ML20093H198

Preliminary Review & Evaluation of Millstone-3 Probabilistic Safety Study

ML20093H198
Person / Time
Site:	Millstone
Issue date:	12/19/1983
From:	Khatibrahbar, Ludewig H, Pratt W BROOKHAVEN NATIONAL LABORATORY
To:	NRC
Shared Package
ML20093H124	List: ML20093H138 ML20093H170 ML20093H184 ML20093H198
References
NUDOCS 8410160199
Download: ML20093H198 (56)

Text

c a . o

,a

. ,, e ( -

( _

BNL-NUREG.

D INFORMAL REPORT LIMITED DISTRIBUTION PRELIMINARY REVIEW AND EVALUATION OF THE MILLSTONE-3 PROBABILISTIC SAFETY STUDY M. Khatib-Ranbar, H. Ludowig , and W. T. Pratt Accident Analysis Group Department of Nuclear Energy Brookhaven National laboratory Upton, New York 11973 December 19, 1983 Prepared for U.S. Nuclear Regulatory Conmission Wasnington, D. C. 20555 Contract No. OE-AC02-76CH00016 FIN No. 3748 0410g[w0 p PDR .

b k_

_ .- _ .___ . - - - _ . . - . .~

1 .. - -

( .

ABSTRACT A preliminary Level 1 review of tne containment failure modes and con-sequence analysis in the Millstone-3 Prooabilistic Safety Study (MPSS) is pre-sented. The review identifies the major features of the plant as they relate to risk assessment, including comparisons to tne Zion and Indian Point stud-ies. Future plans and a list of preliminary questions is also included.

t iii I

u.

,  ! ( -

• ~

, CONTENTS Page ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii LIST OF FIGURES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii

1.0 INTRODUCTION

. . . . . . . . . . . . . . . . . . . . . . . . . . . . I 1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Organization of the Report .................. 2 2.0 P LANT DESIGN AND FEATURES . . . . . . . . . . . . . . . . . . . . . 3 2.1 Pl a n t De s i g n . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.2 Comparison to Zion and Indian Point Plant Designs. . . . . . . 5 3.0 PRA REVIEW. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 3.1 Anal yti c al Method s . . . . . . . . . . . . . . . . . . . . . . 11 3.2 Containment Event Tree and Accident Phenomenolo 12 3.3 Containment Matrix . . . . . . . . . . . . . ......... gy. . . . . . . 13 3.3.1 Release Category M1A. . . . . . . . . . . . . . . . . . 15

'3.3.2 Rel ease Category M1B. . . . . . . . . . . . . . . . . . 15 3.3.3 Release Category M2 . . . . . . . . . . . . . . . . . . 15 3.3.4 Release Categories M3 and M4. . . . . . . . . . . . . . 15 3.3.5 Release Category M5 . . . . . . . . . . . . . . . . . . 15

3.3.6 Release Category M6 . . . . . . . . . . . . . . . . . . 15 3.3.7 Re l e a s e Ca t eg o ry M7 . . . . . . . . . . . . . . . . . . 15 3.3.8 Release Category M8 . . . . . . . . . . . . . . . . . . 15 3.3.9 Rel e a se Category M9 . . . . . . . . . . . . . . . . . . 16 3.3.10 Release Categories M10 aad Mll. . . . . . . . . . . . . 16 3.4 Ex t e rn a l Ev e n t s . . . . . . . . . . . . . . . . . . . . . . . . 16 3.5 Uncertainty in the 'C'-Matrix. . . .............. 17 t

3.6 Acc i den t Sou rc e Te ra s . . . . . . . . . . . . . . . . . . . . . 18 3.7 Of f-site Consequence Analysi s. . . . . . . . . . . . . . . . . 23 3.7.1 Off-Site Consequence Analysi s . . . . . . . . . . . . . 23 3.7.2 Plume Characteristics and Uncertainty . ...... .. 24 2.0 5 '.'.* "< A R Y . ............................. 45 t

O v

l. (. -

CONTENTS (Cont.)

l 1

Page 4.1 Results of Level 1 Review. . . . . . . . . . . . . . . . . . . . 45 4.2 Sugge sted Future Wo rk. . . . . . . . . . . . . . . . . . . . . 46 4.3 Additional Information Needs . . . . . . . . . . . . . . . . . 46 4.3.1 Analytical Model s and Phenomenology . . . . . . . . . . - 46 4.3.2 Uncertainty Analysis. . . . . . . . . . . . . . . . . . 46 4.3.3 Source Term and Site Model . . . . . . . . . . . . . . . 47 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 1

e s

O vi ,

_ . - - - . . = . - , . - ....._..a..

t. (,

\ -

- LIST OF FIGURES Figure Title Page 2.1 Schematics of the containment cooling system in the Millstone-3 . . . . . . . . . . . . . . . . . . . . . . . . 6 2.2 Schematics of the lower reactor cavity in the Millstone-3 . 7 3.1 The MPSS computational approach . ... ......... . 26

(

O vii

~ ' * ' ' ~ *

' ->w - - , , _ _ , _, _

. 5

,e

,  !. (

\

. LIST OF TABLES Table Title Page 2.1 Comparison of design characteristics . . . . . . . . . . . . 8 3.1 Summa ry of computational tool s . . . . . . . . . . . . . . . 27-3.2 Summary of containnent event tree time frames and nodal questions. . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.3 Assessment of MPSS event tree nodal questions. . . . . . . . 29 3.4 Notation and definitions for release categories. . . . . . . 30 3.5 C-matrix (internal initiating events). . . . . . . . . . . . 31 3.6 Notations and definitions for plant states (internal). . . . 32 3.7 Simpli fied ' C'-matrix for MPSS . . . . . . . . . . . . . . . 33 3.8 Simplified ZPSS containment matrix 'C' . . . . . . . . . . . 34 3.9 Simplified BNL containment matrix 'C' for ZPSS . . . . . . . 35 3.10 Release category summary . . . . ... . . . . . . . . . . . . 36 3.11 Interfacing LOCA - V sequence. . . . . . . . . . . . . . . . 37 3.12 Early overpressurization release . . . . . . . . . . . . . . 38 3.13 I sol a ti on f a il u re . . . . . . . . . . . . . . . . . . . . . . 39 3.14 Intermediate and late overpressurization failure . . . . . . 40 3.15 No f a i l . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 3.16 Summary of evacuation schemes and their probabilities. . . . 42 3.17 Subjective discrete probability distribution for site consequence uncertainty evaluation . . . . . . . . . . . . . 43 3.18 List of DPD runs perforned ................. la l j O

viii

s. t

( .

t s _

. 1. INTRODUCTION

1.1 Background

, After the accident at Three Mile Island, the Nuclear Regulatory Commission (NRC) recognized the need to reexamine the capabilities of nuclear power plants to accommodate the effects of hypothetical severe accidents oeyond the design basis. This reexamination included consid' e ration of potential design modifications to mitigate the consequences of these degraded and core melt accidents.

The Zion and Indian Point power plants were chosen to initiate this ac-tivity because of the large populations surrouncing tne two sites. The . con-cern aas :nat due to tne :roximity of these two sites to high population den-sities, tney could comprise a disproportionately high component of the total societal risk from U.S. commercial nuclear power programs.

As part of this continuing effort, programs to evaluate the risk from

. plant sites situatea near high population centers have been set in motion, in order to introduce design modifications and mitigation features, which can reduce the public risk.

1 Probabilistic Risk Assessment (PRA) studies have been undertaken by a number of utilities [1-33 and reviewed ay Brookhaven National Laboratory (BNL) under contract to the NRC. BNL was also actively involved in prepara-tion of a preliminary report [4] (NUREG-0350) whith represented tha staff's initial contribution to tne understanding of severe accident progression and mitigation.

~

This report is a creliminary evaluatian of the containment failure modes and consaquence analysis af tne Millstan: Lrii t 3 Procabilistic Safety Study (MPSS) completed by hortneast Utilities in August 1983.[5]

c.

1

. - - - ~ , , , , , , , - , - , . - , - r,e, , , - -

'( ( .

\..

1.2 Objectives

~

The objectives of this review report are to provide the NRC staff with a preliminary (Level 1) review of the MPSS as part of a broader objective in-volving an in depth review and evaluation of the technical basis for the sub-ject PRA, which will be performed in the coming year. In particular, core

~

melt phenomenology, containment response, containment event trees, release categories, and site consequence models are to be examined.

This Level 1 review, which was performed over only four weeks duration, highlights important features of the plant design and the NPSS as compared to the PRAs of the Zion [13 and Indian Point [3] facilities.- The report also provides an initial assess.aent of the PRA method, validity of major assump-tions, and relevance and adequacy of conclusions.

Areas needing further verification and study are identified, and finally, questions for the applicant or Licensee pertaining to the Millstone Unit 3 are addressed.

1.3 Organization of the Report A brief review of the Millstone-3 design and features is presented in Chapter 2 along with comparisons to Zion and Indian Point Plant designs.

Chapter 3 contains the preliminary assessment of the Mill stone PRA.

Specifically, analytical metnods, containment event trees, accident phenome-nology, containment matrix, uncertainty analysis, accident source terms, and off-site consequences are reviewed.

In Chapter 4 tne results of this preliminary Level 1 review are summa-rized and areas needing further study are also highlighted along with need for acditional information and questions to tne applicant or tne Licensee.

O 2

= , - - -

'. ' , ,6 g

_ l

\ _ _ _ .

2. PLANT DESIGN AND FEATURES In this section, tnose plant design features tnat may be important to an assessment of degraded core and containment analysis are reviewed. These im-portant features are then compared with the Zion and Indian Point facilities in order to identify commonalities for benchmark comparisons.

2.1 Plant Design Millstone-3 is a four-loop Pressurized Water Reactor (PWR). The core and reactor coolant systems are of the standard Westinghouse design, wnile the major calance of plant systems and the containment design are of 5 tone and Webster design.[5]

Major characteristics of tne plant are a 3411 MWt (1150 Mue) core power reactor employing tne Westingnouse 17 x 17 core design. The reactor coolant system is a four-loop configuration with U-tube recirculating steam genera-tors. The emergency core cooling system consists of 4 accumulators containing 6358 gallons of water each, which are designed to discharge when the reactor coolant system pressure falls below 600 psia, a safety injection system which draws water from a 1.2 million gallon refueling storage tank and is delivered to the reactor coolant system via either the charging pumps, high head safety injection pumps or low head safety injection pumps. The long-term core cool-ing is handled via a compl etely independent recirculation cooling system (whose major components are shared with the recirculation spray system) which consists of four (4) pumps and four (4) neat excnangers which are cooled by the service water system.

The auxiliary feedwater system also provides a core cooling function by remoeing neat fro.n :ne .ES af ter reactor snutoonn via the steaia generators.

This sjstem, anicn consists of two (2) 5: percent enotor driven pumps and one 3

_- - . . - ~

.. s .

$' ( (. -

(1) 100 percent turbine driven pump takes suction from the condensate storage tank. .

Containment cooling following an accident which initiates the ESF signal is accomplished via two completed independent spray systems. The quench spray system draws from the refueling water storage tank while the recirculation spray system draws from the containment sump (see Figure 2.1). Together, the systems are designed to reduce the pressure in the containment to a subatmos-pheric condition (normal operating state) within approximately one hour for design basis accident sequences.

The containment geometry design in the area underneath and around the reactor vessel precludes water from entering the reactor cavity area until a major portion of the Refueling Water Storage Tank (RWST) has been exnausted via tne quench spray system (see Figure 2.2). This is referred to as a dry-cavity configuration. The same geometry is expected to preclude the disper-sion of core debris from the reactor cavity to the general containment area following postulated failure of the reactor vessel during core melt sequences.

The cavity area geometry also would preclude the establisnment of effective convective air currents between the cavity and general containment area for

! heat removal of core debris in the reactor cavity area. The containment de-sign also includes a permanent seal ring between the reactor vessel flange and the biological shield walls, which would prevent introduction of water into the reactor cavity from either break flow or spray flow in the area of the reactor vessel or tne refuelir.g cavity. The containment building basemat and the internal concrete -structures are composed of basaltic-based concrete.

l C

4

= C

~

\ .

- . 2.2 Comparison to Zion and Indian Point Plant Designs l

. Table 2.1 sets forth the de' sign characteristics of. the Zion (Units 1 or '

' ~

. 2) and the Indian Point (Unit 2) facilities as they compare to the tiillstone Unit 3 plant.

It is seen that the -three plants are quite similar in containment build-ing and primary system design while they differ markedly _ in containment cool-ing mechanisms and lower reactor cavity configuration and chemical composi-1 tions of the concrete mix.

As concrete is heated, water vapor and other _ gases are released. The-initial gas release consists largely of carbon dioxide, the quantity of wnicn depends on the. amount of calcium carbonate in the concrete mix..

~

Li:aestone concrete can contain up to 80% calcium carbonate by weight, whicn could yield up to 53 lb of carbon dioxide per cubic foot 1of concrete. However, basalt-based concrete contains very little calcium caroonate and would not release a significant amount of carbon dioxide.[4]

These innerent design differences a're expected to ' alter the course of the accident sequences; in particular, following failure of the reactor vessel, where the containment pressurization is significantly influenced by the debris bed coolability and water availability.

The absence of fan coolers in the Millstone plant can also effect the ac.

cident progression and containment pressurization effects.

2

?

5 3

\

I

.~

t  :

( ._

~  :

- - . . l ENCLOSURE BUILCt NG OUENCH SPRAYS (2)

I f T.* I' aGud) _

I RWST d d c. a i 4

1 SLCRS I ll lSkRVICE WATER g] . RECIRCULATION

.. .SFRAYS (4)

To I i i STICK i SUMP -

I Figure 2.1 Schematics of the containment cooling systems in the Millstone-3 c

i 6 .

...a c- .( .

n.

i a

4 9

o e

M A NWAY . fj p*

% il [j f$

i Sh; Hivessci.3

.- L- +,.

o, ,e i.i.e.

.,3Q,9):

.a #

W/////ffffff/ff/ffMffffffn//ffh .

1 i

i Figure 2.2 Schematics of the lower reactor cavity in the Millstone-3 O

O kk 7

, _ , . , , , . - - . , _ , - - - . .-r - ,-., , . , .. -e .

-a ~~. - -.

0:  ;

[. . * - ({ '

- 1

~~ '

- Table 2.1 Comparison of design characteristics Zion Design Parameters U0i t Indian Point Mill stgne IL4J Unit 2[3,4] Unit 3L5]

Reactor Power [MW(t)] 3250 3030 3411 C0tlTAINMENT BUILDING:

Free Volume (ft 3) 2.72x106 2.61x106 2.3x106 Design Pressure (psia) 62 62 59.7 Initial Pressure (psia) IS 14.7 12.7/9.1 Initial Temperature (*F) 120 120 120/80 PRIMARY SYSTEM:

Water Volume (ft 3) 12,710 11,347 11,671 Steam Volume (ft 3) 720 720  ?

Mass of UO2 (1b) 216600 216600 222739 in Core Mass of Steel (lb) 21,000 20,407  ?

in . ara Mass of Zr in (lb) 44,500 44,600 45,296 Core Mass of Bottom (10) 87,000 78,130 87,000 Head Bottom Head (ft) 14.4 14.7 14.4 Di amete r  :

Bottor Head (ft) 0.45 0.44 0.45 I inic< ness 1

e 3

p7 ---y , - , < - - - - - -

, -er-g

l

.- t .

l Table 2.1 Comparison of design characteristics (Continued)

Zion Design Parameters Unit Indian Point Mill stone 1 Unit 2 Unit 3 s

RESIDUAL HEAT REf40 VAL EXCHANGERS (HX):

Yutal Rated (Stu/hr)- 5.6x107 6.16x107 7.05x107 Capacity (2HX) (2HX) (2HX)

Total Primary (lb/hr) 3.9x136 2.88x106 1. '38x 106 Fl ow Total Secondary (lb/hr) 4.96x106 4.92x106 3.3x106 Flow Primay Inlet (*F) 137.5 135 120 Temperature Secondary Inlet (*F) 107.1 88.3 92.2 Temperature CONTAINMENT BUILDING COOLERS:

System Fans Fans Sprays Number 5 5 f 2 Quench '

I (2 Recirculation /

ACCUMULATOR TANKS:

Total Mass of (lb) 200,C30 173,000 348,000

• ia

.- te r Ini-ial Pressure (psia) 565 665 665 l Temperature ( F) 150 150 80 l

l 9

, - - - - .-, , , . .,,.-.n -

~

- ..__.___..__._____._n. _

.:' t ( .

Table 2.1 Comparison of design characteristics (Continued)

Zion Design Parameters Unit Indian Point Mill stone 1 Unit 2 Unit 3 REFUELING WATER STORAGE TANK:

Total Mass of (lb) 2.89x106 2.89x106 107 Water Initial Pressure (psia) 14.7 14.7 12.7/9.1

. Temperature (*F) 100 120 50/40 REACTOR CAVITY:

Design Dry Dry Dry Concrete Material Limestone Basaltic Basaltic l

O 10 ,

. . i (

3. PRA REVIEW In' this section a brief review cf the Millstone Unit 3 Probabilistic-Safety Study (MPSS) is presented. Specifically, the computer codes .and cal- i culational metnods used to carry-out the degraded core and containment re-sponse analyses are identified. Where possible, parallels between tnis study and other existing PRA studies are set forth. Finally, the relevance and validity of the conclusions is addressed.

3.1 Analytical Methods k

A brief description of the computer codes used to perform tne transient degraded core and containaent response analyses is provided in tnis section.

Table 3.1 summarizes the code package as applied to various pnases of tne accident. It is seen that the MARCH code is used to model the core and pri-mary system behavior and to obtain the steam and water energy releases for (1) the entire transient in tne case of non-dispersal accident events and (2) un-til the vessel failure for the dispersal scenarios. These mass and energy re-leases form the input for the other computer codes used to evaluate the con-tainment response for the non-dispersal cases (see also Figure 3.1).

For sequence classes in which the reactor coolant system remains at an elevated pressure until the vessel failure (dispersal cases), the MODMESH code is used. Tnis code calculates the steam and hydrogen blowdown from the reac-tor vessel using an isothermal ideal gas model. The water boil-off from the i

4 reactor cavity floor is modeled using a saturated critical heat flux correla-tion. Additionally, the accumulator discharge following depressurization caused oy tne vessel failure is also considered.

For tne non-ccolacle cecris cea anc ccre-ccncrete interaction, tne INTER subroutine of MARCH is replaced by tne CORCOM-A001 code, modified by Westing-house.

11 d

- - , - - - , , . - . - - . e - - - - , - - .~. _----,e c- .

1

~

' -i ( .

~

The output from MARCH or CORCON is used es input, after preprocessing by M00 MESH, to the C0C0 CLASS 9 code. The C0C0 CLASS 9 code replaces the JiACE sub-routine of tne MARCH code. In C0C0 CLASS 9 code, the containment steam / water:

noncondensibles, and the sump water are modeled by' a single volume. The code also includes a structural heat transfer model, hydrogen combustion, and ca-pability for containment heat removal through containment sprays and sump re-

~

circulation systems, as described in Section 4.3.2 and Appendix 4-E of the MPSS report.[5]

Fission product transport and consequence calculations are performed using CORRAL-II and CRAC-2 computer codes, respectively. (See Section 3.6_ and 3.7 for more details.)

This preliminary review of the approach used in the MPSS for quantifica-tion of core and containment response is directed to a review of the consis-tency of the approach. However, detailed verification of the results obtained in the MPSS cannot be made at this stage, and is tnus deferred to a later date.

3.2 Containment Event Tree and Accident Phenomenology An important step towards the development of the containment matrix in-volves the quanti fic.ation of branch point probabilities in the containment event trees. The probabilities depend heavily on the analyses of degraded core pnenomenology and the containment building response described in Sections 4.2 through 4.7 of the MPSS.[63 In tne MPSS[5], tne containment event tree is divided into six distinct ti.ne frames, anich represent the time phases during an accident event in which potential containment failure is considereo. Table 3.2 summarizes the six time frames along witn tne corresponding containment event tree nocal ques-tions, as reproduced from the MPSS.[5]

9 12 l

,,,~g - - - - - . . - , . , , . , - , , . . - , . . , . , , - --,

t .(

,. l j

Detailed assessment of the nodal questions and the assigned probabilities cannot be made at this time; however, a preliminary evaluation of .the nodai l questions as they relate to Zion (ZPSSE13) and NUREG-0850[43 is _ made in Tabl e 3.3.

A node by node comparison between the MPSS and ZPSS is not possible be-cause of the differences in the plant designs and containment event tree structures. However, in arriving at nodal probabilities, significant credit has been taken for:

1. Core-wide incoherencies during meltdown progression as attributed to the recent TMI-2 neat-up calculations, also identified for ZPSS.E13

2. Reduced energetics, as a result of in-vessel core debris-water inter-action, leads to low probability events early in the accident.

3. Successful quenching of debris bed as a result of high pressure dis-charges following vessel failure.

Therefore , it is essential to review tne event tree structure, and its associated nodal questions and quantifications as they affect the overall risk before a complete assessment could be made.

3.3 Containment Matrix (C-Matrix)

The sixteen nodes in the Millstone-3 containment event trees were out-lined in the previous section. A negative response at any of seven nodes (CII, CI2, CI3, CI4, CIS, CI6, and BM6) in the containment event trees result in failure of tne containment building Dy a variety of failure modes. Eacn of these failure modes results in a particular radiological release category.

For those paths that do not have a negative response at any of the seven noces, tne patn will eventually result in no failure of the containment. The containment event trees, tnerefore, link camage states to a range of possible 13

- . , . ~ s. - -- ,,

( . ,

containment failure modes via the various paths through the tree.

For a given

-tree, each path ends in a conditional probability (CP) of occurrence and these cps should sum to unity. The quantification of an event tree is the process by which all the paths are combined to give the conditional probabilities of l the various release categories. In the MPSS, thirteen release categories were used for the quantification process as summarized in Table 3.4. Note that one of these release categories (namely, M12) correspond to no containment fail-ure. Fission product release for this category would, therefore, be via nor-mal leakage paths in the containment building.

The quantification of the MPSS containment event trees was a significant task, and it was necessary to use a computer code, ARBRE, to group the various ,

path probabilities into the thirteen release categories.[5] However, the containment matrix 'C' is a concise summary of the quantification process.

Taol e 3.5 is a reproduction of the 'C' matrix for the MPSS.[5] It lists tne conditional probabilities of the release categories given the plant damage state; with the plant damage states defined in Table 3.6.

A simplification to the C-Matrix is ootained in Table 3.7 by disregarding all of the very low probability values. This simplification is not expected to influence the risk calculations as discussed in [6].

The comparable simplified C-Matrix for ZPSS is reproduced in Table 3.8 while Table 3.9 lists the ZPSS C-Matrix as determined by BNL calcula-tions.[43 Due to the distinct differences in plant designs and progression of the accicents, an exact correspondence in release categories of ZPSS and MPSS can-not e :nade. However, similarities in tne release categories are identifiea in Taoles 3.3 ana 3.9.

14

. _ _ , . _ _ . _ _ __,.-_.__m.__. _ _ . - _- _r,,_--y, , _-_ _ .

~

. .. l c' t

( ,

3.3.1 Release Category M1A The conditional probability and plant damage states are identical in the three C-matrices. .

3.3.2 Release-Category M1B Unique to MPSS and not identified as a failure mode in ZPSS.

3.3.3 Release Category M2 This release category was identified in ZPSS, bt.'t neither the BNL study nor MPSS results seem to indicate M2 as a significant contributor.

3.3.4 Release Categories M3 and M4 These failure modes were found to se insignificant for both plants.

3.3.5. Release Category M5 Given the plant damage state SL, the probability of this release category is calculated to be about 0.01 in MPSS and it was found to be insignificant in the other studies.

3.3.6 Release Category M6 C This category is found to be only significant for MPSS.

3.3.7 Release Category M7 This release category applies to plant damage states with insufficient or no containment heat removal systems operating. The relatively smaller proba.

I bilities calculated for plant damage states AE, AL, SE, and SL are a'ssociated with tne difference in cavity concrete structure. In Zion, the limestone con-crete with high calcium carbonate content causas high CO2 /C0 releases, and thus nigner containment pressures.

3.3.3 Release Category PS inis fail;re mode aas founc to ce insignificant in ZPSS, MPSS, and tne BNL study. donever, in tne BHL stuoy for Zion,[6] this release category was C

15 O

+- -

- found to influence the overall risk calculations, and thus it needs to be -

further assessed for MPSS.

3.3.9 1 Release Category M9. .

This failure mode applies to plant damage states AEC' and SEC' which are LOCA's with early core melt in the absence of recirculation spray system in MPSS; however, this was not found to be a _significant contributor in the Zion plant primarily, due to smaller amounts of the wcter available leading to less steam generation and overpressurization.

3.3.10 Release Categories M10 and M11 This failure mode could potentially impact all damage states. For plant states without containment heat removal and early core melt (namely, loss-of .

ECC injection), there would be limited water in the reactor cavity and thus a potential for basemat penetration. However, containment failure would occur prior to basemat penetration and thus a higher probaoility is ass'ociated for release category M7 for three damage states. All other damage categories wMl have significant quantities of water in the reactor cavity.

It must be noted tnat impact of basemat penetration on risk is believed to be negligible and thus this failure mode can be neglected.

~

In general, it i s found that containment integrity in Millstone Unit 3 can be assured only if both the containment recirculation spray and quench spray systems are available. Of the two, however, the long-term heat removal capability of recirculation spray system is more important. In all instances, i hydrogen generation by molten-core-concrete-interaction and likelihood of hy-drogen burns were found to be nign.[5]

3.4 External Events In MPSS, containment response to accidents initiated by external events  !

(fires and seismic events) are also considered.

e 16

. - - - - -, .,_-y ,, r -__ --m--, m- , _ . . . - - _ _ ___y.y_.,,., ,, -_, ,-,,__m, - , - , , , , --.-#___.o.c.

. ma,-

, .: ' t.

( _

The external containment event trees make use of the same event tree structure -as is used for the internal initiating events. The impact of dif-ferences in event sequence course is accounted for in the assignment of the split fractions and uncertainty assignments.

Containment thermal response analyses were not performed for externally initiated events, but rather, engineering judgment has been used in the as-signment of each accident sequence to a particular release category as de-.

scribed in Section 4.7.5 of MPSS.[5]

3.5 Uncertainty in the C-Matrix The containment event tree quantification described in the earlier sec-tion was Dased on the assessment of point-estimate probabilities for the split fractions at various nodes of the containment event tree. In order to account for inherent uncertainties associated with phenomenological questions, the Di screte Probability Distribution (DPD) methodology was implemented in HPSS.[5]

The DPD is described in Section 4.7.4 and Appendix 4-N of the MPSS. The distributions are constructed based upon the following cri'terion:

1) Definition of a reasonable upper and lower range of the nodal prob-abilities which represent an upper and lower 95 percent one-sided confidence level .

2) A weighting factor for the point-estimate, upper and lower range values resulting in a three-interval DPD.

These DPD's are tnen propagated tnrougn tne dominant patns of the con-tainment event trees using DPD arithmetic.

l Tne MPSS results for tne containment uncertainty analysis snow tnat the '

range of uncertainty is low for the more probable sequences (higher point-1 estimates) as compared to tne less probable sequences. '

17

, -- --- g --y. - - g - , , -- - - , , --

. 4..

., ,., 4 3.6 Accident Source Terms In this section the approach utilized to determine the fraction of fis.

sion products originally in the core and leaked to the outside environment will be outlined. The fission product source to the environment as calculated by this approach will be compared with those for similar plants. The calcula-tions to be included in this comparison are those done for the Zion and Indian Point Probabilistic Risk Assessments, (ZPSS[1] and IPPSS[3], respec-tively), the Indian Point Study carried out for the NRC and presented as tes-timony at the Indian Point hearings (IPS)[7], and finally, releases deter-mined for the Surry plant using the metnods proposed oy tne Accident Source Term Program Office ( ASTPO). The first three calculations are Dased on the methods used in the Reactor Safety Study (RSS) and published as llASH-1400;[8] the last calculation is based on more mechanistic methods which form the basis of the revised source term arid is published as BMI-2104 Volume -

~

1.[93 8 As in the RSS, the CORRAL-II code is the most important tool for deter-mining the fission product leakage to the environment. This code takes input from the thermal-hydraulic analysis carried out for the containment atmos-phere. In addition, it needs the time dependent emission of fission products.

The fission product release is divided up into the customarily used phases, i.e., Ga p , Pel t , and Vaporization releases. The time dependence of these pnases is determined by the core heatup, primary system failure and core / con-l crete interaction times. In all, thirteen releases were determined ranging ,

from tne containment bypass sequence (V-sequence) to the no fail, sequence.

These sequences are summarized in Section 3.3 and tna results are snown on Table 3-10. j 18

-,,, -- -w - - - - -

-n-~

,. ,... ( { .

l Four of the thirteen releases outlined above are based on the RSS re-leases. M-1A and PWR-2, M-10 and PWR-6, and, M-11 and PWR-7 are all identical  :

in both fractional release and timing. The release M-18, which corresponds to I

a steam generator tube rupture,. is determined by dividing PWR-2 or M-1A by ten. Noble gases and organic iodine are not subject to this reduction in re-lease.

The energy of release for the overpressurization failures (M-2, M-3, M-4, M-5, and M-7) are high in comparison to the corresponding releases re.

ported in the RSS. The effect of a high plume energy is that the plume is lifted higner into tne atmosphere and in most cases the fission products are spread over a larger area. The net effect of this is that potentially more people will be exposed, however, the concentration of the dose will be lower.

Thus, any consequence which is characterized by a dose threshold could be lowered by increasing the plume energy. This general trend could be affected

, ~

very strongly by population distribution,e tihe protective action taken , and weatner.

Finally, a release path customarily included in Probabilistic Risk As-sessment is missing from the list. This is a release characterized by a steam explosion.

It is not clear how much it would contribute to risk. Steam explosions are very remote, and the early overpressurization failure release, M-2, has a similar release character as a steam explosion, except the Ruthe-nium (Ru) and Tell urium (Te) releases are low compared to the customarily assumed release fractions for a steam explosion.

In order to allow for uncertainties in the transport of fission products, a similar netn0d using Discreta Procability Distrioutions (UPD's) described earlier in Section 3.5 was used. This method assumes tnat the RSS metnod of c'

19

,. . - - ,.-_ .__-_y

~

{ .

determining fission product release fractions, has a relatively large uncer- '

tainty associated with it, and thus at major steps in the sequence probability distributions are assigned to the phenomena taking place. These distributions are combined to give a probability distribution for each release category.

There are three major phenomena in the transport of fission products from the core to the environment where there is considerable uncertainty surround-ing the attenuating mechanisms. These phenomena are:

a) Deposition and holdup of fission products in the primary system,.

b) Agglomeration and settling of aerosols in the containment building, and c) Attenuation of fission products as they pass out of the building into the environment.

Of the three phenomena mentioned above, only the first two are considered

~

in the MPSS.[5] -

In the MPSS, the major reldhses kre broken down into their contributing plant damage states, and distributions are assigned to the abovewnentioned phenomena as they occur in eacn damage state. These distributions are then combined to form a damage state distribution, which are further combined with otner damage state distributions to result in a OPD appropriate for a fission product release.

These distributions are a measure of the uncertainty of a particular release. They also give an indication of how different the RSS based release is from the best-estimate release.

Tables 3-11 tnrough 3-15 show comparisons between selected releases ccm-puted for tne MPSS, tne ZPSS,[13 IPSS,[33 IPS,[73 and those determinea for tne Ab?30.[93 ine first four analyses all used similar methods anc were 1

based on methods outlined in the RSS. The determinations for ASTP0 were basea a

20

_ - - - - ,,.-e.. .---m---

. .. _c w_

-s t

.. (. -

on more mechanistic methods and used an improved and extended data base.

Table .3-11 shows the release fractions, timing, and uncertainty ' vector and associated probabilities for tne interfacing LOCA (V-sequence). Comparing the first two columns (MPSS release and ASTP0 release), it is seen that be-

sides the noole gas release fraction, there is a substantial reduction in fis-sion product release when using the ASTP0 metnods. This is particularly true for the Rutnenium (Ru) group which shows a reduction of approximately 25; the Tellurium (Te) and Barium (Ba) groups are reduced by 6. The uncertainty mul-tiplier used in the MPSS peaks at a 50% reduction = in the release fractions.

Furtnermore, it has a 17% weignt for no reduction and 28". weight for a reduc-tion oy a factor of four. A comparison with the last column, whicn corres-ponds to the IPSE73 analysis shows that the fission product releases are comparable with the RSS determined release fractions. However, .the energy of the release i~s auch higher in tne HPSS case, the ratic being -40. ~

Table 3-12 snows 4 comparison between the MPSS, ASTP0, and ZPSS for the early overpressurization release. It is seen that the M-2 release, unicn cor-responds to the MPSS analysis, is equal to or lower than the ASTP0 release fractions for the volatile fission product groups (Xe-Kr-Te-Sb). However, for the less volatile fission products (Ba-Sr-La) the MPSS releases are higher than the ASTP0 releases. The uncartainty factor for this release is seen to imply a substantial reduction in fission product release fraction. -In com-parison to the ZPSS release fractions, very small differences are seen. Fur-thermore, tne uncertainty factors for the ZPSS also imply a reduction in re-lease, out to as large as for the MPSS. Finally, the energy of the release for tne *?53 is ccc: para:le to tne energy of release for the ZPSS.

l .\

l i l

21 i a

1

, -g= ..

_ .m

.( '

~'

Table 3-13 compares tne releases for an isolation failure. M-4, the re-lease determined for the' MPSS, is compared to two releases determined for the IPS.[7] The release fractions determined for the IPS corresponded, to leak-age through openings 8" and A" in diameter, while the M-4 release corresponds to leakage through a 6" diameter hole. It is seen that the M-4 release is comparable to that corresponding to the 8" diameter hole determined for the

+

IPS. The uncertainty multiplier for M-4 is seen to peak at a 50". reduction in the release fractions.

Table 3-14 shows a comparison between MPSS releases M-5 and M-7 with com-paraole releases determined for the ZPSS,[13 IPPSS,[3] rps,[7] and tne ASTPO.[93 These releases all correspond to intermediate or late (8 hr-20 hr) overpressurization failures. It is seen that the fractional releases com-puted for M-5, M-7, ZPSS, IPSS, and IPS are all similar. The largest discrep-ancy occurring for iodine, which is much- higherf for the ZPSS than all the

~

others. Heever, in comparing these releases with- the ASTP0 determined re-leases, it is seen that the ASTP0 releises are substantially lower, except for noble gases which are comparable. This large reduction is attributable to tne improved modeling of attenuating mechanisms in the ASTP0 method. It is seen from the uncertainty vector that a large weight is placed on reducing the source terms, particularly for M-5 and M-7. There is also a large difference in the energy of the released plume for the MPSS results and the remaining calculations. The MPSS energy being four times higher than the IPS result and approximately twenty five times higher than the ZPSS and IPPSS results.

Table 3-15 shows a comparison between the M-12 release from the MPSS, tne no fail release from the IPS[7] anc tne no fail or SB release used in tne ZPSS[1] and the IPPSS.[3] These sequences are all based on a leak rate of '

l 22

. ~ ~n ,_e

' (. (. .

.17,/ day. 'It is seen that the M-12 release fractions are comparable to the ZPSS and IPPSS release fractions. However, they are all substantially lower in comparison to the IPS release fractions. . Since all the fission product re-lease fractions are quite low for this sequence, the differences between them, measured in terms of consequences, will be small.

3.7 Off-Site Consequence Andlysis The off-site consequences due to airborne fission . products were deter-mined using the CRAC2 code. The output of these calculations is in tne form of conditional cumulative probability distribution functions (CCDF's) which form the basis of the S-matrix. The S-matrix is used to determine tne overall risk of the plant. The CRAC2 code requires input for site specific data (pop-ulation distribution, economic parameters, topographical), nealtn data (dose conversion factors for latent ar.d early consequences), meteorological data

~

(wind rose, wind speed, rain), plume cnaractefistics (isotopic content, pnysi- ,

~

cel description), and population response ~(evacuation parameters). Of all the data which is outlined above, we will discuss only the plume characterization, as it affects the uncertainty of the analysis through the fission product re-lease fraction multipliers discussed above, and the evacuation model input.

3.7.1 Evacuation Model A summary of the evacuation schemes is shown on Table 3-16. Tha evac-uation schemes used in this study are divided into two categories, i.e, gen-eral and seismic. The general scheme is represented by two schemes, depending on the weather. Normal weather is expected to occur 88% of the time and ad-verse weatner occurs the remaining 12%. Thus, these two scnemes are weignted l

in tne ratio of .88: .12. Tne normal evacuation scheme allows for a speed of i

10 mpn and a delay time of .92 hr, unile the one corresponding to adverse i

t l C>

l 23 l

ry . .; _ _- n n

( .

weather reduces the speed to 7.5 mph. In comparison, tne parameters used in '~

the IPS[73 for these parameters .are 1.5 mph and 2 hr, respectively. This.is a large difference, considering the sensitivity of early health ef.fects to these two parameters.

In the case of the evacuation scheme corresponding to seismic conditions in the MPSS, the speed is reduced to 2 mph and the delay time increased to 3.38 hr. In comparison, in the IPS,[7] no evacuation was allowed following

~

a seismic event. Furthermore, the plume shielding factors were set to unity (no shielding) since the population may be outside (destroyed dwellings).

The remaining evacuation scnemes (S1-S6) are a series of evacuation schemes used for releases M-1A and M 4, since they were determined to be par-ticularly sensitive to evacuation schemes. Evacuation speed and delay times were varied between 1.2 mph and 10 mph, and the delay time was varied between

.92 hr and 2 hr. A probability asstaciated with each of these scnemes is shown

~

in tne bottom row. It is seen .th'at the general scheme with normal weather has 0

approximately a 40% weight. All these probabilities were determined by sub-jective judgment.

3.7.2 Plume Characteristics and Uncertainty A rigorous treatment of uncertainty of this analysis is not practical.

In the MPSS, it is pointed out that some of the data and/or models are either in state-of-the-art or because of tneir firmness, have relatively small bands of uncertainty. The overall uncertainty from the CRAC2 code will thus only be treated as an uncertainty in the dose delivered to an individual. This uncer-tainty will be measured by changing the isotopic content of the plume. In a nanner similar to tnat used above to esti: ate uncertainty associated with pri-mary system holdup and containment building attenuation, a DPD is defined 24

, _-- ,. . - - , , ..._7 , . . - - , , ,y.- ,

- .. - n= =.

._ m .. .

(

I. -

which characterizes the uncertainty in dose delivered to an individual. Table 3-17 shows the variation in source multiplication and appropriate probability.

It ~1s seen that a small probability of doubling the source exists. However, the largest probability indicates that the source will either stay the same or will be halved (80%). A reduction by an order of magnitude is also given a small probability.

This DPD is then folded into the individual release DPD's outlined above and results in an overall DPD for each release. It is seen that. the overall DPD will now have a finite probability of doubling the source strength in se-1 lected cases. Table 3.18 shows a list of tne CRAC2 calculations which were carried out. It is interesting to note that tne releases M-8 through M-12 were all carried out with a multiplication of unity, thus ignoring any uncer-tainty in the CRAC2 calculation. Tnese are, however, low consequence se-quences and ignoring the uncertainty will have small effect on the overall risk. The overall uncertainty determination is made by carrying out CRAC2 calculations for the various fission product sources appropriately modified by the multiplication factors and weighted by the corresponding probabilities.

O 25 l i

, ,_y .- -,w-r --

P---*- -W '***9-3 "*"T "

• • • • '~

a .._t-

. --=s y.

. .: ~

- W -cr

, .- t

~~

• A55/[:;ER01

. 55/E

ERGY t:..rai LilLASE pet E MC IN)D:IC511 OK Kt4359 i 07: Tait.::f t;T e ...KL trJiL6FF e vtSSEL FA!LliRE e C0':T;!:."f t.T P(tl T(t)

. . *::I .I; t:!'

. VESSEL BLO!; Dour. SECI""E

. ; ri ::ti 5,.rr *

. CA!!iv i::TER co!L- $.Y.f..I4I'fT. 'I.' .E;J,R gpp  :. : - ....$s t

C.*.

:. *.: i .!

5:.rE..;;c: s -

. s -

/ \

s

/

1 ASS lit:iRCi  !

RELEASE /

/ \

/

/

/

,'g'"...... . . . .

~ ~g

CORCO :..".001 ' " ' .i ,'.'."'

~ ~

fi,::I L.''

1. CC::0[:;5ABLE

' l:CCL TABLES a CCRE-CC:.CRili II;iEP3.C T!0.*.

. C0 TUSi! Lit 643 GE *;CI'.1 T 10.'.

s S 1 C:.t! C L:.i * *. T 1 *> .

Figure 3.1 The MPSS computational approach.

26

______ , - , - - - ^ ~

t

'. (

..  : x _

Table 3.1 Sunnary of computational tools Computer Code Accident Phase MARCH 1. Non-dispersal Events - Total transient

2. Dispersal Events - Initial blowdown, slump, and vessel failure MODMESH 1. Non-discersal Events - Interface to other codes

2. Dispersal Events - Discharge and scatter, cavity coil-off

~C_0RCON-M001/W Core-concrete interaction for dry cavity C0C0 CLASS 9 Containment building pressurization and hydrogen combustion CORRAL-II Fission product transport CRAC2 Consequences C

27

..- ( .

Table 3.2 Summary of containment event tree time frames "~

and nodal questionsL5]

Time Frame I: Accident Initiation -.< t < Core Degradation 7

CIl - Is the containment intact?

Time Frame II: Core Degradation < t < Significant Debris Accumulation

~~

in Lower Plenum NB2 - Does the hydrogen not burn?

CI2 - Does the containment remain intact?

Time Frame III: Significant Deoris Accumulation < t < Vessel Failure in Lower Plenum CD -

Is the core melt incoherent?

NB3 - Does the hydrogen not burn?

CI3 -

Does the containment remain intact?

~

Time Frame IV: Vessel Failure < t < Complete Depressurization QUE - Is the core debris quenched?

NB4 - Does the hydrogen not burn?

CI4 - Does the containment remain intact?

Time Frame V: Complete Depressurization < t < 4 Hr* After Vessel

- Failure CD5 -

Is tne debris coolable?

NB5 - Does the hydrogen not burn? -

CIS -

Does the containment remain intact?

Time Frame VI: 4 Hr After vessel Failure < t i Dne Day CD6 -

Is the debris coolable?

NB6 -

Coes tne nydrogen not burn?  !

C16 -

Does the containment remain intact? l SM6 -

Joes tne basemat remain intaC*?

l l

l

' l

• It should take 3 hr to coil-off the accumulator water from the lower '

reactor cavity.L5] l l 4 28 e - , .-. - --

~ ~ _.q

. .. - I f

,'  ?- N ( .

Table 3.3 Assessment of MPSS event tree nodal questions Node Evaluation CII Identical to Node A of ZPSS and in agreer-:nt with NUREG-0850 NB2 Similar to Node B of ZPSS and differences with NUREG-0850 will not affect the probability value CI2 Similar to Node C of ZPSS and the assessment is appropriate CD Identical to Node 0 of ZPSS; however, large incoherencies assumed as compared to NUREG-0850, and NUREG/CR-3300 comments apply NB3 Identical to Node F of ZPSS. H2 generation is equivalent to 20%

Zr/H 2O reaction compared witn 100% in NUREG-0850 CI3 Similar to Hoda E of ZPSS ano tne probabilities seem to be rea-sonable QUE The high probability assigned to the hign pressure discharges (small breaks and transients) need to be assessed in light of recent ex-perimental measurements NB4 Adequate, except for the core and no quench where further assessment is needed CI4 The arguments seem to be valid; however, further calculations are needed to ascertain the assigned probability with maximum uncertain-ties regarding H2 CD5 The probabilities assigned to the nodes need to be verified in light of large phenomenological uncertainties on debris bed coolability NBS Similar to Node (0) of ZPSS, and the assessment is reasonable CIS Due to the strong dependence of the containment failure probability on pressure, an audit calculation is needed to confirm the pressure peaks CD6 Same as CD5, where the success probability needs tu De verified NB6 Reasonable C16 Similar to code (R) of ZPSS and in agreement with NUREG-0350 Ba6 Adequate O

29 l

-- _ ---= m . .

k -

Table 3.4 Notation and definitions for release categories Release Category Descript. ion H1A Containment Bypass, V-Sequence M1B Containment Bypass, SGTR M2 Early Failure /Early Felt, No Sprays M3 Early Failure / Late Melt, No Sprays M4 Containment Isolation Failure

_ MS Intermediate Failure / Late Melt, No Sprays M6 Intermediate Failure /Early Melt, No Sprays M7 Late Failure, No Sprays M8 Intermediate Failure With Sprays M9 Late Failure With Sprays M10 Basemat Failure, No Sprays M11 Basemat Failure Witn Sprays M12 No Containment Failure 0;

1 30 1

.t v ,  ;,'

,i i ,l i

141; V6it ti t!I .!g{ ]

t' f ,]b; "

! $ 1 4 .

1 ) ) ) ) l ) ] ) ) ) ) ) i ) ) )' ) ) ) ) l

- 9 5 5 2 5 1

1 5 ? S l 8 2 5 1 l L 2 5 5 5 1 L "<

L

( ( ( ( ( ( l ( ( ( ( ( ( . ( ( ( ( ( f 2 g.64 6 1 0 5 4 9 h 6 l b 6 1 4 8 '

0 9 4

• 0 0 8 lt 1 l -

l M

1 9 2 1 1 O. O. 94 2 1

. 1 1 9 4 5 O.

1 1 9 2 1 1 1 2 9 3 I 1 1 1 9 3 1 - - ,

% ) ) ) ) ) ) ) ) ) ) ) l ) ) )

2 4 1 2 4 1 2 4 1 2 4 I 2 4 1

( ( ( ( ( ( ( ( ( ( ( ( ( ( (

- 9 7 9 - 9 7 9 - 9 7 9 - 9 4 9 - - - 9 4 9 - -

l - 9 7 8 - - 9 7 8 - 9 6 8 - - - 9 6 8 - -

l - - 9 7 e. - - - - - -

i f - 4 9 9 - 4 9 9 - 4 9 '9 - 4 9 9 - - - 4 9 9 - .-

) ) ) ) ) ) )

2 1 1 1 3 3 2

( ( ( 1 ( ( (

4 - - - 3 - - - 3 - - - 6 - - - 8 8 3 - - - - .

o 3 - - - 1 - - - 6 - - - 9 - - - 8 8 8 - - - - .

l i

f 9 -'--

- - 1

- - - 4

- - . 1

- - - 9 9 9

) ) ) ) ) ) ) ) ) ) ) ) )

4 1 4 4 1 4 4 1 4 4 4 4 4

( ( ( ( ( ( ( ( ( ( ( ( (

- 8 8 9 - 8 8 9 - 8 8 9 - 8 - ) - - - 8 - 9 - -

- 1 9 1 - 1 9 1 - 1 9 1 - 1 - 9 - - - 1 - 1 - .

9 - - - - - - - - - - - .

H - 1 9 1 - 1 S 1 - 1 9 1 '- 1 - 1 - - - 1 - 1 - .

) ) ) ) ) ) ) ) ) ) ) ) )

4 4 4 4 4 4 4 4 4 4 4 4 4 s ( ( ( ( ( ( ( ( [ ( ( ( (

t - 7 7 3 - 7 7 3 - 7 7 3 - 7 - 3 - - - 7 - 3 - -

n 8

- 3 7 6 -

3 7 6 - 3 7 6 3 -

6 - - -

3 - 6 e 1

- 6 6 6 - 6 6 6 - 6 6 6 - 6 - 6 - - - 6 -

v 8 e

g 1 ) ) ) ) ) ) ) )

n 1 1

1 1

1 1 1 1 I i ( ( ( ( ( ( ( ( (

t 0 - - - 9 - - - 1 - - - 5 - 0 - 9 9 7 - f t - -

a 9 - ' - - 4 - - - 9 - - - 9 - 9 - 0 8 9 - 9 - - .-

i l

l t i 2 - - - 3 - - - 8 - - - 7 - 9 - 9 9 8 - 9 - - -

i i

n ) ) ) )

1 1 2 3 l ( ( ( (

0 - - - 8 - - - 7 - - - - - - - 0 - - - - - - -

a 1 - - - 3 - - - 1 - - - - - - - 1 - - - - - - -

n t - - - . - - - - - - - - - - - - - - - -

r I 6 - - - 5 - - - 6 - - - - - - - 1 - - - - - - -

e t ) ) ) ) )

n 3 1

3 3 4

i ( ( ( ( 1

- - - - - - - - - - - - 8 - S - - 0 b - 8 - - -

x - - - - - - - - - - - - 4 - 3 - - 1 3 - 7 - - -

i

- - - - - - - - - - - - 9 - 6 - - 1 4 - 6 r 1 t

a ni ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) )

C 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

(

- -(

-( - - - (

-( - (- - (

- - (- - - - - -( - - -( -( ( (

e 1 ( ( ( ( ( ( ( 1 0 0 ( 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - -

5 l 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - -

it 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 - -

3 e ) ) ) ) ) ) ) ) )

l 4 4 4 4 4 4 5 5 5 b - - - - - - - - -

( ( ( ( ( ( ( ( (

a -

- -- -- -- --- -- -- -- - - - - - 0 0 0 2- 4 1 9 9 9 - -

T 9 9 9 - 1 0 9 9 9 - -

3 H

- - - - - - - - - - - - 2 2 2 2 - 2 1 9 9 9 - -

) ) ) ) ) ) ) ) ) ) ) ) )

4 4 4 4 4 4 4 4 3 4 4 4 4

(

- (- -( -( - (- - (

-( - ( - (- -( ( (

(

1 0 0 0 1 0 0 0 5 0 0 01 - - - - 4 - - - - - - -

2 tt. 9 4 9 0 9 4 9 9 4 9 - - - -

1 - - - - - - -

H 2 2 2 2 3 2 2 2 1 2 2 2 - - - - 2 - - - - - - -

) ) ) ) -

4 4 4 4 _

( ( ( ( _

- -- -- -- -------- --. --- - - - - -- --- -- -- -- - 0 0 -0 -9

t -

9 9 9 - 0

' l i -

f - - - - - - - - - - 1 1 1 1 - 1 et

- -- .-- - - - -- -- - - - - - -- - - - -- - - - - - --- - 0 . .

I - - - - 1 .

y _

s p *. i t

o e J

t t

J C

C 'C E

t t

[ C o C : '. C C C

• C t C C C L D 5 A 1 A 1

J; t A,

t A A A t L E t

! s ! i t t t t

i s s s t t

'5 E E E T I 1 E

I v s 2

O

l

. (c .

Table L 3.6 Notation and definitions for plant states (internal)

Symbol Description AEC Large LOCA, Early Melt AEC' Large LOCA, Early Melt, Failure of Recirculation Spray AE Large LOCA, Early Melt, No Containment Cooling ALC Large LOCA, Late Melt ALC' Large LOCA, late Melt, Failure of Recirculation Spray ALC" Large LOCA, Late Melt, Failure of Quench -Spray Al Large LOCA, Late telt, No Containment Cooling SEC Snall LOCA, Early Melt SEC' Small LOCA, Early Melt, Failure of Recirculation Spray SE Sna11 LOCA, Early Melt, No Containment Cooling S'E In-Core Instrument Tube LOCA, Early Melt, No Containment Cooling SLC Snall LOCA, Late Melt SLC' Small LOCA, Late Melt, Failure of Recirculation Spray SLC" Small LOCA, Late Melt, Failure of Quench Spray SL Small LOCA, Late Melt, No Containment Cooling S'l In-Cere Instrument Tube LOCA, Late Melt, No Containment Cooling TEC Transient, Early Melt TEC' Transient, Early Melt, Failure of Recirculation Spray TE Transient, Early Melt, No Containment Cooling V2EC Steam Generator Tube Rupture, Steam Leak, Early Melt V2EC' SGTR, Steam Leak, Early Melt, Failure of Recirculation Spray V2E SGTR, Steam Leak, Early Melt, No Containment Cooling V2LC SGTR, Steam Leak, Late Melt V2LC' SGTR, Steam Leak, Late Melt, Failure of Recirculation Spray V2LC" SGTR, Steam Leak, Late Melt, Failure of Quench Spray V2L SGTR, Steam Leak, Late Melt, No Containment Cooling V Interfacing Systems LOCA l

l l

ej 32

+ , . , . - , - . , , , . - ,

__ .__.-,-m_ =_ _ __ -,.~, . = _ __= _; - - = - -

wmm

, c.

i. -( ( .

Table 3.7 Simplified 'C' matrix for MPSS Plant State M1A M1B MS M6 M7 M9 M10 M11 M12 AE 0.62 0.29 0.09 AEC 0.05 0.95 AEC' 1.0 AEC" 0.99 0.01 AL 0.54 0.35 0.11 ALC 0.05 0.95 ALC' 1.0 ALC" 0.99 0.01 SE 0.06 0.89 0.05 SEC 0.05 0.95 SEC' 1.0 SEC" 0.99 0.01 SL 0.01 0.79 -

0.20 SLC 0.05 0.95 SLC' 1.0 SLC" 0.99 0.01 S'E 0.99 0.01 S'l 0.99 0.01 TE 0.90 0.10 TEC 0.05 0.9 TEC' 1.0 TEC" 0.99 0.0 V f.0 V2 1.0 l

1

)

l 1

33

-o,-.-,,_. --

. _ - - - - - - - - _ , _ - - - . . _ _ . -=_.---: _-___

t. ' ' ( ,

Taole 3.8 Simplified ZPSS containment matrix C Release Catecory Pl ant State Z-1(M2) 2(M1A) 2R(M7) 8A(M12) 8B(M12)

SEFC -

1.0 SEF SEC 1.0 SE 1.0 SLFC 1.0 SLF 1.0 SLC 1.0 SL 1.0 TEFC 1.0 TEF 1.0 TEC 1.0 TE 1.0 AEFC 1.0 AEF 1.0 AEC 1.0 AE 1.0 ALFC 1.0 ALF 1.0

_ 1.0 AL 1.0 i i

VE 1.0 l C

34

s

( ( .

Table 3.9 Simplified BNL containment matrix C for ZPSS Release Category Plant State 2(M1A) 2R(M7) Z-5(M8) 5(M10) 7(M11) 8A(M12) 8B(M12)

SEFC 0.02 .1 0.88 SEF 0.4 .1 0.5 SEC 0.02 0.88 SE 1.0 -

SLFC 0.01 .1 0.89 SLF 0.01 .1 0.89 SLC 0.01 .1 0.89 SL 1.0 TEFC 0.02 .1 0.88.

TEF 0.4 .1 0.5 TEC 0.02 .1 0.88 TE 1.0 AEFC 0.02 .1 0.88 AEF 0.4 .1 0.5 AEC 0.02 .1 0.88 AE 1.0 ALFC 0.01 .1 0.89 ALF 0.01 .1 0.89 2LC U.01 .1 0.39 AL 1.0 VE 1.0 l

35 l

!t , 7t ji ,

p  ; sf i jjkj 1 b;r ibI,

~ .

i 3 3 7 8 5 7 8 3 4 3 3 3 3 - - - -

E E E E 'E E E L

a E 4

E 4 X K 7 E K 4 2 I I 2 I 2 2 6 8 5 6 8 2 3 2 2 2 2 - - - -

E E u E K E 2 X E S

E 4

E 4 M E I

E 9 K I B R 2 r 6 7 5 6 7 3 2 2 2 2 2 2 - . - -

S 2 - - - - - - -

- - - - E E E E E E E a

B E E E 7

E 8 X S $ X I 2 9 I 2 b 5 6 3 5 7 S 2 - - - - -

- 2 2 5 5 5 3 E E

- 3 E E E 9 e X 0 0 0 0 0 0 I 1 I 2

• T 0 s

e u

l a b 6 4 5 6 -

5 - -

V R 2 - - -

y - 5 - 6 6 6 5 5 3 E E E E E r s n C s

0 E

S 0 0 0 0 0 0 l 2 8 I I a

n e i T n e 5 6 u c r 2 2 2 3 3

3 4

s r B 7 - 5 5 2 E

E E E E E E E u - E 9 G 2 B 2 6 y o I 0 7 0 0 0 I I S

r o

g t e

e a m 3 3 3 3 3 3 5 6 t 3 3 3 3 3 - - - -

i a t - -

E E E E E E E E E E c E s I O K E 7 S S E 6 6 6 7 6 2 2 9 e t n

S i a o r 3 3 e P K

- 9 9 7 8 9 9 9 9 9 9 3 -

E l

e X e

0 0 0 0 0 0 0 0 0 0 0 E I R

0 e

)

r 1 s y h 6 6 . 6 6 6 6 6 6 E E E E E 6 6 a g / E E E E E 3 e re ut 0 0 0 0 0 0 0 0 A l

5 9 7 5 4 4 2 2 A A e n 8 2 2 N N N

- e R E ( 1 1 4 4 5 2 2 l

b a

T n

_ e o )

s i a t s 5 5 5 5 0 0 0 e ar hr 0 0 0 0 0 5 l

2 0 0 0 0 0 0 0 5 e u ( 1 1 2 2 1 1 R D e

m i

T e g )

s n s a i r 2 5 0 1 1 0 0 0 0 0 0 e nr h 0 0 l t 1 0 0 0 4 4 6 4 0 0 0 0 e a 1 1 2 0 8 RW e .

i m

T )

-e. s 5 a-t

r. 5 5 1 0 2 3 3 1 5 0 0 0 5 f

l u. (f 2 0 6 0 4 4 0 4 1 5 5 0 e t 2 1 2 2 9 9 R S y

r o

g 0 l 2 e A D 7 8 9 1 I 1 3 4 5 6 t

a 1

I

-  ?. - - - - - - - -

M n

E M M H H M M M M 1 1

H H n

- M

w,nw _n

. , . = . .= - n - .-- = . -

. . . -mm

.' */ ( ( .

Table 3.11 Interfacing LOCA - V sequence Sequence BMI-2104 V-Sequence IPS[73 M-1A Surry V+a Xe-Kr .9 1.0 1.0 I+01 .707 .2 .7 Cs Rb .5 .2 .5 Te-Sb .3 5(-2) .1 Ba-Sr 6(-2) 1(-2) 6(-2)

Ru 2(-2) 7(-4) 2(-2)

La 4(-3) 2(-3) 2(-3)

Release Time (br) 2.5 2.0 Warning Time (hr) 1.0 1.0 Duration (hr) 1.0 1.0 Energy (Btu /hr) 20(6) .5(6)

Probability V

1 .17 1/2 .55

. 1/4 .28 1/10 -

1/100 -

37 ,

m . _ - - _ = - - - ,~ _._-

. t ( ,

Table 3.12 Early overpressurization release

~

Sequence Bt11-2104 -

TMLB'-6e ZPSSE13

. M-2 Surry 2B Xe-Kr .7 1.0 .9 I'01 .505 .7 .707.

Cs-Rb .6 .6 .5 Te-Sb .2 .5 .3 Ba-Sr 7(-2) 1(-2) 6(-2)

Ru 2(-2) 8(-4) 2(-2)

La 3(-3) 2(-3) 4(-3)

Release Time (nr) .75 2.5 tlarning Time (nr) .2 1.0 Ouration (br) 2.0 .5 Energy (Btu /hr) 150(+6) 250(+6)

Probability U

2 - -

1 .25 .25 1/2 .60

. 1/a .25 1/10 .5 .15 1/100 -

p 38 l

____._._..,,,m.,,___, _. __

., c c .

Table 3.13 Isolation failure Seauence M-4 IPSE73 IPS[73 (6" dia.) (8"dia.) (4"dia.)

Xe-Kr .9 .989 .7 I+0I .206 .28 .26 Cs-Rb .6 .48 .26 Te-Sb .5 .36 .21 Ba-Sr 7(-2) 5.5(-2) 2.9(-2)

Ru 5(-2) 3.2(-2) 1.8(-2)

La 7(-3) 4.9(-3) 2.8(-3)

Release Time (br) .2 Warning Time (br) -

Duration (hr) 2.0 Energy (Btu /hr) 70(+6)

Probability U

1 .4 1/2 .6 1/4 -

. 1/10 -

1/100 -

39

, e i ( .

~ . ,

Table 3.14 Intermediate and late overpressurization (No Sprays)

Sequence MPSS MPSS BMI-2104 IPSI 73 ZPSSIll IPPSS(3)

M-5 M-7 (Surry) TMLB'-5 (2R) 2RW*

Xe-Kr .9 .9 1.0 .96 .9 1.0 0l+I .016 .015 1(-3) 1.05(-1) .7 9.3(-2)

Cs-Rb .5 .3 8( 4) .34 .5 .26 Te-Sb .5 .3 7.0(-4) .38 .3 .44 Ba-Sr 5(-2) 3(-2) 3(-5) 3.7(-2) 6(-2) 2.5(-2)

Ru 4(-2) 2(-2) 1(-6) 2.9(-2). 2(-2) 2.9(-2)

La 6(-3) 4(.3) 9(-6) 4.9(-3) 4(-3) 1.0(-2)

Release Time 8.3 20.1 13.0 10.0 12.0 (nr)

Warning Time 4.1 16.0 8.0 9. 0 11.0 (hr)

Duration .5 .5 .5 3.0 2.0 (br)

Energy 450(+6) 540(+6) 98(+6) 20(+6) 19(+6)

(Btu /hr)

Probability U

2 -

1 1.0 .1 .3

.5 .2 .55

.25 5(-2)

.1 . .64 .11 .7 .15

.01 .31 .89

• Sum of multi-phase releases.

40

- - - ~ . . - . . . - . _ . . . ..__ .

- . _ - - __... _a=n.----. , ,._ n

. e . .

'.: ( , L Table 3.15 tb fail Sequence ZPSS[7] Ippss[3]

M-12 IPS(7] 88 88 Xe-Kr 1(-3) 5(-4) 2.7(-2) 2.7(-2)

OIFI 1.5(-5) 5(-6) 2.0(-4) 2.0(-4)

Cs-Rb 1(-6) 1(-5) 8(-7) 8.0(-7)

Te-Sb 9(-7) 1(-5) 1.5(-7) 1.5(-7) 1 Ba-Er 2(-7) 1(-6) 1(-7) 1(-7)

Ru 8(-8) 1(-6) 3(-3) 3(-8)

La 1(-8) 2(-7) 3(-9) 3(-9)

Release Time (hr) .5 2 2 2 Warning Time (br) -

1.0 1.0 1.0 Duration (br) 5.0 8.0 10.0 10.0 Energy (Btu /hr) - - - -

Probability V

2 - -

1 1.0 1.0 .5 .5

.5 .4 .4

.25 -

.1 .1 1

.01 -

c>

41 ,

c.

Table 3.16 Summary of evacuation scheines and their probabilities CENERAL SEISMIC SPECIAL IREATHENT FOR M1 AND M4 At1ALYSl$ CATEMRY 52 53 54 55 56 E :cuatic,n Scheme 1 2 3 51 Non-Seismic Hon-Scismic Non-Seismic Non-Seismic Non-Seismic Hon-Seismic I;itiating Event Non-Seismic Hon-Seismic Seismic Any Any Any Any Any Normal Adverse Any Any Weather Condition R::dius of Evacuation 10 10 10 10 10 10 Sector (Mil 10 10 10 Rr'lus of Evacuation 5 5 5 5 5 5 Circle (MI) 5 5 5 Distance traveled 15 15 15 15 15 ty evacuees (Hil 15 15 15 15 a

1.2 3.0 10 1.2 3.0 10

[cacuation Sreed (Meh) 10 7.5 2 Octay Time 0.92 2.0 2.0 2.0 0.92 0.92 3.38 0.92 0.92 before evacuation (Ifr) 0.07 0.19 0.39 0.05 0.14 0.16 Probability 0.88 0.12 1.0* I _

~

• Probability is I.0 for Evacuation Scheme 3 f f the release is from a seismic induced event. Otherwise it is zero.

Also, the probability of Evacuation schemes I and 2 and 51 through 56 will be zero for seismic initiated releases.

O

_ _ _ _ _ _ _ L

' -.' *(*

( ( .

Table 3.17 Subjective discrete probability distribution

.for site consequence uncertainty evaluation Release Fraction

. Adjustment Factor

• Discrete Probability i

2 0.10 1 0.35 0.5 0.45 0.1 0.10 1

• Adjustment factor of 1 is always used for noble gas releases.

1 1

i y

O 43 A

(

('

Table 3.18 List of 0P0 runs performed Source Term Multiplier

• 2 1 1/2 1/4 1/10 1/30 1/100 Release Category -

M1A X X X X X X X M2 X X X M3 X X M4 X X X X MS X X M6 X X X X X M7 X X X X X X X M8 X 1

i M9 X M10 X M11 X i

M12 X

• ttultiplier of noble gases remains 1.0 for all runs.

C 44

. x

. . ( (

{ _

4.

SUMMARY

4.1 Results of Level 1 Review The preliminary evaluation of MPSS indicates that the study is of a high quality in botn technical content and the material presented.

The major conclusions of tne study concerning accident phenomenology, ac-cident sequences and release catagories, source term, . and site consequences appear to be generally valid.

Comparison of Millstone-3, Zion, and Indian Piont plants shows signifi-cant similarities in, plant and containment designs. However, variation in containment heat removal and cavity configuration and construction influence the accident progression. With the specific cavity design employed in tne Millstene-3 reactor, it should provide for higher assurance of core debris re-tention during high pressure discharges, while in a design such as Zion, de- '

bris removal is nearly certain during high pressure discharges l The point estimate release fractions used in the Millstone Probabilistic Safety Study are comparable in magnitude to those used in the RSS, IPPSS, and 4

ZPSS. In those cases wnere comparisons can be made to tne more mecnanistic

, source term study being carried out by ASTP0,[93 it was found that the MPSS releases were either higher than or similar to the ASTP0 release fractions.

In tne case of the early overpressurization release, the release fractions were found to be similar, while for the intermediate or late overpressuriza-tion failure, the ASTP0 release is found to be substantially lower. In tne case of tne interfacing LOCA sequence, the ASTP0 release is approximately half tne 1P55 celease. It was found tnat the energy of release was substantially nigner in the MPSS than in all tne other studies.

O 45

~

..t

( ( .

=w.

Detailed evaluation of the accident quantification is not possible at tnis stage, and is thus planned for the future, when audit calculations will be performed in order to verify the plant response and accident progression paths, and therefore, the site consequence and risk evaluation.

4.2 Suggestions for Future Work

~

In order to check the methods used in the MPSS, it is proposed to analyze the late overpressurization failure sequence. Ir. this sequence, the contain-ment building is calculated to fail after approximately 8-15 hours. The t1PSS uses release fractions based on the RSS methods and then modifies tnem by mul-tiplication with uncertainty factors. Since, this modification implies a sub-stantial reouction in this sequence, it would be a good candidate for an audit calculation. Furthennore, the effect of containment leakage ratner than an abrupt failure could be determined using this sequence.

4.3 Ouestions and Additional Information Needs 4.3.1 Analytical Models and Phenomenology

1. Which version of the MARCH code was used for the analysis?

2. What are the implications of the single-volume containment model used in C0C0 CLASS 9 code?

3. Is the steam velocity low enough to limit Zr oxidation by H2 blanketing?

How is the velocity estimated to be several cm/s?

4. What is the implication of including the mechanical erosion process during molten jet-concrete interaction?

4.2.2 Uncertainty Analysis i 1. How important is the range of nodal probability?

2. Does a nodal probability naving a range of 0.99 to 0.9999 have any meaning?

C 46

i j ,:g ,, ( (.

What is the sensitivity of the final outcome to the values of the

3. -

weighting factors and the probability range?

4.3.3 Source Term and Site Model _

1. What is the reason for the higher energy of release of the plane for tne overpressurization failures in comparison to tne otner studies?

2. The evacuation model following a seismic event does not seem to account for the fact that there would be substantial damage to buildings, roads, .

etc.

i O

l 47

.')..-

1

( c .

REFERENCES

1. " Zion Probabilistic Safety Study," Commonwealth Edison Company, September 1981.

2. " Limerick Probabilistic Sa fety Study," Pniladelphia Electric Co.,

September 1982.

3. " Indian Point Probaoilistic Safety Study," Power Authority of the State of New York and Consolidated Edison Company, March 1982.

4. " Preliminary Assessment of Core Melt Accidents at the ' Zion and Indian Point fluclear Power Plants and Strategies for Mitigating Their Effects,"

IlUREG-0850, Vol .1, november 1981.

5. " Millstone Unit 3 Probabilistic Safety Study," Nortneast Utilities, August 1983.

. .:'x

6. W. T. Pratt, et al . , " Review and Evaluation of the 7.'.on Pronabilistic Safety Study, Vol. 2, Containment and Site Consiequence Ar.alysi s ,"

NUREG/CR-3300, Vol . 2, July 1983.

7. Di rect Testimony of J. F. Meyer and W. T. Pratt concerning Commission Question 1, Indian Point Hearings, Docket Numbers 50-247 and 50-286, 1983. ,

8. U.S. Nuclear Regulatory Comi ssion , " Reactor Safety Study," UASH- 14']0, NUREG-75/014, October 1975.

9. J. A. Gieseke, et al., "Radionuclide Release Under Specific Lu8 Accident Conditions, Vol. 1, PWR Large Dry Containment Draft Report," ?at'alle -

~'

Columbus Laboratories, BMI-2104,1983.

I C

48