3, TASK-TM PROC-830731, NUDOCS 8410190279 | |
Download: ML17298B370 (439) | |
Text
DEVELOPMENT OF
. THE COMPREHENSIVE PROCEDURE GUIDELINE FOR CORE DAMAGE ASSESSMENT TASK 467 Prepared on Behalf of THE C-E pWNERg gpoUp JULY 1983 8410190279 8410i7 PDR ADOCK 05000528 E PDR POWER SYSTEMS COMBUSTlON ENGlNFERINQ, lNQ
1
=GAL NOTICE lRED AS AN ACCOUNT OF Y(ORK SPONSORED y, INC. NEITHER COMBUSTION ENGINEERING
,S BEHALF:
CANTY OR REPRESENTATION, EXPRESS OR
'.IRANTIES OF FITNESS FOR A PARTICULAR UTY, WITH RESPECT TO THE ACCURACY, 5S OF THE INFORMATION CONTAINED IN THIS F ANY INFORMATION. APPARATUS, METHOO,
!IIS REPORT MAY NOT INFRINGE PRIVATiLY 1
STIES WITH RESPECT TO THE USE OF, OR FOR 8E USE OF, ANY INFORMATION, APPARATUS,
!D IN THIS REPORT.
DEVELOPHEllT OF THE COl1PREHEi'(S IVE PROCEDURE GUIDELIt(E FOR CORE DA'1AGE ASSESSllii/T TASK 467 Nay >o83 Prepared on Beha1f of THi C-E G"llERS GROUP by C-E PCAER SYSTEHS CGl',BUST IOll El(G IHEiR IllG, I tlC.
ABSTRACT The pur p ose off thi s task is t.
~%
rovide procedure cuidelines w h'ic can be used .
under post accident plant conditions to determine mine ih e d egree a'nd type of reactor core various dama, g e chemistr'nd frcm ph
'l
.he h measured fsic fission produc. isotopes parameter measurements and from readily available to the plant operators. Implementation of this task assumes assur tth att project specific implementation of the hUREG-0737 Iten> II.B.3 requirerents for Post Accident Sampling Syste...s have been met. The task is divided ivi e into t a twot phase h progran.
The first phase of this progran is the preparation of a guiCeline for core damage assessment to serve in the interim to the preparation of the cc;.,prehensive procedure. This irst phase wi 11 d e t emine core damage assessment based only on the radiologic 1 analysis of samples obtained frcm the reactor coolant , ccn.ain...en'uildirg sump, and the ccntainment building atmosph re.
The second phase will d et mine core danace assessrent based on a cc'prehensive evaluation of data on plant n ccn d't ition. The information available trcn all potential indica ions will be factored into the final s irat es.irate. These indications include the core exit therroco errocoup 1 e ."
t erperatures, reactcr coolant and contairment atmosphere hydrogen cence cncenirations, , andd containment radiation dose rates. The implere. n t aiion o; both phases is required to comply with the >lRC criteria. This reporr. provi provid es th t e resu ts of both phases of ef ort on this task-
TABLE OF CQ.)T itS Section Title Pace Abstrac-Table of Contents 1.0 Introduction
- 1. I Background 1.2 Plan for Core Damage Procedure 1-2 1.3 Development o. the Interim Procedure Outline 1-3 1.4 Development of the Comprehensive Procedure 1-4 Outline 2.0 Catecori=ation of the Extent of Core Damace 3.0 Establish;..ent of the Basis for Core Damage Assessment Using Radiological Data 3-1
- 3. 1 Basis for the Selection of Characteristic Fission Prccuc s 3-4 3.2 Basis for Identi ication of the Source of the Release 3.3 Basis or the Determination o the quantitative Release of Fission Products 3-14 3.4 General Considerations on the Limitations of the Procedure 3-20
TABLE OF CO'!TE 1TS (Con"'d) 5ec ion Par e 5.0 Establishment of the Bases for Core Damag Assessment Using Core Exit Ther...ocouples and Other Instruments 5-1
- 5. 1 Pressure Indication 5.2 Level Indi cation 5-3 5.3 Prediction of Fuel Clad Rupture Based on Core Exit Ther...ocouple Temperatures 5 6.0 Establishr'ent of the Basis for Core Oamaoe 6-1 Assessment Using Dose Rate Measurement Inside the Contair,;,.ent Building
- 6. 1 Analysis o In-Containment Dose Rate 6-3 o.Z General Disc ssions on the Limitations of 6-11 the Procedure 7.0 Pe(erences
TA"L=" CF CC.'tT"-.'ITS (Cont'd)
Pace List of Aocendixes Appendix A NRC Guidelines for Core Damage Assessment Procedures A-1 Appendix B.Q Analytical Derivations 8-1 Appendix 8. 1 Derivation of the Transient Power Correction cua ti on or Source Inventory
-, 8-2 Appendix 8.2 analytical Derivation for
,h Core Heatup and Oxidation Analyses 8-?
Appendix C.Q Procedure Guiceline for Assessment of Core Da.-..ace Usinc Radiological Analysis of Samples Appendix C. 1 Examole Use o- the Procedure C-24 Appendix D.O Procedure Guideline for Assessment of Core Oamace Using Hydrogen D-1 Appendix O. 1 Example Use of the Procedure 0-21 Appendix E.Q Prccedure Guideline for Assessment of Core Oar...age Using Core Exit Temperatures E-1 Appendix E. 1 Example Use of the Procedure E-9 Appendix F.Q Procedure Guideline for Assessment of Core Damage Using Radi'ation Dose Pates Appendix F. 1 Example Use of the Procedure F-10
(Contend)
TA L= QF CO;aT"- ITS List of Ficures 4-1 Percent Cladd Oxidation as a Function of Temperature for Linear Ramps From 417'F.
4-2 Subcooled Inlet Flow Rate Required to i'!aintain Coolant Level Above the Core as a Function of Cecay Time 4-3 Coolant Density Distribution During Boilorf Coolant Level as a Function of Time Curing Boilof , 34CO ".~t Series, 12CO psia 4-5 Coolant Level as a Fore:ion of Ti-..e Curing Boilof., 3'00 f'.wt Series, 1".. De ay Power 4-6 Vaxi .um Core Te...perature as a Furcticn of Time After Start of Core Uncovery.During Boilof, 3400 i".wt Series, 12GO psia 4-7 Haxirum Core Te...perature as a Function of Ccolant Level During Boiloff 34GO i lit Series, 1200 psia 4-8 fiaximum Local Clad Oxidation as a Function of Time A ter Start of Core Uncovery. During Boiloff, 3400 l',~t Series, 1200 psia 4-9 Haximum Local Clad Oxidation as a Func ion of Coolant Level During Boiloff, 3400 Pet Series, 1200 psia 4-10 Distribution of Rod Radial tiuclear Peaks
-11 tlaximum Rod Temperature During Boiloff as a Func- unction of adial Nuclear Peak, 3400 i"~t Series, 1200 psia, Power
T~SL 5 OF C:;7--:iTS (Cont g)
Pace List of Fioures (Cont'd) 4-)2 Maximum Local Clad Oxidation Ouring Soiloff as a Function of Radial Nuclear Peak, 3460 i",ut, 1200 psia, 2" Oecay Power 4-38 4-13 Percen- of Rods Embrit.led as a Function of Core Oxidation Ouriro Soiloff 3400 l'.ut Series, 12CO psia 4-40 4-)-'ercenr. of Rods E."..brittleo as a Function of Core Oxidation Durinq Soi 1 of;, Oecay He't = 2."..
ercent o- Rcds With Oxidation Embrittlement as a unct'n Func"icn of Total Ccrc Oxidation 4-43
~-,6 Clad Rupture lempel arure as a Func ion of Clad 01 ard Helium Fill Cas Te..perature as
+r~'1tl~'ressure, a Function or Pressure, 2?CO ';ut Series 4-'6 4-1? Percent o, Fuel Rods Above Clad Rupture Temperature as a Function or Core Clad Oxidation, 3400 Put Series, 1200 psia, 2>> decay Pcwer
-18 Ccolant Level as a Function of labor."..ali.ed Tire With '(arious Inlet Flew Rates 4-52 4-)9 Soiloff '>lith Inlet Flow to '!ain ain Steady GC" Coolant Te. perature ard Inlet Flc'u Pate as a Func-'ion or Inle" Leve'team n et Temperature, at 12CO psia, 3400 "ut Cla s, 1".. Cecay Power 4-53 4-20 Peak Ter;.perature as a Functicn of Time, l!AA? Code Results for Heatup After Total Core Sloudown 4-56
'"BL- 0F C0 T~<TS (Cont'C)
Pago Lis. o Ficures (Cont'd) 4-21 flA'P Code R esvl,s,or Boilort Following Slowdown and Re=ill es 1 Peak Temperature and Level as a Function of Time 4-58 4-22 Ccmpariscn of t',AAP Code With Simplified Calculation Pesults 'I High Pressure Boilcff 4-61 A
-c~ ier pera;ure as a Func:ion of Time, Ccmoariscn of 'MAP Code and Simplified Analytical Results 4-62 4-2~ Tem"eratvre as a Function of Radial 'iucI ear P ea,k, Ccmparison C .".
of l".AAP Code ard Simolified Analytical Results
~-c .axirum Local Clad.Cxidation as a Pure icn of Radial livclear Peak, Ccmparison of ViAP Code and Simplified dna ical Analyytsca Results 4-64 4-25 Hydrogen Prodvction Rate From Alumirum and Zinc as a Func l 'cn
~
of Temperature for Calvert Cli s Units 1 and 2
~ 4-69 4-27 Hydrogen Prcduction Rate Frcm Aluminum and Zinc as a Function ot temperature for Palo Verde nuclear Generatina Sta ion 4-70 4- 8 hydrogen Product;cn Rate From Aluminum and Zinc as a Func ion of Temperature for St. Lucie Unit 2 4-71 4-c"9 Hydrooen Production Rate From Aluminum ard Zinc as a Func ion Func.>on of Temperature for SG"GS Units 2 and 3 4-?2 4-"0 Hydrogen Production Rate Frcm Aluminum and Z'nc as a Func ion of Temperature for WPPSS 4-73
30 / (B2'".0)/s f-> >
Pace List of Ficures (Cont'd) 4-31 Hydrogen Production Rate From Aluminum and Zirc as a Func.ion of Temperature for Waterford Unit 3 4-32 S ecific Radiolytic Hydrocen Procuc:ion as a Functicn'of Tire i>me emperature as a Function of Time After Start of Core Urcovery Ourirc 0. 0 rt 2 LCCn, E h
~verace nsse:,.oly Resoonse in 2700 l'.~t Series 5 ercen: of Fuel Pods 'Pith Rup:ured Clad as a Pure .on of l1axi."..um Core Temperature r >r ca 0-x~
typic 1
analys is tol Post ncc'.ce. ~ Dose Ra e 1ns'i c~ > ~ ~ ~ ~ I Cont'r.."..ent 6- ~
2 6-2 Typical analysis for Post Accicent Dose Rate Inside a Cylindrical Containment 6-'
1.0 BACi(GRGUti0 The tlRC instituted the DURING-0737 (Reference 7.1) requirements as implementation of the Post TNI Action Plan in November 1980. Among these was the requirement for a design and operational review of plant reactor coolant and containment atmosphere sarpling system capabilities under accident conditions. The quantitative review criteria were, in general, beyond the capabilities of existing plants. The industry expended substantial efforts to develop the post-acciden sampling systems and equipment necessary to meet the review criteria. The implerentation date for operating plan s was .
January 1, 1982 and for other plan s was four months prior to achiev'ing five percent power during precperationa1 tes ts.
In ~larch 1982, the 'lRC issued a clarification (Reference 7.2) to llURFG-07:7 providing guidance for preparat>on of a procedure to assess core damage, Appendix A. As stated in this clarification, none of the near term operating license applican s had been successful in providing an acceptable procedure.
As a consequenc, each near term operating license applicant has a condition whi ch may res tri c t pcwer opera ti on. kdditionally, the t(RC stated that a final procedure for estimating core damage may take approximatel 12 th .
Therefore, the ilRC stated its willincress too accept an accep a int interim procedure.
d The interim procedure in con'unction with a firm date for th e ffinal procedure 1
would be used to remove the pcwer restric-ing .licens d t . Th
clari i c~--- . c..- '
o.= .he ""r iU.-.c=-GI~z requirer.ents was s-ated L wi h res"e c". io near erm opera..'nc license applicants. A similar licensing condition nay be anticipated by ""crating licensees as the llRC beains schedulino their review wi h respect to 'lUREG-0737.
1.2 PLAll FGR CORE Oni'iAGE PROCEDURE Combustion Engineering, in conjunction with the C-E Owners Group (CEOG), has implemented a two-phase program to prcvide procedure guidelines for assessina core dar.".age follcwirg severe accidents. This. repor. is the final product ef ort. first o'hat two phase The phase was the interim procedure guideline required by the t(RC for assessing the extent of core damage by utilizing cnly radiological analysis of sar;.ples obtained rcm the Post Accident Sar;.pling System (PASS).. These samples are 1) coolant fron the Reactor Coolant Sys:em
/~ ) ah
- 2) ecol ant, rcm the ccr,cair,ment oas from
~
C'PCS),
building sumps, and 3) con<airmen. building atmosphere. Such saroles are available frcn a Post Accident Sampling Syster.", which has the functional capa b i ities required 1'"'a by Section II.B.3 of t(UREG-6737.
The seccnd phase of the CEOG prcgram provides comprehensive procedure guidelines for utilizing chenistry data fron the PASS and also frcm other commonly available instrurentaticn in addition to the radiological data employed in the interin procedure. The PASS chemis"ry dataa >nc include u e h y d rogen ccncentrations and total cas content in the samples. Other instrumentation includes RCS pressure, Core Exit Ther;.ccouple empera ures, and (CET) temperatures containr,ent radiation levels. The final report frcm this second phase is a comprehensive procedure guideline which utilizes all the PASS sample data and 1-2
0 other instrument indications to provide several co.-..plementary estimates of the extent of the core damage. The plant personnel will interpret these damage estimates 4
in combination .vith their knovledge of the particular plant and accident scenario and their prior training to arrive at a judgement on the extent of core damage.
1.3 DEVELOP/ lEilT OF TH Iis TER I'l PROCEDUR OUTLI~'lE There are three factors considered in the interim procedure which are related to the specific activity of the samples obtained and are employed to assess the degree and type o core damage. These are the identi ty of those isoto"es which are released, the respective ratios of the specific activities of those isotopes, and the percent of the source inventory of the time of the accident which is observed to be present in the samples.
The llRC guidelines for preparation of this procedure define ien categories of fuel damage intended to address uel integrity for post accident sampling.
These ten categories are characteri ed according to the anticipated mechanism of fission pioduct release from the fuel. Each mechanism of fission product release is then charactelized oy the identity of characteristic fission products present in a given post accident sample. This identity may be used to make an initial categorization of the type of core damage. The selection of the representative fission products is described -in the folio ving sections.
There are two sources of the fission products released by the fuel. These are the fuel pellet and the fuel gas gap. The presence of a fission product in either source is a function of the fuel history, the diffusion oroperties of 1-3
the isotope and the half life. The relative ratios of the quantity present for an isotope of a given element will differ between the fuel pellet and the
-;-=1 gas gap. The type of fuel damage, determined initially by the identity of the characteristic fission product, is then confirmed by calculating the isotope ratios and comparing them to analytically determined standards for the pellet and gas gap. The source of the release is added identification to the type of core damage.
The deoree of core damage is expressed in terms of the percentage of the total core inventory available for release. The specific activity o the measured samples is compared to analytically determined curves for the specific activity at the sample loc tion as a function of the total core inventory available for release. The assumptions used to describe the progressive damage expected in fuel rods during core melt accidents and the distribution of the fission products within the fuel rods under normal operation is based upon the material prepared for EPRI through the IDCOR Program, Reference 7.3.
1.4 DEVELOPflEHT OF THE COhPREHel(SIVE PROCEDURE OUTLiHE There are three factors considered ir. the comprehensive procedure which are related to the chemistry and physical parameters of the samples obtained and are employed to assess the degree and type of core damage. These are the hydrogen gas content of both the reactor coolant and containment building atmosphere, the reactor coolant temperatures as measured by the core exi t thermocouples, and the radiation dose rates measured in the containment building atmosphere.
The tlQC guidelines, as previously discussed, define ten categories of fuel damage intended to address fuel integrity for post accident sampling. These ten categories are characterized by the temperatures ach'.eved on the fuel cladding sur,ace. Characterization of core damaoe in terms of fuel clad surface temperature permits the use in this procedure of two parameters which are measurable following an accident. These are the amount of hydrogen gas within the containment building and the core exit thermocouple temperature.
The amount of hydrogen gas measured within the containment by the PASS is correlated to the extent of chemical oxidation of the fuel cladding which occurs at elevated surface temperatures. The maximum coolant core exit te:.".perature measured by thermocouples is correlated to the percent of fuel rods with clad surface temperature above that which is considered o be a threshold for clad rupture due to gas gap overpressurization.
The information obtained 'from hydrogen measurements as an indic tion of fuel cladding oxidation is more applicable within the fuel overheat category of core damage. 'Within this category the clad surface temperatures are sufficiently high to result in the production by oxidation of measurable quantities of hydrogen but are below tnat which results in fuel clad material melting. The hydrogen gas measurements are obtained by the PASS. and analytically corrected to account for the presence of hydrogen gas produced in sources other than uel clad oxidation. The measurements are used to obtai n the total amount of hydrogen produced from fuel clad oxidation. This value is used by procedu> e to estir ate the extent of fuel overheating according to the percent of fuel rods which have been oxidized beyond the limits for continued structural integrity.
1-5
The information obtained from reactor coolant core exit temperatures is most applicable within that category of core damage which addresses cladding
- failure..he mechanism of core damage measurable by core exit temperatures is the rupture of the cladding as a result of high temperature overpressurization of the gas found within the gas gap region of the fuel. It will be shown that this mechanism results in the rupture of a significant fraction of the fuel rods in the core prior to temperature reaching fuel overheating and the onset of oxidation initiated clad failures'he reactor coolant core exit tempera-tures are measured by thermocouples located within the reactor vessel.
Analytical determinations of the radial distribution of the core exit tempera-ture under the most general cases of post accident conditions are described.
The measurements obtained following an accident are surveyed to identify the maximum channel exit temperature. This value along with the analytically determined radial distributions of temperature are used by the procedure to estimate the extent of cladding failure. The estimated damage is the percent of the fuel rods in the core wnich have clad surface temperatures above the established value for rupture due to gas gap overoressuri zation.
The dose rate inside the containment building is a physical parameter which also may be used to characterize the ten categories of core damage. As previously discussed, with regard to the Interim Procedure, these categories were characterized according to the mechanism for release of fission products from the core. The identity of each of these mechanisms of release may be determined by the presence outside the core of specific characteristic fission products. The dose rates inside the containment building are dependent upon I-o
the quantity, identity, and distr "ut on o; -.hose fission ~roduc s "ere=or characteri=aticn of core damage in terms of the mechanisrs of release per~its m the use in this procedure of he measured are~ r~d'-CI ion
~ dos e t to assess rate .
core damage.
The intor.",.ation obtained from area dose r~-es within 'n thee con co t ainment m '1 building b
as an indication of fission prcduc . release is rost applicable within the cladding failure and fuel overheat categories of damage. The application of the fission produc release mechanisms to core d rage assessment is the same as that used in the Co..prehensible Interim Procedure. The difference, wnen applied to the Procedure, is ".he physical parameter being measured. In the Comprehensive Procedure the mechanism of release is correlated to the reasure...ent of; t..e ar>>d the are r~diaiion dose rate rather than to the reasurerient of samp e speci;ic ac.ivity. The use of two different reasurements =or the evaluation of the sare physical parameter is a reans to recuce i"e uncertainity in the assessment of core damage.
1-7
2.0 Cni -'.C.~..~i.C.'i OF i'""" '(T"'fT OF CC"" r'"<iC" The tas~ 1 of applying post acciden sampling system data to assess the condition or a reactor core following an accident requires some description of the relevant conditions. A wide range of accident types and sequences are possible. Therefore it is not appropriate to at. empt to employ specific accident scenarios in the development of such a procedure for core damace assessment. However the end product statement concerning core condition should be capable of describino tt e ther...a) hydraulic and material proper-.es of the degraded core to the extent prac ical for the imple...entation o. hat in or...ation in emergency decisions. The statement of core condition snouid be in ter...s of defined categories which are commonly understood bu. at the sare te do not imply quantitative assessments which are beyond the accuracy of the data evaluation. The Rogovin Report, Reference 7.4, catecori:es coro damage into four major types as ollcws; ro ,uel damage, fuel clad"inc failures, fuel pellet overneating, and fuel pellet melting. Consis.ent with these catecories the t(RC guidelines =ur:her delineate each of the three later categories into initial., inter..ediate, and major thereby assessing the exten:
of each type of'amage. A rationale is then required to describe the resulting ten categories in ter.,s of those physical conditions o the core fm-
~hich measurable data may be obtained.
Independent of the acciden- scenario the s ~ ri of ar- a cora e d core condition deorad in a ressuri "ed Mater React r is the result of a ther...al inbalance betweon the hea. generated in the fuel and. the heatt roroved rcm '."
..ove frcrI ...e core ecol ing
" or the '
water. Core heat re...oval and coolant heat re...oval ar> re .wo th principal 1 a-ety Functions activities of the reactor operator f 11 owing
. an accident.
2-1
The events following this initiating condition as they relate to he thermal and material state oi the core have been the subjec. of a nurber of analy-ical yL1 ca and experimental evaluations. In order to define the physical parameters across the spec:rum of core damage it is necessary to first assume that the accident is allcwed to prcgress through that spec.rum and then to select an analytical model to predict the resul ing conditions.
Particular accident scenarios could be postulated for which changes in the system pressure, the time period of core uncoveryy, a nd . e ra t e "h o f uncovery'ould result in a final core condition anywhere within th e range or spectr;m of core damage. However as stated previously, this discussion does no. no assume any par.icular acc dent scenario. Accident progression from initial fuel Carage thrcugh to the eventual condition of major fuel pellet melting is assumed only to allow correlation of the physical parameters anticipated ihrcuch .he progressive core degradation to the ten selected cate",cries of core damage.
ihe model selected to describe the progressive material interactions ard darage evpected in fuel rods through the spectrum of degraded core corditions is that described by EPRI through the IDCOR Program, Reference 7.3. That
'repor. provides a r odel which is the result of a state of the art evaluaeva uation ion of a number of independent analytical and evperimental works. It is reccgnized that a defi'ni .i ~e model Cd'or progressive core degradation has not been developed. However, the results <f the IDCOR Program are widely accepted and will therefore be employed as a basis for this procedure 2-2
The procression of c or damage, which begins wi I 0 a
~ loss o= i,e equi'ibrium in the reactor core heat balance, for ur"o thee purposes of this report is taken o "e as fol lows. The centerline temperature o fa iue I rc d wil deperd upon its power density, the ther...al conductivity of th e fue I , t h e gap conductance between the fuel and the cladding of th e ro d andd theh conditions of the surroundin g coolant. C enterline temperatures are in the range of 2200 to 3:CO'F ror nor...al operating conditions. F o 11 owing an accident the core may j
not be able to re 'ect the stored energy pIus the fission product Cecay hea-rc' .he cladding surface due to the initiatina loss of heat balance between t..e rod and .he coolant. The surface temperature of ".he cladcing increases, possibly result ng ',p in tern or a rg i 1m boiling of the reactor coolant.
The fuel temceratures continue risinc, follcwing a I oss or= coolant I accident which uncovers the top of the core sine e sieam ccolirg of the uncovered por.ion o; the fuel is not su icient to re...ove thee d ecay heat unless
~ there is a larce temperature difference between he clad and sl s >>mn~ Our ng Cepressuri=ation accents,, a pressure di e'er n.ial exists between he as "r~'e '. in t..e -uei rod gap and the reactor "
olin. pressure e ctor ccool-n- which may cause the cladding to burst. The cladding burs- can be expected to occur in the e.."erature ran ~
e of 14CO to 2CCO'F depending upon the amount of fission gas and prepressur i"ing helium in the fuel rod, t1 e reactor vessel pressure, the rate o emperature rise, ard thee tire ai .emperature, Reference 7 " CI a d b urs may occur at temperatures as low as 1000'F when hi
~ en high dirrerentiaI pressure is ccmbined with long duration at temperature. The clad bursturs resu 'ltts in the release of volatile fission prcduc.s preser.t.. wii, within in tth.e oas gap and to a lesser extent within the fuel pellet surface. Clad rupture C ces not occur uniformly across the core because oi the radial variat '<<.' on in fue f I rod peak clad temperature.
2-3
0 As the core beccr...es uncovered the steam surrounding the fuel rod oxidi-es the zirconivm present in the exposed length o the cladding. A chemical byprccvc yprccvc.
of the reaction is the production of. hydrogen gas. The oxidation is an exothernic reaction whose rate is dependent upon the surface temperature o=I the cladding. The exothermic reaction provides an additional heat source which serves as a catalyst to accelerate the rate of reaction. This reaction the cause o. the 4's rise >n tuel temperature above 22CO'F. Durinc the later stages of core uncovery the steam rising from the lcwer regions of the core can be censured by reaction with the cladding in the upper regions.
Oxidaticn o; the zirconium present in the cladding causes embrittlerent oi the with subsequent degradation of structvral integrity. At some tire Curine the accident the core may be reflooded and cooled or he reactcr ccolant pu-..ps may be started causing a pressure transient. The emori: ed fuel cladding would fracrent as a result of either thermal or pressure shcck.
this increase in fuel surface to volure ratio would increase the release r -.
of fission procucts.
Above the temperatvre rance oi 2000 to 2:"50'F, general lat ice robility exists in the fuel allcwing fission products to dif,vse to nore stable thermodynanic--
states. Atcms which do not react with the VO or any foreian material in the pellet will diffuse frcn the interstitial location to either a micrcbubble or metallic phase. At approximately 24~0 F, the ission products including noble gas, cesiur. and iodire will be released from the V02 grain bcurCaries. At te..peratures above 2450'F, the fission cas micrcbubbles include vapori ed cesiun and iodine.
2-4
Above 3250'F the Eirczioy cladding makes. Endothermic reactions occur between molten Kircaloy and 2r02 and the dissolution of U02 by molten Zircaloy. The release of fission products by diffusion from U02 grains begins to occur at a.
rapid rate. The diffusion process is continuous but the rate is not significant at lower t mperatures. The liquid formed as a result of these endothermic reactions flows throuoh the fuel rod gap and continues to dissolve the U02 fuel.
Those material in.erac ions and damage expected in fuel rods accompanying prolonged core uncovery relevant to the tlRC categories of core damage are su;..imari=ed in Table 2-1. As described above, a temperature range is associated with each physical condition. The mechanisms of fission procuc-release frcm a fuel rod which has been burst are related to the fission prcduct volatili:y and di fusion transport properties. Both of these aroe te.-perature dependent. Tnerefore each of the ten categories oi core dar.:ac~
can be charac.er:ed by the type of fuel darage, the corresponding emper';ure range, and the rechanism o ission product release. The charac eri-ation of the categories is su~ari= d in Table 2-2.
Therefore he cc."bir'ation o Tables 2-1 and'2-2 provide the definition and physical conditions or each o the ten llRC Categories of core damage which are employed througncut the subject procedure. This provides the reauired definitions for cod.;,.on understanding o7 the end produc" statement of core damage assessment.
2-5
0 TABLE 2-1 Prcgressive i'late. ial Interactions and Darage Expected in Fuel Rods During Core t1el t Accidents Types of Fuel Dar'ace Ter,oerature 'F
- 1. Ballooning of Zircaloy cladding > 1300
- 2. Burst or Zircaloy cladding 1300-2GCO
- 3. Oxidation of claCding and hydrogen generation 1600
- 4. E...brittle..ent of uel rod cladding by oxidation > 2200
- 5. Fission Produc. uel lattice mobility 2000-2.""0
- 6. Grain boundary diffusion release of fission products > 2450
- 7. I".elting of met llic Zircaloy 3250
- 8. Fission Product Di.fusion from U02 Grains < 3450
eutectic > 3450
- 10. I!elting o, U02 50eo 2-6
Table 2-2 C>aracter'":..'Cn of llRC Catecor;es cf Fuel Dar'ace tiRC Catecory of ".echani sz of Temperature Fuel Dc@ace Release Rance F
- 1. t(o fuel damace
- 2. Cladding Failures
- 3. Intermediate Cladding Clad burs" and 1300-2000 Failures ~ dif usional gap release
- 4. Hajor Cladding Failures
- 6. In ter ...edi a te Fuel Grain bc"ncary > 2450 Pellet Overheating dlttus ion
- 7. liajor Fuel Pellet Di ~ ;usicnal Release < 3450 Overneating frcm UO 2
crains
- 8. Fuel Pel let .".el t
- 9. Inter.-..ea ate Fuel :scape frcm,"..ol ten Pellet "el t fuel
- 10. l1ajor Fuel Pellet I!el 2-2
3.0 ESTA9LI H'.-.'IT OF THE "-AS."-5 FOR g<,"E .;.iAGE ASSE<<.iEi1T USl;/
RAO IOLQG? CAL OATA The purposes for performing core damage assessments are first to assess the effectiveness both of the reactor operator actions and the automatic engineered safety feature systems to mitigate the consequences of an accident and second to assess the potential for subsequent release of radioactive material to the environment. Section 2.0 of this document described core damage in terms of the material interactions and s ructural integri y ex"ec- d in fuel rods experiencing uncovery and and the conseouent progressively increasin g fuel temo. e r ature. Based upon the stated. purposes for core damace
.assessment it is appropriate to Cefine the cateoories of core damage for use in thi 5 procedure in terms of those phys ical parameters re1evant to the release ov radioactive material. The postulated scope of core damace enccmpasses a broad spectr.m of physical conditions. Therefore, i: "ecc."..es necessary to measure as many parareters as possible in orCer o defire the loocation a of the core within that spectrum of damage. Additionally, to obtain a workable procedure it is necessary to limit the definition to those physical parameters for which measurable data may be obtained using the Pos- Acciden-Sampling System.
specific conclusions This reans that those parameters may be drawn with respec o are selec.ed for wnich cord idion and ore con core f which for the variations in the accident scenario have a minimum in luenco on ha conclusion. Wherever possible, the ccnditions which influ inf uence the measurement of a given parameter are identi ied.
3-1
Ai ihin these cri:er a the core damage categories are defined in ter.-..s of he source of fission product release, the mechanism o, fission product releas~
and the quantitative release of characteristic fission products expressed as a percent o.C the theoret>cal source inventory. The mechanism of fission produc 4 release is identified through the presence of characteristic fission produc s in the sample medium. The source of fission product release is identi ied throuoh the relative ratios of the isotopes of a given fission produc-. The quantitative release is determined by calculation using the concentration reasured in the sample and tabulated theoretical source inventories. Each of these selected physical parameters are quantified in terms of measurable data fn each case however there are conditions which may influence the accuracy or 144$
limi ..he val]Jybidity or the measurerent.
~
The following sections describe the technical basis for the selection and use of each of these parameters ircluding the ccnditions which may influence the accuracy of their measure...ent.
ihe objec:ive of the subject core damage assessment procedure is to achieve an 1
evaluation of t.e radiolooical data within suf=ic:ent :en accuracy ac to determine the.
existing core condition in terms of the ten de=ir::" i 'r.::". ca-categories describeo in Section 2.0. The followirg table provides the cri eri a b y wh ic h eac h category is evaluated with respect to the three physical parameters selected above. By procedure the plant personnel will use the measured radiological data ioo determine each physical parameter, locate the parameter within the table and then use the table to state the core condition n terms of the corresoonding defined categories.
(-)
3-2
Table 3-1 Badiol~oical CIiaracteristics of BBC Categories of Fuel Damage Release of Characteristic IIBC Category of tiechanism of Source of Characteristic Isotope Expressed as a Fuel Damage Release Beleas<< Isotope Percent of Source Iriventory
- 1. t(o Fuel Darirage Ilalogen Spiking Gas Gap I 131, Cs 137 Less tlian I Trarrrp tlrarr iurir Bb 130
- 2. Initial Cladding Gas Gap Less than 10 Fail<<i e
- 3. Inter rrrediate Clad Oirrst and Gas Gap Xe 131m, Xe 133 10 to 50 Claiiiiing Failure Gas Gap Diffusion I 131, I 133 Release Hajor Cladriing Gas Gap Greater than 50 I'd i 1 ur<<
- 5. Initial Fuel Pellet Filel Pellet Cs 131, IHl 88, Less tllan 10 Ovei liea t ing Te 129, 1e 132 Grain Boundary
- 6. Intermediate Diffusion F<<el I'el let, 10 to 50 Fiiel P<<llet Ove l'hea t, i rig
- 7. Ilajor Fuel Pellet Di f fiis iona 1 Release Fuel Pell<<.t Greater ttiari 50 Over liea t.ing FI olll t)02 Gr'a iris
- 0. Fuel Pellet Ilelt Fuel I'eilet, Less thail 10
- 9. 1<<terrriediate Fuel Escape fr'm Ilol ten Fiiel I';llet Oa 140, La 140 10 to 50 Pellet !1el t, Fuel La 112, Pr 144
- 10. Hajor Fuel I'ellet Fuel Pellet Greater than 50 11e 1 t
Core damage will not:ake place urifcrmly among all the fuel rods. Uniform fuel rod damage .hrouchout a given core would in fac be an unrealistic assumption due to the radial variations in fuel rod peak cladding terperatures. therefore, when considering the total effect of the damace on all of the individual fuel rods, the core damage assessment procedure yields a ccmbination of categories which may exist at the time a given sa...pie was obtained. As an examole, he analysis of a given sample may indicate the presence of -both ( 1) ission procuct isotopes charac ieris erist ic of grain bourdary b
di ~ ~ us i on in a quantity equa 1 to 25 percent of the fuel pe 1 1 et irventorv ard
,:ssion prcduc. isot"pes charac. eris.ic of cladd;no burst role>>se in C
a can 1 Jecual .o'GO percent of the fuel gas cap inventory. fn this example
- he core dar..ace assessment would be intermediate fuel pellet overne>> ing i i h c"ncurrent major fuel cladding failure.
CRASIS FOR S="L="CT1G>l OF C:":nRACT"-FISTIC FiSSiO'1 PRO"UC ~
The mechanism of fission product release frcn a d anaged
. d rueII rod is identified thr.uch ihe presence o- characteristic fission products in the sampl~ me"iu~
A survey has been corn..p 1 e.edd .o determine ihe fission prcduc. isotopes which characterize a given mechanism of rele>>se. These isotopes are c'rosen o determine the decree and type of core damage. Specifically the isotopes are seiec.ed .o dif-erentiate between the three major types ypes o f core damage-claddirg failure, fuel overheat, and uel melt. The criteria for selection of the isotopes includes half life, the quantity p"esent in the core, the rate a' which .hey reach equilibrium in the core inventory with respect to fuel burnup, the degree to which their presence in a sample represents a specific type ot core damace, detectability usino standard semiconductor and 3-4
r'ul tichannel anal v" er and the amount o<<..niques within of infer...a tion available a
on o "l; postulated fission producc~ mixture i,eir chemica'ehavior. b ml x iJre,e The fission products selected all have radioactive half lives of suf ic'er.t duration to ensur that the'i g will be present in quantity and time period following an accident to allow detection i' a d ana an l..
ysis. Another A important related fac.or is the history of the fuel prior to c 1 adding d rupture. The physical properties of the isotope determine .the rate at which a speci 'c
..c isotope inventory approaches equilibrium in th e coree as a func:ion of coree burnup. Implementation of th ihe sub'ect procedure under pos" acciCen.
ccnoitions necessitates sim.p lif
'cation or data analysis whenever possible.
Therefore ana jwtical corr'ection 1 0T reasured data to a standard core bur"u" is ~ ~
not Cesirabl e. Seleec.ion of ronitored isotopes which reach radiological equilibrium quickly within the fuel cycle eliminates this concern. The physical parameters of influence to this selec-ion arre iso.epic ; knife, fission product yield, cross section for loss Cue to neutron absorp-'cn ard decay chain branching frac:ions.
To implement p .. . e ec..'onn cri.eria, tthe select the isotopes selected are divided in-o > n.o
.~o groups. The first group includes those isotopes with hal lives between four hours and ifteen days. These isotopes are used to assess the damage condition for cores that have been iona in operational a given cycle for more than thirty days. These isotopes reach radiological equi 1'b i rium levels in the core after thirty days of operation. The second cup inc grcu 1 u d es those isotopes with hal~C 1'ives between one hour and 'wenty four ur hoours. inese >soto es are used to assess the damage c"rdition for cores that havie ~een operational in a oiven 3-5
cycle;or fewer than .hirty days. This group is used in determining core damage early in a given core cycle, but has the limitation that sampling and analysis rust be completed within a few hours followingg >, he acc'dei entt to avoid the loss o data by isotopic decay.
The selec:ion of fission products by detectability is a very practical criteria in the implementation of the subject procedure. Numerous factors influence the ability to sample and detect specific isotopes . Reliability of the sampling is hampered by rapidly changing plant conditions, equipment imi..ion, ardd lack 1 or operator familiarity with rarely used ana)ytical procedures. Ch .,'emis .s are required to exercise considerable caution to minimi the spread of radioactive materials. Sample's have the pote., i'1 of being ccn:aminated by numerous sources and may not result from a uni=or-.
distribution or.,the r 'ampled medium. Cooling or reactions may take p 1 ace in . ~ .e lorg sampling lines. inere-ore the results obtained may not be represent t>ve of the plan . condition ition. Plan.
Pl conditions, radiation exposure, and time reuire...ents may prohibit multiple sa...ples and reduce statistical reliabili .y Speci,ic criteria for detectability of a 'on pro d uct in fission a given sample is based upon the capabilities of typical semiconductor detectors employing multichannel. anal y sis of the fission produc gamma enercy spectrum. 'These criteria include the principal fission product decay eneroy, the prosence o" oiher h isotopes with similar or masking decay energies and n ih e success o> such measure...ents in experiments conducted by C-2 and other er reporte repo d measurements.
3-6
lss cn prccvc isotopes as being rre"re - n~~~-.-'
e..ta. ve o" speci ic
~
~
t/pes o core darace and with respect to the chemical behavior is based upon a survey of th pub isl ed literature. The
availab'1'"
i i a i i.y o- ata on their reports which were of specific contribution are the IDCGR Draft ra t F>nal Final Repor.,
Repor" Reference 7.3, and the Rogovin Report, Reference 7 4 Th e specific 'f i criteria to select isoto"es as indi i dica iors of the type of core damage is their respective volatility.
The category o- no ccrc damace is characteri-ed by t-h e release e o- fission prcdvct"t~ -~
.hrcuch .he rechanis"..s or spikin '
an d t ramo ur a nium fission..
. Re ."r c:"r ccolant system pressure i te......peraiure, and power transients may resul in i"c're spikinc. Iodine s p'iking is identified by a rise in reac:"r c"olant lodire o >re ccrcentrations during the pericd rrcm 4 io S h hours arter .he transient. Ti:e iodine concentrations can be bovrced by ; a va v 1 ue o, "."GO t;.-..es the e"vilibrivm levels Curin raul ted condi i'ons svcn as a steam cener tor tv"e r "-"re wi..cu any tuel cladding ailure. Spiking is ident:fied . ie "y a decro~se ec. ease in reac.or coolant concentration subsequent to th e spi~e peak at a rate eqval to the system purification half li,e.
The categories of core darace. ident fied ie as cladding railure are characterized by the release of fis 'ssion prcduc s throuch the mechanismsII of b urs" and gas cap diffusion. The chcharacieris.ic ,ission produc I s are .,4 e nc bl e gases and halogens, wnich becauuse ihey are volatile can migrate qu'c.'1 :: y th rcugh the fuel pellet and g~as g a p or release following cladding rupture. Tnese e o ati?e in .he te...perature range (1300-1800'F) acce"-ed accep.ed as claddirg bvrst tern"era-ures.
1 . ~ en -h
'~hen t .e claddirg ruptures the entire amcunt of noble fission gases previcvs) y acc a "r.
rulated in the plenum and open voids in the 3-7
fuel will be assumed to be released. :his amount can rance up to 25". of the ion g ha f ife 1 1 f fission gas iso.cpes dependirg on power history. Cesium and iodine are also released when the claCCing "uptures but the quantity carried out with the vented gas is considerably less than that for the noble fission gases. ihe initial release of cesium and iodine deends upon the fuel temperature, the volume of gas vented, and the amount of cesium and iodine
'ia1 in initially1 the fuel gap. The di, fusion release o. the remaining halocens in the gas gap is a slow process in the cladding burst failure temperature rance.
The c . cories of core damace identified as fuel overheat are charac.erizeC b the release or radioact;v>ty through grain boundary diffusion and bv di,usion
.r m within .he l02 grains. Grain boundary diffusion begins above 2450'F.
ihe moCerately volatile isotopes of cesium, rubidium, and tellurium aro charac. eristic of this ype of damace. The IDCCR report estimates that 20.'; of the total initial fuel inventory of 'stable and iona lived cesium would be released at temperatures consistent with orain boundary di==usion Di- usional release of these isotopes from wi~hi n th e UO grains >ncre ses 2
rapidly beyond this temperat re and the rate is a subsequent .unction of temperature.
At grea.er tempera inures (2:"0-3~50'F) reac ions begin between the solid U02 and solid metallic zircaloy, melting of the control rods materials and melting of the zirconium. This is the onset of the categories of core damage identified as core melt. At these temperatures, greater amounts of tellurium are released. 'Alkali metals such as barium volatize as well as rare earths ard ac".nides such as lanthanum ard protac .inium. Tne amount and type of isotopes released is dependent on the extent of fuel fragmentation.
3-8
Table 3-2 Selec-ed Isotopes or Core Oar.ace Assess;..ent Categorr o Selec:ed Fuel History Principal Core Inventory Core Oa.-..ace Isotcoe Hal c Li Grouoina Enarov ~~<ev Orcer or lfacnit Ce Kr 87 2 0. 403 l(~7)
Clad Xe 131m 12d 0.164 5(-5)
Fa i lure Xe 133 5.4d 1 0.081 1(-.S)
I 13'. 1 0.364 7(+7)
I 132 2h 2 0.955 1(.8}
2'. h 182 0.53 1(-B)
I 135 '6.8h 2 14 1('8>
Fuel Cs 13'b Zy 1 0. 605 2(-7}
Overheat 88 ZGl 2 1.86 >(~
Te 129 7Ca 2 0.445 2( 7)
Te 1"2 78h 1 0~ 4 ( -)
Fuel 'lelt Sr 89 7d 0.91 l(-B)
Ba 1 "0 12.8d 0.537 1(-B) 1>>'2.
La 140 La 1-'2 40h cCa 1.596 0.65
(-8) 2(-.8)
Pr 17.~a 0.695 9(-.7) 3-9
Based on these criteria Table 3-2 provides a list o- the isotopes selected for analysis in the subject procedure. The isotopes are grouped according to the type of core damage their presence represents and according to their use with respec: to fuel his.ory prior to the accident.
3.2 BA !S FOR IDEl)TiFICAT!ON OF THE SOURCE OF "E RELEASE The identification of the source of the fission produc. release is useful in determining the'xtent of damage which may exis . in a core follcwing a given accident scenario. For a particular accident the radial variation in peat,
'uel
,uel cladding te...perature can be significant. Therefore accident scenarios can "e postulated in which a limited number of fuel rods may experience fuel pellet overheating while the majority of the fuel may not reach the 18CO'F temperature recuired for cladding burst. During such an accident the icentity and cuanti y o ission products detected in reactor coolant samples is insur-,icient information to ceter'iine the type of damage which has occurred.
The added information needed to evaluate the accident is the sour"e of the detected ission products. Specifically t is necessary to determine wnether the ission produc:s have been released from tne fuel rod gas gap or from the fuel pellet. This determiration can be performed using the relative ratios 0 the isotopes of a given fission product.
During the fission process the relative ratios o the isotopes of a given'ission product will remain cons: nt. The value of the ratio is dependent upon the material being fissicred and the energy o. the neutron which induces the fission. Each isotope has its own characteristic half life. Therefore 3-10
the ratios or the isotopes will vary as a func:ion of t'me following their production. If it 's is assumed a that the only loss term frcm the fuel rod is Cue to deca y of the isoto pes e then an equilibrium c"ndi:ion is reached in which ".he production of the isotopes wi 11 equal their loss Cue to dec y. Under equilibrium conditions a fix d inventory of the isotopes exists within fixed the producers fuel rod. The assumed condition is practical for selected fission prcduc s which are products of a limited number of precursors and whose isotopes have small neutron absorption probabilities F t,ese t fission products the f'r relative ratio of their isotopes within th e fue 1 pe 11 et can be considered a ccnstant ~hen the reactor has operated for or su c ent time for equH ibrium to have been reached.
Ouring power operation the central temperatur e o f a fue 1 rod is sicnifican ly hicher than that of the fuel rod gas gap or cl a Cd'ing sur;ace. Thus a Iar"e temperature gradient exists across the fuel pellet. Such te.,pera>ure gradients cause substantial mioration of vol a t'1 i e ission proCuc s i, they are unhampered by chemical reaction within ~ hee pe 1 1 et. Th ThOSe iSSiOn prCCuC.S which migrate along that gradient and re c11 "l.e gas gap will consisi of material which has existed in the pellet or sufficient tire for "h1S ih15 migration to take place and ther fore may be considered to consis" cons1st or of the older collec.ion of material. T he relat1ve rat1os of the 1so.opes of f1ssicn proCucts foundd in the cas cap is therefore di, ferent rom that ound in tth e fuel p ellet b ecause ihe raiio
+ 4 varies as a function of time following production. inus, theoretical calculations may be employed to det rmine 3-11
typical ratios for isot"pes of a fission produc. in a given region of the core. Comparison o. the ratios obtained .rem sample data with these calculated values determines the source of the fission produc. releas ..
The fission produc.s iodine and xenon were chosen for use in this procedure by employing the criteria for selection of elements for which the assumption of equilibrium conditions is practical. Table 3-3 provides the results of theoretical calculations of the relative ratios of the isotopes o these ele,,ents when found in he fuel pellet, and in the gas gap. The calculation of the values fourd in the fuel pellet employed the ORIGIN computer code for analysis of fission product inventories. The calculation of the values ound in .he ho cas cao employea .he AhS 5.>>'tandard assumptions for the percent of the -uel pellet fission product inventory which enters the gas gap region o a rod in a fuel asse...bly with core average burnup. The values are stated as a range rather .han a specific value. The range is employed to account or iraccuracies inherent in the calculations and for the di.ferences in ccrc design among the C-c. llSSS's.
3-12
TAoLE
~my m q aPg a
~R qq +0yl P ~C~ ~
ACTI'/ITY RATIO Ill ACTIVITY RATIO ISOTOPE FUEL PELLE Ill'/E.'lTORY GAS GAP ItlVE!lTCRY Kr 87 0.2 ( 0.001 Xe ?3?a 0.003 O.CO?-0.603 Xe 133 1.0 1.0 I 131 1.0 1.0 I 132 O.G?-0.05 I 133 2.0 0."-'.0 I 135 0.1-0.""
,'lcb?e Gas Iso:cce Inventory se 13 nventory Icd:re Isc:cce Inventcrv
- i i~i inventory 3-13
3.3 BASiS FOR THE DETER.<IHATIOil OF THE QUANTITATIVE RELEASE OF FISSIO;I PROQUCTS The quantitative release of characteristic fission products is expressed as the percent o the source inventory at the time of the accident which is observed to be present in the sampled media and therefore available for irr'ediate release to the environment. The initial source inven:ory is theoretically calculated for equilibrium conditions. Prior to use, this value is correc:ed by procedure to describe the fission product inven.ory at the t',"..e o; he accident. The value of this inventory is dependent ucon the source of the fission product release which, as described in Section 3.2, mav be ei:.".er the fuel rod gas gap or the fuel pellet. The reason to define the quanti:y of released ission pr"cuct as that which is observed to be present in .he sampled media is a consequence of the limits on the present caoability to predict fission product transport and of the use of this infor.-..ation.
~na ir 'tical 'l models, or -,ission produc. tr nsport,ollowing release from a degraded reac:or core are not Cefini ively developed. Realistic estimates ard data;rcm actual accident case studies indicate .".at a smaller percent of the fission products is released -o th environment than is anticipated by the Regulatory models. This is explained by retention of otherwise volatile species within chemical reactions occurring in the degraded core, by oxidizing reac:ions occurrino within the water inventory present in containment, and finally by the plateout of non volatile species on containment building surfaces with subsequent reevolut-'on into volatile form. he information on condi.ion is required to rake real d'ore istic assessments of:he potential for radioac ive releases at the time of an accidert. These assessments should no 3-14
be based upon analytical mccels developed for worst case licensing evaluations Therefore , t .e 1
q~anti .ative assessments are defined n "erms of the amount of fission products ;..easured in samole fluids which are available for trans~or-to the enviror..-..ent.
This distinction is best explained by example. Consider the case in which measur ed samo les of .he h con.ainment building at'osphere and reactor coolan ant indicate that 20 percent of the I-131 isotope calculated to be in the cas aap is now fourd in the sampled luids. This does not indicate that 20 percent of the fuel rods havee be~ e n ru.ured. A greater number 'ay be ant.'cipated to have fa i led. Thi s nu.ber -b c nno e determined because the effects of oxidation with:n the core and plateout are not analytically kncwn. Therefcre it c~n only be statedte :hat.,'". t 20 pere r ~ nt of ihe gas gap source inventory is available for release to the en~iror....en'. Using the core damage characteris: cs deiineo
>n iable 3-1 this would indicate Intermediate Claddirg .""ailure.
The anal itic i al odels used .o determine the fission product source inventor;es are well defin d 'efined or equilibrium nor.-.,al operating conditions. The fuel pellet inveniory for he selec.ed characteristic isotopes are provided in Table 3-a T ese values were calculated usirg the ORIG"-ll computer code, Reference 7.6 ~ f with the assumptions of 2 ye r core average burnup and 100 percent power operation. The corresponding fuel rod gas gap inventories are provided in Table 3-5. These values were calculated with the assumotion of equilibrium dif,usion C
rates based upon average values predicted by AtiS 5.4 Standard
>>odels. The vaiues are expressed as ;he gas cap fission produc: inventory of all rod" s in' the aier ge .el asse...bly times :he number of fuel assem asse blies in the core.
3-15
TA8LE 3-4 EQUILI"RIU~ COR Il(YE !TORY OF CH'"'C:ERIST:C F'.SSIO'I PRODUCTS PLANT CLASS, iMWT ISOTGPE 1500 25".0 2550 2700 2!?15 3350 3500 Kr 87 1.8(7) 3.0(7) 3.1(7) 3.2(7) 3.4(7) 4.?(7) 5.4(7)
Xe-) 31H 2.9(S) 4.5(s) 4.6(5) 4.9(S) S.2(S) 7 0(s) 8.2(5)
Xe-133 1.5(8) i.'(8) 1.5(a) 1.5(8) 1.6(S) 2.0(a) 2.4(S)
I-131 4.8(7} 7.2(7) 7.3(7) 7.6(7) 8.0(7) 9.9(7) 1.2(a)
I-i32 ?.G(7) 1.0(8) 1.0(8) 1.1(8} 1.2(8) 1.4(8) 1.7(c)
I-133 1.5(a} i.4(a) i.s(a) 1.5(8) 1.6(S) 2.0(8) 2.4(a)
I-135 8.6(? ) 1. 3(8) 1. 3(8) 1.4(a) i.s(a) i. (8) 2.3(-)
Rb-88 2.9(7) 4.5(7) 4.8(?) s.o(7} 6.8(7) 7. (7)
\ ~(a'.4(7}
Sr-c9 3.9(7) 6.'(7) 6.1(7) 6.6(7) 7.0(7) 9.4(7) 1 Te-i29 1.6(7) 2.3(7} 2.4(7) 2.5(7) 2. 5(? ) 3. 1(7) 3. 7(7)
Te-.'32 7.0(7) 1.0(8) 1.0(8) 1.2(:) 1.3(a) 1.4(8} 1.7(8)
Cs-134 1.1(7) 1 9(?) 1 2(7) 1.3(7) 1-8(7) 2.4(7)
Ha-140 8.0(7) 1.3(8) 1.3(8) 1.4(8) 1.5(8} 1.7(8) 2.1(8)
La-i40 8.4(7) 1.3(8} 1.3(8) 1.4(8) 1.5(8) 1.8(8) 2.1(8)
La-142 1.0(8) 1.5(8) 1.57(8) 1.6(8) 1.7(8) 2.2(8) 2.6(8)
Pr-144 6.5(7) 9.1(?) 9.1(7) 9.6)7) 1.0(8) 1.2(8) 1.4(S) 3-16
TABLE 3-5 EQUI ! BR!L,'0 GAS GAP I1'tVE (TORY QF CATARACT"--"'ST',C F SSIQPRQQ PLAl<T CLASS. ".WT 150702= 1500 253Q 2550 2700 2515 35Q 3200 Kr 87 3.6(0) 6.1(Q) 6.3(0) 6.5(0) 7.Q(Q} 9.5(0) 1.1(1)
Xe-131M 2. 7( '} 4. 2(4) 4. 3(4) 4. 6( ') G( '} 6.5{4) 7.7(-')
Xe-133 1.3(7) 1.2{7) 1.3(7) 1.3(7} 1."'(7) 1.8(7) 2.:(7)
I-l~l 4.-'(o) 6.6(6) 6.7(6) 7.0(6) 7 (6) 9.0(6) 1.1(7)
I-!32 4. 9(3} 7. Q(3) 7.0(3} 7. 7(3) 8.-'(3) 9.9(3) 1.2(')
1-133 4 4(6) 6.2(6} 6.7(6) 6,7(6) 7.1(o) 8 9(5) 1.-(2 )
I-135 7.0(5) 1.1(6} 1.1(6) 1.1(6} 1.2(5} f(c) o(
3-17
0 ine t Lbuiated
~
values of fission product source inven-oryJ are fo r ecu'.'1 ibrium nor...al operating conditions. The required information is the source inventorv at the time of the accident. Therefore, these values must be corrected to account -or the history of the core up to that tire. The specific parameters which must be accounted include the core power level and average fuel burnup.
To account for variations in core power level under the condition in which the pcwer has been maintained for suf icient time to allow the charac. eristic fission produc to reach equilibrium requires only a simple power ratio.
Within the accuracy of the sub'ec. procedure it; is established that a time period of 'al; livYes 1 l5 suf icient to achieve equi ibrium conditions.
1 For those pcwer histories in which ecuilibr ium conditions do not exist a transi~.-
analytical correction is provided in the procedure. Oerivation o= the transien. correction equation is provided in Appendix B.l.
I Ir'pie...entation o-. the subject procedure under post accident conditions necessitates simplifica:ion of data analysis whenever possible. This c n be achieved .hr ugn appr"priate selection of the characteristic, ission produc.s as described in Section~c.ion 3.. thereby avoiding the need for use of the trarsient pcwer correction equation.
"Th e characieris ic fission products are divided into two groups based upon their respective half lives. Under those condi ions in which core power level has been maintained constant for a period of t>me greater than 4 days but less than "0 days ys . h en th e rission products in Group 2 should be employed =or analysis. Under those conditions in whichh the 3-18
core power level has been maintaire 'ntaired ccnstant fcr a ..-..e percd greater than 0 days tl en the fission p r oduc.s in Grouo 1 should "e e.i. loyed for aralysis Proper selection of he fis'ssion product Grcuo resul=s in equilibrium inventcries which do not re quire uiro the transient analytical correction.
Selection of the a pp ro p 'e ria iission product grouo re uires a Ce>erzination o=
the period of constant core pcwer. >. in Within th e accuracy of= tl e subject prcce- e-Cure, the acceptance criteria for ccnstantn power is a variation of =10 percen-frcm the ti;,.e averace va1 ue.
The final aral y ti'cal 1 correc .
~ 4 ons which mus ma e tto be r.aC t,e t i ss i cn pl ccuc release determ>ration are the correc.. on or the samole,",.easured saiue .o account for reactor shutdcwn and pressure decay frcm the ti;,.e thee sano and the of the anaiy correc:ion ed samole
.ion for san 1 e was analyzed back to or thee ddiffererce and n ~h
= e att o; the fluid prior .o
'., he e between the te.ceratt i;..e
.,ur re."..oval of r~
frcm contairment.
The Post Accident Sampiing Sys em loc~- for
= liquid 1'cations and casecus sampies are anticipated :o be Cif-.erent -,or each plant o obta>n .he,.os. acc ra,e
~ ~ 4 assessment of core Ca.-..age, it is recessary rec ~ to sample and analyze radicnuclides frcm at least the p rinci "al lcca.ions ial 1 which include tl e reactor.or coolan coo an.
system, the containment buiidinc u-,.p, and su;,. the containment bui ldirg atmosphere. Other samples ma J be
~ b .ar.en Cepercent upon system capabili"ies The measured specific act vity of each ac sama le is related to the total quanti ty
'ht s
at each s a..p m le ocaticn.
1 The sum o- these cuan 'es is en cons>derec to be t"e tot~i t..e total quanti y available for reiease to the env enviror-. i ror-..en t. Typica ly 1 3-19
several hour s are reouired .o recirculate, obtain, and analyze a sample frcrm each location. Therefore, Tf. the sample location to be used during the initial phase o. an accident shculd be selected based on the type of accident in progress. i(nowledge"of edge o a specific accident scenario is not required. The initial sample location can be selected based upon known pressure, tempera:ure and level indications obtained from the plant control room. A list of the appropriate initial sample location is provided in Table 3-6 for various accidents should the t scenario be known and for various plant conditions should the scenario be unknown.
The me sured values obtained frcm the Pos Accid en t S amp 1'ing System are expressed as the s"ecfic ac.ivity of the sarple fluid. To obtain the total quantity of the fission p r cducis d at each location i t is required to knew t,".e quantity of sample fluid at that location. This infor...ation is cbtained fr"m the control rocm and includ 'ludes the reactor coolant system pressuri=er and reac.or vessel levels , the rre >-c.or coolant pressure and tempera:ure, the ccntainmen. buildinc sump e el, ard .he containment buildinc pressure and les 1 This is tf the sare information which is used to select the 4'emperature.
initial sample location.
3.4 GE FERAL CONJS iDERATIGiiS 0'i THE L1t'1ITATIOtlS OF THE PROCEDURE Considering that ideal ccnditions will no'xi - -h e i exis-.t subject b procedure is th measuremeni based u p cn the 4 of as -,any parar.et~rs as poss'bl i e.. Th e core damage assessr ent procedure is anticipated to yield a combination of categories which may exist at the tim 'e a given sc pie is .a~en. individual reasurements may cpear to be contradictory. The user is required to exercise krowledgeable 3-20
~udge...ent in he interpretat on of the liinitat;cns of the procedures capability to evaluate a oiven pieco o n or:.ation. There ar nurerous 7
sources of error in the interpretation -ion oof such infor.ation including .he
'on o, iss.'
determination
~ product inventory th e ro
. Ce l s or fiss on prccuct transport out of- :he ..
" , "the system
" core, ca pa abil'ran i ity to obtain representative sarples, and the system capability to analyze the samples.
3-21
Thl)l.E 3-6 SAHPLE LOCATIOffS APPBOI'RIAIE It)ft COBE I)hffhGE ASSESS) lEffT ACC IDEffT SCEINRIO I:ll0lfll RCS IIOT LEG RCS PRESSI)lt I ZEB . 'tff ff'l Coll TA IffflrftT Coll Thl ffflcrfT flOSf'ffEftE Sf IUTOQWf(
COOL IrlG SYSTH)
STEhll GEffEBATOR SECOfll)hltY Smal I Oi eak t.OCA, lteactor I'ower >1l, Yes Yes Yes Sma11 Break LOCA, Beac tot I'oozier < I'X Yes Yes Yes SnialI Steam Line f)reak Yes Yes Large Oreak LOCA, Reactor I'o)ver >IX Yes Yes Yes Yes Large f)reak LOCA, fteac t,o) I'oiler < IX Yes Yes Yes Large Steam Li>>e f)reak Yes Yes Steam Ge>>erator Tube ftuf) tui e Yes Yes
~
9
TABLE 3-6 (Cont. )
SA11PI.E I.OCATIOIIS API'HOI'RIAII: fOB COBE MI1AGE ASSESSIIEI(T SIIuTOOWII STEAN ACC I OCtIT SCCIIAR 10 ACS IICS COIITAIIIWCIIT C011 Th IIIIICIIT COOL IIIG CCIIEIIATOn IIIII:I;01111 IIOT LEG PIIES IJIIIEEII SlllIP AllIOSPIIIBf; SYSTEtl SI:COIIIIht<Y SIS hctoated Yes Yes Yes Alarm on Containment Ani liii>>g 14dia tion 110>>i tot Yes Yes Alarm o>> CVCS Le tdovin Aadia1.io>> Honitor Yes Alarm on Containment Building Somp Level Yes Yes
I e in -r - -r-c-cur- :5 base~ on -he conparison betwe-1 -easur-d samp>e da d'or a obtained unCer "ost accident conoitions and analytically Cetermired values or
>n~en.ory
~
'ource a ~ >>he tl.-.e o. the ace>dent. Therefore, >>he princic:pa"al consider -.;on is the rodel of the c::=. ac.erist c, ission products in the =uel -uel prior to cladding rupture. The two significant factors are the fuel power history and the power density. The fuel pcwer history determines the fission product inven ory in he fuel pellet. The power density determines the fission prcduc. migration behavior within the fuel.
Calcula:icos o=; .he uel1 pelle inventory under the eouilibrium ncr.".,al o"eratinc cond:ticns using the GR!6="li computer coce yield reliable data.
Par .-..etric evaluaticns o= the acceptance criteria for determinirg if the pcwer is:cry sat s=ies equilibrium condi ions based upon the hal life of the charac. eristic fission product are accurate to wi>>hin 10 p~rcont Therefore,
~or >
this>>ec..nique is ccnsistent wi>>h the intended purpose.
Calculations o the gas gap inventory is less reliable. Fission product rigration to the gas gap is dependent upon local power density fuel burnup fuel rod te.,perature gracient, and chemical reac ion ~ith other fissicn products or with the cladding. The gas gap inventory can differ greatly anong
.he individual fuel rods in the core. Therefore the procedure dces not a..emp. to predic a speci-ic number of fuel rod failures but ccmpares the cuanti.y o; fission products released against the entire core gas gap inventor' The core average cas gap invertory can be calculated wit" greater rel iabil ity.
3-24
~ nu.. er or c.her ac.crs:r.;luence the rel >abri i,y
~ ~ o- the ch equi s t ry sar;,o i e s upcn which the pr"cecure is based. Reliab i 1 i.y '1
" is in; luerced . t;, ab i by the 1 '.y i:y to cbtain representative s r.".pies due to incc-..pie-.e .ixira
. ixir of t"e t" fiss:cn prcducts in the large liquid and gas volures, equip;,.ent linitations, ano lack ot opera. or familiarity with rarely used prc ce~d ures. Th e accuracy achieved in the radiolcgical aralyses are also iniluenced ence b y a nu;.. .-.b er of iactcrs. The equip;..ent employed in the analysis ray be sub'ec:ed to hiah levels of radiaticn exposure over ex ended periods o i t'ice. Ch emists
.. are required to exercise ccnsicerable cauticn tto ;,.ir i"'-o 'he
...'..I~e . e soreado d d'.
oi= radicact;e r;.arer als.
Sarcles have:he "c:ent;al c= "
ct ~eing center.:;nated by r .-..erous scurces and -..ay not resul:;r".". a uniic~ distributicn or the s -.p'e f'u'C. C ol irg or reactions cay take place in '..".e lore sal-,.pie lir. s. Tnererore, the resul:s obtair.ed may rot be representative oi plant cc"di-'cns. To ."..ir.;ai=e these er-,ects ".ultiple sar.pie analysis over an extenced ti."..e period is e".."'o e~
nddi ilcnal ly, upon cc."piet cn Q t! e seccrd "nase 0 his tasK r ce~'I ~s will be available to assess core da;..ace using the balance oi plant irc'c~- 's which include core exit te...peratures, the quantity o, nrdrc~en rele sed =rcn zirccnium degradaticn, and ccntainr..ent rad;ation loni ters.
s a result or these ccr.sicera.icns, the assess-.ent of core damage is cr'yI an esti;..ate. The techniques employed in this procedure aro on y accurate to locate the core ccrdition within ore cr vore o= ..e Pe teen ca t egcries of core dar;.age characteri:ed in Table 3-l. Hcwever, tnis is suificient accuracy ro allcw plant cpera:ors to .-..ake inior;..ed decisicns under <os- "os accident plant ccrditions.
3-2.5
4 4.0 E T"BL -:-:." .'fT QF Tr c "ASIS FOR CORE O><<gG"- pSc-"Sci1r~(- U<" G HynqC -.I E.~SURE.".E.'I T This sect on discusses the sources of hydrogen gas during severe accidents, tl e amount of hydrccen predicted in the coolant and containment and the relation between the measured amounts and the core damage. There are multiple sources of hydrogen during severe accidents, including oxidation of zirconium in the core, oxidation of various containment materials and radiolytic decomposition of water. The discussion evaluates each of these sources and presents ;..ethods or det ruination of the amount of hydrocen which is generated by core oxication.
The amount o hydrocen generated is related to the category of core damace through analyses of selected accident scenarios. The analyses predic. the procression of core heatup and oxidation throughout the core durinc core uncovery accidents. It is shown that the amount of local oxiaa t cn is rela:ec to the local temperature and therefore to the time of clad rupture. Also, when the amount of local oxidation exceeds the oxidation threshold for clad embrittle,.ent, clad fragmentation occurs. By summing the local oxidation throuchout the core, the total core oxidation and hydrogen generation is obtained. This total is related to the type and amount of clad damace and therefore to the ten categories of damage.
Clad damage determined from hydrocen is expressed in two ways - as the nurb~r of,uel C
rods which are ruptured ard as the number which exce d the oxidation ri:tlement threshold. or rods with ruptured clad places the e... Tne nu.--oer
0 damage estimate in one of categories 2 through 4 of Table ~-1, which is he equivalent of Table 3-1 for radiological charac. eristics. The number of rods embrittled represents the nu'ber which structurally ail~ and which release fuel pellets into the coolant. The analyses indicate that the oxidation embrittlement:hreshold is reached at about the same time as the clad melt temperature of 3350'F, placing embrittled rods in the equivalent categories 5 through 7 'or fuel pellet overheating. The measurement of hydrogen is not t
utilized to place the damage in categories 8 through 10 for pellet melt because definitive relationships are not available between the amount of hydrogen produced and such severe core degradation.
The placement o; a given reasurement of hydrogen generated into a speci "c Carage category is dependent on the scenarios analyzed to relate da-.age to
'd i oxidation. Selection of base conditions for analysis is discussed in the following section.
BASiS FOR S"-LiCT10i( OF HYORCGc.'< TO ASSiSS CORi DANGi Hydrogen is produced in the core by the oxidation of zirconium in the =uel cladding and other fuel asse..bly cc'ponents according to the reac ion.
2 H 0 + Zr irO2 + 2 H 2
4-2
Table 4-1 Clad Damage Characteristics of ltAC Categories of Fuel Damage HAC Category Telllpel a t.lire l tee lian i sm Cliaracteristic Heasurement. Percent nf of Fiiel Damage Aanr)e ( F I 0i ()dlll3rQ lleasrrrerrie>> t. Aange Darrrage lhids
- 1. llo Fuel Damage ~?50 I!one Less Than I
- 2. In i t. ia 1 Cladding Aopture Oue to Haximrrm Core <1550'F* Less Tlian 10 Ia i lore fias Cap Exi t
- 3. Intermediate 1200-1000 Overpr essuriza tion Thermocouple <1700'F* 10 to 50 Clarldi>>g Fa i lure Tenipera tore ttrjor Cladding <2300'F Ia i lirre <2i Crea ter Than 50 Oxidation
- 5. I ni t ia 1 Fuel Pellet Loss of Structural Amount of Equivalent Core Less Tlian 10 Ovei liea ti>>g Integrity Oue to llydr oge>> Gas Ox i(la t ion Fuel Clad I'r*orlrrcerl <3$
- 6. Iritermediate 1000-3350 Ox idat. loll (Erl<<ivalent to <IBX I <<el Pellet, i Oxidation of Core)
Ovei liea t, i>>g
- 7. ll,ijor Fuel Pellet <65K Crea tei Tlr(rrr 50 Overliea ting
", Oepenris on Reactor Pressure arid Firel Orrrrrrrlr. Vairres Give>> for Prr.shirie <1200 psia anrl IIurnrrp >0.
0 ibis reaction always exis s in the presence of wa-er but at nor;..al
~ -a o pe ra tl ng temperatures the rate of reaction is very small. Hfdro g en is also genera <ed radiolytically and the amount usually in the pr..'mary system masks the hyCr-cen generated by the oxidation of irc"nium, nor...ally discernible by measurement so that clad oxidation is5 no-of the hydrogen concern"ratioon.
the coul valent normal oxide thickness accumulated on the cladding over the 1
1 no>.
'l Typ i ca 1 ly, entire li,e o> the fuel is about 0.0004 inch or 1.5 o; the original clad thickness.
The nor.-.al maximum temperature reached at I he clad a =
sur;ace is 1'.
limiteC ': to a few Cegrees above the saturation temperature by nucleate boiling. At 2"50 ps'.a ys and the typical pe k heat flux, the dens-Lottes correlaticn predic: maximum sur-.ace temperatures of about 6 decrees above the saturation temperature of 653'F. rwo abnormal situations may be hypothesized which cause hicrer te...perature and hicher clad oxidation: 1) Oeparture from Nucleate Boilina (Dl>B) mignt occur while the core is covered with coolant; and 2) the core mav uncover and hea- up because the result ng steam cooling is inadequa"e dna equa .e.
In the tirst situation, O'8 may occur Curing transients which are initiated from outside the Limiting Conditions or Opera"ion (LCO, t ransients for which multiple failures occur, or transients for which deviations from the assurptions in the Reactor protection System setpoints occur. A sur...ary of event types which can result in D>lB is included in Reference ?-?. ORB is localized and is expected to "e of shor: duration. Ter>,peratures ar~e below about llCO'F and:he total hydrcgen generated by oxidation is too s..all to be c"served. A detailed discussion of fuel rod behavior Curine C'l8 is given by
er Rererence ". 7-0. The PAS
' is no in ended for assessment of damace caused b such events, except >>hat if clad rupture shculd accc.-..:any D s9 an incroase in coolant activity might be observable.
The second situaticn which can cause high clad temper.ture is the main subjec-of this sec.ion. In this sit"
't.a .ion, the reactor is tripped and pcwer is fission product decay only. The fuel is adecuat e Iy coo I e d as I ona as the core is covered with fluid, even without primary co o I an t pump.".. flow. In order to uncover the to p of the ~ e core, over 70. of the primary coolant mass must be Ios:. This can occur only ~ith the catecory of even v n s k ncwn as 'oss L of C"olant ~cciC n.s (LCCA). prese events are divided a-..or.a three grouos:
large breaks, small breaks, and ccmplete loss or he~t sir" Larce breaks are .'" ar. c.e. z d by very rapid blcvdcwn to con"air.;,en: proc~"r~ resul virg in '.otal core uncovery, follcwed by rapid rerlood by the Safe:v j Injection ank (SIT) flew-, and Tank continuation of heat removal by i:PS I ard LPS; f1cws. If ref e lood ood does d no. occur the core will heat up adiabaticai'iv. I, rerlood does occur, but continued ccoling flew does not occur, the coolant progressively boils of= and uncovers the core again, similar to bollof durin a small break, but at a lcwer pressure. eaks are Small break c..arac.eri-ed h by rapid blowdcwn to saturation pre<sure at "h temperature of the seccnc ry side or the steam generators. This is followed by continued reactor coolant flow out4 ofI "he break and by doer~~ r as rg pressure. Both are dependent on break size. If more than 70" .of the ccolant is lost, the .oo of the core uncovers. Design ~uncticninc of t'.".e:".iPS
pressure falls lcw enough, of ref lood by t.".e SiT, would recover or preven prevent uncovery of'he core. In the unlikely event :iat they do not function unc.ion, the "1 e coolant boils o f and the core progressively urcovers. Extensive discuss-.'on of various small break scenarios is given in C="-II' an I'
=i.-l<~, prepared CE" -.or the GEOG (References 7-9 and 7-10).
A complete loss of heat sink results in heat up of he prirary s -, "h consequent pressure increase until the Power Operated Relief Valve (PORV) and/or safet y valves es on the pressuri=er open, releasing primary coolant. Ther ar Therea .er, er the scenario is similar o a small break, except that core urcovery occurs at a much hicher pressure. The accident at THI-7 was essentially the same as this scerario c"-..bired wi . e er-ects o- various with th operator ac ions. The preceding discussion touches on the variety of conditions which might acccmpany core uncovery and the consecuent heatup and core damace. Essentially all these conditions are more severe than those calculated .or Cesign Basis Accidents and approach the conditi-.-s called Decraded Cor~ re ~ Substantial equipment mal,funct'ons and opo~a-or rrorss are erro a required to achieve .hese conditions. In order to es ablish a procedure for damage assessme t, one set o o'ne core cond .icns is selec:ed as a basis ~or -ua, ii th relationshipp between
, ~he a...cunt o hydrogen generated and the core damage.
O.h O.her possible scenarios are evalua: d qualitatively relative to this base. 4-6
h
~s stated previous Iy, the core is adequat~l --' ed -fol 'l cwing ie y cco >
reactor:r.'" < t;ons for eva u<< 1 Ua tion of damage are independent of that a -- o;= -the scenario port;on .
which. prec ces es core uncovery. All scenarios leadi, g i damage di;fer ;..ostly in the rate of core uncovery. For a small break t core uncovers by ak, the boiloff and possibly the effects of misoperation of prirary ry pumps. For larcer breaks, the core uncovers by blcwdown. Since the most s genera 1 situation to occur is that equivalent to boiloff th e following assumptions are made for the base conditions:
- 1) ihe core unc"vers by boiloff at constant prossur~
- 2) After boilof; down t.o a given coolant level belcw he top of "e tenina
'he even: is ed b y rapid, ccmplete core recoverv wi 0 rela='ittle addi:ional clad oxidation a-ter the minimum level is a:: inca.
- 3) The prccecur ;or es e .i;.,ating core
-'---'o damage is implemented after core y.
recovery. The ti;.. ti...e it takes to obtain and evalua e the O'S" a...ples is lono compared to he lithely duration of core urc"very. 'lever:heiess
%~ g Q~ ~
procedure may be used with hydrocen samples obtained earlier. Hcwever .ci,eier, these earlier scmples sar. would probably provide a lower limit es i'ate of cor~ ~darace, since the hydrccen contained within the voided current cor pri'ary system would not be sampled. 0-7
C i e progression o- core uncovery d;ffers rom these assumed ccndi ions l then tht e amoun. o,4 core damace in-erred frcm a given measurement of hydroce.. may be biased, depending on the uncove.y rate. Two examples are: A large break LGCA may cause rapid total core uncovery followed by almost adiabatic heatup of the urcovered core. In the absence of steam to oxidize the zirconium, hydrogen generation is limited, but fuel overheat ard melting occurs if cooling is not restored. A subsequent measurement of hydrocen yields an underprediction of core damage for this case.
- 2) ~ s.,all br ak LOCA uncovers 'more slowly when sore liquid enters the reactor vessel frcm the E...ercency Core Cooling System (""CCS) or ,rom run"ack of condensed steam via the hot legs. The h dro g en is g en e r.ated by -.cr extensive oxidation along a smaller length of fuel clad;near the top of the core. A measurement of total hydrooen generation underestimates the extent of local oxidation on the damaged rccs.
The base conditions selected for analysis to prepare this damage assess."..ent procedure are boilo;f withcu inlet flow. For a measured total amount of c"ro oxidation, the base conditions yield a lower limit estimate of the number of damaoed uel rods La .er sec.ions of this presentation qualitatively relate .he h results of other scenarios to the hydrogen cenerated during boiloff ana irdicate what additional instrument indications may be utilized. In conclusion, the use of hydrccen provides a good indica",'on o damace for boilo, cond',.ions and a lcwer limit irdication for other sce..arios.
~yp 'QRr >, '8 z 4.2 CL 0 1 01 qyD RCGUCTIQ:<
In th>s sect~on, the relaticnsh>p'p i s es.a lished betueen measured hydrccen a .d core damage ror the base case conditionn o- oilor. <<t constant pressure. Simplified analyses are used to demonemonstrate the relations among the parameters and the applic'bilit 'tg o- tti e analyses to all C-E Cesicned reactors. Host previous analyses of ccrc uncoverv were dor onee for
- f. speci.:ed 0esign 9asis Events with an "lRC approved Evaluation I"cdel o veri= riry ih a~ h e imits t of IOCr~=0.-.6 are satisfied. Those regulatcry d esign b ases 1 mits are a peak clad temperature o f ~~~ocF, maximum 2200 local clad oxida ion of= I I -' an total ccrc
'x.Ca ~ on of I . These limits and the capabi li'.yy o= ...e analytical roCels aro
~nacequate 'or prediction of the severe core ccrditicns for which :he PASS is Cesicned. enc,o ihe relationships among clad dara ge, h y d rogen generar.icn ard Hen .
other parameters are obtained with new an a 1 yses. Thhey~ are less det iled, less rig;d analytical rocels intended to give most probable or "es es-'-..a e resui s with more severe core ccnditions. Th e results are c"ns Cer d adecua:-
.or .he sccce or damage assessr:,ent needed durirg an r cr ss.
acciCent in prccross A Cet<<iled analysis o these "C ecraded core" corditiors uould requ'e state-of
'he- .he-art computer codes wnich properly mcdel all th e in t erac. ve physical phencmena. Such analyses the darage assessment are beyond the scope of procedures provid e d h erein. 'h effort for preparation e
analyses 1 e...ployed o= or
'n in the development of this prccedure are suffici icien t .o provi'C e a b'asis b of decision on for implementation or the s>t emercency ac .'on ion p an. A few' etailed analyses 1
are included herein as a qualitative overc> overc, ec< cn the general pred ctions of the simplified analyses uhich are the ~ e ases bases , orr -" the C<< Carnage assessment procecure. 4-9
Hvdrocen 'Generation Correl ation The fiss on produc. decay power produced in the exposed length of core above the '-phase level is partially removed by the flowing steam cenerated by he covered length of core. As the core uncovers, less steam is generated and at the same ti'e a longer leng"h of fuel is exposed requiring cooling. Consequently, the temperature of the exposed fuel rise rises as th e core uncovers unti 1 it is high enough to cause significant oxidation heating. As the fu e 1 clad temperature rises the zirconium oxidation rate increases ihe oxida '.on reaction is exothermic, and the ccrbinaticn of poor cooling ing, 'ission ,ission "rcduct decay power and an exothermic reaction causes an acceleraticn in the fuel rate of temperature rise as temperature ircreases. Hydrogen gerera:ion is therefore sicnificant in the predictions of the rate of Carace. essentially every calcula:,ion of clad oxidation performed with LOCA analyses at C-c is done using the Baker-Just correlation, which is the approved 'ethcd in ihe h licensing evaluation model .. I is based upon li...i:ed early experirental Cata and is del iberately b 1 biased .o assure conser;atively high predictions ot o= clad oxidation at terperatures up to 22OO'F. Oxidation rate data which are currently available with steady state and transien t ns en'mperatures are reviewed in Reference 7-11. It is concluced that the Baker-Just equation yields substantially higher oxicaticn than ac ual at stead ea y st - , s ate, an d t'll higher still h wnen comp;red with transient experimental da a. B ase d on the th steady data available, .he referenced " Oc<en, c d au...or a 0 ! recommends a correlation to replace Baker-Just. 4-10
0
- e. ab I e urcerta>>ty re,.a ins, even in .i c cken corralation, because tl exper mental Cata do not represent ac:"r o"e.a:
reac-- nc and ac dent conditions. A pressure enhancement t effect, i ., 'oror examexarole, a is reported o ; . '. which. increases es thee oxidation ra:e at hi'g'h rass5ures for ter.".peratures up -o "COO' Al so, rar;.p heating experiments yield lower wer oxidation than t -: .;, trans ent ca 1 cu1 a.ions at i ons whi w ic. ch utilize correlations derived from steady s- " s ate ox idat'.on d data. For O'AR type transients and linear ramps, the i. e evperimentally ex reasur d oxidation may be -'.G. to F5" less than the equivalent ca 1cula.ed oxi Cat, ci GA. GA A
-' of ti-..e funct;cn ~
correiatior.s. The
'n
ccr.:parison of the percent
'-.. is civen of ecuisal ent Ficure ~-.1 c 1-ad or tra cri'g in al clad thicxness is a typic i vaiue of 1
thickness oxidi:ed Bakk r-- 'us-
'u ar -,e ar" -".e as a ckann Cck 'rc.-.
~t temperatures above about 2020'F 2 -> " k . c"rreiat'.on
, . e Ocxen "racicts .. pro- o-gressiveiy smalier oxidation. var.".p e, wnen .he clad temoera: re reaches For evar.".pie 1 2500'F during a linear ramo ta' t,oerature rise, 2aker-Just "reoic:5 -' "."re mor e oxidation than Ocken.
Recall that the objective here is to rela ie ""a ..e P'~S s S measure..ent of he to ~ amoun. o hydrccen generated .hr"ughou the course of an a cci'C ert to an assess.-..ent of coree Ca...age. Ca...a The aroun-
- otf oxidation calcula.ed at aAy any ins"-n"As tant is not so ir;.portant as the relative dis .ri'b u t ion ot oxidation ar.'.our,ts on claCding throughout the core because he "iota "1 ." - is amount available by the ,e measurer.ent of hydrogen. Tha rela",'sese d'is.r;bution, for a civen total amount, has less calculationa) uncertainty ~ hann the ~ ~ e loc'1 value 1 as a function oi <il e ~ Hl so, the intent .'s .o pr"viCe a rea 1 1i 5 ~ c assess.,er.: to serve as a ba515 f<<er-er,r a ncy plan act ors rather -qaqn =- asis -or plant design and c=nsing. Tne 9aN.er-Just eguat on a5 disc"ssed above is vora ~ .ore appropriate app for 4-11
FIGURE 4-1 PERCEiVT CLAD OXIDATIONvs TEAIPERATURE FOR LINEAR RAte1PS FROi%1 417 F
~~
20 OCKEN CORRELATION 1'/SEC 10 r 10'F/SEC ~ I +5 20 C 0 1.0 U 0.5 BAKERJUST CORRELATIOiV r 10'F/SEC r 0.2 r 1500 2000 2500 3000 ATTAINED TEi>IPERATURE ( F) ON RAMP INCREASE 4-12
e conservative design and licensin g ac tiv1t,es, Therefore th;5 procedure for damace assessment utili=es h ydrocen dr r,'.easure.ents "ased uoon "he 0cken values of par r.eters in the usual equation for ox'd"ion: claC ox: a alon: w2 -8/RT A e Zr where: w = Equivalent zirconium mass oxid -ed per i ize " uni- area A Zr
- 3. -. 3 x 10 5 (mg Zr/cm 22 ) /sec 8 = 140.6 (kJ/mol)
Temperature ('k) 1.c87 cal/mol Time at te...perat.re T (se .) The oxidation p rod'dic.ed by the previcus e "a='on
='on can c "e expressed as a percentace or the original ciao thickress, <ress, fn "-nclish units the cquisalent oxidat cn thickness is ~ r ~ <Q4Q/ I ~Sr
~here: x = percent oxidation of th l a d thickness ar = Clad thickness (,"..) T = Temperature ('R) iime at tempera.'e (sei 4-13
Oxidation cf 2ircaloy causes embrittlement such tha" s rac- uree s may occur u-on r rerlood and quench. The amount of oxidation in the above equation is expressed as the equivalent thickness of clad that would be convert>" to ox'ce if all the oxyc e n absorbed b and reac==- with the clad were converted t stoichicmetric -irconium dioxide. As se~n previ 1 y, th e oxi d a"ion rate is
"'us strongly dependent on temperature. Test data (Reference 7-12) for specimens slowly cooled to 1520'F and then quenched, indicate the best estimate amount of oxidation to cause embrittlement is 28" for oxidation at about 2600'F and greater than 2S at lower temperatures. 8elow about 1800'F data indicate essentially no erbrittle...ent for tires up to 5 hours. These da .a are su.-..imari:ed as, ol cws:
1 Time at Temoerature Temcerature ('F) Pec ired to "..;.brittle (5ec.) 1880 10,CCO 2060 2,OCO 2240 joo 2420 2CO 2600 20 When the clad is rapidly quenched all the way from the oxidation temperature e...brii.lement occurs at 20" equivalent oxidation for temperatures above 2500'F and at higher values for lower temperatures \ ~ For ihee anal yses ana h, here, the h va ue of 20, is selec.ed as the equivalent oxida ion reouired to embrittle the clad 14
Ine i{RC cat~r o r'c . s o.= = .e',
~ ~ ~
dar;.~ce are charac.eri- c in Taole 4-! acc"r" ~ the extent of ..el c! d e.;.brittle...ent due to ox cation. The correlation developed by Ocken is employed to relate t.. e~u'"al~.ent- ;ass the equ::a .-ass of zirccniun in the fuel rods which c....as ha b o.n oxidi-ed per be unit sur;ace area to the terper~'uro
.er.". er~ uro of the clad sur;acece and ... tir;.e at that ter:.perature. Hcwever, the core da.",.age assess;..en: prccedure is based upon a -.easure...ent of ti;e quantity hydrogen produced during the accident. The quantity of hydrogen produced in the fuel bg t k e oxidation of the irconiua found in the cladding and o-"er o..er ..po e ..s asse."..blyy cc.-.ponents is ass e.~.e P i.o 4e stoichic."..etric accord no to the chenicalof'alues re c:icn discussed~ssed in e .Icn . 1. Usirg this chemical reac-'cn an" t"e k"" ~n -or the plant s=ec;r'.c m me" "e -ourd in -he ~"
c ar.:;.v o~ e rc"n
~ AH ~
u-.. I eV ~
~ f t.".e quant't es of iydrccen src~n in Table 4-2 are calculated as a urc-ic" o-
- ..e percent or to: 1 oxicat.'cn assu;..eo. The purpcse 0- this table is to de...ons.r .e the sicniricarce or ti:e quantity of hydrocen pr"cuced V in deer c r aci'den s. L ire su"'ec= procecure is ince enoent of acc;cent scerario =nc
...r rore degraded core events are pcs.uiate J:n chico substan" 1~'rac" c" s or ..e .o.al inventory o- zirccnivri is oxidized. Because the reaction is assLt ed .o be soichic",.etric ti e quar t; ty of hydrccen produced is linear 'Ii "h the percent of zirccniun ox d zed. Tr. erector a sincle plan. seci ic value is er..ployed -,rori Table 4-2 to relate ti;e quan i y of hydrogen pr"duced o -1 e percent o- zirccniun oxidi= d. That value is the quanti y of h y d rogen prccuced per percent of zirccniun oxidized listed in the first colum o= -i e
Tr'BLE 4-2 gUnllTITY QF HYDROGE'( RESULTIiiG FRCs'1 OX IOnTIGH OF ZIRCCNIUN IH THE (FT AT STP) PERC"-lli OF Z'. RCC>l IUN OX I 0 I Z" 0 10" 50~ ICO". Calvert Cliffs Units } E 2 4.23 x 10 3 4.23 x 10 4 2.1'1 x 105 '3 x 1C" Palo 'lerde ')ucl ear Generatinc S:a:ion 5.65 x '0 5.65 x 10 2.83 z 10 o ~ oo x 10 St. Lucie Unit 1 4.21 x 10 4.21 x 10 St. Lucie Unit 2 4.64 x 10 4.64 x 10 2.32 x 10 <.64 x 10 SOl(GS Units 2 8 3 5.07 x 10 5.07 x 10 2.53 x 10 5.0? x 10" 3 'rlPPSS 5. " x 10 5.65 x 10 4 2.83 x 10 5 5.65 x 10 waterford Unit 3 5.04 x '.0 5.04 x 10 2.52 x 10 5.04 x 10 4-16
0 2 2 enalvs js o r~ol>wt J eve) The raie o, level dr " and ".e lcwest )eve) at .=;..e~ ='.e~ ".v,.e two-pnase c"olant in the core are he .-..os. 'i-icant para...eters in Ceteminirg "4e core heatup sicnifica and subse"uent oxidation of tl e fuel ue cc)ad a . It is shcwn in this secticn that the coolant level c'n be predicted c e as a ffunction of tire fcr :he case of ~ boiloff with no inlet lcw. It r-"~ shcwn that nor."..ali-~d re su is is ffurther 1 OT this analysis are app') icable o a all C-"-c reactors. Af.er Af > react". tr o, .he core is equate)y aC~~ ccc)ed to preven" oxiC'.icn un-' ~
.he lejel Crc"s "e)cw the tcp of:he ac:ive c.ive rue). i 1 ':.". ..r~~ 'r/i<<h the core " s: c"v r~~
at steacy stat .:.e Fission prccuc: decay pcwer is e~uai o ".o s"-. pcwer to ra i se any ini et ~a ter to s a.ura ion plus the pcwe. to va."cri=e ) ic io saturated tean. The aroun t of core inlet flew requireo to ."..aintain a s covered ccrc is Cee> ~ nc n. on the Cecay power level, the pr~~~ur~ a"c:.-e inlet t~...perature. Ficur~ i g, gives the requir eo flew ra > ."
~ - .~ kaeoo ..".o corn ...o cors covered as a furc icn of Ceca ecay tire, over a rance of corcitions fcr .he C-~
desicned reac.ors. The basis or the clad ad Ca".ace Ca,".. assess-..ent procedure which "cs- u ) a -ees Cegl aded core conditicns assuries zero inlet cco) c nlet ing flew. This is ecuivalent to 4 unlike)y events of inccperab)e ciCS and chargino flow or the loss of all e Cwater with a cense uen- '.en'igh pri;ary pressure above the,'-,'PS1 +hu o = head. Until the core unccve. s, all decay Ce ~ power aces to heatirg and vapori:ing reactor ceo)ant. ~hen the coo)an level drops be)ow the tcp of the active fue) , t..e or..'cn ihcs no of t"e ac=;ve = i ue) be 1 cw ph e
- t. e coolant p 1 eve) i s aCecua te ly
FICiUltE 4 2 SUOCOOLEO INLET FLOW ItATI: ftEQUtftE() TO MAINTAIN COOLANT Lf'VEL AB()VE (:OftE vs DECAY TlhlE 30 FLOW HATE IS GPM AT 100 F PER 1000 MWT OPEftATIIJG POWER ( OP E HAT IN G P OWE H, MWT) 51ULTIPLY GPM OY:
'o 200 c (3 FOR INLET AT MULTIPLYGPlvl IIY I
0? I RANGE OF FLOW RATES 100 F 1.0 FOII I'BESSUflES f-ftOM 200'F 1.1 1ra TO 2ra00 f'SIA 300" F 1.3 0 400 F 1.6 j 100 Z CI 0 0u CQ K K U? LL K D 0 D 0 x o z0 C) x LA C4 100 200 1000 2000 50i) 0 10,( DE(.<'!>IE SEC
c"ol ed while ge..crating saturated steam. c.oled The uel above the ccolant level heats up and'uerheu" t s .. e s.earn. Figure 4-3 is a sc4e..a
~ ... ~ 'c o"f i e core and Ccwncc;,.er regicns within the vessel shcwing the r.:ancmeter effec> as the cor~
bo>ls off. f Ste~mm for...ation St in .he covered portion of the core swells the volume of core ccolant, producing a higher effective ccoling level than would otherwise exist for the same mass of wa-or w'thwi . e vessel, and a hicher level in the core than in the Cownccmer. The ti.".,e variation o level is obtair.ed frcn at a h eat andC ' ."..ass b balance i. e on the covered lenc:h, L, of the active fuel length L as follcws An inlet flew is incluCeC 1 C>~ or analytical cc."..pleteness and later cc."..parisons. A heat balance cr the liquid in:he core is: Cecay pcwer beicw Power o vapori:e Pcwer to heat two-phase level Iicuid + subccoied inlet to saturaticn P o OH zh) L
= M ' ( " )H fg + M.
in (H f - H. ) in yieldino: P o OH (t)/L - W.in (H,f - H. ) in (t, ~) s g A,",.ass ".al'nce cn the fluic in .he vessel is:
0 FIGURE 4-3 COOLANT DEiJSITY DISTRIBUTION DURING BOILOFF (0 05 FTZ LOCA) p = 1.21 LBill/FT HOT LEG p = .69 LBh1/FT3 COLD LEG C TOP OF COR~ T"IO PHASc= LEVEL p"-50.1 LBib1/FT c 0 BOTTOi'~1 OF CORE CORF DENSITY, LB%1/FT3 4-20
Rate of charge of Inlet mass Stean exit liqu d ."..ass in flow ra e flow rate vessel pA d2. at =Min
- A(i) s Combining these equa ti ons y i el ds:
dl P o GH M. in (H + H, - H. )
+
at pnLa pnh-fg ZG The solution =or initial conditions x(o) = L is:
=K2(1-e K1)+e u,.(Hf + H. - H,.n) where:
0 pALH 1 ~uH P = Operating power (B/hr) OH = Oecay heat ,raction, assumed constant W. = !nlet flow rate (Ibr'/hr) Hf = Saturated liquid enthalpy (B/ibm) 4-21
H = Latent heat of vapori=ation (3/lbm) H. = Svbcooled inlet enthalpy (B/Ibm) p = Liquid density (9/Ibm) A = Cross sec.ional area of ccolant .'n core and dcwncc'er (ft ) z(t) = Level in core>> (ft) at time t t = Tir.e from start of uncovery (hr) L = Active core heicht (ft) When the inlet Iow rate, in' is W. zero the co i t s an . K22 is zero.. Equation E (4-2) indicates that .he fractional level, z/L, is then de e..cent on only t"e h d "..ensionless ratio t/K . For al I C-e. desicnea reactors the value of K 1 is
=2" of the averace value.
1'ithin Hence, when the core uncovers by boiloff the level, as a frac .'on of core heic ht, varies with t;..e in the same way for all C-i desicned reac:ors. When ~he te core is covered, the total cecay power is conver:ed to steam and t"e i,e rate o- level drcp is fast. When tt e core is partially covered, only the covered length of fuel procuces steam and t.e ra-e oi I evei rop i ' " is sicwer. The fraction of the core uncovered (not the lencth in feet) at any time after the start of core uncovery is nearly the same for all the plants, at a given decay heat fraction or decay tir,".e. Hence, a lorcer core uncovers a longer length of uel in a given tir,.e. This does not necessarily mean higher te...era.Jr s in the uncovered portion, since the covered portion is also longer ar "- produces a larger steam flew rate to ecol the uncovered I~ ngth. - Two-phase swell of i.h e level in the core is neglected, resul ing in a slower t an ac"uaI prediction o- core level drop. 4-22
-o.-.e ir por ant assu-ot.ons in the derivations are phvs.c llv significant. The two-phase level within the core is un'= uni-on across the entire c"re, independent of fuel asse...blv ppower. . . This'are assu-..ption is ."..aCe in the vore Ce~ailed cc-,puter codes. It is equivalent to assuning sc.-..e cross-flow ard nixino o liquid below the two-phase leveel from f the lower power asse.;,blies to the hicher pcwer asse..blies .o acc.~odate their higher va ori-ri-a iion rate.
The axial power distribution is assurred uniform Typica d istributions wi h center peaks cause faster level drop for cool an t eve 1 1 s above the elevation of the axial peak because a greater fr ac .'on of the decay "ower is "r"Cuce -etcw ced be the two-phase level. caked Peak dis.ri utions would cause slcwer level drop for ccolant levels below the elevation of th e pea<. ~' t is shown la er that substantial core damace occurs only ai <<r "f ~ .e 1 eve 1 has h dro:oed rore than halfway in the core. Hercoc, ffor the purposes of darace asses .-..ent, tl e e===c= . of axial p ewer 'ribu.ion dis- on the coolant level is consiCereC a seccnc orco. effect. he density of the two-p'se - h luid in the core is less than the dens'.ty of he liquid in the dcwncc."..er. ~
'hen a given arount of liquid in the core is vapori: d, the core level drops r,ore than the dcwnccrer level. In the analysis above, these density di erences are ignored. Consequently, the aralyses nigh: predict a slcwer than ac:ual rate of cor uncovery. The di er nce is dependent on the relative sapor/liquid dens-'ty r t.'o ard is thererore sr.;aller at higher pressure. awhile these di .erences nay af e-t h 4-23
level at any given instant, they are not considered sicni;icant cn >n in the
~,e establishment of the relationship between he aI;cun- o~ core da.. d ..
amount of hydrogen generation. Parameters or the 3400 t'.wt class are used orr some exarpl es oi coolant level sore examp drop as a func ion of time. Equation 4-2 wi h ero in zero e t flcw, is plotted in Figure 4-4, at 1200 psia and or three values of decay ecay power. pc e F or ~" ddecay power (about 2 hours decay time) and with no inlet flow, the 3400 t'.~t class reactor uncovers 50".. in 13 minutes. At 2" decay power (abcu. 23 min. decay t -..e) it takes half as long for 50" uncovery. Ficure 4-5 shcws the level vs. tire at 1:. decay power for three values of pressure. At 300 psia it akes 21 min. to urcover to 50 and at 2500 psia it ake s 6 .' min. i h ese results are
- typoical ror T al 1 C-"
-E designed reactors within a time scale variaticn of +12" to -~,
I Ql which is the rance of variation of the constant, K, from the value 'or the 34CO Ywt class; Thi s rance of error is considered sma11 enough to permit tini ~ Ion o .he examen o core Camace within the ten cateccries cerined in Table 4-1. 4.2.3 Core Heatuo Analyses d'or The ob'ective of the core heatup analyses is to predict tl e distr',bution of clad temperature and clad oxidation during an event which urcovers the core. ihe distribution of clad temperature is used to establish a relationship
- between the maximum corre exit .herr:,ocouple temperature anC the minirum number of rods which have rup.ured due .o gas gap overpressuri"ation. This relaticrshi will be used in Section 5 as a basis the use of core exit 4-24
0 F IGUR E 44 COOLANT LEVEL vs TliblE DURING SOILOFF 3400 i43WT SF RIES, 1200 PSlA 100 80
)
I O 60 CD l1 Q 4'2 40 4l 1'o DECAY PO"IER I- 2 so z O 20 .0u 0 0 400 800 1200 1600 2000 Tli%1E AFTER START OF CORE UNCOVERY, SFC I 4-25
FIGURE 4-5 COOLANT LEVEL vs TIME DURING BOILOFF 3400 MWT SERIES, 1 ia DFCAY POWER 100 SO Q 60 C 0O LL, 0 300 PSIA <o 40 1200 PSIA z 1 0 20 0O 2500 PSIA 400 800 1200 1600 2000 Tli'<IE AFTER START OF CORE UNCOVERY, SFC 4-26
ther.-.,oc"upi e data to assess core darace. The clad oxidation dis:ribution is used to establish a t'.. relaticnship'p b e.ween the r.:easurerent of hydrocen cere. ~' ~ ~ by the oxidation and the .",.ini;..ua rut-..ber o= rods h' d which have h oxidi:ed beyord he et"..brittle...ent threshold. Fuel oxidaticn e...brittle.".ent throuch t."e discus-r d el rod
':;
s.ructural integrity has been scussion in Section 4.2.1. 4.2. related to ~n analytical derivation is used which includes the doninant physic" a 'e... p enc;..ena A ppenc cc,",.pu 1 x S to support the ob'ective.
~ 2 E~o1 er code solution enables p
E...oloyr..en.. o. an a This derivation is provided in analytical derivation rather than convenient cc'parison of the s'""'=ic n-a nurser "=1 para...eters aconc all the C-= reac "r ces'cns. 0 " icns. Detailed analyses on each
~ et ent col ".='g e CCI i uratt.icn a r . r~wn -.o be unnecessary within the rocuirr>
accuracy ot the overall PASS darace assess;,ent. procecures. As an overc"ec'<<~n the anally.ical soiut-'or s, sc;..e analyses are dore using the l".AAP cc--pu-or c"ce Here are so-e -'y aiI analytlca ..ui.~ e~ " .rca the cerivatIon in Apcendix A x S.2. S The firs. result is that ce " pcwer deternines the rate of coolant level droo when ur cove ry occurs b i 1 o f= . booilo Hicher power causes faster level crcp, as shcwn previcusly by Figur~ 4-4 The second resul is tha- power determines the rate of te."..perature rise wi ~ i'. igJre i-o shcws the peak clad en"erature as a 'unction of t'-,.e after uncovery starts for three values of ecay power. deca However, at a given level, the te."..peratures ar sare for al;..ost thee sage a range of 1.". to 3:. decay pcwer. This is s"cwn in Ficure he tac= that temperatures at the lower 4-27
0 FIGURE 4 6 MAXII'v1Uir1CORE TEMPERATURE vs TlhlE AFTER START QF CORE UNCOVE R Y DURING BQ ILOF F 3400 i%1WT SERIES, 1200 PSIA 3000 2500 2ol DECAY PO'VER U 1;o 0 LU 'P 2000 LV I 1500 X DECAY TIME AFTER PQViER I;o) TR1P tSEC) 1000 10,300 (2.9 HRS) 1300 220 i:PP 400 800 1200 1600 2000 Tlili1E AFTER START OF CQRE UNCOVERY, SEC 4-28
0 FIGURE 4-7 MAXIi'tlUiblCORE TE NIP ERPTUR E vs COOLANT LEVEL DURING BOILOFF 3400 i'rlbVT Sc RIES, 1200 PSIA 3000 34 I 1'io DECAY PO'eV ER 2""00 U 0 UJ I-cf. 2000 LIJ Ja UJ I-X 1EOO 1000 F00 20 40 60 80 100 COOLANT LEVEL "~ OF CORE ACTIVE HEIGHT 4-29
power are slightly hicher is probablv caused by oxica.ion heating Grea.er reater total reaction heat is added to the fuel rod Curing the lorcer eriod o>> .-'-, required to attain a given coolant level 'en the power is lower The conclusion is that coolant level leaves lesss uncertainiy deter...ining the clad temperature conclusion might di er if other or a given factors (such as uncer ain boilof, HPS i of scenario.
=1 7
tth an time in ow)) cause This Th'i the rate of level drop to be less dependent on core power. A second conclusion is tha-thai wi...ou. inlet ccoling flew, the tire after uncovery to reach high tempera" re is only minutes or a few tens of minutes. This seccnd conclusion was also r.".aCe eviCent in previous studies for the CEOG on the adequacy of the core exit ther...ccouples io provide an advance warning of the approach to inadequa-e c"ro coolirg (Reference 7-5). Ouring typical small break LOCA events it was shc n that the time interval is short sh rom the first occurrence of steam superheat un".i 1 the clad ruptured or exc ded 22CO'F. The amount o f 1 oca 1 oxiCaticn is Cependeni on the magni iude of temperature anc the lercth o '..e
~ i-,.e a te...pera ..re. For core uncovery by boi loff, Figure 4-B shcws the local cladd oxiC oxiCa .ion as a furction of tir e aft r uncovery iaris, siarts
- or three values of decay power. Tne oxidation rate is slew initially, until the ter perature rises above about 1800 F. Therea, ter for a boilof'vont the rate o- oxiCation increases extremely fas". i. >n M'0 ~ a ew minutes, local oxiCation in 're ses i rom a few perceni to well beyond the embrittlement threshold. Fi'g ure 4-9 shcws h .'hai .his rapid increase in oxiCation occurs wnen the coolart level has dropped to abcut 25
- !.
4-30
FIGURE 4-8 h1AXIMUM LOCAL CLAD OX IDATIOi'J vs TIfVIE AFTER START OF CORE UiXCOYERY DUR liXG BOILOF F 3400 ibWT SERIES, 1200 PSIA 20 16 o> DECAY z PO'eV ER O l 10/ cX 12-Q CJ u 0 0
.0 400 800 1200 1600 ZCQQ TIi'iIE AFTER START OF CORE UiNCOVERY, SFC 4-31
0 0
F I GUR F 4-9 MAXIi~IUi'i1LOCAL.CLAD C~" DATION ys COOLAi'JT LEVEL DUB I>iu BOILOFF 3400 MV/T SERIES, 1200 PSIA I t I I 20 16 30I z 0 1" 0~CAY POVIER I-12 0 u u O 8 10 20 30 40 50 COOLAiVT L VEL,;o OF CORE ACTIVE HEIGHT 4-32
These ca i lccula,ons assu..e su iciefii s earl prCuc.on "elow the coolant level to oxidize the up"er portion of the fuel clad. More comprehensive calcula ions succest tha- the th s .e ~.m ray be b c....ol etely corsu;.,ed a 1 ong the lower I engngg"h o= o7 exposed clad thereby limiting the oxidation alono the top of the rod. Fuel will then heatup by decay power alone, with the cld and, later, the fuel being destroyed by melting, It has been repor:ed (Reference 7-14) tha for uncovery by boiloff the oxidation embrittlement o, the clad will have alre~C occurred prior to the buildup of hydrogen suf icient to limit the oxidation rate. Thererore, these calculations are aCecuate for predictirg the rrac=ion of fuel rods which have attained the local clad e,brittlement threshold "u" not -or predic ing the axial d'stribution or ex.ent of oxiCa ion alona ihe "e length of a rod. Axial rlcw or steam 1 and hydrogen tends to 5UDport the usual assumoticns of a c..annel c'lcula.ion which ignores coolant mass interchance amonc adjacent accn~ channels. Therefore th e 1'.'"
~ limi .ing ef ects of steam consumption and nydrogen generation on clad oxidation in a high power channel woulC no significantlv . ...e oxida . on in adjacent channels wi th lower radial peaks, This validates the ~ e calcula.'cn cal ul of ihe radial distribution of tt e rods in the core which may at least frac'ent upon quenching or'ater handling. A prediction of the total .ota damage d configuration requir s additional moCel ing.
4,2. Er-.ects e-. Radial Power Distribution In the previous sec:icn it was disc.ssed that decay pcwer Cetermlned the ra es coolan ~ 'r level drop and peak clad:emperature rise. The magnitude of temperature and the time at temperature determines he amount of loc'1I 4-33
oxiCaticn. This section provides:he bases for the relationship be~weon radial pcwer distributicn and the amcunt of local oxidaticn. The relaticrship is given for several values of decay power and se<dra', values of p<< sure. During an initial core uncovery and prior to substantial core structural damage, the coolant level is uniform across he core. Steam flow rate tencs to be higher in hicher power regions or channels, thereby tending to reduce the dependerce of temperature on the radial pcwer distribution. As the coolant level drops and the heat of reacticn increases, convection ror ovos ~ smaller fraction of power and the rods with higher radial peaks increase aster in temperature. At any instan: .here is a distributicn in rcC te.".ceratures across the core above the coolant level. If reflocc anC core ccolirc shculd "e acccmplished at :hat i'e, there will resul: a racial distr'.bution of fuel rcd rup ures and clad embrittlement after core recoverv, A typical distribution of radial peaking factors is selected =or s e~Cy ~ower operation without CiA inserticn. Figure 4-10 shcws the cumulative fraction of ruel rocs in the core with radial peaks above the value civen on the aoscissa i>>e band o the curve ence-..passes the variations, from burnup only, throuchout a fuel cycle of length 14,000 t".r'0/T -or the sixth cycle of a 27CO Ywt core Five calculational intervals of radial peaks are selected with corresponding percentages of the core as follcws: 4-34
FIGURE 4-10 D ISTR I BUT IOi J OF ROD RAD IAL iVUCL EAR P EAKS 100 80 Q 'A'A g C CD 60 Bz0 cn O~ Og RANGE DURING L'J ) FUEL CYCLE u) D u- O QJ 40 Ou 22 V Mcvu 20
,0 0.6 O.S 1.0 1.2 1,4 ROD RADIALNUCLEAR PEAK
Radial Peak In. erval pere nt ce of Core Ca l cul a t'. on Peak
>1. 3 1.4 1.1 - 1.3 22 1.2 0.9 - 1.1 1.0 0.7 - 0.9 22 O.S <0.7 0.6 This dis.ribution is considered a best estimate for reactor conditions which
,would exist most of the time. It is adecuate for generic calculations which sv"-ort the procedure for darace assessment and which are necessarily per;or."..ed "rior to the ccc rrence of an accident. The peak clad temperature as a func.ion of radial peak during boilof. is plotted in Ficure 4-11 for various coolant levels or times. At time zero the core is covered and the clad temperature is essentially uniform at sa ura.ion temperature. As the core uncovers, the temperature of rods uith higher radial peaks increases aster than the te...perature of lower peak rods. For example, when ".he core is half covered, the temperature is '.175'F on rods with a raoial peak of 1.4 and is 960'F on rods wi:h a peak of 0.6. These temperatures would increase proportionally on all rods, i a non-uniform axial distribution revere used. The same calculations yield the local percen age oxidation of clad thickness as a unc:ion of radial peak for various levels. Ficure 4-'2 shows that the maximum local oxidation on rods with 1.4 radial peak is 2:.l when the coolant level reaches "Q... Qn rods wi:h a radial peak of 0.6, the local oxidation is 4-36
FIGURF 4-11 51AXI'AU'1 RCD TE 1PERATURE DURING BQILQFF vs RADIAL NUCLEAR PEAK 3&0 ilIVlTSERIFS; 1200 PSIA; 2~~ DECAY POPPER 3500 CQQLAiMT LEVEL 3000 22 e J 25CO 0 25 ~ D 2000 30 so C D X 1500 40cl c00l 1000 60Ão 90=,a 500 0.6 0.3 1.0 1.2 RADIAL i~lUCLEAR P'EAK 4-37
FIGURE 4-12 MAXli%1UM LOCAL CLAD OXIDATION DURING 8OILOFF vs RADIALNUCLEAR PEAK 3400 svlWT SERIES; 1200 PSIA; 2 DECAY POV/ER 24 17o' l 20 f o> O 16 C! x 0 O ~ O 12 O 0 D 8 X COOLANT LEVEL
%pof 40;o 0.6 0.8 1.0 1.2 1.4 RADIAL NUCLEAR PEAK 4-38
0 only 1/4~. At a coolant level o; 20~, the local oxiC"'.cn rances frcm 24~ on high peak rccs to abcut '. 1/2." on lcw peak rods. There is a wide variation in oxidation at any instant and :herefore in the poten:ial 'or clad embritt;e-ment. By ccmpariscn with Figure 4--', there is only 4 minutes di,ference between these two coolant levels. The conclusion is tha once oxidation gets goina, it proceeds rapiCly, anc a: any instant there are large variations in the maximum loc 1 oxiCaticn on:.".e fuel rods in the core. The total hydrccen released from the core is Ceter... re@ at eacn ins:ar.: by su;..ming the local oxidaticn along the exposed clad leng:h =or all the radial peaks. At the same instant, the nu'oer of roCs which have local oxication greater than the embrit lemen- threshold is determined. Tne resul s are given on Figure 4-13, as the percent of the nurber of roCs in the core which have at least 20 local oxiCation as a func:icn of the total percent oxidat;on of al 1 the clad in the active core length. Calculations are made for three values o- decay power at 12CO psia. The figure indicates a relatively large increase in the number o, rods e...br',tt',ed ccmpared to the increase in total cor oxidation. The ccarseness of the calculated valves limits the detailed conclusicns which can be r.".aCe. However, it can be concluded that a large fr ac ion of the rods may be embrit:led when a relatively small frac icn of the zirccnium in the core clad is oxidized. Figure 4-14 shcws, with the sa."..e ccordinates, the variations in embr',ttle,ent and core oxidation when the pressure varies -.rom 3GO to 2 00 psia. The 4-39
FIGURE 413 PERCENT OF RODS EiilBRITTl ED vs CORE OXIDATIOIII DURING BOILOFF 3400 iMWT SF R I ES, 1200 PSIA 100 3cs /
~~
CI ~et cv Il'K I"., DECAY HEAT O I-Cl X 0 60 O 0 I 40 D 0 D
/
20 O I <0 I 0 20 40 60 80 100 oio OXIDATIOiIIUF CORE CLAD VQLUiblE
0 F IGUR E 4-14 PERCENT OF RODS EMBRITTLED vs CORE OXIDATION DUR ING BOILOFF DECAY HEAT = 2;o 100 ~O C) bl QP SPO PSIA /5 z 0 r I r C X 1200 2 PSIA 60 r u C r l- r C/) C5 0 4P t 20 O 2500 PSIA ~o 0 0 20 40 60 80 100
;o OXIDATION OF CORE CLAD VOLUME 4-41
0 results are similar to those or variations in decay power, in tat a ? relatively small band ence'passes a wide parameter rance. All the results cbtained for the percent o the rods e...brittled as a function of percent total core oxidation are plotted on Figvre 4-15. This figure incluCes the rance o. l~ to 3" decay heat and the pressvre range frcm 3GO to 25GO psia, Given a PASS measurement of the amount of hydrocen released from the core, expressed as a percentage of the core clad volvre which is oxidi:ed to prcduce it, Figure 4-15 is utilized to est;..ate the "ercent of the fuel rcds which micht fracr.".ent upcn quench and/or later harCl',na. This figure is ircluCed in the procedure for daraae assessment and is considered applicable to all C-E designed reac:ors. 4.3 PR DICTIGt( OF FU" L CLAD RUPTURE BAS 0 Ot( HYDRCG:( PRGGLCTIQ'( P. evicusly, hydrogen prcdvcticn was related to clad temperature and to the r dial distribution of clad temperatures in the core. In this secticn, the criteria are developed which relate clad temperature to the occurrence of clad rup:ure by overpressurization of the gas in the rod. Then the number of ruptured rods is related to the arcunt of hydrogen produced. Thus, the measvrement of the total hydrogen produced may be used to infer the number of ruptured rcds. 4.3. 1 Clad Ruo.ure Cri eria Fuel clad will balloon ard rupture when the interral gas pressure is suf icien.i'y gree er '.han I:e esternal ecol ant pressure. Clad terperature and Q) 4-42
F IGURE 4-15
~o OF THE FUEL RODS WITH OXIDATIOiVEh1BRITTLEMENT vs TOTAL CORE OXIDATION FOR 1;o TO 3'.o DECAY HEAT AND 300 PSIA TO 2 00 PSIA WHEN COOLANT LEVEL DROPS BY BOILOFF WITH NO INLET FLOW UNTIL CORE IS RAPIDLY QUENCHED 100 0
bl 60 A O I-C 60 u 0 40 D O 20 0 0 0 20 40 60 80 100
.o OXIDATIONOF CORE CLAD VOLUirlE
'on at e..."e.ature ar sicni-icant para-..ete. s ',n ce:erminirc the pressure di,.eren.ial. i'~orma1 values of these par meters and values exp d durina typical core ncove",y events are discussed here. The temperature wh<c4 causes clad failure during such events is determined. That rupture te..perature is used in the prediction of the number of ruptured rods by this prcceoure.
C-E fuel is prepressuri=ed, at room temperature, to 380 psio with helium Increasirg the temperature to normal operatinc ccnditions increases the in.erral cas pressure to abou ~ 800 psia. The normal external ccolant pres'.r~ is a"cut 1~00 psi higher than the minimum internal pressure. ~ceo'ulaticn of 'issicr. cas increases the internal pressure but does not cause it to exceed coolant pressure at the end of uel life. The fuel does not ruptur~ dur'ng a cepressuri=ation transient at normal temperatures. -() Typical calculaticns for Oesicn crasis srall break LGCA events yield reactor coolant pressures below the secondary pressure when the core uncovers. Seccrdary pressures may rance frcm about 8:"0 to about 1100 psia depending on the plant. 'rlhen the fission gas pr D ssure is acced to .he helium gas pressure
- a. e eva.ed accident temperature, the internal pressure exceeds reactor ccolan. pressure.
1 I whether or not the clad ruptures depends on the particular ~uel uel rod burnup, on the even: scenario and on the clad material properties. P survey o these factors and how they combine to determine i ~ fuel ruptures is proviced by the CEGG sponsored effort on Ir. decuate Core Coolinc Instrumentation and appears in C=,'(-I:-8 (Reference 7-5). Figur~ 4-i6 4
su...,ari es those ccrclusions. It shows the local clad tern"erature at which upture will occur as a func:ion of the clad differential pressure for a rance of tl e duration at temperature 'rom 600 to 3600 sec. It also shows the temperature as a function of internal gas pressure for new fuel containing only helium fill gas. For example, at 1500'F the internal pressure is 1450 psia in new fuel and increases with fuel'burnup. When the coolant pressure is 1100 psia or less the clad dif erential pressure is at least 350 psid. Ficur~ 4-16 shows that clad rupture will occur in less than 600 sec. Core uncovery is also predicted for complete loss of heat sink events where the external coolant pressure is hicher than the internal gas pressure. Ccolant pressure is go'erned by the primary safety valve setting which exc>>~s nor...al operating pressure. Clad rupture may occur later in uncovery or micht occur by brittle fracture upon clad stress reversal when reflood and sys em depressurization occurs. he cconclusion The ncl is that 4 or the most general uncovery events clad rupture will occur in the rance of cl ad temperature from 1200'. to 1c00 F. Three temperatures, 12CO'F, 15CO'F and 18GO F are selec.ed or-later use in evaluating the number of ruptured rods corresponding to measure...ents of hydrogen and core exit temperature. This range of rupture temperature is one source o- inaccuracy in tte use of the subject procedure. However, the results are adequate to locate the the extent of core damage within the defined categories ci aracterized in Table 4-1. 45
FIGURE 4-16 CLAD RUPTURE TE~rlPERATURE vs CLAD DIFFERENTIAL PRESSURE, AND HELIUi%1 FILL GAS TEi41PERATURE vs PRESSURE (2700 i'AWT SERIES) 1800 1600 HELIUr.> GAS TEi'<IPERATUR E "ii vs PRESSURE 0 QJ UJ LU 1400 UJ c Ch DURATION AT U T E i'ilP 5 R AT U iR E 5 PRESSURE FOR D I- 1200 CLAD RUPTURE C L'J 0 1000 3600 SEC 800 0 500 1000 1500 2000 CLAD DIFFERENTIAL PRESSURE, PSID [OR HELIUi'i1 GAS PRESSURE, PSIA) 4 46
4.3.2 fffer=s of -".'cial power Ois:rib ---r
~
C ad rupture occurs at about 1"00'F = 300 . -.or core boilof events "uo-"ro occurs earlier han he occurrence of 20" local clad oxidation. A relation is made between the number of rods which reach the rupture temperature as a funct',on of the percent of the total core clad zirconium which is oxidi=ed at any instant. Even thcuch there is a wide variation in the rupture temperature with time, burnup, clad pressure differential etc., the uncertainty in the resultant relationship is probably not significant. The temperature rise on a rcd is relatively as: ccmpared to the total core wide oxidation so the te...pera:ure rapidly rises through the rance of rupture -. moeratures. The ';..e at which this occurs varies with radial peak. Figure 4-11 yielcs .he racial peak ;or which the clad ter cerature exceeds 1""CO F, at several t .-..es dur rc core uncovers. this is co;..oined wi th the core distribution of rac al peaks in Ficure ~-!0 ) ll and with the percent of the core clad oxidi:ed at each ti;.e. Results are plotted in Ficure 4-17 as the percent of the total nu;.."er o; uel rods which are ruptured vs. the percent of the core clad z.rconium oxidi ea. ihe earliest possible indica.icn of clad rupture from a measurement of hydrocen depends on the sensitivity of the measurement. Typically, the minimum measurable concentration in the containment atmosphere is 0. 1",. b This concentration is ecuivalen to a total amount of hydrogen ~hic1h 'olume. is prcduced by oxidation of abcut 0.5" of the core clad zirconium. Figure 4-17 shcws that by the time 0.:-.". of the core clad is oxidi=ed during bo'lof= 0'fT, e;e 0. ard !CC:. o .he r"cs are ruptured, depending cn whether the lu Jre tempera:ure is 1cCO'F or 1290'F respectiveIy. 4-47
F IGUR E 4-17 PERC NT OF FUEL RODS ABOVE CLAD RUPTURE TEhfPERATURE vs CORE CLAD OXIDATION 3400 MWT SERIES, 1200 PSIA, 2;o DECAY POWER 1200 F 100 80 D
~e LU + 1500'F C
60 1800 F I RUPTURE TE vlPERATURE
~ Ill u I- ~ VJ ~p C 40
, Jill) oo CO Q u 20 0
~O 0.5 1.0 1.5 2.0 o OXIDATIONOF CORE CLAD VOLUI'IE 4 4S
ihe corclusion is .h..ai- hf h d r --~
. n is noi a sensit>ve "ar"me+e p - er or assessment of smail amounts of clad r rsture. In fac-- ihe opposite is true. If anv indica .ion oi. hydrogen is ob:ained rom the con . ainr' inr;.en a;..osphere which is attributable to core oxidation during a boilo CC event, then a large percentage of the fuel rods are pro".ably rup ured. Hydrogen measurement would be a backup to the more sensitive indication ' t' frcm the f;ssion cas o ra d iation of release frcm the ruptured fuel.
4.4 CCl>FI."> ATIC'1 OF A>>ALYTICAL PR D.CTIG'( li s sec .on two methods are used to veri=y In .his ,y th e previously stat~~ 1 concluscn that the analytical resul "s ~ <<bboilof, rom t<<e analyses yield lower limits esiimatese of clad damage for all scenarios and that the si;.,piified I analyses are acequa- ua .e for this damage assessment procedure. analyses o=I slew uncovery with inlet flow are p r esented in Section 4.4. 1 ara are c";..oared to the previous aralyses of uncovery without inlet flow. Analyses with a stat - =-" state-o>-the-art ccmouter code are presented in Sec:ion 4~4>>2~ Pesults are given f core uncovery by blcwdown ard by iven for boiloif. The rap'd rap. d blowdown resuliss su or- .he conclusion that the procedure yields suppor. low limit estimates of cladd damace 1 d for such accident scenarios. The boiloff co"puter c Qosspu t e r results are compared to previous analytical results to verify the applic bility of the simplified analyses.
Ccmoarisons to Fred':ctions wi h Slow I.'ncoverv This section provides a comparison between the cases of boilof with and without inlet low. Because of the potential variability of inlet flow durinc an accident, it is necessary to know how the damage prediction is affected. Inlet flew causes a slower rate of uncovery. The rate is slower when licuid inlet flow replaces some of the steam flow from the core. Urcovery proceeds until a stable coolant level occ.rs for which the inlet mass flow rate is equal to the ste m flow rate. The height o. the stable level is available from the previous deri sations in Section 4.2.2. The level will rise as the dec y power decreases and of ccurse may vary i= system corditions chance. Eruation 4-2 gives the irac-ional elevation vs. t',me: L
=K'2 (I-e ~'I)+e '1 for long times af:er uncovery, the fract,onal level becomes K, which is defined by:
4-50
Thi s shcws th
~ .ha ~ ..e dec y power cenerated below th s:able coolant level
((2/L)(P DH)) is equal to the power to he t the subccoled sa turat-'on inlet to satur < on and to vapori-e p z i t (M. ( in('g (H- + H Hf - H .' ))} ~ Figure 4-18 shows how the l evel approaches the stable level vs. the normali:ed tir;,e, which is t/!< in Eq"ation Oepending on the lowest level attained, the steam convection cooling mignt or might not be sufficient to keep the maximum temperature of the clad from rising because of the decay heat input plus the oxidation heating. It is est "..ated that even without oxidation heating the clad temcerature is above 1800'F with the steady level at 60". Therefore, the clad will con inue 'o 'ise in temperature even if the level is held constant, when the level is lower than abou. 60". The oxidation below the coolant level is essentially zero. ~bove the ccolar<t level, ihehe local oxidation nay be greater for a given total amount of hycrocen generated than if boilof, proceeded without inlet flow and the sare amount of hydrogen were generated frcm a longer,raction of the core length. Ficure 4-13 which predicts the number of embr i ttl ed rods for a measured total hydrccen generation, yields a lower limit damage estimate when there is some inlet flow to the vesse'l. This same argument may be extended to include the additional oxidation which occurs in scenarios where the refill proceods slowly. ihe er feet of inlet flow on core temperatur~ is exempli=ied e by rigure 4-19, where inlet flew is sufficient to maintain a steady 60.. coolant level The average core exit steam temoerature is plotted as a function of the subcooled 4-51
0 0 0
FIGURE 4-18 COOLANT LEVEL vs NORMALIZED Tlii1E MflTH VARIOUS INLET FLOLV RATES INLET fHg Hin~ K DECAY POl'(ER 1.0 0.3 QA u 0.6 O u 0~a o~ 0.4 I-z o 0.2 0O K2= 0 0.5 1.0 1.5 2.0 2.5 NORMALIZED TlihlE 4- 52
4 FIGURE 4.19 BQILOFF 'VIITH liXLET FLO' TQ v1AliNTAIiI STEADY 60 COOLAi4T LEVEL STEAa'r1 TE~i1PERATURE 5 liXLET FLQIV RAT- vs liJLET TEMPERATURE. AT 1200 PSIA 3400 iVIVTCLASS, 1,o DECAY PQ'VlER 400 360 i C 1600 3CO Pl C CA 4 M Pl 1400 200 I A J 150 1200 100 100 200 300 400 500 600 SUBCOQLED LIQUID INLET TE"IPERATURE, 'F
in << temoerature. The decay power gener t~d by the cove ed 60 o he c r heats the subcooled inlet and vaporizes it. With hioh inlet- temperure
~ . per .are, larger frac:ion of the decay power produces steam and he 5 earn ~ lcw rate is high. With low inlet temperature, a larger fraction of the decay power hea:s the subccoled liquid to saturation and the ste m flow rate is low. A lower steam flow rate yields higher steam superheat above the coolant level.
A saturated inlet flow of 315 GrN at 567'F and 1200 psia raintains a steady level at 60" and yields a core average exit temperature of 1250'F. A subcooled inlet flow of 125 Gr!< at 100'F and 1200 psia maintains the same 'teady evel, but yields a core exit temperature of 1200 F. I, boiloff proceeos with zero inlet floA, the transient .emoerature is g"O'F as the level drops down past the 60 level. Hence, above a given coolant level, the temperatures and oxidation can vary depending on the inlet flew rate. Ouring the core unco,ery period of the TNI-2 acc',ent,, there ~as scme inlet flow. It probably caused the dis-ribution of oxidation to be greater in the upper por;ion of the core than would have occurrec without inlet flow and with the sane total amount of core oxidation. Of course any inlet flow is better than rene. The core damage at Ti11-2 would have been greater for the same duration of uncovery if there were no inlet flow. ine conclusion is that a clad damace assess-,.en utilizing Figure 4-15 may yield a low estimate of the number of rods embri".led if a) there is some inlet flow which slows uncovery and/or b) the refill is slow. 4-54
4.4.2 Cc.".oar.scns witn Al ternate Cc=cu-~ '>r o The PA~P ccmcuter code is a state-of-the-ar ool fcr anal ana yzing very severe z'r postulated core o sc narios.
..e .dcwn scon meltdcwn It is being developed under the IDCCR program sponsor d by the nuclear industry. Thee earlyy port ear p ons of several core r 1'el~ scenarios are reviewed here. These examples are run on the 2700 i'1wi'cot~ re configuration. Only the portion of the accid en t stuart ng;rcm core uncoverv and prcceeding to 20" local oxidation is cons nsi d ere d h ere.... Assess-.,ent of rore severe core damage would require much greator ef ccr.. i h.e e..or: is beycrc the scc" c e o. .his P~ ~ prccedure and is of questionaole value her~ sine~ -re r easurement of the darace would require interpretive alccr -"-.s u-'i='r" ru .'pie sourcesa or recorced ins.rument data. Such Ca:a is rot recor"ed "v the PASS and michtt nott Le a v ailable at all plants. Raciation measure,.ents bvr the PASS can provide one estimate of the extent of sever~ core re't .e irsi case analyzed,wi h,"~AP is a large break LCCA. It is inc'.used to shcw how the damage assessment procedure yields an underpredicticn cf the ex.entt o damace. Lar"e break LOCA events depressurize to ccntairrent pressure with',n tens o'econcs, exhausting the primary system and leaving the core empty ot coolan.. Ficure 4-20 shcws the adiabatic heatup which follcws--
the blowdcwn assuming the "CCS Coes not function. Within about 10 minutes -I.ehe temperature on the peak pcwer rods exceeds the 2ircaloy mel tirg point o~ \ 33:0 .. The fuel s.ructure fails by 4 mel tine rather than oxidation induced e ragmentation.
0 FIGURE 4-20 PEAK TEihlPERATURE vs TliVIE MAAP CODE RESULTS FOR HEATUP AFTER TOTAL CORE BLOWDO'PIN PRESSURE = 50 TO 40 PSIA LOCAL CLAD OXIDATION < 0.001;g OXIDATION LlhkITED BY ABSEiJCE OF STEAM FLOVl RAQIAL PEAK = 1.'20 4000 CLAD iMELT 1.0 3000 D 2000 ly NORMAL INITIALROD AVERAGE TEiblPERATURES 1000 400 1000 200 600 @00 TlihlE AFTER REACTOR TRIP, SEC
t ihe ar" n. of hydrccen ceneration calculated '.or .h'.s examole, prior to rerlocd, is much less than the measure...ent sens'. "..ty .or he ccntairmen: atmosphere samples. uring a subsequent reflccd, a creater amount of hydr==en is generated, wi"h the amcunt being dependent on the speed of the reflood as the core flcods, steam and gas pushed up and cut of the core may cause a rapid r'.se in .he Core exit Thermoccuple (C"=T) terper".ores. Scme indicat cn of -"e
+II speed of reccvery is available frcm the recorced trace of thermocouple temperature. A rapid rise follcwed by quenching to saturation temperature indicates rapid ref locd to abcve the C"=7 elevaticn nd minimum hydrcgen generaticn. Figure 4-15 may then uncerestimate the numoer of rods damac d.
hich (hicher thafl sa ul t ~ on) valid C"=T temoerature, wnich gradually chances over huncreds of seconds, ircicates slcwer refill wnicn is pro"ably accc.-.."anied by oxidat',on anc nydrcgen ceneration. Figure 4-i5 ind',ctes ha- ,. ere is dar aged ruel, bu. .he relaticrship between core car ace and hydrocen generated is urcerain. :.",e hvdrcgea measurement indicates a large fraction of the core zirc"nium s oxidized, say 'ore than 20", then substantial core dar age is cer ain, regardless of the particular reflood scenal 10 ~ The secord case analyzed with l>nP is a boiloff at very low pressure. var;aticn of the previous case. Properly functioning SiT's rapidly recover the core followina a large LOCA. Best estimate analyses indicate little or no core damage when all syste..s funct.on. 1 subsequent action, both automatic and by the opera. or, fails to maintain cont',nuous inlet, lcw frcn the HPSI and I PS: sys.ems, the ccrc ur c"vers again by boiloff at very lcw pressure Ficure v 2$ cives the resul ts -.or a l'~AP calculation of this scenario. 4-57
FIGURE 4.21 i%1AAP CODE RESULTS FOR 801LOFF FOLLOWING 8 LOWDOWN A iND REFILL PEAK TEMPERATURE AND LEVEL vs TIME I DECAY POWER = 2 4 TO 1.6~o PRESSURE = 48 PSIA / CLAD MELT
/
3000 u 0 UJ
< 2CPO RADIALPEAK =
C 1.4 0.6 1000 200 ll 100 Lv 80 60 40 lL 20 400 800 1200 1600 2000 TIME AFTER START OF CORE UiVCOVERY, SEC
>he oxidation at the top of he peak rods reaches abcut 20.o at about the sa"..e time the temperature reaches the clad melting point of 3350'F. The HA'P analytical model assumes no fur.her local oxid"io xi a.ion, b ut" "h e t temoerature continues to rise until it reaches the fuel pellet melting point. Oxidation meanwhile continues at the lower uncovered elevaticns eva icns on th t e maximum power rods and along the other lower power rods until each elevation on e~c'i rod reaches clad meltina. So,.e S m core s ruciuraI rearrangement may occur as melting progresses, which eventually invalidates the analytical model. For this particular scenario, the local oxidation reaches the embrittlement threshold at about the same tir;.e that local clad meltino star.s The "o al hydrogen generation from core wide oxidation yields a low estimate of the nu.-,.ver of rods with embr'.ttled clad which may spill fuel into coolan" thee coo an~. For example at 14CO sec. the coolant level is g~ and,".AAP indic~-~s 6" o- he core clad is oxidi=ed andjor melted, representing so".,e struc:ural damace to about ~tY 7~ oC the rods. At 6" core oxidation, Figure 4-15 indicates 20 to =0". of the rods may be embrittled. The third lQAP case is for boiloff at high pressure. Boiloff at the safety valve pressure setting is pos.ulated for a ccmplete loss of heat sink even-event. The secondary boils dry and subsequently the primary tempera ure and pressure rise. Safety valves open at about 2500 psia and the primary boils o== at he safety valve pressure. A PPAP code analysis of this scenario is available for the 2?00 l'wt class reactor. Results are ccmp~red to simpli ied calculations done on the 3400 l'.wt class reactor. 4-59
0 A Figure ~-c2~ 8) shows he peak temperature on an averace power rod and the coolant level as a function of ti;..e. ihere is good agreement ,or tempera:ure anc 'or level. Lf the clad rupture temperature is 1800'F, the fic indicates that the average power rod ruptures within 16 to 18 minutes of uncovery, when the
.he coolant level i s down to 15".
When the level is 10.>>, the simplified analysis predicts less than 2.". of the core clad is oxidized. At the same time after uncovery starts tNP predic 4 s about 3" of the core clad is oxidized. Ficure 4-15 indicates that between """.. and 22" of the rods exceed the embrittlement threshold when 2~ of the to:al core clad is oxidi:ed, and between 10" and 30" when 3" is oxidi:ed. Both -these results place the extent of damage in the same category and are ade" a:e for damage assessment. ii~P also predicts that clad melt occurs at the time when the local oxidat'.cn reaches about 20". The code limits the local oxidation thereafter. Clad .-.elt occurs at 3350'F and is assumed to block the flow chanrel, thereby limitirg steam flow and preventing further oxidation at elevatiors above the blockaco. e To the extent that this analytical model correc.ly prevents continued local oxidation, the core wide total hydrogen production is limited and Figure 4-1
~ould yield a low esti"ate of the nu,"..ber of embrittled rods.
The las: '1AAP case is for boiloff at an inter-.ediate pressure of 1200 psla psia ~ and is shown by Figures 4-23, 24 and 25. The maximum temperature vs. time is shown ',n Figure 4-23. Within less than a 200 sec. differential, the t<AAP c"ce yields the same temperature as the simplified analysis. Such a shift of 2 or 4-60
FIGURE 4-22 CO'iIPARISOi J OF MAAP CODE )VITH Sls'IPLIFIED CALCULATION R ESULTS HIGH PRESSURE BOILOFF 3000 MAAP 1.08"~ DECAY PO'iVER 0 2550 TO 2490 PSIA RADIAL PEAK = 1.03 2700 MIVT CLASS
~ zaaa 1.0,o DECAY POPOVER I- 2500 PSIA RADIAL PEAK = 1.0 0;.1'iVT CLASS 1000 600 100 80 0
60 I-40 O 0o 20 400 800 1200 1600 2000 TIAIE AFTER START OF CORE UiNCOVERY, SEC 4- 61
F I GUR E 4-23 TEiVIPERATURE vz TIME COMPARISON OF MAAP CODE AND SIMPLIFIED ANALYTICALRESULTS MAAP 3000 2700 MWT SERIES 1240 PSIA 3400 MWT SERIES 1200 PSIA 2500 0 2000 I-D X 1500 1000 500 400 800 1200 1600 2000 TliiIE AFTER START OF CORE UNCOVERY, SEC 4-32
FIGURE 4-24 TE IPERATURE vs RADIAL i JUCLEAR PEAK CO" 'IPAR I SOi I 0 F ii1AAP COD E AND SI i%PL I F I ED Ai I A LYTICA L RESULTS 3500 CLAD .'i1ELT 20oo 51AAP 3000 w/ 22io r 14o'OOLANT
- u. 2500 Sl".1PLIF I ED 0 ANALYSES D
I-r r LEVELS I-Q O, 2000 r D X Q0c/ 1500 30-.o 1000 60' 4 Qol 500 0.6 O.S 1.0 1.2 RADIAL NUCLEAR PEAK
)3
F IGUR E 4-25 MAXli".)Uid LOCAL CLAD OX)DATIOiVvs RADIAL iVUCLEAR PEAK-COi%1PARISON OF ie1AAP CODE AND Sli'i1PLIFIED ANALYTICALRESULTS 24 20 P% 'z 0 SIMP L IF I ED i%1 AAP L ii%1 IT 16 AiVALYSES V/HEN CLAD I'ilELTS 0 9",o Q u
~ 12 COOLANT u LEVEI S 0
i'iIAAP 22'.0 D 140/ r r r r 30-;o 21 ao 0.6 0.8 1.0 1.2 1.4 RADIAL NUCLEAR PEAK
to a..a,n a given .emperature is not so impor:.nt to this procodure 'e as the core wide distribution in temperature and oxidation at the time a given peak temperature occ.rs. Figure 4-24 shows the dis:ribution of temperature with radial nuclear peak at several times or coolant levels. The NAP code ppredic .s a faster ashier uncovery, unc v but C d'or a given peak core temperature, the core wide distributions are essentially identical S'imi ar 1, y, F igure 4-25 shows that the radial distributions of local oxidation are identical The conclusion is that for a given otal amount of oxidat'.on as inferred '~ I rot the PASS hydrogen r.easurements the simplified analytical model yields the correc core radial distribu ion of the oxidation and there=ore of he ru-.."er or rods which are ruptured or are erbrittled. Similarly for a givon ">>k core temperature the simplified analytical model also yields a corr .. estimate of the number of rods with temperatures above a spec:fico clad rupture temperature. 4.5 BASES FOR RELATION(SHIP BFT'~E"=tl Ai~<OUt<T OF HYDRCG""'I t'lE'~UR".0 All0 ANGUS>T OF CORe. OXIDATIOll There are multiple sources of hydrogen gas production inside the containment-- building during postulated accidents in addition to the oxidation of the 2lrconium in the core. These other sources include: the hydrocen gas normally tound in the reactor coolan for corrosion control the oxi "ation o= various metals used wl thin the containment 'I and the radi o 1 y t ic d eccmposition of water. The procedure for core damage assessment using hydrogen gas measure...ent employs correlations between the degree of cladding rupture or
oxidation and the amount of hydrogen produced by the initiating chemical reactions. The hydrogen measurement is performed by .he PASS sys em on samples obtained from both the Reactor Coolant System hot leg piping and from the containment building atmosphere. These measurements of the hydrocen gas do not distinguish between the produc.ion source of that oas. Therefore, the procedure requires a means to determine that contribution of hydrogen to the sample measurement which is produced from sources other than core oxida-tion. This is accomplished through analytical estimates of the produc ion rates. These production rates are sho~n to be source dependent upon either the containment building atmosphere terperature or the fission produc. dis:ri-bu .ion. This section describes in detai 1 the analytical techniques and assumptions employed in the required determination. Each of the sources of hydrogen production are discussed. The reac:or coolant under normal plant operating conditions contains a dis-solved hydrogen" concentration which is present to scavance any oxycen which may be present and thereby reduce the potential for corrosion. This hydrccen is present as a result of the radiolytic decomposition of both the reactor coolant and in some plants the pH control additives and as a result of direc. addition through the Chemical and 'Iolume Con rol System. The nor,.al operating range is between 10 and 50 cc3STP/kg. Therefore, the total quantity of hydrogen present in the coolant for any C-E tlSSS is anticipated to be less than 500 SCF. This value is considered to be a negligible contribution to he measurement of total post accident hydrogen concentration or several reasons. First the presence of this quantity of gas is well known and would not be misinterpreted in a post accident, measurement. Secondly, the contribution of this quantity to the concentration in the containment building atmosphere 4-66
under he assumed cord~..on in which all the hyd.ogen is released is less than 0.2 volume percent. This value is close to the mini.-..ua sensitivit'= 0t "e she typical PASS measurement capability and is bel e cw tth~a o; "...e procedure wnich distinguishes only between the assumed
.'(RC condition is unlikely because the s o 1 u b'1 i '
categories of core darace. i.y o- h y d rogen Also the in water is such that a depressurization to below 100 psig uould be reouired for a cc;.,piete release. Er.:..p ., i the loy ed with',n h containment building are a variety of etals which
. cons i-tu.e a potentially sicnificant s'ource of hydrogenn pro prod uc.icn as a bypr.cuc: of the oxidation reactions which occ r as corrosion The specific metals wnich con.ribute .o this source are principally aluminum and zinc All o her metals known "o .o be an insigni-.icant contribution 're when corpared to these two iso.
Aluminum and zinc are found princ.'pally wi thin the electrical ccmoonents paint, and ga lvani ed steel struc:ures. The corrosion reactions are a resul t o- the chemical environment uncer accident conditions. Independen of :,".o postulated accident scenario, these metals are e'xposed to a borated solution which has been pH adjusted to the alkaline range througn the addition of chemical additives, which contains significant amounts of dissolved oxycen due to the exposure to the containment building atmosphere and which under"oes transient tempera ures that may rance up to 300'F. The rate or hydrooen prccuction from oxidation of these metals is de"en"en" upon many variables which include the surfaco to rass rra .io io of o th e retals, the
~
quantity o; the metals present, the use of protec ive sur=ace coatings on -ie the meta s pr sent, the surfac te...perature, the presence of pH additi es, he 4-67
extent of metal e,.ersion in the borate solution, and the experimentally Ceter-mined reac.ion rate constant. These variables are plant specific. Therefor~ implementation of this procedure requires the development of plant specific analytically determined hydrogen produc ion rates. The production rates. are required to be expressed as a function of the containment building atmosphere temperature. This information has been developed using input data available from the latest revision of each specific plant Final Safety Analysis Repor- Repor. (FSAR). The result is provided in the form o. a graph, Figures 4-26 throuah 4-31, for each plant. A detailed description of the procedure used to calcu-late this infor...ation is provided. These curves may be used directly or each utility may choose to redevelop a given plant specific c;rve based uccn more recen. input Cata. The purpose of providina these grapns is to give exaroles of the analytical techniques and assumptions to be employed. The subject, procedure is a document intended to provide an actual assessment of core damage Yor the purpose of implementing emergency operational cecisions following an accident. Therefore, care should be taken to e...ploy results of realistic or best estimate dose rate analyses rather than conservative infor-mation which may have been developed for such purposes as licensing activities or 'he
.he design bases for equipment. The use of hydrogen production analyses based on conservative assumptions could actually result in a lower than actual assessment or core darage. This is because the conservative assumptions dictate greater consequences for the production of hydrocen than may be actual for a given category of core damage. The measurement of he lower or 1'ealistic hydrogen quan ity would then be corr lated to a lower than actual category or core damage. The purpose of this section is to describe the analytical assum ~ p.ions recommended o be employed in that realistic analytical cevelopr,ent.
4-68
FIGURE 4 26 HYOROGFil PROOUCTION RATE FROM ALUMINUilIANQ ZINC vs TEMPERATURE FOR CALVERT CLIFFS UNITS 1 8c 2 1100 1000 900 800 700 Z C CJ c 600 C z o 500 Q ~ >. 400 1-300 200 100 100 150 200 300 TEi'sIPERATUR E, 'F 4-69
0 F IGUR E 4-27 HYDROGEN PRODUCTION RATE F RQibl ALUMINUlVlAND ZINC w TEMPERATURE FOR PALO VFRDE NUCL AR GENERATING STATION 4200 4000 3800 3600 3400 3200 3COO u 2800 z O 2600 l O 2400 C C E 2200 z ZOOO Q 0 1800
~I
~ O 1600 1400
~
1200 1000 800 600 400 200 100 120 140 160 180 200 2 0 240 260 280 300 TEMPERATURE, 'F 7Q
FIGURE 4-28 HYDROG E'V PRODUCTIOiV RATE FROibl ALUiblli JUi'~l AiV0 Zli JC vs TE'PERATUR E FOR ST. LUG IE UiVIT 2 8000 7600 7200 6800 6400 6000 5600 o 5200 Z o 4800 I o /j D 4400 E 4000 z 3600 C C t
>. 32CO G 2800 ~l 2400 2000 1600 1200 800 400 100 120 140 160 180 200 220 240 260 280 ~00 TEibIPERATURE, 'F 4-71
1200 FIGUR E 4-29 HYDROGEN PRODVCTIOi J RAT FROiil ALUMINUiilAND ZINC vs TEillPERATURE FOR SONGS UNITS 2 & 3 1100 1000 900 800 700 U C/J z 0 C 600
,)
0 500 U O C 400 ~t 0 l-300 200 100 0 100 120 140 160 180 200 220 240 260 280 300 TEMPERATURE, F 4-72
F IGUR E 4.30 RATE OF HYDROGEN PROOUCTIOW FPR ALUI.IINUislAND ZliMC vs TEA1PERQTURF FOR 5'/PP$ $ 20 19 18 14 o 13 Z G 12 u 11 0
- c. 10 z
8 0 6 I 0 100 120 140 160 180 200 220 240 2EO 280 '00 TEi'rlPERATURE, 'F 0 4-73
I FlGURE 4-31 HyDRQQEN PRODUCTlON RATE FROM ALUie1 INU> 4600 7lNC vs TEMPERATURE FOR IVATERFORD UNlT 3 4400 4200 4000 3800 3600 3400 3200
- u. "000 2800 c
2600 2400 z 2200 0 C 2000 ~ c1 800 1-1600 1400 1200 1000
. 800 600 400 200 1pp 120 140 160 180 200 220 240 260 TEiblPERATUR E, 'F 4 74
"e '- iables requi~ed in the analyses are well defined for a specified plant. These variables include the specifics of the metal inven-orv >nven.ory inside the c"ntainment buildirc and the presenc o= aadd'f i ives to adjus the pH to an alkal ine ran ange. . The Th in-.or...ation used in the analyses of the produc"ion rates provided in Figures 4-26 throuch 4-31 ha.as b een o b tained ffrcm the plan.
speci ic FS~R. The inventory of aluminum and zinc expressed by weigh and when asailable surf thee sur.ac to volume ratios are s ated in the FSAR which describes the sys ems for hydrogen control within the containment building Two assumotions were made in the analytical product pr C on rate determination with regard to .he metal inventory. First it is assu-.ed tha . . h ose meta s, such as pain , for wnich th the surfac to ..ass ra &i 4 o is large, oxidize coroletely wit.". n ore hour should the c c"n-'e
- n. irment ui lding at."..osphere temperature exceed 2GO F.
ibis is assumed bec use acc'ate ex-erimental p ' ' ra e cons-a ~ n.sts are not avai'able for larce sur-ace to mass ratios. The conseauence of this assumotion is th: the pr ceCure should not be applied to hydrocen samole measure,.ents taken h'i.hin the rirst hour -allowing an accident. This is consistent ~itn pr~sen-design capabilities of typical PASS systems which require several hours to d'or obtain and analyze a representative sample. A value is provided in Table '-3 for hvdrogen assumed to "e procuced in this manner '.or these plants with specif c FS~R data. Second it is assu .ed tha there is a limit to the tote 1 quantity or maximum yield of hydrocen produced based upon the deplet'.on of the material present. This results from the assumption of ccmplete reaction =or those materials e ia s with lare surface to mass 1 ratio and an assumed surface Ce p leticn '.
.hose materials with a small ratio. A valu e s provided in Table ~-4 for t..e the max'mum m hydrogen yield .or those plants with spec'.fic FSAR data. As previousl sta- ~ .ed, .hese values may be used directly or each utility may choose to emplo a p1 ant specif.c value based upon more recent data.
4-75
I TABLE 4-3 HYD CGE'I RESULT IiiG FRCiM IHSTAttTktfEQUS REACT ICi( 0 ALUNINUN At(0 PLAUDITS HYOROG""!( SCF Calvert Cli fs Units 1 8 2 Palo Verce 'tuclear Generating Station St. Lucie Unit 1 St. Lucie Unit 2 5235 S0.'lGS Units 2 6 3 200 WPPSS 8398 Waterford Unit 3 - - ~ Indicates no instan;aneous reaction was assumed, due to insuf,icient data on surface to mass ratio in the FSAR. 4-?6
TABL"= HY0RQGE') Y1 "I Q F'"Cq 4-'~AX[i"Ui1 PL.~ITS HYCROG-ll SCF Calvert Clii=s Units 1 8 2 tIot Available Palo Verce iluclear Generating Station 200,271 St. Lucie Unit 1 481, '142 St. Lucie Unit 2 292,051 SCllGS Units 2 h 3 90 ~Q/3 'AP PS S 86,6c6 Mater, ord Vni t 3 173,582 FSAR does not provide the in-.orration needed to estmate i"'axirum Hydrogen Yield. 4-77
ihere are, however, several variables which require the use of analytical assumptions. These variables are the extent of metal emersion in the bor te solution and the selec"ion of the reac.ion ra e cons-ant based upon the available experimental data. These analytical assurptions are related because the reac.ion rate constant depends upon the extent of emersion in the borate soluticn. E ch of these assumptions is discussed in detai l. The oxidation reac .on rate is influenced by the availability of oxygen and the builCuo of reaction byproCucts on the surface of the metal. The avail-ability of oxycen is assured by he presence of the containr..ent build',ra a:;..os".here. The builCup of reac:ion byproducts on a metal surface is Ce."en-Cent u"cn the presence or flow of the bora:e solution across that surface. metal sur;ace submerged in a st'gnant pool of water will corrode a: a slcwer rate than a surface wnich is uncerooing a spray of water. This is because the spray flew will remove the soluable reac ion byproCucts and continually ex=cse new metal to the corrosive envircrment. The procedure has been Cefined as app icable in the core damace category of uel overheat. The category of fuel overheat cannot be reacred unless some fraction of the core has been uncovered. Therefore, it must be concluded hat a large r.'.ass of high entha>py reactor coolant has been released :o the containment buildira. This is su.- ficient to assume that the metal surfaces in question are emersed in steam with condensation resul ing in a limited flow of ~ater. ACCitionally,'should it be ac.ivated, the Containment Spray System will further increase the surface flow rate with borated coolan and pH control additives. Independen o a speci ic accident scenario o it is also concluded from the category of core damage that the contents of the Safety Injection Tanks and Refueling Mater 4-78
Tank are introduced into the containment. This aCCiticnal water results in a sufficient level in the buildina sur,",p to subr,.erce some of the metal inventory. Hcwever, a review of the FS~R data indicates that only a small fraction of the a 1 uminum and zirc is anticipated to be ccmoletely submerced in the sump. Therefore, the reccr...endation is that the assured reaction rate constants should be determined frcm expe rir;,enta 1 data whi ch address a r.etal surface at elevated temperature exposed to a continuous flew of borated water with pH adjusted to the alkaline ranae. The reac. on rate equation to be emoloye~ 'h in . e oxidation d of aluminu." anC zinc is similar in fora to the equaticn used tor zirconium oxidation Cescri "eC in Section 4.2. 1. The fon of this equation is expressed as: u = ~ metal where. equivalent metal rass oxidized per unit area per unit time metal experimentally determined rate constant experimentally deter'mined ac ivation eneray ideal gas cons: nt surface temper ture This equation can bee used :.o yield the rate of hydrogen prcduc.ion by assumina the cxidation re c. on to reaction t recede in a stoichiorretric -anr.er. The equation is hen adjusted by .hen a simple unit conversion .o yield th'e standard quan ity o0 f hydrccen produced per unit surface area per unit time. 7g
The two values which mus be determined from experimental Cata are the r te constant and the activation energy. These values will vary for each metal . A survey of the published experimental data was conducted to determine the values reccr,.ended for use in this procedure. This survey identified a number of temperature dependent influences on the applicability, of an exponential rate equat',on. The observed effect of temperature on corrosion reactions is not explicitly exponential as it would, be for most chemical reac:ions nor linear as it would be under physical change. Each are discussed as an indica-tion of the limitations of the analytical prediction and .herefore the accuracy oi the procedure. Ter.perature may affect the corrosion rate through its effec. on oxygen i .usion in
~ the borate solution at the metal surface. The corrosion rate ."..ay increase wi th,temperature. The rate will decrease rapidly to a low value a the boilirg point Cue to the decrease in oxycen solubility.
Temperature may affec: the corrosion rate through its effec. on pH. Because the dissociation of water increases with temperature the pH decreases with emperature. This is more significant in the early time periods o an accident prior to the introduction of the pH additives through the Containr.ent Spray System. Temperature may affect the corrosion rates throuah its effect on the surface films. 1 The temp;rature dependence of the solubility of the protective corrosion byproducts will vary the corrosion rate with temperature unless an "~ aggressive surface flow is present. A change in temperature also may bring about changes in the physical nature or the chemical composition of the. protective byproducts. z) 4-80
fl Heat lux may affec: the corrosion rate. The te..."eratures of the metal sur;ace, the borated coolan , and the ccntainr,:ent building atmosphere may all be dir C erent and each may be anticipated to vary with location inside the containment bui l dirg. Based upon the survey of the available experimental data the . ollowino corre-laticns are reccmmended as being applicable to this procedure for the realistic analytical estimate of hydrogen production rate. For the oxidation of aluminum, the exoer;..ental data provided in ."-.e;e. erc ?-1g was -it to the exponential la e equation 'to yield 16g0 W = 5.9 x 10 e ~T."".} For the oxidation oi zinc the data provided in Reference 7-20 s ates the r~-~ equation to be:
?233 5
W = 2.1 x 10 e ~T-K) where in each case the variable W is the produc ion rate of hydrocen expressed ~resi as s andard cubic,oot produced per square foot of metal surface area per hour. The validity of these selec:ed correlations to provide a realis.ic estimat~ was veri".ed by a series o. parametric analyses in which the cons nts determired e4 frrcm a num er o- ot..er published reports were employed to yield estimates o; thh~ hydrogen h prcduction rates. ihe recc",...-..end equations were 4-81
shown to yield results lo~er then those correlations known to have been developed or the conservative purpose of licensing activities or equipment design. Therefore, the above equations were employed in the calculation of the graphs provided in Figures 4-26 throuch 4-31. The Cecomposi ion of water by radiolysis results in the production of a significant source of hydrogen gas. This is especially true following an acciden. in which fission products become dispersed in the reactor coolant. Under this condition, the radiation shielding provided by the structural materials of the core is no longer present and all of the radiationn energy, energy both beta and cama, is absorbed in the water. The rate oi prcCuction of hyCrccen frcm radiolysis is dependent upon many variables which incluCe reac=or power, fuel burnup, the extent of fission product release to the ccolant, anC tne rate constant for hydrogen produced as a function of eneray absorbed. Each of hese variables can be normalized to yield a gereric procuc=ion rate expressed as a function of reactor power. ihe hydrccen prccuc:ion from radiolysis is dependent directly upon the q nt',ty o- .he -ission prccucts releasec from the core. The convention employed to express that quantity is the same as that described in Section 3.3 for use in the interim Procedure. The quan.itative release of fission products is expressed as the percent of the source inventory at the time of the accident which is observed to be present in the sample media and therefore available for immediate release to the environment. The reason f >r this c"nven ion is the limit on the presen capability .o predict fission product ranspor ou of the core following an accicent. Defining the cuantitative 4 c2
release in this wa y d'ces not '. ply a quantitative kncwledce of the r.echan'.sm or transport phencmena of that release. The radiolccical charac:eristics cf the .'lRC Categories of Fuel Oar;age presented in Table 3-1 remain ~ plic~ble -o the Ccr<<prehensive Prccedure. The radioloaical characteristics are expanded upon in Table 6-1 to include the distribution of the fission procucts inside Luild' the containr,:ent Luilding. This dis .ribution is an est'.mate based upon tlRC Regulatory Guide assumptions. The core damac e assess<<.ent procedure usina hycrocen measurements is apolicable in .he ~lRC catecory of -uel overheat. Table 3-1 identifies the source of fission prcduc: release for this cateccry to bee t"ee
' ue pe lllet.
1 The source ~ inven.ory is calculated using the techniques descr bed in Section 3.3. Th~ source inventory includes all fission prcdvc.s ana is not limited to the lis= provided in Table 3-4. The source inventory is expressea as an enercy scvrcescvr"~ term in Tabl e 4-5 f for hydrocen h produc=ion forlowirg major fue'1 overheat.
/alues are Values provided, or enercy deposited in the coolant .rcm scvrces mixec with the coolant and rcm sources in the fuel rods. The fraction o- the tc- l.
garrma eneray deposited in the coolant frcm sources in the .uel rccs is analytically determined usina a Non e Carlo radiation transpor. rcdel us inc the gecr;etry of C-F. designed fuel asse.,blies. ihe rate cars;ant for hydrogen produced as a unct'".n nf enerrv absor b(.~ <<% 's I an p,i .<<.en~a I 1'/ (i '>rr..inr (:. v<< ye.
~ Pa<ad ( ccn a 5L I ~
experimental data, the value rep(~".~d in Reference /-r.. hT 0.3 l<<alee!plus os hvdfo.'e <roduc "(I ocr 1CO ev . r e::e .y absc b:.'-
- h. pro( e~ '.r".".r:r.cion ra.-.:
~.-. <<a ~ "alys'"
ys,". ~ i,e ra~iolysis of ':a'r -". a closed system 4-83
TABLE 4-5 E!'ERGY SOURCE T ?NS FOR HYDROGc..'l PRCCUCTICll Il( CG'iTAINME."JT FOLLCWlliG MAJOR FUEL OVE."HEAT Enercv Oeoosited in Coolant frcm Sources Nsxea wstn tne Coolant Time After Shutdown Seta Gamma Enerov Enerc'I!eV/Matt-sec) (hr) (Yev/Watt-sec) 1 2.26 (+9)" 6.55 (+9) 10 1 17 (~9) 3.20 (-:9} 100 8.61 (+8) 1.10 (-o) 10CO 4 55 ( 7) 4.53 (.7) Enerov Geoosited in Coolant frc.-.. c rces >n tne rued .-.Gas Frac.ion of the To:al Time After Enerc s Deoosited Garma Ene re@ in Fu e l Rod Shutdown in Coolant Deoositea in Ccolar.: (hrs) ('!eV/Matt-sec) 1 01 (yo)x 0.076 10 9.25 (-8) 0.054 100 7.'72 (+8) 0.072 1000 2.'8 (-:8) 0.066
- Indicated powers of ten.
4-84
36ga(8"GS)/si-65 includes both:he prccucicn d rec .-..binat on oi '.he byproduc: c s. Ther rore the rate c nstant is kncwn to exhibit i. -. ssc;..e temperature de"endence because or the indirect eiiect it has on th..e solubili y oi the gas byprcduc:s. As temperature increases, the gas solubil't i y wi'll decrease and thereicre, the ra te o C r ecc.p ~ nation w> 1 1 decrease. Th e value 1 will be assumed to be a constant within the accuracy oi this procedure F The specific radiol tic hh ydrogen production rate calculated with the assu-.," 'ons descr'.bed above are provided in Figure 're 4-""
-~c, in the " form .. or a . .
gra"n. Two gaea.this func-'.cn curves are shown in figure because u e th e h y d rccen procuc:icn is a oi the fissicn p roduc release which is dependent upon the degreo o= core darage. The two curves represent the results cr th e ca cu ation 1 1 'or .-..a Jor and initial fuel overh using ihe radiolooical charac:eris:ics oi these ca.ecories described >n Table 6-1. The user oi ihis is pr"c < procedure is recuired to rely upon an estirate oi core damace cbtained frc~ onee oi or:he" ot.er orccedures provided in this docu;..ent to select the appropriate curve. The curves in Figure 4-~2 are applic ble to all plants because the values presented are ncrmali"ed to pcwer level. Coirection for variation in reactor power his ory prior to the accident is required because of the result upon the source urce fiss' ission pro uc inventory. The correction is us d ".o adjus d'ed i the measured sourc inven-or- o th e ccrrespondira value had the p lant b e n o operating at 4 full power. The analyticall y d etermined hydrocen p rodu c.ion ra,es which are used to assess core da a;ace are ca'cula<<ed assumin g ull 1 pc~er equilibrium scurce inven-ories Th e t ec h niques e...ployed in this correc:ion aree <<h .he same as :hose discussed in Sec .'io 3 ..3. Th ey are 4-85
F)GUA E 4.32 SPECIFIC AADIOLYTICIIYDAOGEN PAODUCTION vs TIME 13 12 MAJOA FUEL OVEAIIEAT 11 CV 10 0 9 U D D 0 0 lC Ul (3 7 0 K O INTEAMEDIATEFUEL OVEAIIEAT X 6 U I-gC 0 0 4 U U 3 Ll3 INITIALFUEL OVEAHEAT 100 200 300 400 r~OO GOO 700 . 000
", IIOUAS
the COndi iCnS that: equilibrium source inventory is reached af er 30 days of ccnszant pcwer oper zicn'he eqvilibrivm value is directly proportional to the product on rate expressed as i r power; reac-or ~ and that constant pouer operation reans no chance greater than - 10 percen . The result is that a sir,:ole ratio of the power may be employed :.o o bt ain . e full power equilibrium correc:ion. Corstant reac:or power operation is not no anticipated. Therefore, engineerinc judcement is required to determine the reactor pcuer uhich is most re resen".- t've o. .he fissicn prcduct inven cry prior -o the ac cident. cxpl ' Ce er...i-nation ot the representatise value would re~uir~ d ~~'1 e d ccm"uter cede e.ai ai d o-i speci ic analysis. y Hcwev
. e er, ercine ring jvdgerrent e.played wi h ..e t
guidelines is suf icient to yield a resul within the accur'cy o'his procedure. These cvidelines are as follows: ihe averace The pcwer during the 30 day tire period is not recessarily the mos. representazi se value for correction to eqvH ibrium corditions 1 iions ~ (2} The last s cu r levels at wnich the reactor operated shculd ueich -..or~ pcuer heavily in the judcen:ent than he earlier levels. (3} Continued o"erazi e aiicn for an ex.ended period should 'ueich rore heavily in the judge.".ent than brief transient levels. 4-87
0 0
5.0 ES-;,.-.L SH..EHT OF HE BASES FOR CO?E DAi'AGE ASSESSilE!1T USI"" CO?E EXIT THER)'.OCOUPLES A:>0 OTH" R 11(STRUl'HEI'1TS The objective in utilizing the Core Exit Ther'ocouples (CET)is to obtain another redundant indication of core damage. The CET can supplement the radiologic and hydrogen meaurements and can provide a damage assessment for events which are less severe than those which generate measurable amount of hydrogen. Two additional instruments indications, the pressure and the coolant level above the core are used to supplement the use of the CET in the determination of core damage. The CET provide the primary indication of core heatup following the start of core uncovery. They function best for slow uncovery by boi loff. The pressure indication supplements the CFT in providing information on the probable rate of core uncovery and therefore on whether the CFT temperature is a valid representation of core temperature. Also, the pressure indication is used to reduce the uncertainty in the prediction of clad rupture temperature. The rupture temperature is a function of the differential between the internal gas pressure and the reactor coolant pressure during the period of core uncovery. The tir e period of core uncovery might be questionable when only the CET temperature is utilized to determine it. By using the reactor vessel level indication, another estimate of the uncovery period is obtained. The level provides th time p riod when the pressure should be observed and also can help in interpreting the transient behavior of the CET temperature. These three indications, pressure, level and CET temperature are discussed in the following three sections. The end result is a prediction of the number of 5-1
ruptured fuel rods, some interpretation on the probable accuracy of that pr"diction, and an assessment of clad damage expressed as one of the NRC categories of damage given in Table 4-1. 5.1 PRESSURE I flD I CAT ION The rate of pressure decrease indicates the relative size of break for LOCA events. A large break LOCA causes the pressure to fall from the normal operating value to near containment pressure within about l00 sec. or less. This indicates complete core uncovery by blowdown, and is an indication that the C:i temperature rise is probably much less than the core temperature rise. The damage assessment procedure using the CET could greatly underestimate core damage. A small break LOCA causes the pressure to fall to within 100 pst above the secondary pressure, remain constant for a period of time which is proportional to break size and then to continue falling (See Reference 7-9). This indicates relatively slower core uncovery by boiloff, although functioning of the ECCS may prevent core uncovery. The CET respond best to such slower uncovery events and the CET temperature may be given greater weight in the core damage assessment. The prediction of clad rupture using the CET temperature includes an assumption on the rupture temperature. This rupture temperature varies substantially with individual fuel rod burnup and system pressure. In order to remove some of the uncertainty, predictions are made for three postulated 5-2
0 rupture temperatures which are related to the system pressures. The pressure indication during the uncovery period enables the choice of the rupture te .perature which yields the best damage prediction. 5 LC'lEL IHO I CAT IOiN The coolant level in the reactor vessel above the top of the core is used to estimate the period of core uncovery. Knowledge of when the core is uncovered is used in two ways in the orocedure. First, it determines the ti;e period during which the pressure should t be observed as dicussed in the preceding section. Second, when the core is uncovered the CET temperature should rise. If it does not rise and fall according to the core uncovery period then some interpretation of the validity of the CET temperature as an indicator of core temperature is required. Hence, the level indication can help to reduce the possible uncertainty in the assessment of the core da...ace obtained with the CET temperature or at least can suggest that there is a potentially large underestimate of the predicated damage. The relationshios between CET temperature and level are discussed in the following Section 5.3. There are two co;;ercially available reactor vessel level systems available. tleither indicates all they way down to the top of the active fuel, so some procedure for extrapolating the recorded traces of level may be needed in the plant specific procedure for core damage assessment in order to obtain a good estimate of the core uncovery period. Such plant specific details are out of 5-3
scope for this generic procedure guideline. Owners of C-E designed reactors with the C-E level monitoring system can obtain a performance evaluation during accidents from Reference 7- 15, sponsored by the GEOG. 5.3 PREDICTION OF FU"L CLAD RUPTURE BASED ON CORE EXITI THERYOCOUPLE TEr~PERATURES A performance evaluation of CFT's during core uncovery is provided for C-E designed reactors in CE-HPSD-212 (Reference 7-16) prepared for the C-E Owners Group. That report should be used to supplement the presentation here. It is assumed here that a Core Exit Thermocouple instrumentation system as required in Regulatory Guide 1.97 is available. In addition, the system cavabilit must allov the user of the procedure to obtain the maximum CET temperature and superheat as a function of time during the uncovery period. Plant specific implementation may require additional hardware and/or instruction on how to obtain these data from the instrumentation system. Given a system, several considerations on the performance of the CET during core uncovery are presented and are related to the interpretation of CET temperature as a measure of core clad temperature. The CET's are located depending on reactor design from 10 in. to 26 in. above the top of the active fuel. They do not monitor fuel temperature. They do monitor the temperature of fluid above the exit of the core which is liquid normally and steam and/or gas during uncovery accidents. In order to infer core temperature from the CET indication, there must be a flow of heated fluid from the core past the CET elevation. During complete core uncovery or
temporarily follcwing any rapid partial urcovery, .here is no low to couple the core and CET te..peratures, so the core heats up without a CET indication Ourirg slower uncovery by boilof=, the superheated steam flowing up fro~ rom "he
<he core provides effective thermal coupling and core heatup is indicated by tl.e rising CET temperature. The information which is obtained from the CET record for both of these two cases is discussed in the followina.
I n the first case, while the core is uncovered, there is no thermal coupling but some'useful infor...ation ray be obtained from the CET emperature record tempera"ure during the recovery period. Ouring a rapid recovery, the liquid entering the bottom of the core pushes the steam and gas up and out of the core, past the CET. This may cause a sudden, shor: duration peak in the CET temperature when it follows a deep uncovery. It is an indicator of more severe prior core heatup than would be inferred from the earlier CET temperature record. Tese data from LOFT and Semiscale LOCA tests which are summarized in Reference 7-16 confirm this behavior. The conclusion is that such a scenario yields a CET temperature record which ha,s some characteristic behavior that i ndi cates the potential for higher core temperatures and therefore greater core damage than this procedure would predict. In the second case, when there is thermal coupling between core temperature and CET temperature, the effectiveness depends on the particular CET installation configuration and on the time rate of steam temperature chances 8oth of these contribute to the amount of time lag in th e CE i response. Some 5-5
na ytical calculations of the ti-e res"onse of C in C E desicred reactors rra the LOr i and Sem>scale facial >t>es, which are su;,.marqzed >n Re-,er~nco 7- ~ . For the C-E desicred reactors, this lag varies with re~c-or des, t abou~ 6 minutes. abou It has the effect of reducing the peak temperature recorded'rom the CET and possibly of shortening the apparent duration of core uncovery. For example, Figure 5-2 shows the calculated steam temperature as a function of time C or a 0.1 ft small break LOCA and the delayed CET response temperature. The conclusion is that the CET recorded temperature represents a lower limit for the peak steam temperature and could be substantially lower than the peak core temperature. Therefore, this procedure will yield a lower limi: prediction of core damage. Typical analyses of Design Basis small'break LOCA predic. continuously decreasing pressure as the core uncovers. This can cause a delay in the apparent tire of uncovery as indicated by the CET temperature. The reason is that the satura:ion temperature is also decreasing along with the pressure decrease. There= o r e, the level could drop >nto the core and cause superheated steam at the top of the core but the temperature of the steam increases more slowly. The conclusion is that it is desirable to also trend the superheat derived rom. CET temperature and reactor coolant pressure. The superheat of the CET also has a disadvantage for indicating the time of core un'-overy. This disadvantage ccmes from the location of the CET above the top of the core ccmbir ed with the thermal lag in the temperature response of the CEi in a steam environment. The CET is at saturation temperature until 5-6
FIGURE 5.1 TEMf'EBATUBE vs Tlt)E AFTER STAB I OF COBE L)f)COVERY DURING 0.10 f'T~ LOCA, AVEIIAGE ASSEME)LY II I:-SI'()flSE II( 2700 ILliVT SL'HIES 1600 1400 FUEL CLAD TEMP EIIATUB2
/'L 1200 CORE EXIT STEAM TEMPERATURE a
Uj K D 1000 <C CL lO 800 / r STEAM TEMPERATURE IN lIIPI'LcII I'l. LEWUf.l
~
600 COBE EXIT TIIEBIIIOCOUPLE TEIiIPEIIATUIIE 400 200 4()0 (I()() ann 1000 1200 1)00 TIILIE AFTEfl START OF COAL UNCOVEBY, SEC
the level drops below the C:-T loc tion. Thereafter he CFT e,."er"ure remains about the same or changes slowly whereas the saturati "n te~"erature is fal ing 1 wi th "ressure. ence, the C="T indicates superheat and a = lse indication o- core uncovery. Fortunately, these effects are usually srall ard last only several minu es in the desion basis s s LCC ~ ana yses, until the core uncovers and temperatures rise subs-antially It wou ld pro b a bl y be important only in the less severere seer!ariasseer! r where there is some question about whether cr not the core started to uncover. Ourirg reco:ery, the CRT rapidly quenches when the coolant level rises above the C":T elevation. The C"-: ind' d'ca.ion drops to saturation tempera:ure or lower, de"endinc on the rra e o coolant flow. The conclusion is that if the C=i indicate superheat, they are certainly uncovered and the ccr~ is ei-her unco;ering or is about to start uncovering. If ih e C="
=i are su c"oled and a '.dic . d or i level is irdic"od trey sued nly go frcm su"erhe--s 't o suvccu > ~ r) ine/
y and the core are covered. The CET are oenera1 1 d esigned .o unction or accidents which are less severe than the Tl!I-2 ac";cent. c-'dent. They have good accuracy up to readings about 1650'F and c n .rend up to about 2"00'F. The thermal-hydraulic functional design objectives for the C:-T and
~
other inadequate core cooling instrumentation are provided in C""-llPSD-199. (R; (Re.erence 7-1I } for the CEOG. A ccparison with the bo.. m en.ry CEi syste.;.s C:--liPSD-171 (Peference same as ( for top rounted on the 5-SO e 7-1S).. 7 1S) C""T up design and on elaine Yankee is provided in Their functional capabilitJ is essentially the to core temperatures of 2300'F
ln order to utilize the CET recorded temperature, three relations are necessary. First, the peak recorded CET emperature must be related to the peak core temperature. Second, the peak core temperature must be related to the distribution of temperature in the core. Third, the core temperatur distribution must be related to an assessment of core 4 damage. These three relations are discussed in order. The relationship between the peak core temperature and the CET temperatur~ is ind',cated by an example. Analyses of Design Basis small break LOCA events on the 2700 l'.~t class indicate that when the core ore pea~ pea' c ad t temperature during an event is in the rance 1500 to 1700'F, the maximum subchannel exit steam temperature is about 2:"O'F to 100'F lower, respec.ively. The magnitude of the CET temperature would be 1 o w er than h .he subcharnel steam temperature because of the ther;..al la~g and s. am cooling above the ac:ive fuel and because the CET sees steam -rcm an array oi subchannels. The averace temoerature from the array is lower than the peak subchannel steam te...per ture. The C"i temperature lags the steam temperature by up to about 6 min. When the duration ov uncovery is long cor pared to the lag time and the temperatures vary smoo.hip, ~he Cci temperature is closer to the steam temperature. At i ...erature, the peak core temoerature about 1600'F steam tern"er attained during Design Basis "-vents could be within about lCO'F above the peak steam temperature. The corclusion is that for smooth CET temperature records which-peak near the upper limit of the CET range o good accuracy at 16""O'F and -"or uncovery durations of 20 min or loncer the peak core tern emperature e t attained is within a few hundred degrees above the peak recorded C c.
"i t empe ra ture.
Sec"ndly, the peak core temperature is related to the dlstribu ion of te...perature on all the rods in the core by the analyses described previously in Section 4. An exarple of how the desired rela ionship is developed is given as follows. Assume the peak core terperature is 1950'F, as shown in
~ ~
Figure 4-11 when the coolant level is 30 . The figure shows that all rods with radial peaks above 0.75 have tempera ures above 1500 F. From Figure 4-10, about 94" of the rods in the core have radial peaks above 0.?5. If clad rupture occurred at 1500'F, then a peak core temperature of 1950'F is an indicator that 94 of the rods are ruptured. In this manner, the third relation between peak temperature and core damage is established. Successive values of peak core temperature are used and yield various percentages of 'rup ured rods. If the rupture temperature is 1800 F, another set of values is obtained. The results of this calculatioral routine are shown on Ficure 5-2. Three values of rupture temperature are selected for plotting, 1200'F, 1500'F and 1800 F. As discussed in Section 4.3. 1 on clad rupture criteria, the clad will rup.ure when held a temperatures in this range for about ten minutes, depending on the clad differential pressure. For new fuel, these rupture te.peratures correspond to coolant pressures of 100 psia, 1100 psia and all-pressures respectively. In utilizing these curves in the procedure, the first step is to determine the coolant pressure during core uncovery, as discussed in Sections 5. 1 and 5.2. The pressure determines which of the three curves best represents the clad rupture conditions. Use the curve corresponding to the measured pressure or 5-10
FIGURE 5.2 PERCENT OF FUEL RODS',V1TH RUPTURED CLAD vs MAXli%1UiMCORE TEi'vlPERATURE 100 ~I D 80 1200 F ~ VJ CLAD RUPTURE LQ C ~ TFi'lPERATURE I 60 1- w W VJ Q~ 0 '- 1500' ao Q u- O 1800'F O~ l- ) 7 UJ ~ 20 UJ 6 0 1200 1400 1600 1800 2000 2200 MAXli'i1Ui'r1CORE TEiilPERATURE
hicner. 0bserve the r;.axe;.un C=T emperature during ',",e vnc"very period. The
.--p ra " s at leas . this high, and could be con>> derably hicher dependinc cn the scenario as discussed perviously. E'nter Ficure 5-2 i "h peak ..
te,."eratvr e a ndd readd '..p rom ihe appropr iate rupture curve the percentace of rods which are ru pt~ r ed. d This is considered a lower limit estimate of the percent o clad ruptures. Table 4-1 then yields the tiRC category of core damage corresponding to the amount of clad failure. I is evident from Figure 5-2 that the number of ruptured rods is very I sensitive to the coree pe ~kk te...perature, and is even more sensitive to the assvmed rupture temoeratvre. The uncertai nti es in the darace assessment can be reduced with plant specific analyses. At any time during a given fuel cycle, the burr'.v p ~ and the consequent fission gas pressur vary ar...ong the ivei rods. Svrrup is scmewnat reiated to radial peaking factor. exan p le For exam 1 third cycle fuel ha s ..i r;,ax'...um internal pressure at a given te.~e r a iure, "vt it usually does no- have maximum radial peakino actors. Plant speci ic calculations for a given burrup cycle are feasible which could recuce the three curries d'or threo ru".vre temperatures to one curve at a given core burnvp and system pressure. Several curves could be derived for several valves o~ 1 durin the rvel cycles and burnu during o burnup for a set of system pressures. Such curves would reduce the uncertainn y in hee percent of rods which rupture gl Zen ccrc peak ter:peratvre attained dvring a core uncovery accident The cvrves in Figure 5-2 provide a lcirer limit estimate of nv-:ber of the nv.: ruptured rods for a given peak CFT temperature during uncovery by boiloff For ex mple, at an observed peak C"-T temperature of 1600'F and a rupture 5-12
te...perat'e 0< !":CO F, the ficure predic-. ~='-" c s "i ru tut eo e~ rccs ~ <he ~c "~i pea~ core temperature were 200'bovee 'h . e C"- .-.,->>-" C=i te.-.."erat'e, the correct ans;,er would be 75:. ruptured rods. As the temperature ircreases the urc~rtain v decreases. At a peak C":7 temperature ""-""'" of <.= 0 r, over ".0."< of the rods are predicted ruptured with high confidence, regardless of burnup or system pressure. If the maximum temperature CFT exceeds the upper lini t of indic tion at 2 00 F, no r,;ore specific statement about core da mace
. e is available rcm the maximum C";T recorded trace. A core map of all C"c~ .e."..peratures is needed to indicate the nu;..ber off C""
C:"7 above 2 00 F, their radial lccaticns ard tl e racial nuclear peaks at those loca;ions. With these dat, an esti;..ate cf the nu-,.ver 0 o rods l which are above the clad;.el-<<empera e...perat ure or whic.'ave exceeded t.".e oxidaticn e..britt!erent threshold can be r,ade or examp'e i= the CRT ir, locations with core average raoial "eaks exc~~d ee 2 'L "0'"-", at 1- ~'I east v.<< C
+< g) \ ~~
rods in the core "a,e ha;e m '. i "- clad. rel-~d ihe Camace assess;..ent prccedure presen;:c in the appendix dces not irc'ude the utilization of such core r:.aps of C"-7 temperature aandd radial radi 1 pe k. Plan~ speci-<c in;or;..atior, on data ac Jlsl c'l capabilities and,".,ore det'.'led analyses o- more severe core da-..age scenarios are required to extend this prccedure to incluCe the r,:easured distribution o-, core ex i t te"...pe ra:ure. 5-13
6.0 ESTABI ~<i +E.'l7 CF THE 2ASIS FOR CORE ni<p"qc=cc!~ci~T US~~~G DOS" RATE!<EA<L'R~~') Il/SIDE THE COalTAI"Nc'qT -"U>< DI"G m The categories of core damage are characterized in Section 3.0 of this document in terms of those physical parameters relevant to the release of radioactive material from the core into the containment building. These parameters are the source of the fission product release, the mechanism of fission product release, and the quantitative release of characteristic fission products expressed as a percent of the theoretical source inventor;. That information is used in the development of the Interim Proc dure for Core Oamace Assessment. The measurement employed for the purpose of the interim procedure was the specific activity of the samples obtained frcm the reactor coolant and from the containment building atmosphere. A discussion of the uncertainty in these measurements and the 'limitations of their use in core damage assessments is provided in Section 3.4. The same physical parameters may be evaluated using the measurements of the dose rate inside the containment building. The use of two different reasur ments for the evaluation of the same physical parameters is employed as a means to reduce the uncertainty. As with all of the physical and chemical-- parameters discussed in this procedure guideline, the assessment of core damage based on one parameter serves to compliment the assessment based on the others reducing the overall uncertainty. This section describes the basis employed in the Comorehensive Procedure for use of the dose rate measurement inside the contair,"ent building to assess core damage. 6-1
ihe ins.rument emploved in .he measurement is the ~iCe rance ar~~ ra i monitor. There may be several of these ins-rumen~s irs ce t..e
. ume..ts insice "e containment building. They are typically located in the higher elevati s of the buildin in order to measure i e vol the volatile fission produc:s as they disperse thorough through the building atmosphere. The dispersion of volatile f',ssion procuc 5 the building atmosphere is increased by air, low resul inc from operation of the hVAC system and by thermal gradients or s earn flow. Chemi ca 1 s prays are used to remove non-volatile and halogen fission produc s from the a,.os nere.
atmosphere ~s a ccnsequence of the identity o the fission produc:s to which these ins.rurents are exposed, damage overheat. assessment The dose rate measurements within the categories of "iue 1
'1'l' are ros- applicable to core c 1 a dd ing railure fission products of high volati 1't y, speci 'f ica y ncb)e gas ard and fuel halogens, are characteristic of cladding failures. The fission produc-s o=
in?er;..ediate vol ati1 i t c~ sium, telurium, and rubidium, are charac:eristic of ue overheai. Oose rate measurements are not ccnsiCered applicable to the catecories c of fuel meltt because b ihe Cemission product transport uncer this condi.ion 10 is poorly unCerstood and therefore realistic
'ordi dose rate correlations cannot be p erf or,.ed.d The reasure...ent of sarple specific actlvi y is rore more ird;ca.ive o the ruel reit condition. For these reasons the core damage assessment using dose rate reasurements will be limited to within the uppe r range of fuel overheat.
The ccrvention of this document is to express th re 1 ease or fission products as the percent of the source inventory a'e
. . e iime o- "h ihe accident wnich is 6-2
observed to be +"re< e en'n the sa..pled or "easured ;..ecia. The radiologic C. arac erist cs of tthe.e 'iRC ilRC ca.egori
> es o, fuel Cazace are de f'ned~ in Tab I 25 V and 6-1.
6.1 ANALYSIS QF IN-CCNTAIic~~NT GOS" RATa Tne procedure =or core damage assess-,.ent is b ase d upon theh ccrparison between dose rates reasured following an accid'n t an d ana 1 ytically deter...ined values of the realist c or besti esiir.:a.e Cose rates that ~ould correspond to the . spe s l ic cateccries of cor~ C C arace. ihe radiation dose r tes insice the contair.-..ent buildino ffol 1owin g an acciden: are dependent upon many variables .which include reac:or power p *, b rr fuel burnup, contair.;.,ent building ceoretry, the iCentity and quantity of the fission products released -,rem ,rem the i.e core, core .he he use of chemical s p ra y s for ~h ihe removal o; air orne fission pr".duc:s, anC wehe
~
location within the building at which they are reasured. 'ch o.;he e variables are plant specific. There~ore, ir.;plerentation of this "rccecure requires the Cevelopzent of plant specific analyt cal ly Ceterzined Cose ra-~s which is beyond the scope of this dccu,".ent. Such plant specific Cose rat~s e may already be developed on a case basis for other purposes. However, the subject procedure i's a d docua' .en'ntended to provide an actual assess-..en of core '" Car.,age g ror tthee purose r" of i-,pie.".enting emergency operational decisions
~olio ollowing i an acciCent. Therefore, care should be taken to employ tt'e resu'-s of realistic or best est'zate dose rate analyses rath. er ~h an conservative inforzation wnich may have been developed for such purposes as licensing ac.iii.ies or the design bases for equip,"..ent environ."..ental qua >-1catlon.
qualification The use o- Ccse rate a n alvses based on ccnservati:e assuzpt ons could actua ac ual.l . 6-3
T,)l)le f) -l Radiolo ic Characteristics of llRC Cate pries of Frrel Dama e llRC Category of llechanism of Source of Percent of Source Inventory Distribution of Fission Frrel Dama e Release From Core Release Released to Containment. Prnrhrcts in Containment
- 1. llo Fuel Damage lla1ogen Spiking Gas Gap Less than 1 Airborne Tramp Uranium
- 2. Init-ial Cladding Gas Gap Less tlran 10 Airborne Failure
- 3. In terrrredi ate Clad Orrrst and Gas Gap 10 to 50 Airborne Cladding Failure Gas Gap Diffusion Release
- 4. Hajor Cladding Gas Gap Greater than 50 Airborne Failure
- 5. Initial Fuel Pellet Fuel Pellet Less than 10 Overheating Air bor'ne:
Grain Boundary IDOL lloble Gas Diffusion 25'f. Il<r 1 ngen
- 6. Intermediate Fuel Fuel Pellet 10 to 50 Pellet Overheating Plated Out 25K llalogen I'l, Solids
- 7. llajor Frrel Pellet Diffusional Release Fuel Pellet Gr.eater than 50 Pellet Overlreat.ing From UO> Grains
result >n a lower than ac.ual assessment of o c"re -.
- ~ C a.-ace. this Th is bec use the conservative assumptions dic.ate greater conseq onseque..ces .. e ,or he release o=
fission prcduc.s than may be actual for a iven category gisen c off core damace. d Tne measure...ent of the lower or realistic dose ra t e wou ld then h be corr lated to a lower than actual category of core damage. The purpose of this section is to describe the analytical assurptions reccmmended to be employed in that realis.ic analytical development and to provide an example of the resul s o= such an anal y sis. lant speci Plan ic implementation may require the use of assu;..pt;cns di,.erent from the assu-..ot'.ons those employed in th e exar used and the basis examole shown is as ceneric as or s'ble. ibl each , ill wi p1 e.. b e d erined. Tnerefore, Th Plant speci;ic implementation The all o; soecif,c "e p os may per.or...ed by analytical correction of the example dose rates rather .han an i nCepenCen t ca 1 cul a t i on. A number or the variables required in the analyses are well Cefined. These variables are the desicn reac:"r e c-r ower,
~ ecuilibrium core inventory of fission proCuc=s, containment bui ldirg gec"..etry, and the location of the wide rance area radiation monitors used to ;,.easure he dose rate. There are however several variables which require the use of analytical assumptions. These variables are the quantitative release of fission prcducts frcm the core uncm the Cefined corditions oi fuel cladCing failure and uel. overheat; thei r gecmetric distribution inside the containment building, and the consequences o h chemical he sprays for the reduc.ion o, airborne fission products. These analytical assumptions are discussed in Ce~.a i'1 b e 1 ow. Add i ti ona111ly, a method is reauired to correc. c. .he h me sured dose rates for power history prior to the acc',Cent because of th e result 1' upon the source fission product inventory.
6-5
The dose rate measurement is dependent directly upon the quantity of the clission products released from the core. The convention employed to ex"riess that quan.ity is the same as that doser'.bed in Section 3.3 for use in the Interim Procedure. The quantitative release of,ission produc:s is expressed as the percent of the source inventory at the time of the accident which is observed:o be present in the sampled redia and therefore available for immediate release to the environment. The reason for this convention is the limit -n the present capability to predict ission product transport out Q. the core -ollowing an accident. Cefining the quantitative release in this way does not imply a quan itative knowledge of the mechanism or transport phenomena o that release. ihe radiolooical charac-eris-'.cs of the 'iRC Categories of Fuel Carnage presented in Table 3-1 remain applicable to the Ccmprehensive Procedure. The radiological charac:eristics are ex"arceC u".cr. in Table 6-1 to incluCe the distribution of the fission products irsiCe the containment building. This distribution is an estimate based upcn .'iRC Regulatory Guide assumptions. The categories oi core damage idertit'.ed as cladding failure are charac-eri-ed in Table 6-1 by the release of fission procucts through the mechanisms of clad burst and gas gap di= usion. The charac. eristic fission products are the noble gases and halogens. These fission products are released in highly vola I i le chemical species. The dose rate analysis for the categories o, cladding failures will assume that the fission products remain airborne. quantitative release cr iter',a to distinguish the extent of fuel claddin damage are also provided in Table 6-1 exoressed as a pere nt of the source inventory. These values are the recc;..mended assumptions for use in the 6-6
analvtica 1 de e~'nation. e source inventory in this case is the ec ~ '. 'ory. ilibrium cas ggapp inventor . This inventory includes all noble gas and halogen fission products and is not limi ~> to the list provided in Table "-5 in wh>ch sn which .h.he c>>aracteristic
>> ~ " lsotct es are ild entitied for use in samole analysis. The implementation of these de fined criteria 'se in the form of a procedure require the calculation of thee d ose rate as a function of tire .or
".he two'ases of 50 percent and }0 percent er assumed quantitative release rcm the source into the containment buildi'ng a t,mosphere. The categories of core damage ii ie identified as uel overheat are characteri=ed in Tablee 6-} - by the release of iission prcduc s thrcuch grain boundary diffusion and by dif;usion frcm within the UO 2 grains. The characteristic fission prcCucts are cesium, rubidium, and n iellurium. The fission procuc=s released in this catecor include the less volatile chemical species driven i /en offT b'".e 0 i..e hich te.-..oeratures in addition d 'n to increased quantities of the hi ghly vol a V e species discussed with regard to cladding failure. Therefore the ~ose ra e analysis will assu-.e that the d'- is .ribution includes both ai rborne dispersion and sur=ace pplateou t 'de insid the containment building. The reccr. ."end assumptions are that: the air 'rborne dispersion will include 100 percent o, the noble gas and 50 5 erc nt of the halogen pere fission products which have been assumed to be released , and .he pla eout will include 25 percent of the halogen and } pere r ~ n. o tt e solid fission produc s which have been assured Q be released. ese values These va u represent the percentace of that which has be~nen I released and arenot not theh percen.age of the source inventory. The ~ercen-age of th e source inventory which is released for eac h category is a charac:eristic defined in Table 6-> 6-7
>he airborne dispersion is assumed to be hcmogeneous and the platecut is assumed to b e on the walls or the containmenr building. The quantitative rel ase criteria to distinguish the extent of,uel cladding are also provided in Table 6-1 -1 expressed as a percent or the source inventory. These values are the recon;ended assumptions for vse in the analytical determination. The source inventory in this case is the equilibrium core inventory. This inventory inclvdes all fission prcduc.s and is not limited to the list provided in Table 3-4 in which the charac. eristic isotopes are identified for use in sample analysis. The implementation of these defined cr> cri er)a ria in the ,crm o, a prccedure re uire the calculation of the dose ose rate ra as a ,unction cr Cqg time for the two cases of 50 percent and 10 percent assu-.ed qvanti a-ive release frcm the source into the ccntainment building. Operation of the Containment Spray System has the efiec- oi reducing "e air"orne concentration or the halogen fission prcducts. Evaluation oi ihi s errec. requires assumptions concerning he distribution oi chemical speciesc arorg .he h halogen fission products and the efiiciency of the spray to remove each rcm the atmosphere. HRC Re ulatory Guide 1.4 assumptions are re c...
~ ended f the h d;s.ribu~icn d'r o chemical species. Tne efficiency o. the c..emical spray is expressed as the deccntaminaticn factor (OF) after two hours o operation. It is assumed that those fission products remaining airborne ar.er two hours will be unariected by ccntinued operation of the spray. These assump ions are explici ly stated as follows: 91 percent oi the halogens are or the elemental species which has a 2 hcvr DF ~~ 20; 4 percent of the halcgens are of the organic species which has a 2 hour DF of 1 and 5 percont are or the par:iculate species wnich has a 2 hour OF or 1. The ccmbined 6-8
effec: is tha: af:er two hours of containment sprav c"er zion }3 5 percen" o= the oricinal quantity of halogen remains air'r ~ ~ ~ r y~r essed differentlv :he halogens are assu",.ed to be reduced by a - faac:or o f 7 ~f ~r wo hours of contair..",.ent spr y. The validi y of these assumotions were evaluua t e d b y" a series of parametric dose rate calculations per-or'ed with th e h a 1 ogen removal fac.or varied between } and 20. The result indicated thaat increasing the re,,oval factor to 20 (assuming all halo g ens to be elemental species) would lcwer *the dose Qse late ra ~ measure'd ive hours after the accident to one-thi'r d or i,e P va ue calculated 1 wi th a removal factor of 7. Th Therefore, the assumptions used for the ef;ic.'ency of the contair.,"..ent spray is a significann - aactor in the h accuracy of the procedure to estimate core Camace. Additionally, it ;..ust be assumed that the distribution of chemical s pecies is hcmogeneous thrcuchout he ccntairmeflt atmosphere and that the radiation ronitors are located a e in a hc-..cgeneous region which is subjected to the spray. Correc.ion of the measured dose rates for varia"ionn inn reac.or r a power historv pr or to the accident is required bec'use o; the result upon the sourc~ fission product inventory. The correction is used.to adj'u s-. th e measured Cosy rate to the correspond>ng value had the plant been operating at full power The anal y-' ticall C et rained dose rates which are used orr c"r. comparison a to assess core Carage are calculated assuming full pcwer equilibrium source inventories The techni q ues er'o p yedd in this correc.ion are the same as those discussed ir Section 3.3. The are t,.e ccndi 4.ions that: equi libr',um source inven,ory is reached after 30 Ca s of cons. ons.ant power operation; the equi ibrium value is 1 6-9
diroctl ". y por.,on re + ' '.e production rate ev"ressed io ag as reac or:; r power; andd that constant power operation means no change greater than =10 percent The result 4
>s that a simple ratio of the power may be employed to obtain the full . ~et equilibrium correction.
Constant reactor power operation is not antici pa t e d . Th erewore, engineering judgement is required to determine the reactor power
\ ~ which is er wn>c >s most mos representative of the fission product inven ory prior to the accident Explici determination of the representative value would require detailed comouter code analysis. However, engire ring jucgement e...ployed with the aid o, speci;ic guidelines is su>ficient to yield a result within the ac"u-acy of this procedure. These guidelines are as follows:
()) P ihe averaoe power during he 30 day time period is not necessarily the most representative value for correction to equilibrium conditions. (2) The last power levels at wnich the reactor operated should weioh more heavily in the judcement than the earlier levels. (3) Continued operation for an extended period should weigh rare heavily in the judgement than brief transient levels. ihe case consider d ,or the example is a 3800 1'.wt plant. The dose ratos e are provided for two cQn .ainmeni building geometries: the spherical containrent 6 has a volur e of 3.3 x 10 cubic feet and a surface area of 1.2 x 10 square eet, .I e cylindrical containment has a volume of 2.7 x 10 cubic feet and a 6-10
sur;ace area o; '..2 x 10" square feet. Tne assu;..pt cns include an ecuil ibri .m two year burnup at ul I reactor power th
~ .e re...oval of airbcrne iodine bv -".e ccntasr,.",.ent building spray sys em su .s in results an averace reduc: on by a fac="r OT ? two hours af:er spray initiation , an d 1 oc tion of the dose rate r.:easurement is tcp centerline of thee ccntainrrent building. Tne resul ts of:he exa;..ole analyses are provided in Figu ree 6-1 - an d 6-2. These ca 1 eclat>ons were per;or...ed using the point kernal tech nique to determine the ga.....a flux.
Several cc'puter ccdes are available to per form this analysis. 6.2 GE lERAL OISCUSSIONS OH THE LliiITATIGWS OF THE PRCC='""."".-" This procedure is limited in applicability t o th e .i~ " C catecor es or fuel cladding and fuel overheat. Th e proc dure is linited in accuracy tc tlat 0-the assumptions rade in the analytical dose rate de er~irat'on. Tr.ese consicerations have been discussed above. '" Ad d > 1 nal >g, ...ere are linitat;.".s in the radiation monit 'ring system capability to obtain representative remeasure...ents which should be considered in the deternination of core Camace using measured cose rates. This procedure relic'es upcn radiat'on dose rate measurements taken frcn one or more r,;onitors loc"ed insid the ccntainr.ent building to determine the total quanti .g of ission prcducts released frcn "he core I si and ~ .ererore available for release to the environr:.en The amcun Th of ission prcducts present at ti.ee location of the monitors ma
~ . y be changing rapidly due to transient plant ccnditicns.s. Ther~r Ther ~ore multiple r.easurements should be cbt>ired wi'%in a m~n>...un .ire per',cd ard ~hen possible under stabili:ed plant conditions
FIGURE 6-1 TYPICAL ANALYSIS FOR POST ACCIDEVT DOSF RAT INSIDE A SPHERICAL COiVTAIiVA1ENT 0 g( Og C( Vp, 7p P~ C( O~
/r~
/~p
/p (G~ "f(
Cg 7g
/p/ /g 10 100 Tl 1E POST ACCIDEiVT HOURS SPHERICAL CONTAINi%1EiVT
FIGURE 6-2 TYPICAL AiVALYSISFOR POST ACCIDEiVT DOSE RATE IiVSIDE A CYLliVDRICALCOiVTAliViblEiVT 1x10 n7)
/~
1x10 cg C lg 0/ Op Pp,
/p C~ P
( Og Cr
/~
Og 0@ C'~ ")p A'/ 1x10" ( 7Q P/
~
g 1 x10" 1 10 100 1000 Tl'IE POST ACCIDEiVT, HOURS C Y L I i"J D R I CA L C 0 N TA I iVi /I E iV T 6-13
amp es obtainea Curing rapidly changing plant n con d.- co itions srould not be weighed heavily into the assessment of core Ca@ace. g The re 1 i ab i I i t of t. the me'.'.
~ r d dose rates is influenced by red a number of factors which include: the abilit too cbtain representative measurements Cue to incomolete mixing ',g of the~ e m measured redia; equipment operation in a harsh envirorment; and operator familiarit wi. rarely 1 used procedures.
6-14
7.0 Pcs ~ C 7.1 Clari;ic".ion of T.'ll Ac.ion Plan Require:-..ents l<UREG-0737 dated llove...ber, 1980. Post Accident Sarpling Guide for Preparation of a Procedure to Es imate Core Damage, US ilRC. (Included here as Appendix A) 7.3 Release of Fission Products Fre~ Fuel in Postulated Degraded Core ~, 1 Accidents IDCCR Subtask 11.1 Draft Final Report Cated ~u!y, 19"-2. n Repor" o the Cc;aission and to the Public, liRC Special Irqu',re Grcup. ititchell Rcgovin Director 7.5 C"=.'i-158-P Evaluation of Instru;..entaticn for Detection o; InaCecua:e Core Cooling in C-E:VASSS. l1ay 1981. 7.6 ORIG:-.'i Isotope Generation and Depletion Code Oak Ridge 'laticnal Laboratory CCC-2'17. 7.7 EPRI i(P-1022, "Require."..ents or Analysis of Transien Fuel Rod Behavior During Design Basis Events," Prepared by Cc"bustion Engineering, Inc., for EPRI, Parch, 1979. 7-1
7.8 EPRI fop jcco ~~r pc+mrs A cec+ Post-D."S Operation ', r Light ~ater Reactors," ",olu;..e 1, Prepared by Cc'bustion Encine rirg, Ir c., for EPRI, Aucust, 1981. 7.9 C ",- -, ia-114-P 1 Amenc... n -P, Review oi Small Break Transients in Combustion Er.oineering t)uclear Steam Supply Systems," Prepared bv Cor;,bus.i on Enoineering, Inc., for the C-E Owners Group, July, 1979. 7.10 CE!l-115-P,G"Response to tiRC IE Bulletin 79-06C, Items 2 and 3, for Combustion Encireering nuclear Steam Supply Systems " Pre~ared by Combus ion Er.oireerirg, Inc., for the C-E Cwners Grcup, Aucust, 1979. II A 7.11 H. Ocken, ~n Improved Evaluation 5!odel for Zircaloy Oxidation," i.uclear technology, Vol. 47, February, 1980, pp. 343-357 7.12 H. H. Chuno and T. F. Kassr er, "Embrittlement Criteria for Zircalo Fuel Cladding Applicable to Accident Situations in Light-"ater Reac:crs: Su.....ary Repor t " hUREG/CR-1344, A<cL:-79-48, January, 1980. 7.i3 P,. K. Cole, Jr., "Generation of Hydrcgen During the First Three Hcurs of the Three labile Island Accident," r",UREG/CR-0913 t SAtlD79-1357, July, 1979. 7-2
HS~C-29, Ihe Retarding Erfec. of Hydrccen on Zircaloy Oxidatiorn 1 July, 1981. "Analysis of HJTC/R'lL,'iS Perfor-.ance Ouring Accicent Ccnditiors " Prepared by Combustion Engineering, Inc., for the C-E C;~ners Gr"up, Graf Issued April, 1983. II
-E i S0-212, Pertomance Evaluation of Core Exit Ther..ocouples as Inadequate Core Cooling Instru."..entation," Preared by Co.-..bus.ion Engineerirg, Inc., for the C-E Corners Grcup, Harch, 1983.
C-E 'iPS0-199, "Generic Ther'al-Hydr ui c Func.ional Desicn Gb'ectives for Inadecuate Core Cooling Instr.;.entaticn," Preparec bv Cc;..bust on =ngireering, Inc., for:he C-E C~rers Grcu", Ora;t Issued, ".arch, 1983. C-E iiP 0-171, "Cc".oarison of Bottca-i!ounted Core Exit arC In-Core Ther.-,.ccouples to Cetect Inadecuate Core Cool',ng," Prepared by Co;..busticn Ergineering, Irc.. .or the C-E Owners Group, inarch, 1983. "The Corrosion of Materials in Reactor Containrent Spray Soluticns" Griess and Bacarella !nuclear Technology 'lol. 10, April 1971. Bill -ilUR G-24532 "Hydrogen Release Rates frcm Corrosion of Zinc and Aluainun", Hay 1978.
Fle-cher, '4. D., Bell, 'A. J., Varchese, R. T. an4 Gallagher, J. L;, "Post LCCA Hydrogen Generation in P'AR Contair'ents," lluclear Tec.",rolocv, "olu-..e 10, April, 1971.
APP= iDI X A t/RC POST ACCIDK)IT SwllPLIaiG GUIDE FOR PREPARATIGlt OF A PROCE"URE TO EST"!ATE CORE D~lQGE
GU a 0~ t'OR P.".iPr.Rnl I C>i Ot' PPO<="4R TO EST I'lAT CORE Onl'lAGE The major issue reraining to complete our evaluation of flTCL's for ccrpliance with the post-accident sar pling criteria of tlUPEG-0737 is preparation of procedures for relating radionuclide concentrations to core dar.:age. To Cate, none of the applicant's has been successful in providing an acceptable procedure. As a consequence, each tlTOL has a license ccrdition which may restric. power operations. One of the contributirg factors in the applicant's slow responses to this item is their confusion on exactly wha. to prepare. The attachment is intended to provide informal guiCance to each tlTOL applicant so that their procedures, when prepared, will aCCress the core damage est.,"..ation in a manner acceptable to us. I ~e P antici" te that preparation of. a final prccecure for es imatirg core damage may take approx',mately 12 r;.onths. Therefore, we are willirg to acceot an interim procedure which ocuses on fewer radionuclides han are inCicated in the attachment. The interim procedure in conjunction with a firn date for he final procedure would be used to rerove the power restricting license condition. The primary purpose in preparing a procedure for relatino radioruclide concentrations to core damage is to be able to provide a realistic estimate of core damage. 'Ae are primarily interested in "eing able to differentiate between four major fuel conditions; no dar.age, cladding failures, fuel overheating and core melt. Estir'ates of core damage should be as realistic as A-2
possible. lf a core actually has one percent cladding failures, we do not want a prediction of fifty percert core melt or vice versa; extremes in either direction could significantly alter the actions taLen to recover from an accident. Therefore, the procedure for estirating core damage should include not only the measurement of specific radionuclides but a weighted assessment of their meaning based on all variable plant indicators. The following d'scussion is intended to provide general quidance pertaining to the factors which should be considered in preparing a procedure for estimating core damage but is not intended to provide an all inclusive plant specific list. The rationale 'or selecting specific radionuclides to perform "core damage estimates from fission product release" is included in the Rogovin Report (page 52:l through 527, attached). Basically, the Rogovin Report states that three majoi fac:ors must be considered when attemptina to estimate core damage based on radionuclide concentrations. For the measured radionuclides, what percent of tne total available activity is released (i.e. is only gap activity released, is sufficient activity released to predict fuel overheating or is the quantity of activity released, only available through core melt?
- 2. !'hat radionuclides are not present (i.e. radionuclides will, in all probability, not be released unless fuel overheating or melt occurs).
The absence of these species bounds the maximum extent of fuel damage. A-3
0 Hhat are the ra:ios of various radionuclides species (i.e. the gap activity ratio or the various 'radionuclides may differ from the ratio in the pellet). The measurement of a specific ration will then indicate whether activity released came from the gap of fuel overheatinglmelt. In addition to the radionuclide measurements, other plant indicators may be available which can aid in estimating core damage. These include incore temperature indicators, total quantity of hydrogen released from zirconium degradation and containment radiation monitors. I~hen providing an estimate of core damage the information available from all indications should be factored into the final estimate (i.e. if the incore temperature indicators show fuel overheat and the radionuclides concentrations indicate no damage, then a recheck of both indications should be performed). Consistent with the categorization of fuel damage in the Rogovin Report, the four major categories of fuel damage can be further broken down, similar to the following list, consistent with state-of-the-art technology. The suggested categories of fuel damage are intended solely to address fuel integrity for post-accident sampling and do not pertain to meeting normal off-site doses as a consequence of fuel. failures. Ho fuel damage
- 2. Cladding failures ( <10 ')
Intermediate cladding failures ( 10 -50 ).
- 4. ."1ajor cladding failures ( >50.")
- 5. Fuel pellet overheating ( <I0..)
- 6. Intermediate fuel pellet overheating (10;.'-50::).
A-4
7 ~ (~ a jot uel pel let over" ea ing (>50')
- 8. Fuel pel'(et r,.el-:,nc ((>0~)
- 9. Inter;..ed'.ate;uel pellet;..el tina (10.",-50,",).
l0. Hajor fuel pellet melting (> 0:.). Because core degradation will in all probability not take place uni=o~iy "e f>nal categories will not be clear cut, as are the ten listed above. Theretore, the preparation of a core damage estimate should be an iterati;e process where the first determiration is to find which ot the four major catecories is indicated (for illustrative purooses, only radicnuclide corcentraticrs ~ill be considered in the following exarole, bu. as indicated above, the plant specific procedure should include input frc~ other plant indicators). Then proceed to narrow down the estimate based on all available data and knowledge of how the plan . systems function. Kx .-..o 1 e in a given acciden operation, there is 70".. clad failure, sicnifican fuel overheatirg 'and one fuel bundle melted. Utilizing the iterative process. First Calculate the maximum fuel melted by arb r rily at:ributing all activit, to fuel melt (under these ccnditicns, five to ten rel ted 'bundles may be pred>ct d). (heretore, the worst possible condition is fuel pellet meltinc. A-5
Seccnd Calc late the maximum fuel overheated, by arbitrarily attribut'.ng all ac ivity to fuel 'pellet overheating (under these conditions, major fuel pellet overheating is predicted). Third Calculate the maximum cladding failures, by arbitrarily attribu.ina all ac.',vitv to claCding failures (under these conditions, greater than 100" fuel claddirg Carnage is predicted). At this point i is obvious tha- major cladding damage is present and that a large amount of fuel pellet overheating has occurred with the. potential for sc;..e miror fuel pelle: melt:,ng. Fourth Check for the presence of radionuclides which are indicators of fuel pellet melting and overheating. In his ins ance, obvious of overheating wi 11 exi s ~ along with trace indicators of potential pellet melt. Fii;h Based on the radionuclide indicators of fuel pellet overheating damaae (confir.-..ed by incore temperature) make an estimate of how much fuel A-6
overe ed. Thss resul ~s 11 sn all probabsl> ) dna,c te ma]or fuel pell overreatirc. Sixth Subtrac the ac.ivity estimated from fuel pellet overheating, plus the ac ivi ty at.ributabl e to 100" gap rel ease from the total activi ty found. Thi s will result in a negative number because the contributions from overes.i.-..ating cladding damage. ( 100" versus 70"..) and fuel overheatino (major versus in:er;..ediate) will exceed the activity contribution from one melted bundle. At this point, knowledceable judgement must be employed to establish the bes: estimate of core damage. Although all damage could be attributable to cl Cding damage and fuel pellet overheating, the trace of radionuclide irdicators o uel pellet melt indicate the possibility of some fuel meit nc. ased on ~rowledge or co. e temperature nrem 1CG; claddira 1 ~
~
damage would ', exist variations, wi *thout it significant fuel is highly unli<elv tha t mel ting. x Al so, so-.e of the activi y attributed to fuel pellet overheating must be assoc'.ated iwi h the amount of fuel pellet melting which is indicated. Therefore, the bes estirate of uel damage would be that intermediate fuel overheating had occ rred, with rajor cladding damage and the possibility of minor fuel pellet, melting in ore or two fuel bundles out of 150 fuel bundles." The above example is obviously ideal and makes the major assumptions that:
e 0
A. The radionuclide/s monitored are at ecual corcentr tions in all fuel rods. In actually, at no time will all r dionuclides be at equal concentrations in all fuel rods. Sec use the time to reach equilibrium or each radionuclide is different, due o their hichly variable production and di.ferent decay rates. Some isotopes will approach equilibrium quickly, while others never reach equilibrium. Therefore, it is necessary to factor in reac.or power history when determining which radionuclide is optimum for ronitoring in a given accident condition. probably the optimum radionuclides for estimating core damage will vary as a function of time after refueling and based on power history. S, Ecuilibrated samples are readilv available from all sample loca ions at
~
the ins ant of samolina. Considering the large volumes of liquid and vapor spaces that a leakage source migrates to and mixes with, for other than very large leaks, it will take many hours or even days to a Ih pl g t\ equilibrium c"nditions at all sample locations. C. Maximum core ceoracation occurred prior to initiation o sarolino. Unless total coolirc is los , core degradation can be anticipated to progress over a period of hours. Thus, there is not a given instant when sampling can be conducted wi th posi tive assurance that maximum degradaticn has occurred. Considering that i "eal conditions will no exist then, procedure for estimating core damage shculd be prepared >n a manner that the e;tects of variables such as t me in core life and type of accident are accounted for.
C t"e
+h proc Cur -or es.i-a -'rg c-r C--.co should inclde the determinati n of both shcr and long lived casecus ard non-volatile radionuclides along wi.h ratios or apprcpria e species. Each separate radionuclide analyzed, along with predicted ratios of selected radionucliCes would be used to estimate core dar..age. This process will result in four separate estImates of core damage, (short and long-lived, gaseous and non-volatile species) which can be weighed, based on power history, to determine the best estimate of core damage.
The post-acc Cent sar.".pling system locaticns for liquid and aaseous samples varies for each plant. To obtain the most accurate assessment of core damage 'I it is necessary to sar.pie and analyze radicnuclices frcm each of these locations (reactor ccolant, containment at.-..osphere, containment sur..ps and suppressicn pool), then relate the measured cor:centration to the total curies for each radicnuclide at each sample lcc".ion. These ;..easured radionucli"e concentratior,s need to be Cecay ccrrec:ed to the est r.".ated t'.--e or core damage. Their relaticnship o core damage can be obtained by cc,"..paring the total quantity and ratios of the radior uclides released .~i th the predetermined radionuclide concentrations and ratios which are available in the core based on power histcry. Assu"..ir.g one hour per sample location to recirculate, cbtain and analyze a sample frcm each loc tion it would ake hours to perform each of those ar.alyses. Based on the above rationale, the final procedure for estimating core damage using measured radionuclide ccrcen rations ~ill probably rely only on one or t"o sample locations dur'.ra the ini ial phases of an acciden . The optir,",um A-9
radioruclices for estimating core damage will also in the shor erm, "e or. term based on rec n. pcwer history. i(hen equilibrium conditions are established at all sample locations, radicnuclide analysis can be perfor;,.ed to obtain a be:: r esti;.,a:e of core damage. The spec',fic radionuclides to be analv-~d under equilibrium conditions ray be different than those initially analyzed Z because of initial abundances and di, erent decay rates.'he specific sample loca.ions to be used during the initial phases of an accident should be selected based on the type o accident in progress (i ~ er for ~ O', a small liquid line break in the primary contairment would release only small quan'ities o volatile species to the dry well. Therefore sampling the cry well first would not indicate the true magritude of core damage) For th. same small break accident, i pressure is reduced bv ventina safety valves to the suppression pool, then the suppression case of a small steam line reek, without venting safety valves to the, suppression pool, the dry well may be the best sample location. To.account for the variations in prime sample locations, based on type of accident, the procedure should incluce a list of primary sample locations. This list should include both a prime liquid and gaseous location and state the reasoning used to determine that these locations are best. Additionally, the procedure should address other plant locations which can be used to verify that the sample locations selected are best or the speci ied accident ccrdition. A-10
Finally, tt e pr"cedvre should irc"rpor.te <
4 APP=":NOIX 8.0 A'lALYTICAL DE."; I "ATI C.'<'
0 RII"ATICll OF THc, TRAHSIE )T POtliR CGRR""CTICls E(UPTICK( FOR SCURC""
~ lggt ~ I I OQY
0 'RI A I iCss 0 In" I wnl/S I ".f I PG )Eq (p 0 r ~ G" a '~ pp'~ pR SDURr-For those plar t power histories in uhich equilibr',um conditions do no" exis-an analvtical correction is provided in the procedure. The mathematical model used to calculate the quantity of fission products in the core fuel pellets as a;unction ov time involves a group of linear, first order differential equations. These equations are obtained by applying a mass balance for production and removal. The terms for fission prcduc produc.ion include direc. fission yield, parent fission produc. decay, and neutron activation. The ter...s for fission product loss include decay, neutron acti sation and escape to the coolant. Each equation in the group is expressed as follows. d;( o: " <' -1'1 'x-1 k~k' where the sariables are def'.red as follows. Fuel pellet fission product inventory, atoms Average fission rate, fission/t'.ut-sec Fission product yield,, ract:on Core power, i'.r't Decay constan:, sec a = llicrosccpic cross sec:ion, cm Esrape rate, coe f fi ci ent, sec Branchirg fraction Tire, sec 8-3
and wnere the subscripts are defined as ollcws. Isotope
'I Precursor to isotope a or decay Precursor to isotope 2, 'or neutron ac.ivation Within the accuracy of this procedure, the terns for ission product production by neutron ac ivation and for fission product loss by neutron activation and escape to the coolan are insignificant. The equation then beco-...es as follows:
dl), QC ( )(Y )(p) + ('.
~ ~ ] ~. ] ) 'i ~
1
~
A ~ I ~ Additionally, it can be assu;..ed tha. the terms for procuction are both linear with respect to plart power. Tnerefore, the equation becc;,.es as follcws. where (G)(P) is the procuction ter~ wnich is linear with respect to power. The solution of these equat'.ons are of the followina.iorn.
repres nis ih Quant) 'j Qf f) s,'cn prccuc "~r- s s prc4uceu Gul' ti;..e, t, while the reactor is at power p~ " ~ re a=.
~i s""e i ice at .er t,.e r eac.cr > s shutdown, the fission products which re.".,ain are as follcvs.
where h t. = the tire bet'.veen the end of period j and the tire of reactor shu td c:vn. ~he equaticn .vhich expresses the total fissicn products;vhnich rerain af er rruit;pie ti-..e pericds of different pcwer levels is as;ollc;vs. The power correcticn factor then beco;,.es as follcws. 0 tl (t} 0 Pcver P zjPi(le aj)e zj t( (t) 100'ower 100 (1-e ~ "j) Within ihe accuracJ of this prccedure and under the ccndition in which the total period of operation is greater than our radicactive half lives the pcwer correction is as fol ious. 8-5
l,3>4x J
'. 593 A
0 (I e j. j) e j Power Correc-ion Fac.or = jco 8-6
0 OERIVnTIOlf OF THE CCRC. HEATUP AlcD OXIDATION) RELATIC.'lSHI?S 8-7
Pl~~"1 0 1 ~ I
'nl ~C.~ ~ ~ ~ A+ +I C. In CG...."nTUP n>lQ C)(lDATxP.'I "" > --r ~
ic:> sic. I hl 5 v p oendi' x nrem s,nt~ .'..e ..a.'..~..a=ical derivat.on of the hea up and oxidation relations which ar~ a Js d .o c~.~'n .he inal results which apcear in the damage assess-...ent procecure. Knowledge o the derivation in this Appendix is not essen.ial 1 to pro"er implementation of the procedure, but it r.".ay help to unders.and d d the physical phenomena and thereby enable a better judcement on the amount and configurat on of core damage for an event which deviates subs an-tially rom the usual predictions for core uncovery events. The Cel lvatlon presented here is similar to that presented in Reference i-l3. .The derivat'n applies to the uncovered portion of a fuel channel durinQ boilo;; at corstant pressure and core decay power, A charnel ray be ccnsidered to be a fuel assembly or a single fuel rod and associated flow su"chanrel. The only dif ererce among channels (prior to c 1 a d ba llooning and rupture) is the pcwer level, ~h'ch is directly pro"or ional na o ..h e ra d 1 "ia'ucl ear peakinc ==ctor. It is as su."..ed that the normal ized radial peak pea(, is.ribution is the same for dec'y pcwer ard -.or normal operation and is equivalent to typical core wide distributicns with all CPA's withdrawn The stearI'enerated below the coolant level is assumed to remain in the sane same charnel above the ccolar,t level. The mass flow ra e is direc'1 s rec. y p proportional to the radial peak. I the total core steam flow is "fS -or 0 channels, the flow in a charnel with radial peak FR 'R is !('S F';i. Oecay power per uni lerg.h 8-8
- o. channel is the core average value times the radial peak q F An energy balance on a un'it length of channel, above the coolant level is 15 writ wl veen
')
en assuming a ho'ogenized rod of uel and clad at uniform temperature. Time rate of change Heat convected in Decay of sensible heat in minus heat + power fuel and steam convec.ed out .q,(.p ) ~+ PgA CQTg)
= .,' 9 - [., 9 9 . ( . )) +
Fpd (l) where: Total fuel rod specif';c neat (-"/f - 'F) R Rod temperature ('F) ti."..e after star. of core uncovery (hr) Pg Steam density ( ibm/f ) A Charnel flew area (ft ) Steam specific heat (9/ibm - 'F) Stean temperature ('F) Core total steam flow rate (ibm/hr) Fp Radial peaking factor Nu."..ber of rods or channels in core Height above bottcm of active length ( t) Core average decay linear heat rate (B/hr - ft) Several physically based assumptions are maCe. In the first tern, the heat, capacity of the steam is very ruch less than that oi,the fuel, so the second term in the parenthesis is neglected by cc'parison with the first. Physical 8-9
properties of fuel and stean are assuned consant The par ial derivative with height beccr..es ~('< T )/32 at any instant of t-'-..e, and because density changes are neglected his reduces to W aT /aZ. Frc"... tt e previous equations for coolant level, subst tuting quation (4-2) in-o ('-1) yields an expression fol thhe s.ean fl flow,,",, as a func.ion of tirre after uncovery starts: (2) Cc.-..bining (1) and (2) yields: aTq K3
+>>L ( 2) e- t/K1l o R d (3)
Ot pn ol hR lR where: PA C FR 3 tl l')R C Assure that the stean attains the rod temperature instantly, which is equivalent to an infinitely large equivalent surface heat transfer coefficient. The rod ac uallv rises above the steam ter:p rature, so Equation (3} resul ts would be low or te."-peratures below the rapid oxida ion temperature range. The heat input fron the exothernic oxidation reaction is added late~. 8-10
The par.ial differential equation is reduced to an ord nary d',fferential ~
'quation by substituting for time after uncovery, t, the time interval, t~,
from the ti-,.e the coolant level drops past a given eleva ion, l, to the current time when the level is a. From Equation (4-2), t is: Z/L-K t* = t + K 1 ln ( 1- i'2 ) Using: and taking acc"un oi the assu;..p:ion that TR
= T, Equation (") becomes:
FR qdg:.R cR vAL (2lL - ~ ) 2 3 This represents the time variation of the rod temperature at a given eleva ion as a furcticn of ti.",.e after the level drops past that elevation. It accounts .. for convection of some of the decay power by steam and .or the increase in sensible heat of the rod by the remainder. Assume that all heat of reaction by zirconium oxidation goes to raising the rod temperature. Oxidation heating becomes significant above about 1800'F, which is attained only for low coolant levels when the steam flow rate is
relatively low. 'ijith lo;I steam flow, cooling by convec ion and thermal radiation to steam is less e fec.ive, and a smaller fraction of the total decay power and reac-ion heat is transferred to the steam. There,ore, the addition of reac.',on heat as an adiabatic te.;perature rise is a fair assur ption when the core uncovers by boi loff. Reaction heating is expressed by q ,, the equivalent linear heat rate caused by oxidation. The rate of temperature rise is q react./HR CR, and is added to Equation (5). The reaction heat rate is expressed as the time rate of conversion of the mass ot :irccnium into oxide per unit length of clad times the heat of reaction: d ~ 2 2
~
react a: Zr 4 o i react Assume oxidation cn only the outer clad surface, 0 . This yields an 0 a'ppropriate esti..ate for total core oxidation prior to substantial core damage, but may uncerestimate the local oxida" ion on both sides o thinred clad at the location of clad rupture. Let x be the equivalent percentage of tt e original clad thickness, <r, which is oxidized. Then the preceding ecua ion becomes: re~c. Zr 0 dr o 1GO uH reac. dx a The equivalent oxide thickness, x, is given as a unction af time and temperature - by the equation in Sec ion 4.2. 1. The resulting equations for clad temperature and oxidation are: 8-12
p dl FR qd/;.1R C --c0 ('r/100) aH'eact dx at i'
'in K 3 1 t /K R R pnL (i/L - K )
2
' K 2 (100) 2 A ?r t - 8/R(T+ 060) ar 2 pZr These equations are normalized by substituting:
t' t"/K '1 T' (T - T(o))
')
K..'iZ C, (" Yielding: dT't' + K4 K dx V3 5 OC'
'e K~
1/(T' /K, + K<) (g) These ecvations gi re the clad temperature and local oxidation vs. tire for core vncovery by boiloff with so;..e inlet flow. 'Ahen the inlet flow is zero,
The ec a:ions are sol~ed numeric lly wi:h . ro inlet flow and for / values of the cons. n:s, K, evaluated for the 3'CO '.wt class In order o cc,are he re'tl ts d'or various reac.ors .he c rs ants are defined in Table 1 and rumerical valves are given in Table 2 ,or a selected decay heat fraction of 1" and for two pressures. The table shcws that .he variation of the values among the C-E desianed reactors is relatively small and is much less than the difference in values at 300 psia and 2=00 psia. In other words, the di ",erences in the fuel heatup behavior amoro the C-e, desicned plants for a boiloff event are much less e h~an th e d i.-erences
.-, to be ex"ec:ed just because of the effect of possible pressure di,fererces during two alternate accident scenarios. Therefore, the 3400 t'.wt class is evaluated over the rarce of 300 to 2500 psia ard ircm 1;. to ";., decay heat and <<e relative dis ributicn of oxidation is used in the procecure for darace assessment on all reactors. The range of error in rcd ced b y ih'is ecislon is c"nsidered su=ficientlv small to allow iden-'=ic .;.ic"icn -icr o ~ho ext nt o-. core damages within the ten cateccries defined in Table 4-1.
8-14
Table ! Definitions of Constants in Hea uo Ecvations Constant Definition K1 pALHf P H W ~ (Hc + Hc H. ) P Dii V
~ \3 pAcFp o R K,4 FR "~ K( Hs(,
1+ (L/L I gi(H - - n fg Hff in )
~
Y HD dH o p 7,r reac: dr~
~
1GO K1 FR qd K6 100 Zrr dr 2 pZr
'FR K) qd Ks T(o) + 460 8/R 8-15
Table 2 Constants ln Itea tbsp fgiia t inns-Comparison of Valnes for C-E oesi ned Reacto> s Vari ahl e Ft. Calhotin Calvert Cl i f fs SntiGS-2 S-An (ft) 10.67 11.4 12.5 12.5 (ft ) 60.1 107.3 109.7 118.5 qd 9 IN (8/hr-ft) 205 212 . 181 182 8 P Dii 0 IX (8/hr) 5.12 x 10 9.21 x 10 1.16 x 10 1.30 x 10 tl 23408 38192 51212 56876 D (ft) 0.442/12 =. 0.0368 0.440/12 = 0.0367 0.382/12 = 0.0318 0.0318 zr (ft) 0.032/12 = 0.00267 0.026/12 = 0.00217 0.025/12 = 0.00208 o.noro8 HR CR (8/hr-ft) 0.0474 0.0179 0.03'6 0 0356 3 Z (l m/f ) 407 407 407 407 6,II (8/ I bm) 2?70 2770 2770 2770 A> (ibm 2
/ft -hr )
4 5020 5n2n 5020 5020 8/R ('R) 30429 304? 9 30429 30429
0 e
Table 2 (Cont'd) Constants In tteatu E nations-Com arison of Values for C-E Desi neil Reactors Variable Ft. Calhoun Calvert Cliffs SOttGS-2 S-AO Pressure (psia) 300 2500 300 2500 300 2500 300 2500 p (ibm/ft3 ) 52.9 35.0 52.9 35.0 52.9 35.0 52.9 35.0 Hffg (8/1 bm) 808.9 361.6 361.6 808.9 361.6 808.9 361.6 C g (8/ibm-'F) 0.565 0.612 0.565 0.612 0 '65 0.612 0.565 0.612 KI 0.536 0.159 0.561 0.168 0.506 0.150 0.488 0. 144 K2 V A R I A 0 I. E It I T lt I tl L E T F L 0 lt R A T E K3/FR 1.62 1.16 1.75 1.26 1.80 1.29 1.75 1.25 K4 V A R I A 8 I. E 'lt I T Il I tt L E T F L 0 lt R A T E K5 FR 0 . 0294 0.0993 0.0?18 0.0737 0.0240 0.0811 0. 0247 0.0836 6 7 3.53xlO 7 7 7 K6 2 . 2 Ax 1 0 6.75xlO 3.67xlO 1.08xlO 1.05xl0 7 3.41x10 I.OIxlO7 K7/FR 0.0762 0.0226 0.0825 0.0244 0.0845 0.0250 0.0819 0.0242 K8 0.0288 0.0371 0.02AA 0.0371 0.0288 0.0371 0.0?88 0.0371
APP"";"OIX C.O PROC:-""UR"" GU IO"=L.".lE FOR ASS" SSl" l)7 OF CGR" OAl'AGi US IllG RAD ICLCG ICAL nl<ALYS IS OF Snl'APPLES
PACE 1.0 Purpose C-4 2.0 References C-4 3.0 Definitions 4.0 Precautions and Limita ions C-5 5.0 Initial Plant Condition/Symptoms C-6 6.0 Prerequisites C-6 7.0 Procedure C-7
- 7. 1 Record of Plant Condition C-7 7.2 Selection of Sample Location C-7 7.3 Sample Analysis C-7 7.4 Temperature and Pressure Correction C-7 7.5 Decay Correction C-8 7;6 Iden.-',=ication of the Fission Product Release Source 7.7 quantitative Release Assessment C-8 7.8 Plant Power Correction C-10 7.9 Assessment of Core Damage C-11 LIST OF E fCLOSVRES Erclosure 1 Radiological Characteristics of HRC Categories of Fuel Damage C-12 Samole Locations Appropriate for Core Damage Assessmer,t C-13 Recur d c f Sample Spec i; ic Activity C-14 Density Correction Factor for Reactor Coolant Temperature C-15 C-2
LiST Cf E 1CLQSU."ES (Ccn'd ) PriG Record of Sample Temperature and Pressure Correction Record of Sample Decay Correction Record of fission Product Release Source Identi fi ca ti on C-18 Record of the Release quantity C-19 Containment Huilding Sump Level C-c,0 Erclosure 10 Record of Steacy State Power Correction C-21 1 Record oi Transient Power Correction C-ZZ Enclc ure 12 Record of Percent Release C-3
1.0 PU??QSE This pr"ce ure s to "e followed urger --s" ~cc "'n- plant con-ions 0 deter.-;ice he y"e and ce reo 0 reac."r c"re damace
~
which may have occ;rred by using -ission proouc= isotopes measur d in samples obtained free the Pos . Accicent Sa~-..-, '..".g System are three factors considered in this pr cecure which are (P'SS'here related to the specific activi tv of the samoles. These are the identity o. those isotopes which are released frcm the core, the respective ratios of the specific activi tv oi those isotopes, and the percent of the source inventory at the ti-..e of the accident which is observed to be present in the samples. The resultina observation of core damace is described by one or more of the ten categories of core damage in Enclosure l. 2.0 REF" Re..'iCES 2.1 Develocment of the Comprehensive Prccecure Guideline for Core Darage Assessment, C-E Cwners Group Task '67, ".ay lg63. 2.2 Post Acc Cent Sampling System 0 er :ing Pr=cedures. (Plant specific Cocu.-..ent) . 3.0 0 "F I.'I I T I C.'lS 3.l Fuel Damace: For the purpose of this proc cure fuel damage is defined as a procressive failure of he ma;erial boundary to prevent the release of radioactive fissicn prcc.c:s .'nto the reactor coolan star.irg wi;h a penetraticn in the "ircalc; claddirg. The type o-. fuel damage as Ceter...inca by this proce"ure is reoorted in ter.-.s of four major cate"ories wnich are: no c-.-,.ace, cladding failure, fuel overheat, and fuel melt. Each o; these catecories are characteri-ed by the ider.ti y of the fissicn procuc=s rele=-s d, the mechanism by which they are released, ard the source inventory within the fuel rod frc~ which they are released. The decree of fuel damage is measured by the percent of the fission prccuce source inventory which has been released in:o fluid media and thererore available =or immediat release to the environment. The degree of fuel darace as determined by this prccedure is reported in terms of three levels which are: 'iritial, intermediate, ard major. This results in a total often possible cate"orgies as characteri.ed in Enclosure 1. 3.2 Source Inventory: The source inventcry is the total quantity of fission prcducts expressed n cur es of each isotope present 'in either source; the fuel pellets or the fuel rod gas oap. C-4
4.Q P" "A JT:C!l r'li0 LllltTATjQ,",S 4.] The assess;..ent of core damace obt ined by using this procedure is only an esti;..ate. The techniques er.:olored in this procedure ar only accurate to locate the core condition within one or more of the 10 catecories of core damace described in Enclosure 1. The proce- . dure is based on radioloaical data. Other plant indications may be available which can improve upon es imation of core damage. These incluCe incore temperature indicators, the total quantity of hydrogen released from zirconium degradation and containment radiation monitors. 1ihenever possible these additional indicators should be factored into the assessment. 4.2 This procedure relies upon samples taken from multiple locations inside the containment building to determine the total quantity of fission prcCuc.s available for release to the environment. The amount oi fission proCucts present at each sample location mav be changing rapidly due to transient plant conditions, Therefore, it is required that the samples should be obtained within a minimum t't"..e ericd and if possible under stabilized plant conCitions. Samples obtained Curira rapidly changing plant conditions should not be weighed heavily into the assessment of core damage. 4h ~ V A nu.".ber of, actors influence the reliability of the chemist y samoles upon which this procedure is based. Reliabili'ty is influenced by the ability to obtain representative samples due to incc delete mixing of the fluids, eouipment limitations, and lack Gi
~
operator familiarity with rarely used analytical procecures. Tne acc;racy achieved in the radiological analyses are also influenced by a nu;..ver o; fac:ors. The equipment employed in the analysis .-..ay be sub'ected to high levels of radiation exposure over extenCed per'.cds o time. Chemists are required to exercise consiCeraole caution to minimize the spread of radioactive materials. Samples have the poten ial o- being contaminated by numerous sources and they may not result from a uniform distribution o the sample fluid. Coolino, or reac ions may take place in the long sample lines. Therefore, the results obtained may not be representative of plant conditions. To minir,ize these ef,ects multiple samples should be obtained over an exterded time period from each location. C-5
4 5.0 Ili'.T! 'L O'T CG.'lOITIOl)S Jdi,'O Sy;:,PTq.)S This procedure is to be employed for analysis of radiochemistry sar;.pie data uhen it is determined tha. a plan. accident with the potential for core damace has occurred. The following is a list cf plant symptoms to assist in this determination. This list is no. a ccmplete representation of all events which r,",ay cause core damace. One or more of these syr;,ptoms may exist at or before the ti'e the sample is cbtained. Under these conditions, sampling should be per;ormed using the Post Accident Sampling System. 5.1 High alarm on the containr.ent radiation rroni tor. 5.2 Hi gh a 1 a rm on the CVCS 1 e tdcwn radi a ti on moni to r. 5.3 High alarm on the main condenser air ejector exhaust radiation monitor. 5.4 Pressurizer level low. 5.5 Safety Injection System may have autcmatically actuated. 5.6 Possible high quench tank level, temperature, or pressure. 5.8 Possible noise indicative of a high enercy line break. 5.9 Decrease in volume ccntrol tank level. 5.10 Standby charging pumos energized. 5.11 Urbalanced charging and letdcun flew. 5.12 Reactor Coolant System subcooling low or zero. 6.0 PRc.RECU IS I 7""S An operational Post Accid nt Sampling System with the capability o obtain and analyze the identity and concentration o. fission product isotopes in fluid samcles wnich have the potential to be hichly radioactive. The system should meet he requirements of tlUPEG-G?3? Item II.B.3, Reference 3. C-6
0 7.0 PRCC'U."E 7.1 Record he follcwing plant indications. 'e use of transient cord ticns the values should be recorded as close as possible to the ti-...e at which the radiological samples are cotained frcm the Pos-Acc;dent ampl ino Sys erl. 7.1.1 Reac cr Coolant System: Pressure PSIG Temperature OF Reactor Vessel Level Pressurizer L vel 7.1.2 Containment Building: Atmosphere Pressure PS;G A:,"..osphere Tempera ture Of Su..p Level Prier 30 days Pcwer History Po'~el, Pe~~ent I'uraticn. Oavs 7.1.4 Time or Reactor Shutdown Date I 1",e 7.2 Select the most apprcoriate samole locaticns recuired ;or core da.ace assess-,en usirg the guidelines provided in Enclosure 2. 7.3 Obtain and analyze the selected samples for fission prccuct speci;ic activity using the procedures for Pos- Acciden" Sample System operation described in Reference 2. Record the requireo sample Cata for each selected samole. Enclosure 3 is provided as a worksheet. All of the isotc"es listed in the enclosure may not be cbserved in the sample. 7 ~ 't Correct the measured sample specific activity to stancard temper-ature and pressure. IIgO>>>> ~ I ~ ~ This ster is recuired only if it is not included in the procedures for Post Accident Sample System Operation, Reference 2. 7.4.1 Reac ol coolant liquid s roles are correct d =or system temperature and pressure using the factor for water density provided in Enclosure 4. The ccrrection factor obtained frcm the enclosure is divided into the measured value to obtain the densi ty corrected value. 7.4,2 Containment building su'p sar>>ples Co not require correction for temperature and pressure within the accuracy of this procedure. 7.4.3 Containment building at-.osphere gas samples are corrected using the following equation. C-7
p T + 460 sPeci-ic, c.ivi "j(ski} = spoci;ic 'c:ivi:i' ( ) x ( 0 0 2 where: Tl, Pl = i"ieasured Sample te.-..ger ature and Pressure recorded in step 7.3. T2, P2
= Standard ter perature, 32'F and Standard Pressure 14.7 psia.
7,4,4 Enclosure 5 is provided as a worksheet. 7.5 Correct the sar.pie specific activity at STP for decay back to the tir.".e of reactor shutdcwn which is recorded in step 7.1.4 using the follcwing equation. Enclcsure 6 is provided as a worksheet. A A o -At where: A = the specific ac.ivity ~( the sample corrected back to the ti."..e of reactor shutdown, " '/cc. the rreasured speci.ic ac:ivity, " A = /cc.
= the radioactive decay ccnstant, 1/sec. = the tire period frc~ reac.or shutdcwn to sa.-..ole analysis, sec.
7.6 Identi icaticn oi the Fission Prcduct Release Source. 7.6. '. Calculate the followirg ratios for each noble oas and iodine isotc"e only using the spec',fic activities obtaired in step 7.5. Erclosure 7 is provided as a worksheet. I Noble Gas Ratio = tioble C 5 Isotope Sr eci, ic Activitv Xe 133 Specific Activity Iodire Isotcce Scecific Activity Iodine Ratio = 1-131 Speci-ic Activity 7.6.2 Deter>ine the source oi release by cc;., aring the results obt>>ned to the predicted ra ios provided in Enclosure 7. An accurate compar-ison is not anticipated. Within the accuracy of this procedure it appropriate to select as the source that ratio which is closest to the value obtained in step 7.6.1. 7.7 Calculate the total quant'-y of ission products available for release to the environ"-en 'nclosure 8 is provided as a worksheet. 7.7.1.1 If the water level in the reactor vessel recorded in step 7.1.1 indicates that the vessel is full, the quantity of fission products found in the reactor coolant is calculated by the following equation. C-8
Tot 1 r~ctivi:y (C', ) = n ("'cc) x RC Volu;..e where: A = the speci;ic activity of he reactor coolant sample correc.".d
'Pl to tice of re v ctor s shutCcwn ob \ ained in s \ep s/
RCS Vo 1u;,.e~ .. = th.e full reactor coolant system water volume correc. d to stardard temperature and pressure using Enclosure 4. I the water levels in the reactor vessel and pressurize r recor Cod e
. p . 1. 1 indicates that a steam void is present in the reactor vessel, then the quantity of fission products found in the reac."r coolant is again calculated by step 7.?. l. l. However, it must be recognized that the value obtained will overestimatee th e ac t ua l q unti" '.y rre eased. Therefore, this sample should be repeated at 1
such time when the plant operators have removed the void frcm the reactor vessel. If the water level in he reactor vessel recorded in step 7. 1. 1 :s belcw the low end caoability of the indicator, it is nc" possible to determine the quan:ity of fissicn produc:s frcm this sarpi ample b ecause the volu;,.e o r water w in the reactor coolant system is unkrown. Urcer this condition, assess;;.ent of core damace is obtained using the containment sump sample. The cuantity o fission products found in the containmen buiidin" sur.:p is determined as follcws. The wavel volue in the containment buildirg sump is determined from the su"...p level recorded in step 7. 1.2 and the curve providea in . The quantity of fission products in the sump is calculated by the follcwing equation. Total Actisity, Ci = A (" /cc) x Surrp Volume where: A = the specific activity of the contairment sump sample corrected to the tire of reactor shutdown obtained in step 7.5., " /cc. The quantity of ission products fourd in the contain;,.ent building at."..osphere is deter.".,ired as follows. The volume of cas in the containment buildina, at the t
., is correc:ed to standard temperature ard pressure usin u ing the ollcwing equation.
Ggg,(piU p (~-,p) Ggg yp)pp ~21) P + P T + 460 ( 2 T1+ 460 1 P2 C-9
where: T1, P1
= Con.ainmen: ntI..ospnere te.:perature and pressure recorded in s.eo 7...
1 T2, P2
= Stardard temper'ture, 32'. and Stanc rd Pressure 14.7 psia ~
7.7.4 The total quantity of fission produc"s available or release t" :he environment is eaual to the sum of the values obtained frcm each sample location. 7.8 Plant Power Correction I The quantitative release of the fission products is expressed as the percent of the source inventory at the time of the accident. The equilibrium source inventories are to be corrected =or plant power history. 7+8+ 1 To correc- the source inven.ory or the case in which plant power level has remained constant for a period greater than four radio-active half lives the following procedure is employed. inclosure 10 is provided as a worksheet. 7.8.1.1 The fission products are divided into two grouos based upon the radioactive half lives. Group 1 isotope are to be employed in the case where core power had not changed greater than =10 percent within the last 30 days prior to the reactor shutdown. Grcup 2 isotopes are to be employed in the case ~here core power had no-chanced greater than =10 percent wi hin the last 4 days prior o the reac:or shu.ccwn. 7.8.1.2 The following equation ray be applied to the fission produc. Grcup which meets the criteria stated in 7.8. 1.1 only. Power Correction Factor = Power Level for Prier 30 Davs Group 1 i00 Pcwer Correcticn Factor-Pcwer Level for Prior 4 Qavs Group 2 100 7.8.2 To correct the source inventory for the case in which plant power level has not remained constant prior to reactor shu.dcwn, the followino equaticn is emoloyed. The entire 30 days pcwer history should be employed. inclosure 11 is provided as a worksheet. Power Correction Factor r.. p. (led tg)e lh to t J 100 C-10
'<<here P. = steady reactor pc<<er in period t.j = duraticn of period j t.0 = time ircm end of period j to reactor shutdcwn 7.9 Comparison oi Measured Oata with Source inventory The total quantity of fission products available for release to the envirorr..ent cbtained in step ?.?.4 is compared to the source inven-tory correc ed ior plant pcwer history obtained in step 7.8.2. This cc'parison is r,".ade by dividing the two values for each isotcpe ar.d calculatirg the percent of the corrected source inventory that is ncw in the sampled fluid ard thereiore available for release to the envirorment. Enclosure 12 is provided as a worksheet.
7.9 CCRE Di'iMAGE ASSES M":HT The ccr'elusion on core dar..aae is made usir.g the three par reters develcped above. These are:
- l. Identi-ication o, the fission product isotopes which most characterize a given sample, step ?.3.
- 2. Identiiicaticn oi the source of the release, step 7.6.
- 3. quantity o- the fission produce avail ble ior release to the er.vircr.;..ent expressed as a percent oi source inventory, ste" 7.9.
Kro<<ledgeable judcement is used to compare the above three para-rreters to the definitions oi the 10 HRC categories of fuel damace found in Enclosure l. Core damaae is not anticipa ed to take place uniior-.ly. Therefore when evaluating the three parareters listed above the prccedure is anticipated to yield a cor;bination of n
.o oi ihe .0 ca.egories deiined in Enclosure 1. These categories will exist simultaneously.
The .ype oi core dar ace is described in terms .of the '10 ilRC ca=egories defined in Enclosure 1. The decree oi core damaae is escribed as the percent oi the fission products in the source inventory at :he time of the accident which is no<< in the sampled fluid and hereiore available for release to the environr.ent.
EtlCLOSllRE 1, Radiolo ical Characteristics of tlAC Cate pries of Fuel Dama e Release of Characteristic tlRC Category of t<echanism of Source of Characteristic Isotope Expressed as a fuel Oama e Release Percent of Source Inventor
- 1. tlo Fuel Damage llalogen Spiking Gas Gap I 131, Cs 137 I ess than 1 Tramp llranium Rb 98
- 2. Initial Cladding Gas Gap Less than 10 Failure
- 3. Intermediate Clad Durst and Gas Gap Xe 131m, Xe 133 10 to 50 Cladding Failure Gas Gap Diffusion I 131, I 133 Release
- 4. tlajor Cladding Gas Gap Greater than 50 Fa i lu> e
- 5. Initial Fuel Pellet Fuel Pe 1 Ie t Cs 134, Rb 88, Less than 10 Overheating Te 129, Te 132 Grain Boundary
- 6. Intermediate Diffusion Fuel Pellet 10 to 50 Fuel Pellet Overheating
- 7. tlajor Fuel Pellet Diffusional Release Fuel Pellet Greater than 50 Overheating From U02 Grains
- 8. Fuel Pellet t1elt I uel Pellet Less than 10
- 9. Intermediate Fuel Escape from tlol ten Fuel Pellet Da 140, La 140 10 to 50 Pellet tlelt Fuel La 142, Pr 144
- 10. )lajor Fuel Pellet Fuel Pellet Greater than 50 tie 1 t
where.
. = steady reac:"r power in period j
- t. = dura ion of period j t.0 = time from end of period j to reactor shutdown
? 0 Co;..par tson of measured Data wi th Source Inventory The total quantity of fission produc.s available for release to he envirorr.:ent obtained in step 7.?.4 is ccmoared to the source inven-tory corrected for plant power history obtained in step 7.8.2. This ccmparison is made by dividing the two values for each iso o".e and calculating the percent of the corrected source inventory that is now in the sampled fluid ard therefore available -.or release to the environment. Enclosure 12 is provided as a worksheet. COiE D""GE ASSESS.".E'I The corclusion on core darace is rade using the three parame-ers developed above. These are:
- l. ICentification of the fission product iso c"es which most char cteri=e a given sample, step ?.3.
- 2. IC ntif'.cation of the source o the release, step 7.5.
3. enviror.-...en- expressed as a percent 7 0 r < of source 's 4uanti y o; the fission produce available ;or release to the inventory step
< p KnowleCceable juCcement is used to comoare the above three para-met rs :o the definitions of the 10 NRC categories of uel damage found in Enclosure 1. Core damage is not anticipated "o take place unifor...ly. Therefore when evaluo-ting the three parameters listed above the procedure is anticipated to yield a combination of one or more of the 10 categories Cefined in Enclosure 1. These categories will exist simultaneously.
The type of core damage is described in terms of the 10 tIRC cate"ories Cefined in Enclosure 1. The degree of core damace is described as the percent of the fission products in tie source inventory at the time of the acc Cent which is now in the sampled fluid and therefore available for release to the environment.
t EllCl.OSllRE I Radiolo ical Characteristics of t'RC Cate nries of Fuel Oamane Release of Characteristic tlRC Category of Hechanism of Source of Characteristic Isotope Expressed as a Fuel Oama< e Release Release Isotnpe Percent of Source Inventory
- 1. I/o Fuel Oamage )lalogen Spiking Gas Gap I 131, Cs 137 Less than I Tramp Uranium Rb l10
- 2. Initial Cladding Gas Gap Less than i0 Fa i lure
- 3. Intermediate Clad Burst and Gas Gap Ye 131m, Ne 133 10 to 50 Cl adding Fa i lure Gas Gap Oiffusion I 131, 1. 133 Release
- 4. Hajor Cladding Gas Gap Greater than 50 Fa i lure
- 5. Initial Fuel Pellet Fuel Pellet Cs 134, Rb 88, Less than 10 Overhea ting Te 129, Te 132 Grain Boundary
- 6. Intermediate Oiffusion Fuel Pellet 10 to 50 ~
Fuel Pellet Overheating
- 7. Hajor Fuel Pellet f Oi fus i ona 1 Release Fuel Pellet Greater than 50 Overheating From U02 Grains
- 8. Fuel Pellet Hel t Fuel Pellet Less than 10
- 9. Intermediate Fuel Escape from tlol ten Fuel Pellet Ba 140, La 140 10 to 50 Pe 1 1 e t He 1 t Fuel La 142, Pr 144
- 10. Hajor Fuel Pellet Fuel Pellet Greater than 50 Helt
ErrcLostrAE 2 SAIIPtf I.OCATIOIIS APPRnrrrlr TE FOR CORE nAI:AGE ASSESSIIEIIT SI ttr TOntlfr AccloEIIT scEIIARlo ACS COII TA 1 rrf IEIIT COIITA1 IIIIEIIT cont.lr:G I:I:Ot;II IIOT I,FG SIIIII> ATIIOSPIIERE SYSTEM'I I Small Break LOCA, Reactor Pokier >I'X Yes Yes Yes Small ttreak LOCA, Re3ctnl PONer <ll Yes Yes Small Steam Line Break Yes Large Break LOCA, Reactor Porkier > 1K Yes Yes Yes. Yes Large Break LOCA, Reactor Potpie~ <ll Yes Yes Yes Large Steam Line Break Yes Yes Steam Generator Tube Aup r.ure Yes Yes
EilCLOSUR"" 3 R"-C"RD 0;" S'l'.PLE SP""CIFIC 'CT"'I Y Sar.;p] e i'lu."..ber: Loca.ion: Tir.:e of Analysis: Tet"..pera:ure, 'F: Pressure, PSIG: Sa;..pie Ac-.iri".y, " /cc: Kr S7 Xe 13 Xe 133 I 131 I 132 I 1~3 I 135 Cs 134 Rb SS Te 12o Te 132 Sr S9 Ba 140 La 140 La 142 Pr 144
Ell CLOS i.:R-"4 RAT10 OF HgO DEi JS1TY TO HgO DEiVSITY AT 8TP vs TFi'1PERATURE 700 600 500 400 300 200 100 0 0 0.25 0.50 0.75 1.0 PACT~PSTP
E'lCLOSURE 5 R CCRD 0" SAMPLE T" MPE..ATUR" CORR" CTICll Sar;,pie humber: Location: Tire of Analysis: Ter,perature, 'F: Pressure, PSIG: Measured Speci. ic Activi ty Correc:ion Speci.ic Act :i ty Iso:one (Enclosure 3). " /cc Fac.'or 8 STP,
" /cc Kr 87 Xe 131m Xe 133 I 131 I 132 I 133 I 135 Cs 134 Te 129 Te 132 Sr 89 8a 140 La 140 La 142 Pr 144
El(CLOSURE S PECGRD OF RELEASE gJ'llTlTY Reactor Ccolan Ccr tainrr:ent Svr.",p Ccntairrent Total Sa-.pl e eev;,.ber, Sar ole Hurber, At...osphere Sarpl e guanti t j jso:c".e Ci e'(u;..ber, Ci Ci Kr 87 Xe 13}rr Xe 133 I 131 I 132 I 133 I }3S Cs 13>> Rb aS Te '129 Te 132 Sr aa Ba '>>0 La }>>0 La }>>',2 Pr }4>> C-19
E;lCLOSUR= 9 CONTAIN 3ENT BUlLDlNG "/PTER LEVEL vs VOLUillE 25 22 This is an exa."..ole. 0 specific cur;e is required for each nlan:. 20 19 2P PPP 30,000 40,000 50,000 e0,000 V0,000 S0,000 90,000 VOLU lE, FT3
EllCLOSURE: Q RECORD OF STEADY STATE PC',lER CORRECTIOis Sa."..pl e 1'sumber: Loc tion: Steady State 30 Days Power Level: Steady State 4 Day Power Level: Fuel Power Equilibrium Corrected H>story Correction x Source Source 1so;ooo Grcuoirn Factor Inventorv>> Inventor r Gas Gao ln".n or'i Kr 87 6.3(o) Xe 13:m ~.3(-") Xe 133 I.3(7) I 131 6.7(6) I 132 7.O(3) I 133 6.7(6) I 135 I.I(6) Fuel Pellet Inventorv Kr 87 2 3.1(7) Xe 13'.n I ~.6(5) XB 133 I 1.5(8) I I '. I 7:3=(7) 2 1.0(8) I 133 2 I:5(a) I 135 2 1.3(8) Cs 13'b I 1.0(7) 2 4.5(7) Te 129 2 2.4(7) Te 132 I I.O(8) Sr 89 I 6.1(7) Ba I-'Q I I 3(8) La I+Q I 1.3(8) La 1~2 2 1.6(B) Pr Iwh 2 9.1(7) Plant specific values should be substituted here frog Tables 3-4 and 3-5 of Section 3.3. Example here is ror 256O,'iwt class
EHCLOSURE 11 Rr.rC~D Or IRAiISI il I PO"'ER, CO~~
'ampl e ."u.-..ber:
Location: Prior 30 Day Power History: Power ". Dura t'cn, Days Power Correction X Equilibrium Source Corrected Source Isotcoe Factor Inventorv>> Inventorv Gas Gap Inventcrv Kr 87 6.3(0) Xe 131 < 3(') Xe 133 1.3(7) I 131 6.7(6) I 132 7.0(3) I 133 6.7(6) I 135 1.'.(o) Pe 1 let Tni. en Kr 87 Xe 13';. ~.6(5) 1~'.1(7) Xe 133 I 131 I 132 I 133 I 135 Cs 134 Rb cS Te 129 1 5(8) 7.3(7) 1.0(8) 1.5(8)'.3(8) 1.9(7) 4.5(7) 2.<i7) Te '32 1.0(8) Sr 89 6.1(7) Ba la0 1.3(8) La }40 1.3(-) La 142 6(8) Pr 9.1(7) >> See footnote on Enclosure )0. C-22
ElICLO~(.'R-" 12 Re.CORD OF PORC=";l Total guanti .y Power Correc.ed Available For Release -: Source Inventory, x 100 = lso-.ooe (Erclosur 8'I, Ci Ci (E.".closure !0 or 111 Per" ot Gas Gao Inven.orv Kr 87 Xe 131 Xe !33 I !&M I 135 f s'D! Pss 1 l e s' needing orv Kr 87 Xe 13'.' Xe 133 I 131 I 1"2 I 133 I 13: Cs 134 Rb 88 Te 129 Te !32 Sr 8<3 Ba 1-'.0 La 1-'0 La !42 Pr l~>> C-23
EX&llPLE AP~-"'iDIX C 1 X~<!PLE USE OF THE PROCEDURE
r ihe,ollcwirc is an era;..pie of the use of this 'procecuro for assess-.en" o= core carace. The speci ic case sited is for an 'iSSS o- .""0 "...~t c ass. o= th e ".-=0 ihe da;a recorCed cn plan: condition at the ti-.e o= "h i e samp e anaiysss is
, 1 1 as iolicws:
Reac:or Coolan: System: Pressur 1600 PS1G Temperature 300 'F Reactor Vessel Level 100 ".. Pressuri=er Level 80 .". Con.a ir..-..ent Bui i ding: Pressvre 0.5 PS1G
'F Temperature Su;..p Level en'ura 220 21 <<e~>>
Pr'.or 30 Day Pc~er Historv Pc'~e r. Pe rc cn~ ~ Oavs 22 QO 1? 100 'Tire or reac:or shutCo:~n 0100 on 12/25/82 C-25
EX>EHPL E ENCLOSURE 3 RECCRD 0 Sn>'IPLE SPEC IP ~ C ACTI"I Ty Sar:.pl e flu.-..ber: Location:, RCS Hot Leg Tire of Analysis: 12/25/82 0400 Temperature, 'F: 300 Pressure, PSIG: 1600 uc1 Sar.pie ~ctivity, /cc: Kr 87 Xe 13'.a 1(-:2) Xe 133 I 131 1(-:4) I 132. I 133 1(+2) I 136 Cs 134 Rb 88 Te 129 1,(-:3) Te 132 Sr 89 8a 140
'a 140 La 142 1(+1)
Pr 144 C-26
EHCLOZUR"- 3 R Ci.i0 OF Sn .Pl ~ Sr-, ~ tC nCi ~ ~ IiY Samp1 e t(umber: Location: Containment Sump Time of Analysis: 05CO 12/25/82 Temperature, 'F: 150 Pressure, PSIG: 0.5 Sa.".pie Acti vi ty " /cc: Kr 87 Xe 13:m 1(-5) Xe 133 I 13'. I 132 I 133 1(0) I 135 Cs 134 Rb 88 Te 129 1(+1) Te 132 Sr 89 Ba 140 La 1-'.0 La 142 l(-1) Pr la>>'-27
EXAl'.P' EllCLOSURE 3 RECORD 0 S~iiPL SP~ ir IC ~Ci Sa;..pl e t'u.-..ber: Loca ion: Cont a inn:en t At-,.os pne re Tir:.e of Analysis: 0600 12/25/82 Ter..perature, 'F: 220 Pressure, PSIG: 0.5 Sa-.pie Activity, " /cc: Kr 87 Xe 13la Xe 133 1(-l) I 131 1(-1) I 132 1(-3) I 135 Cs 134 Rb 88 Te 129 Te 132 Sr 89 Ba 140 La 1<0 La 142 Pr 144 C-28
E!lCLOSURE 5 RECORO OF Sn~e"PLE TE~vlP RrtTUR CGRR" CTIGW Sar,pl e 'lu,"..her: Location: RCS Hot Leg Ti;..e o-. Analysis: 12/25/82 0400 Te."perature, 'F: 300 Pressure, PSIG: 1600 treasured Speci=ic Activity Correction Isoto"e (Enclosure 3), " /cc Fac+hr Kr 87 Xe 131~ Xe 133 1(-:2) 1/O. 9 I 131 I/O. 9 132 I 133 1(+2) 1/0.9 I 13" Cs 134 Rb 88 Te 129 1/0.9 Te 132 Sr 89 Ba 140 La 1~0 La 142 1/0.9 Pr 144 C-29
EXAllPLE Ei"CLOSURE 5 RECORD OF SA<lPLE TEhPERATURE CC RE 7'0'l itu...gael: Sal".i/i e Loca ion: Contain'ent Su;..o Tire of Analysis: 0500 12/25/S2 Ter.perature, 'F: 150 Pressure, PSIG: 0.5 treasured Specific Act'.vity Correction Specific Activity I50000Q (Enclosure 3l. 'icc Fac.or 0 STP.
" /cc Kr 87 Xe 13ln Xe 133 1(-5) 1(-5)
I 131 1(-2) tl/A I 132 I 133 1(0) 1(0) I 135 Cs 134 Te 129 1(+1) 1(+1) Te 132 Sr 89 Ba 140 La 1~0 La 142 1(-1) Pr 144 C-30
CLGSUR:- 5 R"=CC."-.0 OF SAl:APL"- --'~P"-" -Li"- CO""-C- 0 Sampl e hu.-..her: 3 Loc tion: Containment Atmosphere Time oi Analysis: 06CO 12/25/82 Temper ture, 'F: 220 Pressure, PSIG: 0.5 l!easured Specs-1c Actlvlty Correction Speci-.ic Activity I "oto (Enclosure 3). " /c" Fac-"r 9 STP,
" /cc Kr 87 Xe 131'e 133 1.3 1.3(-')
I 131 1.3 '- 3(--) I 132 I 133 1.3 I 135 Cs 13>> Te 129 tl/A 1(+1) Te l,32 Sr S9 Ba 1~0 La 1~0 La 142 Pr 144
EXP>1PLE The contain;,.ent at;os"here specI=ic ac ivi ies rust be corrected for te..perature and pressure. The correction ;actor calc iat;on is performed as ollows: This value is recorded on Enclosure 5. C-32
EfiCLQSR= RECQRO QF DECAY CORRECT 0 Ti;..e of Reac.or Shutdcwn, Step 7.1.4: 12/25/82 01CQ Sar.",pl e Bur,;ber: 1 Location: RCS Hot Lec Tire of Analysis: 12/25/82 0400 Dec y Specific Activity Oecay Correc:ed Constant, 8 STP (Enclosure 5), Specific Ac '.Vi ty, l:Cl /cc Iso:ooe 1/sec /cc Kr Si 1.5 (--') Xe 13'.~ B.7 (--) Xe 133 1.5 (-o) 1. 1(-:2) 1.1(-2} I 13 9.9 (--) 1.1(=a) 1..(.4) I 132 8. -'-5) I 133 1.1(-:2) 1.2(-'2) I 135 29(-) Cs 134 1.1 (-8) Rb 88 e.5 (- ) Te 129 1 7 (--'-) 1.1(-:3} 6.9(-.3) Te 132 2.5 (-o) Sr 89 1.6 (-7) Ba 140 6.3 (-7) La 140 4.8 (-6) La 142 1.2 (-4) 1-'1(-1) 4.0(+1) Pr 144 6.7 (-4) C-33
EXnllPLE EllCLOSURE 6 RECORD OF DECnY CORRECT'.Cll Tire of Peactor 5"v-"c~n, St p 7.'..4: 12/25/82 01GO Sar..ol e llv,".,her: Location: Containr,:ent Su;.o Tire oi'nalysis: 0500 12/25/82 Decay Speci ic Ac .ivity Decay Correc ed Ccrs:ant, 8 STP (Enclosure 5), Specific Activity, Iso.oce 1/sec /cc yci/ Kr 87 1.5 (-'} Xe 131m 6.? (-7) Xe 133 1.5 (-6) 1(-5) 1(-5) I 131 o o 7) 1( 2) 1(-2) ( I 132 8.; (-5) I 133 '- 3 (-6) l(0) 1(0) 135 2o(5) Cs 134 1.1 (-o) Rb SS 6.5 (-;) Te 12o 1.7 (-;) 1( 1) I.2(-.Z) Te 132 2.5 (-6) Sr So ( 7) E; 140 6.3 (-7) La 140 4.8 (-6) La 142 1.2 (-4) 1(-1) 5.6(-1) Pr 144 6.7 (-4) C-34
EXA'PL E.'iCLQSUR"" 6 R CC.".0 0;- DiC:;V CDRRiCT10:: Tire of Reactor Shutdo'un, Step 7.1.4: 12/25/82 0100 Sample t/umber: 3 Location: Containment Atmosphere Tir.e of Analysis: 0600 12/25/82 Decay Specific Ac.ivity Decay Corrected Constant, 9 STP (inclosure 5}, Specific Activi y, I/snr gcl /cc gcl tcc 1.5 (--) Xe 131m 6.7 (-7} Xe 1.5 (-6) 3( 1) 1.3(-1) 131 ~ '- (-7) 1 3( I) 1.3(->} 1 ~ 2 8.- (- ) I 133 9.3 (-6) 1.3(-3} 1.5(-3) I 135 2.9 (-5) Cs 134 1.1 (-8) Rb 88 6'--') Te ~2o 1.7 (--) Te 132 2.5 (-6) Sr 89 1.6 (-7) Ha 140 6.3 (-7) La 140 4.8 (-6) La 142 1.2 (-4) Pr 144 6.7 (-<) C-35
EXnl'1PLE Oecav Correc-ions RCS Xe13q 1.1(-:2) [1.5(-6)j (3) 36CO 1 )( g) 1.1(-4) [9 9( 7)j (3) 3600 131 1.1(-4) I 133'. 1.1{+2) -- e (3) 3600 1.2(+2) 7 ~ . - [1.7(-4)j (3) 3 00 1 1(+3) 6 a( ") 129 La1g2>> 1 1() .,- [1.2(- )j ( ) oo 4 0(,1) Con.air.;..ent Su'o
- [1.5(-6)] (4) 3600 Xe13+ ~ 1( 5) = 1(-5)
I ~1 1(, - [9.9(-7)j (t) 3600 I ~
) = 1(+2)
I133 1(0) . ,- [9 3(-6)j (') 3600 = 1(0) e 1>>>> g I(~I) e- [1.7(-9)j (4) 36CO
= 1.2(-2) "'1.'2'(-1) - e ' - " -' 5 6( 1)
Con-,a ir..-..en. A:."..os "here Xe1 3. 1 3(-1) - e 3600 1 3( 1)
. e- [9.9(-7)J (5) 3( 1) 1 = 1.3(-1)
I \ ~ 1 3( 3) e- [9--"(-6) ( ) 13 = 1.5(-3) These values are recorced or ""ncl osure 6. C-36
EtsCLOSURE 7 RECORD OF FISSIOII PRODUCT RELEASE SOURCE IDEUTIFICATIOII Sample Ihnuber: 1 Location: RCS Ilot Leg Decay Corrected Specific Activity Ca 1 cul a ted Fuel Pellet Activity Ratio Identi fied ~lsoto o fFnclosure 6)," /cc Isotope Ratio'nventor~ In Gas Gap Source }'r 87 Xe 131m 0.003 0.003 Xe 133 1.1(+2) 1.0 1.0 I 131 1.1(+4) 1.0 1.0 IIA I 132 1.4 0.01 I 133 1.2(+2) 2.0 0.5 Gas Gap I 135 0.17
- Deca~Corrected IInhle Gas Specific Activity Decay Corrected Ye 133 Spt:cific Activity Decay Corrected Iodine I sntnpe Speci f ic Act ivi ty R t Iodine Ratio =
Decay Cori ectetl I-l3l 'pecific Activity
ENCLOSURE 7 RECORD OF FISSIOtI PRODUCT RELEASE SOURCE IDEt(TIFICATIQtl Sample tiumher: 2 Location: Containment Sump Decay Corrected Specific Activity Ca 1 cul a ted Fuel Pellet Activity Ratio Identified P ~ISOtO O (Enclosure 6)," /cc ~ISOtO)O BOtio* Inventor Source Kr 87 Xe 131m 0.003 0.003 Xe 133 1.0 1.0 I 131 1(42) 1.0 1.0 <in I 132 0.01 I 133 1(0) 2.0 0.5 Gas Gap I 135 1.8 0.17
- tioble Gas Ratio- Deca Corrected tloble Gas Specific hcti~vit Decay Corrected Xe 133 Specific Activity Iodine Patio ~DI Decay I
Corrected II 1-131 I I!Oi~ih Spec>fic Activity
fttCI.OStlAE 7 ACCORD OF FISSIOlt PttODttCT RELEASE SOURCE IOEtlTIFICATIOtt Sample ttumber: 3 Location: Containment Atmosphere Decay Corrected Specific Activity Ca 1 cul a ted Fuel Pellet Activity Ratio Identified Isotope JFeclesure 6},'cc Isotope Ratio* 4 In Gas Gap Source Kr 87 Xe 131m 0.003 0.003 XG 133 1.3(-1) 1 ~ 0 1.0 ttn I 131 1.3(-1) 1'. 0 1.0 ttn I 132 1.4 0.01 I 133 1.5(-3) 1.2(-2) 2.0 0.5 Gas Gap I 135 1.8 0.17
- ttoble Gas aRatio =Decay Co}rected Xe 133 Specitic Activity Decay Correcteit Io(tine isotope Specific Activity Iodine Ra t1 0 Decay Correcte<t 1-131 Slu cit ic Act.ivity
EXPiPL E E:;CLQSu.=.E S Re Co~0 QF'gL~iS~ qLi
~ ~ s Y
Reac:or Ccolan- Ccr.- in .. en'ur.:p Con.a i n;..en: To-.a l Sarple lour.ber, Sa--pie flu;,.ber, A:-..osphere Sa-.,pie gu n itj Iso.ooe Ci Ci Hu;,.ber , Ci Ci Kr 87 Xe 131n Xe 133 2.5(+-) 1.1(-2) 6 o( 3) 3.2(+4) 1
'7 1 2.5(-6) 1.1(+5) 6 9(+3) 2.C(-o)
I 132
.I 133 2.8(+4) 1.1(-3) S.O( I)
I 135 Cs 134 Rb 88 Te '29 1.5(-.5) 5(=>) Te 132 Sr 89 Ba 140 La 140 La 142 9.2( 3) 5. 8(-:2) 9.S(+3} Pr 144 C-40
'lolu;..e corrections to Ssr.. PCS {RCS volu'eS = 94GO f" x .87 = 8178 ft = 2.3(S) cc]
- l. 1(~2) " 2.3(8)cc 2.5(-:4)
Xe]33. cc x x 1(-6) pci =
" cc 131'.1(+4) x 2.3(8)cc x 1(-6) pci = 2.5(=6) 133'.2( 2) " cc x 2.3(8)cc x l(-6) pci = 2.8(+4}
Tel o.. 6.9(+3) "'c x 2.3(8)cc x 1(-6) 'ci = 1.6( 6) La,,: 4.G(-:1} " cc x 2.3(S)cc x 1(-6) pci = 9.2(-3) Ccn'ain,.en~ Su o {(3 21 40,GOO f< 1 1(9) cc) See Enclosure 9j Xe...: 1(-5) x 1.1(9) 1(-o) = 1.1(-2) 1(+2) x 1.1(9) 1(-6f = 1.1(+5)
! 1%3 ~ l(0) x 1.1(9) l(-6) = 1. 1(~3)
I 0 1.2(+2) x 1.1(9} 1(-6) = 1.3(-3) Lal.2'5.6(-1) x 1. 1(9) 1(-o) = 5.8(+2) Contairr.,en . iAt'.05 here {volu
~ ~ ..e i STP 7.1 x 10 lp cc x la.7 +
14./ 0.5 492 680 5 3 x 1o cc Xel~~ 1.3(-1) x 5.3(10) x 1(-o) = 6.9(+3) 1,3(-1) x 5.3(lp) x 1(-6) = 6.9(.3) 1.5(-3) x 5.3(10) x l(-6) = 8.0(~1) Ti:ese values are recorded on Erclosure 8.
l N
E':CLOSL'.-",:- 9 CO'IT>I~~~'~>ENT BUILDING 'VJATER LEVEL n VOLUa'IE 24 23 L'J 22 21 20 19 20,000 30,000 40,000 50,00p 6Q,QQQ 70 QQQ 8Q QQp gp ppp VOLU'1E, FT C-~2
EX;.;PLE E tCLOSURE 11 RECCRO OF TRnHSIE!IT PO'hER CORRECTi0iI Sat-,.pie t)v'ber: 1, 2, 3 Location: Prior 30 Oay Power History: Power ".. Ovration, Oavs 22 50 17 ICO Power Pcwer Correction " Eqvilibrivn Sovrce Corrected Sovrce Isoto"e Fac:"r Inventory Inventory'as Gao Inventory Kr S7 Xe 13'.a Xe 133 0. 63 1. 3(7) 8.1(o) I 131 0.63 6.7(6) 4.2(6) I 132 I 133 6.7(6) 6.!(6) I 1"5 F;el Pellet In;entorv Kr S7 Xe 131~ Xe 133 0. 63 1.5(S) 0 ~(C) 131 0.53 7.3(7) 4:6(1) I 132 I 133 01 a.s(8) 1.<(B) I 135 Cs 134 Rb 08 Te 12Q 1.0 Te 132 Sr S< Ba. 1~0 La 140 La 142 1.0 1.6(S) Pr 14~ C-43
~'LE POllFR CORRECT10H FACTORS 0
Power Correction Factor = E P. (1-e j) e j 100 For Xe133 the Power Correction Factor is calculated as follows:
<P(1xt )xt 0 75(1[le5(6)][log(~6)])([1o5(6)][1+6(6)])+60(1[loS(6)][1,5(<6)])
[1 5( 6)][1 7( 6)l) + 100(1 e [1 6( 6)l[1 7( 6)])(eo) 63 Power Correction Factor = --'-
-63,6 = 0.636 1OO The remaining isotopes are calculated in the same manner; the results are recorded in Enclosure 11.
E (A>"ioLc EliCLOSURE 12 R r0~0 OF PE-"CE'iT REL~'<"- Total quantity Po:~er Corrected Available For Release : Source Inventory, x 1CO = Iso:o"s (Enclosure 8}, Ci Ci (En'closure 10 or ll) Percent Cas C o Inventory Kr 87 Xe 131 Xe 133 -'.2(+4) 8.1(5} 0 3a 131 2.5(-.6) < 2(6) 62 1"2 I 133 2.a(-:4) 6. 1(6) 0. -'8 I ldll Fuol "oll~t invnn-~rv Kr 87 Xe 13i.-., Xe 133 2.8(+",} a.4(a} O.CCG3 2.6(-5) 4.6(7) ~ ~ / I 2.'-(--') 1.-(8} Q l11 I 1"". Cs 134 Rb 88 Te 12a 1. 5(+6) 2."(7} 6.7 Te 132 Sr sa Ba }40 La 1'0 La 142 a 8(.3) 1.6(8) O.Q1 Pr 1'4 C-45
E X w"u'i P L E The following results are c"ncluded: (1) \ ihe charac= r..'s",'c -issicn products are r 1" 1 ard Te 129 (2) The sourc of iodine release is principally from the fuel rod gas gap. (3) 62 percent of the fuel rod gas gap 1-131 inventory is available for re 1 ea s e to the env i ronmen t. 6.7 percent of the fuel pellet Te-129 inventory is available ~ or release to the environment. Sased on these three pieces of infor."..ation and the charac; ristics of the ten categories o core damage described in Enclosure 1 the following conclusion is drawn.
== Conclusion:==
ihe core damage is estimated to be 'major Fuel Cladding Failure with concurrent Initial Fuel Pellet Overnea.irg. C-46
APPc.'NOIX 0.0 PRCCiDURE GUIC"='ll FOR ASSiSSl'ti'(T OF CORi Dr'l'QGi US I.'lG HYORQG~.'(
~ ~
0-1
TABL"- 0"- <"'i"-""S PHG>> 1.0 PURPOSE 0 2.0 R="--"-""-"CES 0 3.0 DEFI)( I i IGllS 4.0 PRECAUTION!S 4'i0 LIl1ITATIC;lS 0-4 5.0 It(ITIAL PLANT CCl/OITIOll/SYl<PTCl1S 6.0 PREREgUISITES 0-6 7.0 PROCE"URE 7.1 Record o. Plant Cordi ticn D -6 7.2 Hydrogen SarIpl ino 0-7 7.3 Contain-..ent Hydrocen Generation 0-7 7 4 Padiolysis Hydrogen Generation 0-8 7.5 Core Clad Oxidation 0-8 F 6 Percent of Ruotured Fuel Rods 0-8 7.7 Percent of Erbrittled Fuel Rods 7e8 Procedure Bias Adjustment 7.9 Core Danage Assess-,.ent LIST OF EllCLOSURES Clad 0'".age Characteristics oi tlRC Categories oi Fuel Oarage Encl osure 2 Record oi Core Uncovery Conditions Record of Sarpling Conditions and (!easured Hydrogen Erclosure 4 Ratio of Hater Densi.y at Sarple Terperature to Density at STP-; as a Func:ion of Temperature Enclcsure 5 Rec"rd and Calculation 'r!orksheet for Hydrogen Cenerated in Cor tair-,.ent Plant Spec'.;ic Hycrocen Production Rate in Containr.ent as a Function o= Terperatvre Record ard Calculation Morksheet or Hydrogen Generated by Radiolysis Generic Hydrogen Production by Radiolysis as a Function Tire After Reactor Trip 0-2
T"BL OF CC'lT:-."75 (Con-'d. ) Erclosure 9 Record and 'Aorksheet for the Percent oT Col e Clad Oxsd>-~d 0 Pere nt of Ruptured Rods as a Function o the Percent oi Core Clad Oxidi"ed 1 Percent of B:brittled Rods as a Function of the Percent o Core Clad Oxidized 0-3
1.0 PURPOSE This prcc cure is to be followed uncer post accider plant condi iors o determine the extent of fuel clad dai.aae which may have occurred. It utilizes hydrocen measured in samples obta'.-:d wi th the Pos- Accident Sampling System (PASS). The measured hydrogen is related to the amount of fuel clad oxidation. Clad oxida ion is in turn related to clad damage which is expressed in ter...s of the percent of fuel rods which are ruptured and the percent which are e..brit.led. The resulting observation of damage is described by one or more o the ten categories of core damage in Enclosure 1. 2.0 RE.""=~"."CES 2.1 Development of the Cc."..prehensive Procedure Guideline for Core Oamage Assessment, C-E Owners Group Task 467, llay, 1983. 2.2 Pos \ Accident Sampl inc System Opera tino PrcceCu res . ( Pl ant Speci f i c Oocu;..ent) 2.3 isUREG 0737 Item I..B.3. 3.0 OE." I .'(! 7 I C.'l S Clad Ructure: The fuel clad ruptures wnen he internal gas pressure exc=.ecs tne external coolant pressure ard he clad yield strength is reduc d because of elevated temperatures. Clad rupture results in release of casecus fission products frcm the cas gap and "cssibly sc;..e frag.-.ents o" fuel pellets but does not oiher:iise destroy the struc:ure cf the fuel asseroly. Clad Erbrittle-...ent: At temperatures above the rupture te...perature s>gnat-.icant ox;oation of :;-.= clad occurs. I, the oxidation exceeds the embrittle...ent threshold, frac;..enta:', .". of er:.bri: led clad may subsecuen.ly occur from thermal shock,;".,draulic pressure orces or handling such that the structure o, the fuel assembly is destroyed and substantial fuel pellet ,ragments are released to the coolant. 4.0 PR" CAUTIO"S . 's0 L )'ITATIO"S The assess;..ent cf core dar...aoe obtained by using this procedure is only an est'.-..ate. The technicues employed in this procedure are only accurate to locate the core condition within one or more of the 10 catecories of cor damace in Enclosure 1. The procedure is based on hydrocen data. Other plant indications may be available which can improve u'cn estimation of ;-ore damace. These include radiologic sa.".pie charac eristics, incore temperature indicators, and containment radiation,",onitors. "'henever possible these additional indic tors should be factored into tl e assessment. 0-4
1.0 PURPeS=- This prccecure is to be ollowed uncer post accident plant conditiors to determine the extent of fuel clad damaae which may have occurred. It utilizes hydrocen measured in sarples,obta'.".-d with he Post Accident Sampling System (PASS). The measured hydroaen is related to the amount of fuel clad oxidation. Clad oxidaticn is in tvrn related to c'lad damaae which is expressed in ter...s of the percent of fuel rods which are ruptured and the percent which are e.",.brittled. The resvlting observation of damaae is described by one or r;.ore o the ten categories of core damage in Enclosure 1. 2.0 RE,".ERE ACES 2.1 Development of the Cc'orehensive Procedure Guideline for Core Oamage Assessment, C-E Owners Group Task 467 <'!ay 1983. 2.2 Post Accident Samplinc System Operatina Prccedures. (Plart Specific Ooc ;..ent) 2.3 h i'lUREG 0737 Item Ii.i.3. 3.0 0"=.= I.'l ' I C.'l S Clad Ruoture: The fuel clad rup ures wnen the interral aas =pressure exceeas tne external coolant pressure ara the clad yield s rength is reduced because of elevated temperatures. Clad rupture results in release o- caseous fission products frcm the aas gap anc "cssibly sc.-.,e fraa."..ents of fuel pellets but does not otnerwise Cestroy the struc;ure cf the fuel assembly. Clad E...brittle.".ent: At temperatures above the rupture te.,perature s>gn>->c rt ox aation of t;-.-- clad occvrs. I the oxidation exceeds the e..brittle."..ent threshold, free;..enta:'.". o; e...brittled clad may subseauen.'y occur from thermal shock, h,draulic pressure orces or handling such that the structure of the fuel asserbly is destroyed ard svbstantial fuel pellet ragments are released to the coolant. 4.0 PRECAU IO'lS 0 L Il'IT ETIO.'lS 4.1 The assess;..ert cf core damaae obtained by using this procedure is only an est'...ate. The techniques employed in this procedure are only accurate to loc te the core condition within one or more of the 10 catecories o- core Camace in Enclosure 1. The procedure is based on hydrccen Cata. Other plant indications i".ay be available which can i".prcve u"cn estimaticn of ;-ore damace. These include radiolccic sa;..oie charac eristics, inccre te...perature indicators, and ccn ain;..ent radiation ;,.onitors. 'enever possible these addi:ional indicators should be factored into tI e assess'ent.
4.2 r"~ This pr"cedure relies upon hydrocen sarples taken frcm the c cn.ain",.ent '.o a -.os herre an~ .I e reac:cr ccolant system hot leg. Those samoles p'. may ccntain a mixtvre of hydrogen generated within the core by clad oxidaticn and also hydrcgen rem radiolyt',c dissociation of wat r and oxication o. aluminum and zinc in the containrent. The estimate of c ad damage is influenced by the arovnt of hydrogen generated by ex-core sourc s and by the ability to identify plan condi t',ons conducive to such hydrogen generation. Therefore, a hydrocen reasvre...ent is not a unique indicator of the amount of core clad oxidation. j 4.3 There are large areas c ~ - t of aluminum components in thee con.ainment ui ing of some plants. This aluminum would oxidize rapidly at te...peratures about 200'F and would be consumed within about two hours. The remainder of the aluminum and other oxidizing material react at a rate determined by temperatur and over a longer time. In the procedure all of the shor- term transient hydrocen is aenerated within the irst two hours and is added to tt:e slower acc .-.,u ation as a function of time. Hence, in ccntainments with arce areas of rapidly reacting aluminum, the prccedure is valid ,or hydrogen samples taken after about two hours with temperatures about 2CO'F, or af er the shor-. term oxidation is ccmplete. This proceCure yields estimates of the percentages of fuel rcCs wi:h rvotured clad and embrit:led clad. Sirul tarecus with e.-..brittl inc or the clad, there may be clad melting and pellet overhea ina occurrirg. This prccedure provides an estimate of only tl;e percen.ace of rods which have procressed to at least clac ruptur or c ~d erbrittlement, and does not attemot to predict the phys'.cal configuration of those rods which have prcgressea "eycnd lcca clad 1 frag..en ation. 4.5 Cepending on the accident scenario, a aiven total arount of hydrogen proCuced by oxidation of fuel clad can represent varyira local amounts and distributions of clad Carage. This prccedure attempts to bi s the da'ace estimates such that the results represent lcwer lirit estimates of clad damaae. Actval damaae could be greater, epending on plant specific details and on he acciden scenario. 4.6 This procedure is applicable unCer conditions for which there are no voiCs reasurable by the reactor Vessel Level Monitoring System. It is assumed that if such voics had been found, their reroval would be acccmolished by vsing the Reactor Vessel Vent System as prescribed elsewnere in the acticns to mi ticate the consequences of accidents. Hcwever, if the hydrogen samples are taken under conditions in wnich measur;ble void does exist, a guideline for analysis is provided in the addendum attached to this procedure to estimate the contribution of that source to be added to the total hydrcgen measured. 0-5
5.0 INITIAL PLA'.lT CCaeD I7 iS Ail0 SP":PTC,')S This procedure is to be employed for analysis of hydrogen sample data wnen it is determined that a plan accident wi h the potential for core damage has occurred. The,ollowino is a list of plant symptoms to assist in this determination. This list is not a complete representation of all events which may cause core damage. One or more of these symptoms may exis at or before the time the sample is obtained. Uncer these conditions, sampling should be performed using the Post Accident Sampling System. 5.1 High alarm on the containment radiation monitor. 5.2 High alarm on the C'1CS letdown radiation monitor. 5.3 High alarm on the main condenser air ejector exhaust radiation monitor. Pressurizer level low. 5.5 Safety Injec.ion System may have automatically actuated. 5.6 Possible high quench tank level, temperature, or pressure. 5.8 Possible noise indicative of a high energy line break. a Oecrease in volume control tank level.
- 5. 10 Standby charging pumps energized.
- 5. 11 Unbalanced charging and letCcwn flow.
5.12 Reactor Coolant System subcooling low or zero or superheated. 6.0 PP,ER:-CU IS I-=-S An operatioral Post Acciden Sampling System with the caoability to obtain and analyze the concentration of hydrogen in fluid samples which have the potential to be highly radioac ive. The system should meet the re".uirerents oi tlURiG-0737 Item II.8.3,.Peference 2.3. 7.0 PROC="DURE 7.1 Record the Following Plan: Indicators 7.1.1 Core damage can occur following reactor trip only when the coolant level within the reac:or vessel drops below the top.of the active fuel. Several instrument records are available from which an estimate of the core uncovery and recovery times might be made. The instruments are: 0-6
Reactor '/essel Level l',oni oring System Core Exit Ther.-.,ocouple Temperature Core Exit Thermocouple Saturation t1argin Record da a from these instruments according to the ins ruc:ions on the worksheet of Enclosure 2. 7.1.2 The magnitude of Reactor Coolant System (RCS) pressure durirg th core unc very period can influence the number nf early clad ruptures. Interpret the data from Step 7. 1. 1 to determire the best estimate for the time period of core uncovery and determine the range of RCS pressure during this time period. Record on the Enclosure 2 worksheet. 7.1.3 The presence of some subcooled inlet flow while the core is uncovering can slow the uncovery and cause greater loc l c'lad oxidation for a given total amount of core oxidation, thereo lead;.".a to a greater underestimate o the number of da;.aced rods predic:ed by this procedure. Observe available ins;ru;..ent reccr"s to determine if there was some reactor vessel inlet flo<<during the risino .emperature portion of the core uncovery period. Incluce net flew =rom charging and letdown systems, HPSI, LPSI, spray, etc. Record the data on the Enclosure 2 worksheet. 7.1.v Record the conditions in the containment and the re. c:or c"olant sys:em at the tirre the hydrogen samples are obtained in Step 7.2 folic<<ing. Enter on the worksheet of Enclosure 3. 7.2 Obtain a liquid sample from the RCS hot leg and a gas sa;..pie ,r"m the con.ain-,.ent atmosphere and analyze then for hydrocen corcentration using the procedures =or Post AcciCent Sample System operation described in Reference 2.2. Record the results cn the worksheet of Enclosure 3. Follow the instructions on Enclosure 3 to obtain the total amount of hydrogen measured in units of cubic feet of hydrogen at standard temperature and pressure. 7.3 The total measured hydrogen in Step 7.2 includes the hydrogen generated by three processes: 1) core clad oxidation, 2) radiolysis of water and 3) oxidation oi containment materials such as aluminum and zinc. The amount of hydrogen generated by the last two processes is calculated and then subtracted from the total measured to yield the amount generated by core clad oxidation. Enclosure 5 is a worksheet for calculating the amount of hydrogen generated by oxidation of materials within the containment. It utilizes measured data for the containrent temperature as a function of time up to the sampling time and a plant snecific curve of the rate of production as a function of con ain'ent temperature i n Enclosure 6. Record he data required on Enclosur~ 5 -nd ccr. l th e indicated calculations to obtain the cubic feet of hydrogen at STP generated by containment materials oxidation. D-7
0 'tl 0
7.4 The hyCroc n cenerated by radiolysis is a function of operating power anc cecay ti-e. Record the data required cn the worksheet of Enclosure 7, and utilize the curve of Enclosure 8 to obtain the cubic fee: of hydrooen at STP generated by r diclysis. The appropriate power is determined as follows: 7.4.1 For the case in wnich the operating power is constant or has no changed by more than 10 percent for a period greater than 30 days~ that power is used. 7.4.2 For the case in which the power has not remained constant during 30 days prior to the reactor shutdown engineering, judgement is used to determine the most,representative power level. The following guidelines should be considered in the determination. 7.4.2.1 The average power during the 30 day time period is not necessarily the most representative value for determining radiolysis by fission products. ?o4.2 2 The last power levels at which the reactor operated should weigh more h'eavily in the judgement than the earlier levels. 7 Continued operation for an extended period should weigh rore heavily in the judgement than brief .ransient levels. 7.4.3 For the case in which the reactor has produced power for less than 30 days, the procedure may be employed. However, the estimate of hydrogen frcm radiolysis will be too high and therefore the calculated hydrogen by core oxidation will be too 1cw. Hence an underprediction of core damage may result. 7.5 Enter the amoun.s of hydrogen from Steps 7.2, 7.3 and ?.4 on the worksheet of Enclosure 9. Subtract the arounts in 7.3 and 7.4 fr"m 7.2 as indicated on the worksheet to yield the cubic feet of hydrogen generated by core clad oxidation. Adjust with he plant speci.ic constant as shown on the worksheet to obtain the estimated percent of the core clad which is oxidized. 7.6 Enter the abscissa of the curve on Enclosure 10 with the percent o core clad oxidized from Step 7.5. Use the curve labeled with the pressure closest to but greater than the RCS pressure during the core uncovery period as obtained in Step 7.1.2 and re'corded on Enclosure 2. Read on the ordinate o Enclosure 10, the percent of fuel rods with ruptured clad. Record on the worksheet of Enclosure
- 9. Note that he sensitivity o measure...ent of. hydrogen is comparable to the range of oxidation on Enclosure 10. Hence, small amounts of clad rupture are not reliably predicted by this procedure.
7.? Enter the abscissa of the curve on Enclosure 11 with the percent of core clad oxidized from Step 7.5. Read on the ordinate the lower 0-8
ard u""er values 0 the rar."e inC'icated by the curve for the percent o; fuel rods which have e.,brittled clad. Record on the worksheet of . For a given percent oxidation of the core clad, the lower limit est'.-.,ate of embrittled clad in Step 7.7 is, =or most accident scenarios, the least amount of potential fuel structural failure. c.ual values are probably greater. The upper limit of the range in Step 7.5 may be interpreted as follows: Wnen the pressure during uncovery, frcm S.ep 7. 1.2 and recorded cn , is less than about 1GO psia, a rapid core uncovery by blcwdown is concluded. Heatup with minimum clad oxidation occurs. The extent of potential clad structural failure by melting may be greater than the upper limit of embrittler,"ent from Ster 7.7 as determined by oxidation. Hence, use the upper limit frcm Step 7.7. 4hen there is inlet flow while the core is uncoverino, the rate o= unccvery is slower than assumed in the derivation oi the curves on Erclcsures 10 and 11. For a measured total aroun: oi oxidation. the local percentage oxidation is probably greater along a shorter length of the upper portions of the fuel. Hence, favor the upper limi from Step 7.7. CO".-.E 0'l"AGE ASSESSllEl/T ihe conclusion on core damage is maCe using the two resul.: frcm above. These are:
- 1. Percentace of fuel rods with ruptured clad, Step 7.5.
- 2. Percentace of fuel rods with embrittled or struct rally damaced clad, Step 7.7.
Knowledceable judcement is used to cc pale the above two results to the Cerini:ions o-, the 10 'iRC ca .ecories of fuel damage found in . Core damage does rot take place uniformly. Therefore when evaluat ng damace usino these results, Enclosure 1 may yield a cc.-..binaticn of categories of damace wnich exist simultaneously. 0-9
EflCI.OSUAE I CLAD DAHAGE CllARACTER I ST ICS OF tlRC CATEGORIES OF FUEL DAHAGE t(AC Category Temperature Hechanism Characteristic Heasurement Percent of of F<<el Damage Ran e ( Fj of flamaqe Heasurement Range Dlllllbgt.'o(ls
- 1. tlo fuel Dama ge ~750 tlone Less Than 1
- 2. Ini tial Cladding Rupture Due to Haximum Core <1550'F* Less Than 10 Failure Gas Gap Exit Overpressurization Thermocouple
- 3. Intermediate 1200-1800 Temperature <1700'F* 10 to 50 Cladding Failure
- 4. Hajor Cladding <2300'F failure <2'X Greater Than 50 Oxidation
- 5. Initial Fuel Pellet, Loss of Structural Amount of Equivalent Core Less Than 10 Overheating Integrity Due to llydrogen Gas Oxidation Fuel Clad Produced <3%
- 6. 'Interaiedia te 1800-3350 0xidatioll (Equivalent to Fuel Pellet X Oxidation <18% 10 to 50 Ovei hea ting of Core)
- 7. Hajor Fuel Pellet <65K Greater Than 50 Overheal.ing
'lepends on Reactor Pressure and Fuel Burnup. ~ ' .-. for Pressure <'l200 psia and Burnup >0.'
EtlCLOSURE 2 CC"-= UllCGV~~Y CCl'OIT C"S S ep 7.1.1 Time period of core uncovery. Cc...piete the following table usirc recorded instru.-..ent data. Estimated Es ti-.a ted Ins r ';ent Core Uncoverr Tire Core Recoverv Ti.t.e Reactor 'lessel Level Lower Limit Elevation Lowe r Limit El eva ti on a4'.onitoring System Uncovers, Reccvers. Time Ti".e Core Exit Ther;..occuple Star. of Cont',nuous Rapid Ter.'perature Te.""era ture Rise or Exceed 660'F. Drop to Saturaticn. Time I I TerOera;ore Temperature Ccrc Exi . Ther;..occuole Star. of Superheat. Return to Satura",cn Sa:ura .cn .'9rcin T i-..e or Subccoling. I 1.>.e S:. o 7.1.2 In.erpret above data to obtain best es irate for time pericd o-. ccre uncovery and obtain pressurizer pressure rarce during that period. The superheat derived frcm the therroccuoie temperature
,aro corr sponding system pressure is considered as the best ircicator for core uncovery durirg "oiloff ard shculd be Us d, "ut should be cc."..pared with the o her indicators to help icenti iy pcssible arc;,.alies. The pressure durirg unccvery is used later on Enclosure 8, Step 7.6, to determine the apprcpriate curve ;or assessment of he number of clad ruptures.
Core LIncoverv Cor Recoverv T iI-..e Pressure Ste" 7.1a3 Estira:e vessel inlet flow rates dur rg core urcovery heatup period, up to approxi"atelv the time oi peak core exi. ther;..o-ccuple temperature. tIet inlet flow indicates that procedure may have additional bias which uncerpredic.s clad damage. Charging Flow Rate Letdown Flow Rate HPSI Flow Rate LPSI Flc;I Rate g>> Other Inlet Flows a 0-11
EiaCLOSURE 3 S .".PL!.'lG CO"DITIO:IS All0 l1""ASUR"0 HY"."-CG"= Step 7.1.4 C"..-;-..n the RCS and containment conditiors at the ti."..e of sar-.pling for hydrocen. Reactor Coolant System Contair.;,.ent Sampling Tire At;..osphere Pressure Ds19 Pressure ps 1g Atrosphere Terperature Tenperature oF Has Hydrogen Reccr.biner Yes/ "o Operated Reactor 'lessel Coolan Level Does Pressure or Te.-.."era-ture History Ird cate a Pressuri er Level Hyarogen Burn Yes/ o Step 7.2 Hydrogen Sar.-.ole Data Reduct on. Cont. Sa."..ol e
".ol . "./100)
Cont .".ol . (F'") ( "2 + ~60 } ., tlo ra 1 Ter:.p.
+ 860 )
Ft Hydrocen at STP
$ 92 Ho" Leg Sat;.ole RCS ".ol. Density Ra io 1CC0 F:" Hydr cen (cc/ka 0 S P) " (F:") " (=nclosur ~~
STP 1CCO Total Also record total on irclosure 9. D-12
0 EiICLQSL!RE 4 RATIO OF H~O DEi JSITY TO HgO DEiXSITY AT CTP vs TEib1PERATURE 700 600 400 300-200 100 00 0.25 0.50 0.75 1.0
>ACT STP 0-13
EilCLCSi3."-.E HYDROG '1 GE""~ 3 CC'tep 7.3 Record 1-.- .containment temperature at selec=ed time intervals and calcui't the hvdrocon c~"ora-~< bv ox'c~ io~ o= cort~irr~n materials u.i li=ing the plant-specific pr"c.ction rates from Enclosure 6. 5 ~ Avg. Containment H~ Prc". Rate Time at Star. Interval Temp. Durinc (ft"/hr, H> Produc d = of Intervals Duration (hr) Interval ('.=) Enclosure 6) 2 x 4 Acciden: Starts Sampling Time Long Term Hydrogen Production in Conta'r,"..ent, Total ft 9 STP Shor. term rapid hydr"cen production by contairment a'uminum (Table 4-3, Section 4.") Total Hydrogen Production in Contairment SCF Record total on Enclosure 9 also. 0-14
A E 'CLOSURE 6 HYOROGEN PRCDUCTIOi~h RATE FROsil ALUM'IliMUi'IAi"JO ZliVC vs TENIPERATURE FOR ST. LUCIE UNIT 2 8000 7200 6800 6400 56CO u 52CO This is an ev .",.ole of a plant specific c 4800 cur;e. See Fic"res 4-2o throuoh 4-31. I- fOr Other plantS. (J 4400 a C 4000 zW 3600 O a
>. 3200 G 2800 2400 2000 1600 1200 800 400 0
100 140 160 180 200 220 240 260 280 300 120 TE51P E RATUR E,
EhCLOSURE 7 HYOROGE.'l Gi.'lERATEO BY RAOIC'S!S Step 7.4 Record the .olio'iing data a-.: utili=e the curves of Enclosure 8 to determine tl e hvdrocen cererated by radiolysis. Prior 30 day power history Power, Pe.cent Duration, Oavs Po'ier o use in evaluating long erm hydrocen production by radiolysis = (Full Peuer, l'.ut} x Rear. or Trip Time hrs Sampling Time (see Enclosure 3} hrs Decay Time (Sampling Tin;e - Trip Tir,e} hrs Enter abscissa cn Enclosure 8 with above decay time and read t<<o values of hydrocen produced by radiolysis, one from each curve, in cubic feet of hydrocen at STP per f',<<t opera irc po<<er. I!uitiply by above power and record as follows: Hydrccen Produced Op~rating Total Hydrocen Limi t Curve (SC."-/,":it. Enclosure 8) Po'ier Produced (SCF) Upper Lower Using results from Padiological Oarage Assessment Prccedure est'rate which results should be used; upper limit or major fuel overheat, lower limit for
~
initial fuel overheat or appropriate estimate between the t<<o curves for intermediate fuel overheat. Circle corresponding value of hydrogen above and also record on Enclosure 9. 0-16
.'... (l.jul:l SPECIFIC IIAc':O' TlC I I YOIIO(iEH f'ilODUCTIONvs "IME 13 12 hlRJOII 1'UEL 0VEflllCRT 11 X
u V2 10 0 I- 9 CD D 0 0 K 0 0 'Lu (3 7 0 0 INTEBMEDIATE F UEL OVEAIIEAT x G CD I-
~f 0
0 Cr CD ll CD 3 ul A. tA INITIALFUEL OVEBIIEAT 100 200 300 ll00 NU GDO 700 T ll'1 E, I IOU llS
EllCLOSURE 9 CORE 0>WAGE <SSESS"='lT F";0;l Step 7.5 Hydrocen l'.easured, Step 7.2 Enclcsure 3 SC Hydrocen Prccuced in Containment, Step 7.3, Enclosure 5 SCF Hydrocen ProCuced by Radiolysis, Step ?.4, Enclosure 7, -SCF Subtract S eo 7.3 and ?.4 rcm 7.2 io Get Hydrogen Prccuced by Core Clad Oxidation SCF 0',vice by ( SCF/1".Clad Oxici=ed) = CI
= ".. Core Oxidi=ed S:ep 7.6 Enter abscissa cn ..".closure 10 with "", Core Oxidi:ed" and read ordirate rrcm cur'e iabelec with pressure durirc core uncovery as gi;en cn Erclosure 2, Stev 7.1.2. Record here Percent o; Fuel Rods wi th Ruptured Cl ad Step 7.7 Ente'r abscissa on ="nclosure 11 wi:h a"cve "".> Core Oxidized" and read range 0- values cn orcinate. Rec" r" here Percen= of fuel rccs e...bri: .ed Rance - Upper - l.ewer Step 7.8 Review S:ep 7. 1 and Bases sections to Cetermine which o> these limits is more likely to b representative of the core damage.
Step 7.9 Frcm Enclosure 1 select the core clad damaoe categories based on the above percentages o= rods ruptured and rods embritiled. (~) Plant Specific Factor frcm column 1, Table 4-2 of Section 4.5.
Ei JCLOSURE 10 P RCENT OF FUEL RODS WITH RUPTURED CLAD vs CORE CLAD OXIDATIOiV 1200' 100 RUPTURE TE IPERATURE 80 O C) 1500'F 60 1800 F l-I- 40 C C WHEN THE PRESSURE USF CURVE LABEL@0 C Ii'J STEP 7.1.2 IS WITH TEi'iIP ERATUR E iO
- 20. 4 100 PSIA 1200' 4 1200 PSIA 1500'F 4 16"0 PSIA 1800' 0
0 0.5 1.0 1.5 2.0
;o OXIDATION OF CORE CLAD VOLUs'1E
ENCLOSURE 11
~ OF THL FUEL RODS 'ARITH OXIDATION Ei'.IBRITTLE IENT vs ~
TOTAL CORE OXIDATION FOR 1",~ TO 3 ~ DECAY HEAT Ai JD 300 PSIA TO 2500 PSIA VJHEi~J COOLANT LEVEL DROPS BY BOILOFF V/ITH NO INL T FLO'~t UNTIL CORE IS RAPIDLY CUENCHED 100 UJ l-GO Z 0 60 O X 0 O 0 40 I-C/7 0 20 U u 0 ~O 0 ao 0 20 60 80 100
;: OXIDATIO'J o. CORE CL D VOLU:rIE 0-20
ACO""!lDUl1 TO APP"=.'iDIX D.O EST i'lATIOll OF NiCUtlT OF HYDROGE!t It) PEACTOR '/ESS""L HEAD VOID 0-21
1.0 PUB"-0< =
~he pur".ose o. this adCendum is to prcvide a cuiCeline for an analy.ical prccedur" to calcula e the amount of hydrogen cas con:a ned in a void in the top of .he reac .or vessel. This hydroce."
is added to he measured amount in Step 7. o; the procedure to determine the total hycrccen generated by all sources. 2.0 L lair. TAT i Ql15 2.1 The prererred method of Ceterminina the amount of hydrccen in the primary system is to sample liquid from the hot leg when the system is full. However, if the system cannot be filled, a procedure based on this aCCendum could be used to estimate the hydrcgen whi h s i th e vvessel void and which would not be evident frcm the hot leg liquid sar,:pie. 2~l This cuiCeline aoplies when the coclant level is above the hot leg and the remairder of the primary system is illed. Verification that the steam cenerator tu"es are filled can be provided by the existence of nat.ral convecticn flow in the primary system. If'the ccolant eve is below the hot leg, the guidel.Ines of this addenCu' Co rot apply 2.3 A reactor vessel level monitoring system is recui red which 'can provide the coolant level. The rolu;,.e of the void is obtaired .acr.e b y e a.inc the volume in the vessel above the coolant level to the va ue o; level for each spec',fic reac:or vessel desicn. 2.4 This cuiCel inc proviCes the analytical means for only an est'rate 0" fe hycrcgen ccntained in the void. The presence of other cases
.he inc udirc helium, nitrccen and fisscn product cases will add uncerta nty to the result.
3.0 P."".OC"-DL'2"" The follcwing is a cuideline "or an analyt'cal prcceCure to be fo lowed to es;...ate he a...ount of hydrocen cor. a'red in a d op oT the reac.or vessel. Calculational details ard plant specific information -..us be incluCed to irplerent this. guideline . 3.1 Oe te rmi ne the corditicns of the void as fcllcws: VVoid volu-..e (Ft3x) Cerived rcm measurement of ccolant level T Temperature of licuid at ccolant surface ('F) L '~ater satura'.ion pressure at tempelatul e T p pslt Reacor coolant system pressure (psia) to 3.2 A first approxi--aticn is r'ace assumiro the following: 3.2.1 The partial pressure of vapor in the void is assumed equal to saturation pressure at he liquid er.perature, T . This implies no 0-22
hea;na o- the voiC cas by 'he reac=or vessel walls and head. The are nor;..allv at re c:or cutlet temperature and could rerain above the ter>>perature of the void causing the vapor to be superheat d. 3.2.2 All the non-ccrdensible cas in the void is hydrcgen. This imolies no hellium or =ission pr duct gas from ruptured fuel roCs and no nitrccen frcm aiety Injection Tanks. A seccnd apprcvimaticn which eliminates this assuroticn is given in 3.4. 3.3 Calculate the arount of hydrogen as follcws: p =p tot -p sat' HZ Ft HZ 9 STP this
=
( ( PH I) ,) 14.1 (, TL + 460
)
total hydrocen in Add amount to the Step 7.5 of Appendix D.O.
~ seccrd pproximation can be r'ade in plants with a C-c Cesigned PASS wnich measures both total cas and hydrocen which are dissoi;ed in;he hot lea liquid sample. This aoorcxiration ircluces the follcwinc assumotions regardina the relati;e solubili .es o; '..".e non-ccncensible cases in the liquid.
3~>> j ihe cases are assumed to have the sare values of Henr;"s law ccrstant wnich relates the partial pressure of gas to tt:e arcun- c; gas dissolved in the liquid sa ..pie at equilibrium. 3 ~ >>oZ ;neo .he dissolved cas is not in equilibrium with ti:e cas 'n .he I~ void, the dissolved concentrations are in the sare rela-'.ve pl opol Ion as i= equil ibriun did exist. 3.5 The partial pressure of hydrocen is calculated ~ rcm ard he arount o hydrccen in the vessel head void is calcula ed us irg the equa ti cn abo se in 3.3. 3.5 This procedure can be extended to incluCe soeci>ic values o H n ~
'nry s
law constants but the assumption of equilibriun at the gas liquid interface would still be aues icnable. Also. to utilize detailed va ues of the gas constants, the individual cases in the sarple would have to be iCentified and reasured. This would require addi:icnal measurement capabi i ty. 1 0-23
EXc'llPi APP":<0 IX D. 1 E: ~.."PLE USE OF THE PRGCEDiv'RE 0-24
The folic'piro is g u.',liz ..g ihe a .p'e an eva;,. 1 f ' o .ne use of 4ihe prccedure for assess-...~nt o= clad al arount or hydrogen gener~-ion "h~ T e speci;ic c se ci.e
'+pg4 w',s =or the eZ ~
25c0 \I, g
.wt class ar selected or an accident equivalen- o ha-~ use of da;..ace assess;..ent based on r diolooic charac-or's-ics gu 'h g reac-Qrr. quantities n i in us d in "..e exa."..oie t..e ev ro'e f core =or .er sties ar..d given in Append x .
C.<. This exar.:pie consists of a set of cc'pleted:~or's"eet-prccedure and the acc".-.. anJ'ing figures and plant specific infor."..ation lna ol ...chilon. D-25
EXhtiPLE EtiCLOSURE ]- CLAD DhtiAGE CIIARACTERISTICS OF tiRC CATEGORIES OF FUEL DhiihGE tiRC Category Temperature . flecl~anism fharacteristic iieasurement Percent of of Fuel Aaniage Veasnrenent ~lian o ~Omnia e Ands
- 1. Iio Fuel Damage ~750 tione Than 1
- 2. Initial Cladding Rupture Due to Haxim>>m Core
<1550'F'ess Less Than ]0 Failure Gas Gsap Fxit
- 3. Intermediate 1200-1000 Overpressuri za t ion Tliermocouple <]700'F* 10 to 50 Cladding Failure Tempera tore
- 4. Iia.jor Cl adding <2300'F Fa i lure <2 "l, Greater Than 50 Oxidation
- 5. Initial Fuel Pelle Loss o f Structural Amount of Equivalent Core Less Than 10 Overheating Integi i ty Due to llyih ogen Gas Oxidation Fuel Clad I'rod>>ced <3X
- 6. Intermediate 1800-3350 Oxidation (Equ iva1ent to <]8%
Fuel Pel let X Oxidation Overheat\ng of Core)
- 7. Iiajor Fuel Pellet <65K Greater Than 50 Overheat.ing spends on Reactor Pressure and Fue"i ~';rnup. ~-.is=-. For Pressure <1200 psia and Aurnup >0.
0 E CLOSURE 2 CCR= U"CO"ERY COiiDITIOihS Step 7.1.. ~ Time period of core urcovery. Ccmolete the following table usinc recorded instrument data. Estimated Es tima ted Ins r 'ent Core Uncoverv Time Core Recoverv Ti;..e Reac:or "essel Level Lower Limit Elevation Lower Limit Elevation Yonitoring System Uncovers. Recovers. Tirre 02CO on 12/25/82 Time 02"5 Core Exit Ther.ocouple Start of Cont nuous Rapid Temperature Temperature Rise or Exceed 660 F. Drop to Saturation. Time 02:0 Time 02"5 Temperature "c0 F Temperature "32 Core Exi t Ther.-..ocouple Star. of Superheat. Return to Saturation Sa ur ation l! r"in Time 0205 or Subc"oling. Time 02"5 Step 7.1.2 Interpret above data to obtain best estimate for time period of cor uncovery and obtain pressurizer pressure range during that period. The superheat derived from the therrocouple temper ture and corresponding system pressure is considered as the best irdicator ior core uncovery during boiloff "and should be used. "u. should "e compared with the other indicators to help identiry possible arcr;.alies. The pressure durino uncovery is used later on Enclosure 8, Step 7.6, to determine the appropriate curve for assessment of the number of clad rup ures. Core Uncoverv Core Reccverv T i I..e 0205 0235 Pressure 1CCO goo Step 7.1.3 Es.'.-..ate vessel inlet flow rates during core uncovery heatup period, up to approximately the time of peak core exit ther,.o-ccuple t:..perature. linet inlet flo v indicates that procedure may have additional bias which underpredicts clad damage. Charging Flow Rate 0 @&il Letdown Flow Rate 0 HPSI Flow Rate 0 LPSI Flow, Pa e 0 Other Inlet Flows 0 Q-27
EXAl sPLE EliCLOSURE 3 SA"PLIiiG CO.'lDITIO:(5 Ai'D l',EASU."-.ED H'.";CG"-.'( Step 7.1.4 ("-. in the RCS and cont ir'ent conditions a the time of sampling for hydrogen. Reactor Coolant System Containment Sampling Time 0-'CO, 12/25/82 Atmosphere Pressure p" Pressure 1600 psig Atmosphere Te,,perature 220 F Temperature 300 'F Has Mydrocen Recombiner Ho Operated Reactor Vessel Coolant Level 100 Does Pressure or Teroera-ture History Indica e a Pressuri:er Level BO Hydrogen Burn iso Step 7.2 Hydrccen'Sample Data Reduc ion. Cont. S ."..ole Cont.3Vol. (32 + 460) Ivor...al Temp.) Ft Hydrogen (Vol. "./!CO) (Ft ) + ~60 at STP 6 O.COS x 2.5 x 10 x ~92 -: 580 10.600 Hot Leg Sa.-..ole RCS ".ol. Density Ratio . 1000 Ft Hydrogen (cc/ko 9 STP) (Ft ) (=nclcsure 0) at Sip 1200 x g'GO x 0.9 -: fQQP -10,2CO Total 20,800 Also record total on -".nc',osure 9. D-28
E'I CLGS v'."-."- RATIO OF H~O DENSITY TO HO DENSITY AT AP vs TEi'.IPERATGRE 700 600 500 400 300 200 100 00 0.25 0.50 0.75 1.0 PACT STP
0 E:(Ai t PL Ei'CLCSURE 5 HYC CG>>
~ G i~ti A 0 I >I CC 'T '>'T Step ?.3 <~cord the;"ntairment temperature at seleced t -.e ir,ervals and calcula:e the hydrogen cenerated by oxidat-',on of conte',nment materials utilizing the plant-specific prcc'ction rates from Enclosul e 6o Avg. Containment H2 pr"" Rate Time a Star Interval Temp. Ourinq (ft"/hr, H2 Produced =
of Inter:els Duration (hr) Interval ('F ) Enclosure 6) 2 x 4 Accident Starts OICO 25/50 300 2a20 20/60 260 2~GO 800 01'5 15/60 1SOO PPO 02CO 1.0 2~0 1&GO 1400 03CO 1.0 229 700 ?00 S'mpling Time Long Ter... Hydrogen Produc.ion sn Con.<>rmeni, To.al 6300 Short ter.-.. rapid hydrogen prcduc"ion by containmert aluminum (Table 4-3, Section 4.5} 5300 To al Hydrogen Produc ion in Con ainment 11.600 SCF Record total on inclosure 9 also. 0-30
FIGURE 4-28 HYDROG EV PRODUCTION RATE FROi'~l ALUM'IliJUi'il AiVD ZINC 'rs TE:.'iPERATURE FOR ST. LUCIE UiVIT 2 8000 7600 7200 6800 6400 6000 5600 u 5200 C
- c. 4QOO z
3600 O Q o- 3200 2800 2'00 2000 16QO 1200 800 400 10Q 120 1'0 160 180 200 220 2'0 260 280 300 TEiiIPERATURE, F 0-31
ENCLOSURE 7 HYORCGFoi GFeaERATa-"0 BY RA01QLYS S S ep 7.'ecord he. fbi lc;ving data an"'tili "e the c"rves o "rclosure 8 to deter~ire the hycrcgen cererated by radiolysis. Prior 30 day pc;ver history Power, Percen. Ouration. Oavs 22
}7 100 Pcwer to use in evaluating long tera hydrcgen prcducticn by r diolysis = (25cO) x 0.5 Reactor Trip TiI-..e 0}CO hrs S r;.cling Ti-..e (see inclosure 3) 0~00 hrs Decay Ti;..e (Sa.-..pling Tir,.e - Trip Ti;..e) 3 hrs I
Enter abscissa cn =rclosure 8;vith above decay tif.".e and read two values of hydrccen prcc.ced by radiolysis, are frc~ each curve, in cubic feet of hydrocen a STP "e. ".~t operating pcwer. t'.ul tiply by above power and record as follows: Hydrocen Prcduced Operatirg Total Hydrccen Limit Curve (SC";/.".v . ""rclcsure 8) " Pcwer Prc"uced (SC."-) Upper 0 }280 5}2 Lc;ver 0 } 1280 Using results free Radiolog cal Oa"..ace Assess;..ent Procedure estimate '.vhich results should be used; u"-er limit =or rajor fuel overneat, lower linit for ini ial fuel overheat or ap"rcpriate estirmate between the two curves for inter:-..ediate fuel overneat. Circle corresponding value of hydrocen above and also record on "=nclosure 9. 0-32
0 SPECIFIC BAD IOLYTIC IIYOI)OGEN PRODUCTION vs TIME 13 -. 12-MAJOB FUEL OV EH I I EAT x u u V) 10-z 0 I-u D O 0 K CL CO C1 zW D U O fL 0 LJ O INTERMEDIATE FUEL OVER I IEAT c, W x U I-0 0 K O u u UJ 0 V) INITIALFUEL OVEBIlEAT 100 200 300 noo No 700 7lti1E, ) !OURS
ERAL'APL E EilCLOSURE 9 CORE OAi'QGE ASSESS<'i HT FRCa1 HYORGGE!( HEASUREEt! E!lT Step 7.5 Hydrccen .".easured, Step 7.2, Enclosure 3 20.800 SCF Hydrocen Prcduced in Containr.ent, Step 7.3, Erclosure 5 11,600 SCF Hvdrocen Prcduced by Radiolysis, Step 7.4, Enclosure 7, 130 SCF Subtrac: Step 7.3 and 7.4 fron 7.2 to Get Hydrccen Prccuced by Core Clad Oxidation 0 100 SCF Oivide by (-'.650 SCF/ "..l Clad Oxid',zed) =
= ".. Core Oxidized Step 7.6 Er.:er abscissa on Enclosure 10 with "".. Co r e Oxidized" and read or"inate frc' curve labeled with pressure Curine core uncovery as c ven on Erclos re 2, Step 7. 1.2. Record here Percent o; Fuel Rods with Rup ured Clad ~1CQ Step 7.7 Enter abscissa cn:nclcsure }1 with above Ccrc Oxidized" and read rar.ce of values on orcir. re. Record here Pe. ce..t oi fuel rcds embrittled Range - Upper 22 - Lcwer Step 7.8 Review S:. o 7. 1 and Bases sections to determine which o these limits is r;.ore likely to be representa ive of the core damage.
Step 7.9 Frcn Erclosure 1 se'lec the core clad damaae catecories based on .he above percen. ces of rods ruptured and rods embrittled. The assessment yields ca:egcry 4, .'major Clad Failures with category 5, Initial Fuel Pellet Overheatir.g. Because o slew core urcovery and moderate pressure, lower limit of " rcds e,.."rit-led is selected. Frcm Column 1, Table 4-2 of Section 4.5. 0-34
~ ~ rgb F. ( r.:".
Eii!CLOSURE 10 PERC NT OF FUEL RODS'V~ITH RUPTURED CLAD vs CORF CLAD OXIDATION GO'
~1 2 100 RUPTURE TE IP ER ATUR E 80 C
C 47 C 1500' 60 1800'F I Ch 40 O
"PHEfJ THE PRESSURE USE CURVE LASELED C I~J STEP 7.1.2 IS yyITH TEi'APER~TURr I ~ 100 PSIA 1200' 20 4 1200 PSIA 1500 F ~~ 1650 PSIA 1800'F 0.5 1.0 1.5 2.0 o OXIDATION OF CORE CLAD VOLU'1E
ENCLOSURE 11 o OF TH FU L RODS ~iITH OXIDATION BRITTLEÃcENT vs TOTAL CQ R E OX I OAT ION FOR 1 o TO 3".DECAY HEAT AND 300 PSI TO 2500 PSIA SVHEN COOLANT LEVEL DROPS BY BOILOFF Vr'ITH NO INLET FLOWN UNTIL CORE IS RAPIDLY CUEi".CHED 100 I I-SO 0 GO O X 0 u 0 40 l-C/) O O 20 D LL u O
~O 0
0 20 40 60 80 100 o OX ID TION OF CORE CLAD VOLUi'i>E 0-~6
APP.".'lD IX E.O PROCEDURE GUIDEL IllE FCR ASSESS.","-.'li 0;" CORE Dr"'lAGE USI'lG CORE EXI7 THER,",OCCUPLES
T""-L-" QF CO'lT"""TS 1.0 PURPOS E-3 2.0 REF. RE'ACES E-3 3.0 OEF I fl IT I 0'lS E-3 4.0 PR" CPUT,'ls S nhO L Il'1 I TAT 10.'/S 5.0 Iia IT InL PLANT CCliD ITIO>s/SYt'OPTO"IS 6.0 PP ~P ECU I S ITES E-4 7.0 PROCEDURE LIST OF E'ENCLOSURES .nclosure 1 Clad Oamace Characteris:ics of t/RC Cateccries oi Fuel 7)- Oar.",ace Record of Te,.perature, Pressure and Oa;.,ace Es:i;..ate E Percent of Fuel Rods with Ruptured Clad as a Furc=;cn of Haxirun Core Exit Trer."..ocouple Temperature P Q E-2
0 1.0 PURPQZ~ This procedure is to be followed under post accident plant ccnditions to determine the nv;.,'r of fuel rods with ruptured clad. It provides and estimate of damage up to abcut the tir,:e when the peak core temperature reaches about 2300'F. At that time r.os: o-the rods probably have ruptvred clad but little other structvral dearadation has occurred. Therefore this procedure applies to the . relatively less severe acciCents althcuah it may be used for other accidents to ccntirm that damage exceeds this minimum r.ount. The resulting observation of core damage is described by categories '. thrcuoh 4 of the ten l(RC categories in Enclosvre 1. 2.0 REF""RE.'lC" S 2.1 Development of the Comprehensive Prccedure Guidelines for Core Damage Assessment, C-E Cwners Group Task 467,,'lay, lg03. 2tZ (Appropriate plant specific Cocvment which describes c oabilities and operaticn of InaCequ te Core Cooling Ins;rumentation incluCina Core Exit Ther;..occupies.) 2 ~ J Generic Ther .al-Hydraulic Func ional Qesicn Objectives for Inadeauate Core Coolina Instr,"entation, CE-HPSD-Igg, prepared for the C-E Cwners Group. 0~=1;)r T 3 ' 3 Clad R ": re: Clad rupture is defined as a break in the fuel roC g Ruptvre ray be preceded by ballocning of the clad if the internal gas pressure exceeds the exrernal coolant pressure durirg an accident, and the temperature is higher than norral. 4.0 PRECAU '0'lS AilD L lllITATI 0'sS The assessment of core damage obtained by usina this prccedure is only an estimate. The techniqves employed in this procedure are only acc rate to locate the core condition within one or more of the. 10 c'.ecories of core damace described in Enclosure 1. The procedure is basea on core exit temperature data. Other plant irdications may be available which can improve upon estimation of core dar age. These include radioloaic sample charac. eristics, the total quantity of hydrogen released frcm ".irconium degradation and containment radiat on monitors. 'rlhenever possible these additional indicators shculd be factored into the assessment. 4.2 The assessr ent of damage provided by this procedure. extends up to the time of clad rupture on most of the fuel rods. This time occurs early in very severe core uncovery acciCents. tagore severe core dar,age cannot be quantified by this procedure.
4.3 col equi ther ocQJ e c aC :e..."erature varies with the core uncovery scenario. This procecure applies to slow core uncovery by boiloff of the coolant. or o;her 'ore rapid uncovery scenarios this procedure could yield a very low estimate of the number of rup ured rcds. In c neral, for core uncover. at pressures below about 12GO psia there is high confiCence that at le'ast he predicted estimate of roCs are actually rup-ured. 5.0 I f( I 7 I AL PLi) T CQllO I 7 I Qaa 5 Atl0 S Yt OPTO" S This procedure is to be employed for analysis of core exit, ther...ocouple data when it is determined that a plan I, acciden with t e oo.ential for core damace has occurred. The following is a list I of p'iant symptcms to assist in this determination. This list is not a ccmplete representation of all events which may cause core dar ace. One or more of these symptoms will exist at or before the time the core exit:her."..occuple recorded temperature is utili-ed to estimate damage. 5.1 Hi ch alarm on the containment radiation monitor. 5.2 High alarm on the C'JCS letdcwn radiation monitor. 5.3 High alarm on the main concenser air ejector exhaust radiation monit"r. 5 Pressuri=er level low. 5.5 Sa;ety Injection System may have autcmatically actua.ed. 5.6 Possible high quench tank level, temperature, or pressure. 5.8 Possible noise indicative of a high energy line break. J~0 Decrease in volur.e control tank level. 5.10 Standby cha rg ing pumps ene rg i zed. 5.11 Unbalanced charging and letCown flow. 5.12 Reactor Coolant System subcooling low or zero. 6.0 PP.ER".DU IS IT"=S An operatioral inadequate core cooling instrumenta ion system which includes core exit ther.-..occupies and which can select ard permanently record the hichest ther;.,occuple temperature ior convenient, later inspection. A system which satisfies the requirements for core exit thermoccuples in Reference 2.3 is adequate.
0 7.0 PROC".".>~P c 7.1 Obtain the ollcwing frcm the instrument recordir.as: 7.1.1 Frcm the recardina of maximum core exit thermoccuple temoerature,.as a func:ion of ti;..e, obtain and record on Enclosure 2 the maximum temperature and the tir.-.e it occurs. 7.1.2 From the recording of reactor coolant system pressure as a function of time, obtain and record on Enclosure 2 the pressure during the pericd of maximum thermocouple ter;".perature. 7.2 Select the curve on Enclosure 3 which is labeled wi h a pressure approxirately equal to or greater than the pressure in Step 7.1.2. Enter the absc ssa at the maximum temperature frcm Step 7.1.1 and read on the orCinate the percent of the fuel rods which have rup:ured clad. Record on Enclosure 2. ?.3 The result in 7.2 is probably a lower limit estimate of damace. Sc-...e iudce."..ent on the bias is available as follcws. I ov ~ s This procedure applies most directly for relatively slew core urcovery with a maximum te... erature belcw the rapiC ox cation temperatures at about ]SGO'F and above. A smooth core exit ther.-..occuple recordirg and an uncovery Curaticn cf 20 .-..irutes or loncer are indicators for a gooa predict cn of clad ruptures. ?o3 2~ If the pressure in 7.1.2 drops to less than abou 1CO psia with'.n less than about two minutes o- accident ir.itiaticn, a lar"..e br e~ is i'ndic ted. This causes ur.detectea core heat p follcwed by flasn'.n" durira refill. Depending on the rate o; refill, the ther."..occuple temperature may rise rapiCly then quench <<hen the core is reccve.ed. This prccedure could yield a very low estimate for the percent o-rods ruptured. ?e3.3 If the pressure in Step 7. 1.2 is above abou 1650 psia, it could exceed the rcd internal gas pressure deending on rcd burnup, causir.a clad collapse onto the fuel pellet instead of outward clad balloonirg. The clad ruoture criteria are less well defined for-- such condi:ions, but at temperatures above 1NO .F <<here the highest pressure curve applies on Enclosure 3, clad ailure sufficient to release fiss'.cn gas is likely and this prccedure may be used to obtain estir.",ates of damace. 7~< CCR Dn<'.AGE AS S" S hE'lT Use the percent of rods ruptured rcm Step 7;2 and the clad dar.",age characteristics of Enclosure 1 to determine the tIRC.cateczry of cl aiding a i lure. This procedure vields damage estimates in categories 2, 3 or 4 ~ E-5
e EllCLOSURE 1 CLAD DAtlAGE CllARACTERISTICS OF tlRC CATEGORIES OF FUEL DAHAGE t/RC Ca of Fuel tegory Dama e Temperature Ran< e ('F) llechanism Characteristic
~ll Heasurement ~D Percent of
- 1. tlo Fuel Damage ~750 t(one Less Than 1
- 2. Initial Cladding Rupture Due to llaximum Core <1550'F+ Less Than 10 Failure Gas Gap Exit
- 3. Intermediate 1200-1000 Overpressuri za tion Thermocouple <1700'F* 10 to 50 Cladding Failure Tempera tur e
- 4. tlajor Cladding <2300'F Failure a <2X Greater Than 50 Oxidation
- 5. Initial Fuel Pellet Loss of Structural Amount of Equivalent Core Less Than 10 Overheating Integrity Due to llydrogen Gas Ox i da t, 1 on Fuel Clad Proouced <3X
- 6. Intermediate 1800-3350 Oxidation (Equivalent to <IQX Fuel Pellet X Oxidation Overheating of Core)
- 7. llajor Fuel Pellet Greater Than 50 Overheating
, spends on Reactor Pressure and Fuel Burnup. ", '.44" far Pressure <1200 psia and Gurnup >0.'
/
E'lCLOSURE 2 RECGRQ OF TE,"PEoATURE PRESSUR" QA',)hG QS 'rl ~ 4 Step 7.1 Record the following data Haximum Core Evit Ther'ocouple Temperature oF Ti-,.e of I1aximum Temoerature Reac.or Coolant System Pressure at Above Time psia Step 7.2 Fr"m Enclosure 3, at maximum therr.ocouple ter.".perature and at appropriate pressure read percent of r ptured rods. C/ Step 7.3 Cc.-.,;..ent on prcbable b as of result in ?.2 (see paragr ph 7.3 in text). Step 7.4 tlPC category of cladding failure frow Enclosure ~ E-7
ENCLOSURE 3 PERCENT OF FUEL RODS WITH RUPTURED CLAD vs hIAXli'IVi%1CORE EXIT THERhlOCOUPLE TEAIPERATURE WHEN THE PRESSURE USE CURVE LABELED IN STEP 7.1.2 IS WITH TEii1PERATUR E
~~ 100 PSIA 1200' 4 1200 PSIA 1500' ~ ~1650 PSIA 1800 F 100 C
80 1200" o F 0 CLAD RUPTURE D Tc'riPERATUR E 60 I 0'h C 1500" F C v0 1800' 0 z o 20 0, 1200 1400 1600 1800 2000 2200 51AX I" IUi" 1 COR E EXIT THERiblOCOUPLE TEi'IP ERATUR E
APP"- F0lX EXr'il!PLE USE QF THE P<ct =.~ 'PE E-9
EXAt LPLE The ~ ollcwino is an exa<...p le oo= .he h d'or prccecure -.or assessr:,ent of clad use oi4 the darace by util i=ing the maxi;..u~ temperature recorC~d =ro~ th t er...ocouples. ihe speci ic case cited is or ~h ~ e
~-"=0 '! class of reac.or. ".=0 v:wt 1 Quan ci ties in ti e oxarole aree selec.ed s 1 for an accident equ',valent to that used in exal.pie . Col based on radiolocical characteriscics he 'pp '
nd Given in Appendix C.l. and 's 0 da .aoe asassess;..ent
- 1. Th', exa...pie c nsis:s or a cc'plated;vorksheet given in the proceoure.
E-10
e EXAllPLE E'iC'~.'RE 2 RECORO 0 T "P~~ "R"- PRCSC""c A;,O O "AG- "S 1"""" Step 7. 1 Record D'e following data i'iaxirum Core Exit Ther..ocouple Te...oerature 2000 F Time of llaximum Temperature 02"0 on 12/25782 Reactor Coolant System Pressure at Abore Tin:e g00 osia Step 7.2 From Enclosure 3, at mavimum thermocouple temperature and at apprcpriate pressure read percent of ruptured rods. 05 Step 7.3 Cc.-...-..ent cn probable bias of result in 7.2 (see paragraph 7.3 in tex:). Cool nt pressure <llCO psia for curve, so estimate is low. Unco:ery period iona compared to CET delay time so temperature reoresents steam closely. Rod temperature is actually hicher than CE; e.-.."er".ure, but the estimate would not change much with a Conclude that prediction is a good estimate of actual 300'ncre'se. cl-ad failures. Steo 7.'. ~ Ia I i~C ca tegor: o'l adding fa i lure free Enclosure 4, .".ajar clad -.ailure. .'atecory
APPEflOI< F.O PROC v'Ri GU IDEL IH FOR ASSR l'l"!a7 OF COR" DAl'!AGc. USING RAOIATIOf( 00SE RAT.S
ii~~L. OF CO,li- iS 1.0 PURPOSE F-3'-3 2.0 REF"REtlCES 3.0 DEF I HIT I GtlS 4.0 PRECAUTION'IS AtlD LliiITATIG'lS 5.0 INI I IAL PLn>sT CCIaDITIG>f AI'eD SYIIPTCa'1S F-5 6.0 PREREGUI S ITES F-5 7.0 PRGCEDURE F-6 7.I Record of Plant Condition F-6 7.2 Plant Po.~er Correct on F-6 7.3 Decay Correction F-7 7.4 Assess-..ent of Core Damace F-i L IS OF =.'lC'SUiRES Enclosure I Radiolcoical Charac:eris.ics o'RC Cate ories o; ruel Darace F-o Erclosure 2 Pos: nccident Dose Rate Inside:'re Cert ir..-..ent "=uild:ng F c
1.0 PUR Oc This prccedure is to be followed under post acciCent plant ccndi ions to deter.-..ire the type and dearee of core damace which .-..a have occurred by usina radiation dose rates measured insice the con:airment bvildina using the wide rance radiation moni"or. The radiation dose rate is related to the qvanti ative release of fission prcducts frcm the core expressed as the percent of he source inventory at the tire of the accident. The resultina abservation of core darace is described by one or more of the ten ca egories of core damage in Enclosure l. 2.0 R'EF:RE!ICES 2.1 Develop'ent of the Cc."..prehensive Procedure Guideline for Core Oamaae Assessr.ent, C-E Owners Group Task 467, t1ay lga83. 2.2 Mid Rance Ccntainr.;ent Radiation Dose Rate tlonitor Operatirc Prccedures (Plant Specific Document). 3.0 0" ~'l tTr.0'lS 3.1 Fuel Oamaae: For the purpose of this prccedvr fvel damage is aer>nea as a proaressive failure of the material bovndary to prevent the release of radioactive fission products into the reactor ccclan. starting with a penetration in the zircaloy cladding. The tye of fuel damace as determined by this procedure is repor:ed in terms of fcvr r,".ajor catecories which are: no damace, claddina failure, fuel overheat, and fuel r.elt. The first three catecories are charac.eri ed by the resultina radiation dose rate insiCe the ccnta',n...ent building. Tne degree of core damage is determined by making a comparison between dose rates measured following an accident and analytically determined values of the realistic or bes es imate dose rates that ~auld correspond to the specific catecories of core damage. The decree of core darage as determined by this prccedure is reported in terms of three levels which are: initial; intermediate; and majcr. This results in a total of ten possible catecories as characterized in Enclosure l. 3.2 Source Inventory: The source inventcry is the total quantity of fission procuc;s expressed in cvries of each isotope present in either source; the fuel pellets or the fuel rod gas gap. 4.0 PR CAUT}0<(S niiO L [a'llTA I ICaaS 4 } The assessment of core da...aae obtained by using this procedure is only an estir.",ate. The techniques er.ployed in this procedvre are only accurate to locate the core ccndition within one or rore of the 10 categories of core Carage described in Enclosure 1. prccedure is based on radiation dose rate. Other plant indicatiors may be available which can improve vpon the estir ation of core damace. These include sar.pie radiological analysis, inccre F-3
tempera ure irdicatcrs, and the total cuanti.y of hydrccen released frcm zirccniu-... Cecradation. 'whenever poss ible these adcitional irdicators shculd be fac:ored into the assessment. 4.2 This procedure'elies upon radiation dose rate measur=.:nts tal'en frcm one or more monitors located inside the contain;..ent building to the .otal quanti y of fission products released ,rcm ".Ve 'etermine core and therefore available for release to the environment. The amount of fission produc s present at the location of the monitors may be chanoing rapidly due to transient plant conditions. Therefore, rultiple measure..ents should be obtaired within a minirum time period and when possible under stabilized plant conditions. Samples obtained durino rapidly changing plan conditions should not be ~eighed heavily into the assessment of core damage. 4.3 A number of fac.ors influence the reliability o =the measured radiation dose rates upon which this procedure is based. Reliability is influenced by the ability to obtain representative measuremen:s Cue to inccmolete mixing of the measured meoia, equi".-..ert limita"',cns, and lac.': o; operator far..iliarity with rarely used procecures. additionally the prcceoure relies upon analytically Ceter."..ined values of the oest estimate dose rates tha . are anticipated to correspond to the specific cateoories of core damace. These analytical values are based upon assumptions maCe about the identity and relative proortions of the fission prc"uc s released rom the core and their transort within the ccn: in.-..ent building. Therefore the procedure is only accur te to within the validity of the assu."..pticns. This procedure is limited to the upper bound condit',on cf fissicr product release frcm the core due o fuel overheat. Si.,.u'.t recus with fuel overheat, there may be localized fuel pellet .-..elt.no within the core. The transport of the non-volatile fission products released due to melting is rot known. The Cose rates measured uncer cordi tions of fuel pellet mel ing are anticipated o exceeo ".hose, shown in Enclosure 2, for major Tuel overheat. Hcwever, this procedure Cces rot attempt to identify the extent of any potential fuel melting. 4.5 This prccedure is limited:o the interpretation of the Cose ra e measurement resul inc frcm a mix of ission products. The procedure canrot acc.rately dis.incuish be'.:veen the conditions of fuel claddirc failure and fuel overheat when he resultino dose rates are the same. The proceCure Coes provide an upper limit estimate of the progressive core Ca."..age. Concurrent conditions of claddino failure and overheat should be anticipated due to the radial dis ribution of heat generation within the core. Distinction between the type of core Car age recuires the identifica .;on of the characteristic fission products. The procedure for core damace assessment using radiological analysis of fluid samples is required to explicitly distinguish between the categories.
4I 4.5 This procedure is limi: d in apolic bili y to those ccnditions in which the fissicn prcCuc. ',nventory in the core ha" had suffic'.ent time to reach ecuilibrium. Equilibrium fission prcduct inven or" is a func.icn of re'c.or pcwer and burnuo. Based upon the fissicn prcducts of ccnc rn ecui librium conai t ons are achieved after thirty days of cperat cn'at ccnstant power. Constant power is consiCered . to incluce chances or no greater than 10 percent. The procedure may be used follcwina non-ccnstant periods of oper .ion by usina engineerina judaement to select the most representative power level during the pericd. The prccedure may also be used if the reac.cr has produced pcwer ror less than thirty days, however, the resultina assessment of core damage would be an unCerpredicticn of the ac:ual conditions. 5.0 IN ITIAL PLANT COlsD ITIOtiS Ai'lD SYMPTOMS This procedure is to be employed for analysis of dose rate Cata when it is Cetermined that a plant accident with the potential for core damace has occurred. The follcwing is a list of plant s:mptcms to assis; in this Ceterminaticn. This list is no: a cc;..olete representation of all events which may cause core Camace. Ore ore more o- these symptcms ray exist at or Lefore the time the sam ie is obtained. UnCer these conditions, samolirc shouid be "er;or.-..ec usina the Post AcciCent Sampling System. 5.1 Hiah alarm on the ccntainment radiation monitor. Hich alarm on the C"CS letCcwn radiaticn monitor. High alarm on the main condenser air ejector exhaus: raciaticn mcni or. 5.4 Pressurizer level low. 5.5 Sa ety Injection System;ay have autcmatical ly actuated. 5.5 Possible high cuench tank level, temperature, or pressure. 5.8 Possible noise indicative of a high eneray line break. 5 a Decrease in volume ccntrol tank level. 5.10 StanCby charging pumps energi-ed. 5.11 Unba 1 an c ed charging and letdown f 1 ow. 5.12 Reactor Cool rt System subccolina low or zero. 6.0 PP".".cCU IS An operational '~iide Rance Padiation Dose Rate Monitor System with the c pability to measure the area dose rates inside the containment F-5
buildirg resulting rcm fission products dispursed ir, the build rc at ~ os@here and plated ouz on building surfaces. The system should meet the require.".enzs of Regulatory Guide 1.97. 7.0 PROC"DURE 7.1 Record he following plant indications. 7.1.1 Containment Suilding: Radiation Dose Rate Rads/hr. Time of i".easurement Oa te Time Prior 30 days power history: Power, Percent Dura-.'.cn. Days 7.1.3 Time of reactor shutdcwn Date T i IT.e 7.2 Plant Power Correcticn The measured radiaticn dose raze inside the containment buildiro is to be corrected for the plant po'>er history. A ccrrection factor is used to adjus the measured dose raze to the corresponding value had the plan". been operating at 100 percent po'>er. 7.2.1 To correct the radiation dose rate for the case in which plant po:>er level has remained ccnszanz for a period greater than 30 days a simple ratio of the pcwer may be employed. The reactor power is considered to be constant if it has not changed by =10 percent within the last thirty days prior to the reactcr shuzdcv>n. 7.2.2 To ccrrecz the radiation dose rate for the case in v>hich reactor power level has noz rerained ccnstant during the 30 days prior to the reactor shu dc;>n engineerina, judcement is used to determine the most representative power level. The following guidelines should be considered in the determination. 7.2.2:1 The aver:ge pcwer durino the 30 day tir.".e period is not necessarily the most re r senzazive value =or correction to equilibrium conditions. 7.2.2.2 The last power levels at which the reactor operated should weigh more heavilv in the judgement than the earlier levels. F-6
7.2.2.3 ontinued opera- on ior an extended pericd shculd>>eich;..ore hear',1 in the juCaezent than "rief transient levels. 7.2.3 In the case in which reactor has produced power for less than 0 davs the procedure may be employed. Hcwever, the estirate of core darace obtained uncer this condition may be an under pr diction c-. the actual ccndition. 7.2.4 The follcwing equation is applied to det rmine the radiation dose rate corresponding to equilibrivm ull pc>>er source inventory corditions. Equilibrivm Heasured 1CO Dose Rate Dose Rate Reactor r'ewer Levei The reac.or pcwer level and the resulting dose rate ccrrection factor used above will be the sare for all svcsequent reasuremen:s oi the Cose rate ~ Record these values to recuc the work requirec to evaluate the subsequent measure..ents. 7.3 The decay correc .icn icr the radiati on dose ra te requi res the determination oi the tire duration "etween the reac:cr shutCcwn and the measure...ent of the Cose rat . This is Ccrc sircly using the ti'e o reactor shutdcwn recorded in 'Secticn 7. 1.3. The conclusion on the extent of core damage is made usira the equilibrium dose rate, the duraticn of reactor shutdcwn, and the analytically determired dose rates provided in Enclosure 2. The eauiiibr; m dose rate is plot:ed cn Enclosure 2 as a iunc ion oi t'.';,.e ioilcwira reac:or shutccwn. Encireerinc juc e-...ent is used deter.-. re which ca eccry of core ca-..ace shcwn cn ~nclcsvre 2 is ;.."s: representat ; e of he particular value that has been plotted. The follcwing criteria shculd be considered in the determinat',on. 7.4.1 Ccse rate measure..ents may have been recorced during periods of tr nsient ccrditicns within the plant. l'easureren s rade Curino st ale plant condit ons should weigh r,ore heavily in the assess-,ent, o core da;..ace. 7e~o2 Dose rates sicnificantly above the lower bourd for the category of major fuel overheat may irdicate concurrent fuel pellet melting. This prccedvre ;..ay rot be employed to es ima o the degr. e of fuel pellet relting. 7.4.3 Dose rates within anv category of fvel overheating may be anticipated to incluCe concurrent fuel cladding failure. This procedure ray rct "e vsed to distinguish the relat ve ccntribu ions of the two categories to the total. dose rate. The procedure does give the estimate of the hiahest cateaory of damage. 7.4.4 Dose rates corresponding to the two categories of major cladding failure and initial fuel overheat are observed to overlap on Enclosure 2. The evaluation of other plant parameters may be required to distincuish between them. However, concurrent ccnaitions may be antic'pated. f-7
0 Enclosnre 1 Badiolo ic Characteristics of tlAC Cateqories of Fuel Dama e ttRC Category of t1echanism of Source of Percent of Source Inventory Distribution of Fission Fuel Oaina e Release Frown Core 1/elease Aeleased tn Containment Pro~lucts i>> Containment
- 1. tto Fuel Damage llalogen Spiking Gas Gap Less than 1 Airborne Tramp llranium
- 2. Initial Cladding Gas Gap Less than 10 Airborne Failure
- 3. Intermediate Clad Burst and Gas Gap 10 to 50 Airborne Cladding Failure Gas Gap Diffusion t)elease
- 4. tlajor Cladding Gas Gap Greater than 50 Airborne Failure Initial Fuel Pellet Fuel Pellet Less than 10 Overheating Airborne:
~ Grain Boundary 100K tloble Gas Diffusion 25K llalogen
- 6. Intermediate Fuel Fuel Pellet 10 to 50 Pellet Overheating Plated Out 25K lla)ogen IX Solids
- 7. t1a j or Fuel Pe 1 le t Oiffusional Aelease Fuel Pellet Grea ter than 50 Pellet Overheating From U02 Grains
0 EiVCLOSURE 2 TYPICAL AiVALYSIS FOR POST ACCIDEiVT DOSE RAT liVSID A CY L liVDR ICAL COiVTAli >el
~ 'T 1x106 7Q 1x10~ O~ '( +( 7p C/l Og C
p7 g( 7~ 0 0@ L~ Vp .
+( +~
7Q 7g
~o +g P~
1x10" C'~ 7/( C 7/( ( C@ (~ oo 7/( C~ 1x10 1 10 100 Tli'<1E POST ACCIDEi JT, HOURS CYLIiVDRICAI COiVTAliVA'1EVT F-9
APP~'iQIX F 1 e.X~llPL= US"" OF THe. PROC"=QUPe. F-10
'g
the C 01 Iowirc is an exa.-..pie of t,",e us~ of this prcceCure for assessment co e Ca
~ ~ age. The Cata rec" reed cn plan-. c"ndi tion is as fol 1cws:
Contair.;..ent 2uil dino: 5 l'ea sure."..en,"u;,ber 1 Dose Pate = 1 x 10 Rads/hr Tir.:e = 0300 on 12/25/82 measure.-..ent t(u,"ber 2 Dose Rate = 5 x 10 Rads/hr Tire = 0600 on 12/25/82 4 lg~
'easure.-..ent llurber 3 Dose Rate = 1.5 x 10 Rads/hr Tir.e = 0100 on 12/25/"-2 3
4'.easurer..ent flu."..ber 4 Dose Rate = 4 x 10 Rads/hr Tire = 0100 on 12/31/82 Prior 30 Cav power history: Power, Pere. n: Dura.ion, Days 75 22 r,0 17 100 2 Tire of reactor shutdown: 01CO on 12/25/S2 Step 7.2 As shc'>>n in the recorded Cata, the reac or po>>er has not rerained ccnstant for the 30 days prior to the accident. Therefore engineering judger.".ent is used to deter.".,ine the "ower level e.",ployed in the assess;.,ent of core Car.:age. The cr -erie sta:- d in Step 7.2.2.2 ard 7.2.2.3 are used in 'he Ce e ~ inat,on. Tr. value selecteo is "0 percent. Tnis value is selec=. d "ecause durina the 17 days at that level ",.any f i s s i on procuc ts reach equi ibriun inventory in the core. 1 Ourirg:he 2 Cay i-..e period at 100 percent power, the inventory of al l f i s si cn products increased. The short lived fission pr'cucts;.,ay even have increased to tl eir equilibriu~ corres-cno'.ng to 100 "ercent power. However, rost of the isotc"es with loncer half lives, those greater than 1 day, re"...ain at in:ent"ries closer to that corresponding to equilibri..".. at "0 percen-. power. Also us ng 50 percent power would sc;..ewhat ur Cerest,"ate the fission product source inventory and the resulting core Car..age assess;..ent would therefore be conservative. Step 7.2.4 Usino =0 percent power, the ull power equilibrium dose rate for r,.easurerent nur.:ber 1 is as follows: Equilibrium 0o 5 e R e
= 1 x 1 0 5
x 100 0
= 2 x 10 5
P. a / r d h the re...aining full power equilibrium dose rates are:
~".easure.-.ent U" el 2 = 1 x 10'ad/hr A 3 = 3 x 10" Rad/hr ".easure...ent .'lux,."er 4 = B x 10" The t re duraticn between reac:cr shutdcwn and the reasure."..en of the dose rate =or each case is: i',easuret"..ent timur.ber 1 = 2 hours ~>easurerent i)u-.her 2 = 5 hours t'.easuret;.ent hur;.ber 3 = 24 hours Heasuren.ent Hu."..her 4 = 144 hours Assess;.ent of Core Oa~age ihe eouilibrius full power dose rates andri the duration of reactcr shutdown a. plctted on the following co"yr oi Enclosure 2. The ccrclusion is that core dartace is in the category of Initial;uel Overneat. Because of the ex.ent oi this darace, concurrent fuel cladding ailure is anticipated although it is not exressly distir guished frcn the total dar.:age by this prccecure. F-12
( l ~ ~ lh/ ENCLOSUR = 2 TYPICAI ANALYSIS FOR POST ACCIDENT DOSE RATE I."'S DE A CYLINDRICALCONTAIN'.1E~JT 1x10~ O~ vp 1xl0 C
+(
0 it+
/n / C~
( O~
'7 + /n vg ~ 0<
Pp ( P~
~ogQMBLl H EKG!NE P~
7/( C'~ 1x1 0" C~ O~
/ "Pg
( (r~
",Fg 0/n/
9/( G~ 1xlG 10 100 Tl>'IE POST ACCIDEi4T, HOURS CY LIP'DR ICAL CONTAINMENT}}