ML20215L336

Summary of 870327 Meeting W/Bwr Owners Group/Idcor Re Mark I Containments.Numarc Containment Issues Working Group Efforts Re Response to 15 Questions Discussed.Attendee List,Proposed Agenda & Presentation Matl Encl

ML20215L336
Person / Time
Issue date:	05/06/1987
From:	Hulman L NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:	Floyd S, Fulton J, Thadani M BOSTON EDISON CO., CAROLINA POWER & LIGHT CO., NRC
References
NUDOCS 8705120198
Download: ML20215L336 (61)

Text

,R

1. ' =

+....

^

/ sauss%g '

UNITED STATES

!\ & NUCLEAR REGULATORY COMMISSION

- { e- . ,E wAsHm0 TON. D. C. 20666 y gy <

MEMORANDUM TO: DISTRIBUTION FROM: L. G. Hulman, Chief Severe Accident Issues Branch Division of Reactor Accident Analysis Office of Nuclear Regulatory Research

SUMMARY

OF MARCH 27, 198'/ MEETING WITH BWR OWNERS GROUP /IDCOR ON MARK I CONTAINMENTS The meeting was opened by Messrs. Denton and Bernero, who discussed the background and the nature of the 15 questions addressed to the BWR Owners Group. A previous meeting with representatives of the research comunity was referenced. The sumary of that meeting was identified as available through the Public Document Room. Enclosure 1 is the attendance list for the meeting.

Enclosure 2 contains the proposed meeting schedule and lists the 15 questions.

V. Boyer, Philadelphia Electric Co. (PECo), indicated that the Owners Group /IDCOR were requested to respond to the 15 questione. The responses were coordinated through the NUMARC Containment Issues Working Group of which he is chairman. He indicated that other NUMARC efforts were being delayed to respond to the request for information on the 15 questions, and that the NUMARC working group draft report to the steering conmittee was not expected until mid-May as a result. He suggested that NUMARC would probably not be able to report on their study to the Commission before this sumer. He indicated that the IDCOR (Industry Degraded Core Rulemaking) effort was going our of business. He then introduced the responses, and sumarized his views on the most critical issues and infonnation available (Enclosure 3, The critical issues identified were 1) the progress of core failure, 2)p 2-4). cooling of a core on 'the floor, and 3) core concrete interaction.

R. Diedrich, PECo, described the industry evaluations (Enclosure 2 p 5-8). He t indicated that they were evaluating both overall risks (referred to as bottom l line), and conditional failures. He indicated their conclusion that conditional l failure is sequence and plant dependent, thereby making it difficult to compare j plants in a meaningful way. He also stated a conclusion that the Chicago

Bridge and Iron Company containment study is indicating that the ultimate MK I pressure capability is higher than generally assumed, and that the torus

' airspace is the most likely failure location. He compared the IDCOR~and NUREG-1150 efforts, including the conclusions from both that modifications were not justified. He concluded with a sumary that indicated the NUMARC working l group is studying MK I containments, that he believed sufficient technical bases exist for NUMARC to make decisions, and that cost / benefit comparisons i

, will be made of potential modifications. He indicated studies to date have I l shown no modifications to be cost beneficial. [/

l em =%

e

%b,e I Y. '

2% ,

A s.

f .

K E. Burns, Delian Corp..' discussed the responses to questions 1 and 2 (Enclosure 3, pg-10). He-indicated there were four or five PRA's for MK I plants available that indicate no specific accident type dominates for all MK I's. He, therefore, concluded that the spectrum of potential sequences was important.

He also concluded that there was no mechanistic coupling of containment failure to inducing coremelt. (See Enclosure 5)

^

R. Henry, FAI, discussed the responses to questions 3 through 10 (Enclosure 3, l- p 11-20). The conclusions presented with respect to containment failure were in large measure based upon evaluations of heat transfer in which the contain-ment shell was not postulated to fail by perforation (Enclosure 4). This evaluation was noted as significantly different-from those of the NRC staff and contractors. The significant points of his analysis were: 1) a 12 cm debris t

.t bed depth, 2) water above the debris bed acts as a heat sink with nucleate boiling at the shell surface, 3) the concrete below the debris acts as a heat sink, and 4) the debris bed was assumed to be near the melt temperature. His other main points were:

( ,

} (04)-highpressuremeltshavenosignificanteffectoncoremelt progression, but the distribution of material in the contairment is influenced; (QS)therearenosignificantdifferencesbetweenBWRsandPWRsin meltdown or melt through times;

'(Q6) the debris properties of a " core-on-the-floor" are different, but the behavior is not. BWR's would have more metal with less oxidation; (Q7) water on the drywell floor is beneficial, but requires replenish-ment. (Note that use of the 10COR heat transfer model results i

in no prediction of steel containment liner or downcomer melt through);

(08) drywell sprays would reduce containment challenge, sufficient water to i remove decay heat would be adequate, and sprays can help remove l

airborne fission products. Spray rates in the range of 500 - 1500

!, gpm appear adequate. Enclosure 4 was again referred to for a t discussion of heat transfer and related conduction. It was noted that the IDCOR heat transfer methodology was included in submittals L

l. to the staff, but little feedback had resulted;

~

(Q9) a debris barrier would not be useful, and could result in negative effects; and r-f (Q10) a debris barrier to contain debris in the pedestal area under the vessel was considered detrimental. He suggested that if something

! was done, it would be to allow a coremelt the maximum expansion area and attempt to stabilize it with water.

p /

?

I

. *f

_4-Corzents on a draft summary were solicited by memo date March 31, 1987.

Several informal comments and three sets of formal coments were received.

All were considered in this final summary. The formal comments by Messrs. J. C.

Carter, A. R. Diederich and G. A. Greene are enclosed (Enclosure 6). Copies of this sumary are being furnished to those participants of the March 27 and February 3,1987 meetings.

,' ..G.&Hulman

~ ~, .Chief Severe Accidents Issues Branch Division of Reactor Accident Analysis Office of Nuclear Regulatory Research

/~,

'ub> and Outsticn List 1

k t

j

-_ _ _ _ _ _ _ _ _ . _ .- .J

al, r .

t R. Diederich discussed Q 11. He indicated no analysis was inade of the gap between the drywell and the biolcaical shiefd. However, if the drywell were breached, some fission products might be trapped in the gap in the path to the reactor building through penetrations in the biological shield. (See Enclosure 3, pg 21) The calculations were characterized as conservative because no credit for fission product attenuation was taken for the biological shield area.

E. Burns discussed ventino (Q 12). He indicated venting was a means of preventing uncontrolled releases and establishing a heat removal path as a last resort. Further, venting can be used to prevent coremelts in such sequences as TW. However, he indicated large costs were not justified generally, but plant specific analyses may indicate differently. (See Enclosure 3, p 22)

R. Diederich discussed noble gas venting (Q 13). He indicated such venting as a last resort can reduce the impacts of some sequences, but that negative effects must be considered. -(See Enclosure 3, p 23) He presented a backup slide which showed substantial reductions in doses if releases of noble gases were delayed about 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br />.

R. Henry discussed the use of containment sprays for station blackout sequences in response to Q 14. He indicated several benefits (debris cooling, delay of containment failure, and fission product removal), but eventually containment heat removal is required. (See Enclosure 3, pg 24). He also discussed debris coolability referring to pages 25-35 of Enclosure 3 using inferences from TMI, experimental evidence and analytical assessments. Analogies were also made to debris coolability in coal fired power plants and experience in the steel industry with electric furnaces by several participants.

R. Diederich discussed 015 (See Enclosure 3, p 25). He indicated that the NUMARC evaluation is not complete, but that to date no cost beneficial modifications have been identified.

R. Bernero asked whether modifications such as a more reliable ADS system could help. R. Henry indicated he did not consider such modifications cost l

r beneficial.

The issue of steel shell perforation was again raised. R. Henry again l sumarized the IDCOR view that the carbon steel and heat transfer capabilities as modeled precluded such as occurrence.

V. Boyer concluded by indicating the NUMARC working group report was expected

( in mid-May, followed by a review by a supervising technical comittee. He l

indicated no finn dates had been established for briefing the Comission or the l staff, but any briefings would likely not be before sumer.

o or =

6 Comments on a draft summary were solicited by memo date March 31, 1987.

Several informal comments and three sets of formal comments were received.

All were considered in this final summary. The formal comments by Messrs. J. C.

Carter, A. R. Diederich and G. A. Greene are enclosed (Enclosure 6). Copies of this summary are being furnished to those participants of the March 27 and February 3, 1987 meetings.

, &. G.- Hulman

~ ~, Chief Severe Accidents Issues Branch Division of Reactor Accident Analysis Office of Nuclear Regulatory Research

Enclosures:

1. Attendance List

2. Proposed Meeting Schedule and Question List

3. Owner Group /IDCOR Slides

4. IDCOR Heat Transfer Model

5. BWR Severe Accident Sequence Classes

6. Formal Comments f --

.e, *t ,

s Comments on a draft summary were solicited by memo date March 31, 1987.

Several informal comments and three sets of formal comments were received.

All.were considered in this final sumary. The formal coments by Messrs. J. C.

.. Carter, A. R. Diederich and G. A. Greene are enclosed (Enclosure 6). Copies of this summary are being furnished to those participants of the March 27 and

- February 3,1987. meetings.

L. G. Hulman, Chief Severe Accidents Issues Branch Division of Reactor Accident Analysis Office of Nuclear Regulatory Research

Enclosures:

1. Attendance' List

2. Proposed Meeting Schedule and Question List 3.~0wner Group /IDCOR Slides

4. IDCOR Heat Transfer Model

5. BWR Severe Accident Sequence Classes

6. Formal Comments-4

- 0FC- :DRAA:5 I  :  :  :  :  :  :

NAME~

L gt :  :  :  :  :  :

DATE :5/ 7/87  :  :  :  :  :  :

0FFICIAL RECORD COPY

a :.

Je5 ' f . .

,g ;g ,

Ii' g..

DISTRIBUTION: ..

. Stephen D. . Hoyd,'. Carolina Power & Light,Co.

Jack Fulton, Boston Edison Co.-

Terry.Pickens,' Northern States' Power.Co.

(D.EDiederich, Philadelphia ~. Electric Co.. ~

V. S! Boyer Philadelphia Electric Co.

J.; Carter,1 International Technologies Corp.

-R. Henry,:Fauske & Associates, Inc.-

• E.zBurns,'Delian Corp.

.J. Lang, EPRI

W. ~ C.1 Ham;- House Subcommittee 'on Energy & Power- Staff-'

W. Smith, Bechtel Corp. .

L. Gifford, General Electric

Co.

.H. Spector, N.Y.~ Power Authority, ,

M. Idell 'Public Service Gas & Electric Co.

E. Dluzniewski, GRS

P.LHill, Pennsylvania' Power & Light Co. ~

E. Hughes,-Erin Engineering . .

R. A. Cushman, Niagara: Mohawk Power Corp. '

A. J. Marie. -Philadelphia Electric Co. -i, ,

P.

J. Fulford. .NUS Corp.

R.,A.:Pinelli, GPU Nuclear '

R.:Huston', AIF . -,

M. Thadant NRC:J

F. Eltawila, NRC.

'M.~Chiramal,-NRC J. A. Murphy.-NRC W."Hodges, NRC-

~

'Jocelyn Mitchel1 'NRC K. M. Campe NRC.

Jc E.:Rosenthal, NRC.

-J. DelMedico. NRC C. Ader,'NRC S. Hodge, ORNL,

.A.-Benjamin,SNL-

' O. - Powers ,- SNL':

G. Greene, BNL. '

M. Khatib-Rahbar-BNL R._ Denning,-'BCL

-T. Collins, NRC L.~ Soffer, NRC NRC Z. RosztoczyAUS

t. Schmidt, Corp.

M.;Jamgochian, NRC D. Fadden, INP0-L.-Shao A. Thadani C. Wright, GE '

PDR R. Houston, NRC

'SA Jjle R. Starostecki, NRC

- u% M. Silberberg, NRC c E. Beckjord,-NRC man T.- Murley, NRC F. Miraglia, NRC B. Sheron, NRC j

.H. Denton, NRC-T. Speis, NRC G. Lainas, NRC D. Ross, NRC C. E. Rossi, NRC

-R. Bernero, NRC J. Kudrick, NRC .'

J.;Conran. NRC T. Walker, NRC k..' - . . _ . . .

- - - a

e o

' -I .

ENCLOSURE 1 INDICATE IF YOU ATTENDANCE LIST WANT THE PREVIOU5

SUMMARY

, AND CHECK IF YOU WAMT TO COMMENT ON MTG.

NAME BUSINESS ADDRESS TELEPHONE NO.

SUMMARY

L. G. Hulman U. S. NRC 301-492-7941 X NL - 007 Washington, DC. 20555 M. C. Thadani NRC 301-492-8649 X Tom Murley NRC No Eric Beckjord NRC No T. P. Speis NRC No Z. R. Rosztoczy NRC No Farouk Eltawila NRC Yes Matt Chiramal NRC Yes R. W. Houston NRC No Ashok Thadani NRC No Stephen D. Floyd Carolina Powe~r & Light Co. X 411 Fayetville Street Raleigh, NC 27602 Jack Fulton Boston Edison Co. (617) 849-8912 X 800 Boylston Street Boston, KA 02199 Terry Pickens Northern States Power Co. (612) 337-2037 X 414 Nicollet Mall Minneapolis, MN 55401 Dick Diederich Philadelphia Electric Co. (215) 841-4516 X 2301 Market Street Philadelphia, PA 19101 VS Boyer Philadelphia Electric Co. (215) 841-4000 X 2301 Market Street Philadelphia, PA 19101 i

^^ ' ~ ~ ~

I o

.- -e .

INDICATE IF YOU WANT THE PREVIOUS SUPPARY, AND CHECK IF YOU WANT TO.

COMMENT ON MTG.

NAME BUSTHESS ADDRESS TELEPHONE NO.

SUMMARY

Jim Carter International Technologies (615) 481-3300 X-575 Oak Ridge Turnpike Pak Ridge, TN 37830 Bob Henry Fauske &' Associates, Inc. (312)323-8750 X 16 WO70 West 83rd Street Burr Ridge, IL 60521 E. T. Burns Delian Corporation (408)446-4242 X 1340 Saratoga-Sunnyvale Road Suite 206 San Jose, CA 95129 H. R. Denton NRC .X Mark W. Idell Public Service Electric & (609)339-3073 Yes.

Gas Co.

P. O. Box 570 Newark, NJ 07101 Eryk Dluzniewski GRS 293-4200 X 801.18th St. NW Suite 300 Washington, D.C. 20006 Paul R. Hill Pennsylvania Power & (215) 770-7949 Yes Light Co. X 2N 9th St.

Allentown, PA 18101 E. A. Hughes Erin Engineering (415) 943-7077 Yes 1850 Mt. Diablo Blvd.

Suite 600 Walnut Creek, CA 94596 R. A. Cushman Niagara Mohawk Power Corp. (315)428-7476 X 301 Plunheld Rd.

Syracruse, NY 13203 A. J. Marie Philadelphia Electric Co. (215) 841-6378 2301 Market Street Philadelphia, PA 19101

I 'e INDICATE IF YOU HANT THE PREVIOUS

SUMMARY

, AND CHECK IF YOU WANT TO COMMENT-0N MTG.

NAME ' BUSINESS ADDRESS TELEPHONE NO.

SUMMARY

E. R. Schmidt NUS Corp. (301) 258-5831 910 Clopper Road Gaithersburg, MD 20878 P. J. Fulford NUS Corp. (301) 258-8692 X 910 Clopper Road Gaithersburg, MD 20878 R. A. Pinelli GPU Nuclear (201) 316-7155 X 1 Upper Pond Rd.

Parsippany, NJ 07054 V. M. Campe- NRC No J. E. Rosenthal NRC Yes X

Joe DelVedico NRC Copy of this mtg. summary, please CFtries Ader NRC Yes X

Roger Huston AIF (301) 654-9260 X 7101 Wisconsin Ave.

Bethesda, MD 20814 Dennis Fadden INP0 (404) 980-3219 No 1100 Circle 75 Pkwy.

Atlanta, GA 30339 J. F. Lang EPRI (415) 855-2038 Yes P.O. Box 10412 Palo Alto, CA 94303 J. A. Murphy NRC/RES X37921 Yes D. R. Muller NRC/NRR No W. C Ham House Subcommittee on Minutes Energy & Power House Annex 2 H2-331 Washington, D.C. 20515

i .

. s .

, ' INDICATE IF YOU WANT THE PREVIOUS

SUMMARY

, At:0 CHECK IF YOU WAt:T TO COMMENT ON MTG.

NAME- BUSINESS ADDRESS TELEPHONE NO.

SUMMARY

Jack Kudrick NPC/ DBL /PSB No C. R. Wright General Electric Co. No Suite 201 7910 Woodmont Avenue Bethesda, MD 20014 W. A. Smith Bechtel 7 15740 Shady Grove Rd.

Gaithersburg, MD 20877 Wayne Hodges NRC/NRR 492-7483 This mgt., Yes, No L. S. Gifford General Electric Co. 654-0011 Yes Suite 203 X 7910 Woodmont Ave.-

Bethesda, MD 20014 R. Bernero NRC No X

H. Spector New York Power Authority (914) 681-6994 Please send 125 hain Street all material White Plains, NY 10601 Jocelyn Mitchell NRC (301) 443-7983 No Yes

~

m. .

' EUC.LOSoREd i

, PROPOSEO SCHEDULE MEETING ON MARK I CONTAINMENT Opening Remarks 1:00 - 1:10 Introduction 1:10 - 1:30

-Response to 15 Questions A. Accident Sequence 1:30 - 1:50 Questions 1, 2 and 3 B. Core Melt Behavior 1:50 - 2:25 Questions 4, 5 and 6 C. Effects of Spray 2:25 - 3:00 Questions 7, 8 and 9 Break 3:00 - 3:10 D. Corium Retention 3:10 - 3:25 Question 10 E. Drywell Release Path 3:25 - 3:35 .

Question 11 F. Effect of Venting 3:35 - 3:50 Question 12 G. Cost Benefit Analysis 3:50 - 4:10 Questions 13 and 14 H. Alternatives 4:10 - 4:20 Question 15 I. Discussion 4:20 - 5:00

i .

. 2

~

l i

PROPOSED AGENDA AND DISCUSSION LEADERS *

1. What core melt accident sequences may be expected to be significant in BWRs with Mark I containments? (E. Burns)

2. Do current analyses indicate containment failure preceding core melt?...and causing core melt? (E. Burns)

3. What are the approximate time scales for significant sequences?

e.g., time to core uncovery, time to core melt, time to melt through, time to containment failure. Is this generic or very plant specific?

(R. Henry)

4. Do high pressure melts (ADS _ f ailure) have a significant effect on the physical behavict of the core melt in a BWR? (R. Henry)

5. Do current models indicate substantial differences between PWRs and BWRs in meltdown times?. . . in meltthrough times? (R. Henry)

6. Are the physical properties of the " core-on-the-floor" for a BWR expected to be significantly different than for a PWR? e.g. , thermal conductivity, viscosity, etc. (R. Henry)

7. In a typical Mark I, initiation of drywell spray before meltthrough can cover the drywell floor with up to 1 foot of water before core material begins to drop. Is the presence of such a water layer beneficial?

(R. Henry)

8. In a typical Mark I the drywell spray can distribute up to 20,000 gpm in the area outside the reactor pedestal area. If this spray is operating at the time of meltthrough, can it inhibit corium movement toward and attack of the outer wall of the drywell? Would success be proportional to water flow rate? (R. Henry)

9. Given the presence of drywell spray, would a short diversion barrier which could double or triple the path length to the outer wall significantly reduce the likelihood of liner meltthrough? (R. Henry)

10. If a substantial barrier of refractory character could be provided to hold most of the corium in the ruactor pedestal area, would this be preferred? Would attack of the reactor vessel pedestal be a significant concern? (R. Henry)

11. Is any release attenuation expected from the biological shield surrounding the Mark I drywell?...is it treated in current models? (A. Diederich)

• Discussion leader is expected to initiate discussion on the topic with a 3-5 minute statement, viewgraphs can be used. Discussion leaders may exchange topics by agreement.

d i: ..'

~

3

12. In a typical Mark I containment avai:able er practically adaptable vent paths have an effective diameter of about 10-12 inches which-is sufficient to pass water vapor at 1 to.1% times design pressure equivalent to 1-2% decay heat. What effectmon significant accident sequences can be expected if there are assured means to open this vent path? (R. Henry)

13. Calculations now available indicate that although noble gas doses can be high (see attached Figure) deliberate release of those gases appears to be better to avoid the far greater releases that might occur with an uncontrolled release. Do present models indicate that deliberate venting of noble gas activity may not be justified?

(A. Diederich)

14. To what extent could reliable containment spray alone, without venting, substantially reduce containment failure in the station blackout sequence? (R. Henry)

15. Is there any other practical change to the Mark I containment system which can significantly improve its performance in core melt?

(R. Bernero and A. Diederich) l l

I l

l i

s. e

, . _ .. .. -J..- , :/ .***- t ...L .A )

.__..___-...._]_,.'.

. . . , , , , , g m.. J _ _-.._,.1.,. g. y4 ,.u q . _

r.....

-.- _. L MTES:

- 4 _

s.....

_7_ ._.z_- 1 ) The dose values used are expected mean values adjusted ?

(,, . , _ . :_.,, . -C. p .

for a 3412 MWT BWR. Actual doses received could be '.!

8- - -

J .. . . 1, _ approximately an order of magnitude higher or as low- -

M=e

--- -. - M- - -b_ =.jsj as zero, primarily depending on meteorological and a e..... -- "

accident conditfons. ~

.___-..._x.._._=._-:=_ - .-.

2) The estimated doses from noble gases released at E l-

-- - ~

ground level or elevated assume one hour holdup and _-

2'---.s-8[ , c.g. .g-x

, 7 y

..n

_ - -:..a

3. n decay prior to release. The release is assumed to be n-

% % =: @ _ w 5 % over a five hour period. Greater delay in release can_:+

,'. : :: h:-\f :=gpq--es produce lower doses (e.g., as much as a factor of 1--

i.... p = ::;= -Mg about 30 at one mile for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> of in reactor $jl

. - - . m@h ? holdup compared to one hour). n n=.,w.m+m_-m = _.=

=-

n

, s.--.. _ .12

= . .

_=.gcf '-2

3) The direct breach of containment dose values assume W-

EE. - 2- -- ~T (a) essentfally the loss of all installed safety 5

. - - . t._ _ _ ____1 _ ;- features at the plant (b) no emergency response .-

T '%_- - actions taken, and (c} one day exposure to .:=-

w- radionuclides deposited on the ground.

, m - _q-m ,

..u.

,.--.. m . e

, i. A x. 5 c.p --g,1. . , . r .- g q .g w .. p-y_2 u ; ..s. mu. _ g. ; _

.w- .x. - _ . . . ,m_.3.n.e

.~..n . . - m .,.

,---"5mw. +_

=_-w=._,%c_

mm e h2. -

= m-=n_= _m --

.e-.- - = w.

_-=r .c. -m =_=_ .-

e,: e,2

- n

- m

- gm

=.m

,-- - = E _- E_. __ +y w -;: w=_- __; =_

r.r::= s- w = - .3E-w-m - m.,_e.- m a. . -

n .~, - x

3. %s.2 m. . . .~. s

_.n,. ,_._, ,. m

,--wa-.x - .n v.a -x . .,, _n v n.. n gm.m um

%-- -- _ m1,mf-mm , , m

• W-F J. SrE.?';:ES-F: ^ii -!!!!: -F u.=. . Err!. 4.!iC -

i

, - - -' "t *. J Ff ": '

'?:w. N E=:Ed.=3

= _-g.a=- x-m _-y c_sp p.;-.:g 'iKf7:-- .VsiT_-F-

, w-g.9 .m rg.m. = = =;

,~ " ..= = .. . . . -. _. ..= .:.- e m =_. - , =. _. i. _. _=_ +_ . _%. .;.u. . .a =.. n._ . _.. .-. _ _ _ _ _ . _ _ . .

~

"h. CORE MELT FOLLOWED BY DIRECT BREACH 0F CONTAIMMENT _-: gg(i_-

i

. m _

i 8

g 7__  :- _. - ..p

. - _ w-.- -

s i

I N. . . g .n -. m m .. ~ n ~.

.w- ..wm N'

C-Ch Mif? Vt! ' *mMd.'T. .P T_.NisEJIC, i mI'.J" a V'- ' fa ''-7', . . . . v-x

. v.- - . w .. . . w.e W g,,, fig V .-lF . , = ._t I ; i' s y**' d i i ".V4'ETr .%.FEiiEiE* "'%-E - s 7mW id-==EF. 5i: d'DE: L-U"f-N *=itE- '.t- F 4 r9+F> t.F.4 E g: =5

! I g E i: i.-C-2 3=E .' -5ET*.=m ._d W".~ri"'r N i. .'-?.iE.d i '=V

• ira-'.Yi -i y W .Fi-M -iTt-i.::!Ed -(=-i l .5V

, , , , g @m='=w e._ _ - ',~. ~ ~ f =1'- =Z T{- _ ::L C -i~E-E _2 : .- = Wi .:: -t 5 t '-

i . : ._

R .P.y iA "g %g.rf g g

.m.- _-+. .u . . . g.m s.- t s. w .. m-+ .r. . wno

,,q ,, g e'M_ ag :_pg ,,-

c .. . m m. .e.n S ,3, g,,, 7gg-w pM9 'g, , gy.-] y y f_,4 7,g:-

m n:s E-iidiiic" i' . 'EE ibr =E. E- - niif]Eh' fl' EE3:'pkril5-if:IE L""Ta'5=7:.miEd 1WsD M*-WXf1G:suil uk.dnu S. . :8- ~3: ': --'

C ~'~ ~- b*- 8-2" I E O O '"~53E i8

~

g,,

[: = -'""7-[1-N GROUND LEVE1. REl. EASE OF A06LE MSES

..&..."...,..-_..hb .

,jfj;8 (10 MET l m _ _r= .== _ _. g 1,, 1 , _ .

. . ... . . . _ _, w Wy=WannMace7 1 i6' musemat . ahm .

' e ps'm *i- F re r' --4 '4*... ..M

• m-m m ~ - -- .N.m-m7E'm  % -_f w . a_4 w- 4 , 5

,[.

y Ei4 3.4 W :EU i EMiiEi 5O -.' :a TEI  ;-nim- _ _ . _ ^'

".

• E" UFK:Em "RiMP fI 12 ".N%.P Q ,

--_ =  ;

--~~_

._ g..-__ -

-TEL"=t_m r'.:v.- :::=- 7, w 01-~~-:3 E-*w

.-._ .=.

~~

I ""-- - ~ '

g.

m-- .= .= F- ,- -' 4)

___ye

.mi  ; ,,mg ._ . i., .,

ww a q.. _;91- - g t

4 a

L E,,,EVgED RELEASE OF NOBLE GASES (100 METERS)$

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

M@Q.

~=

Mnd t

= - - . - -. - . '-4-.L.%-*-~,=-e..: ' _.n:'_-

l 3. ~

~~ ~. !. ~ '-- "

- . . _ ..~

I sg

- - -- ::2r - - '-

NininI:'-2 .": 2 Ci 1

t t *,*

.y .

--.?.C .m__-

-=w --.-

m .

l ** . -

__ - " _ -K_ ~

mw :ef;;;-

I I

DISTANCE FROM RELEASE POINT IN DOWNWIND DIRECTION'(MILES) ' N Z- . _ _ . .

.W.- _.

I 1 1 I I l

g' a u g g g I e i

ll g g g g g i (

$ I I h j [ h 7 Y $ N b l . _ -- - ... ... . -

[.

_1

'BWR OWNERS GROUP /IDCOR RESPONSE TO FIFTEEN MARK I CONTAINMENT RELATED QUESTIONS PRESENTED ~TO NRC STAFF BETHESDA, MARYLAND

. MARCH 27,- 1987 3

I U. ~ % <~w,.n-- .m em.....e .- . .. . a , . . - . , - , - - .

e .

l 2 I f

l i

INTRODUCTION 0 RESPONSES FROM BWR OWNERS GROUP AND IDCOR WORK, FACILITATED BY NUMARC CIWG 0 NUREG 1150 NOT REVIEWED O RESPONSE PREPARATION HAS DELAYED NUMARC EFFORT 1

0 CONSIDERABLE EPRI, DOE, NRC RESEARCH WORK UNDERWAY TO FURTHER UNDERSTAND CONCERNS AND TO REDUCE UNCERTAINTY

.T ..

. 3 O MOST CRITICAL ISSUES: RELATE T0: .

o PROGRESS OF CORE FAILURE o COOLING OF CORE ON THE FLOOR o CORE CONCRETE INTERACTION 0

0 SUFFICIENT WORK DONE BY IDCOR FOR INDEPENDENT PLANT EVALUATION e

e a

. +

O CONTAINMENT INTEGRITY IS RECOGNIZED BY INDUSTRY AS BEING IMPORTANT O ~NUMARC WORKING GROUP FORMED 0 COMPREHENSIVE MARK I~ EVALUATION BY BWROG (UNDERWAY)

SPECTRUM 0F CHALLENGES ,

ALTERilATE MODIFIC.ATIONS COST BENEFIT ANALYSIS k.

s

E PERFORMANCE' MEASURES 0- RISK - " BOTTOM LINE" ALLOW COMPARISONS I'DENTIFIES OUTLIERS P

O! : CONDITIONAL FAILURE

,- PLANT SPECIFIC 1 i SEQUENCE DEPENDENT COMPARIS0NS DIFFICULT 4

f i

• * *

• _, 4 -m m e-a+. ---+n

l I

O '-

1 . C MARK I - OBSERVATIONS O BWROG/CBI PRESSURE CAPABILITY ULTIMATE CAPABILITY HIGHER THAN GENERALLY ASSUMED TORUS AIRSPACE LIKELY FAILURE O NUREG-1150 INDICATES SIMILAR BLACK 0UT CDF AT MOST PLANTS 0 HEAT CAPACITY SIMILAR FOR ALL CONTAINMENT TYPES 0 ATWS IMPORTANCE DECLINING IMPROVED DESIGNS (E.G., ATWS RULE)

IMPROVED PROCEDURES (BWROG EPG's)

OPERATOR TRAINING INCREASED UNDERSTANDING 0 STUDIES GENERALLY INDICATE NO COST-BENEFICIAL IMPROVEMENTS

. _.____-_----_--_._--_--__w__--_-_ _ _ _ - _ - - - . _ _ _ - _ - - - - - - _ - _ - - _ _ - ~ - - - - - - - -

. 1 7-f.

I MARK I STUDY RESULTS 0 IDCOR-SPECTRUM 0F SEQUENCES CONTAINMENT FAILURE LATE RELEASES SMALL RISK LOW MODIFICATIONS NOT JUSTIFIED 0 NUREG-1150 FEWER SEQUENCES CONTAINMENT FAILURE VARIES -

RELEASES HIGHER RISK LOW MODIFICATIONS NOT JUSTIFIED l

8

~

CONCLUSION COMPREHENSIVE MARK I EVALUATION 0 UNDERWAY BY BWROG 0 TECHNICAL BASIS FOR NUMARC DECISIONS 1

0 COST-BENEFIT COMPARISON FOR POTENTIAL MODIFICATIONS

,m - - ..,a _ m , - .,,- . . ._.. -.m,, _ , ,, __-,

D

1. -WHAT CORE MELT ACCIDENT SEQUENCES MAY BE EXPECTED TO BE SIGNIFICANT IN BWRS WITH MARK I CONTAINMENTS?

RESPONSE

FACTS ON MARK I PLANTS 0 24 MARK I PLANTS 0 INCLUDE BWR 3 - 4 0 DIFFERENT AEs

~ '

0 DIFFERENT UTILITIES 0 CONSTRUCTED OVER 20 YEAR PERIOD ANALYSIS 0 PRAs SHOW PLANT SPECIFIC DOMINANT ACCIDENT SEQUENCES GENERIC APPLICABILITY 0 IN GENERAL DOMINANT ACCIDENT SEQUENCES ARE NOT APPLICABLE TO ALL MARK I PLANTS BECAUSE OF LARGE DIFFERENCES IN BALANCE OF PLANT AND SUPPORT SYSTEMS p g**

• io

2. DO CURRENT ANALYSES'lNDICATE CONTAINMENT FAILURE PRECEDING CORE MELT?.....AND CAUSING CORE MELT?

RESPONSE

ANALYSES o SOME PROBABILISTIC ANALYSES HAVE POSTULATED SUCH EFFECTS o TREATMENT CONSERVATIVE IN PUBLISHED PRAs o NO MECHANISTIC COUPLING OF CONTAINMENT FAILURE TO INDUCING CORE DELI.

EXISTING BNR CAPABILITY o DIVERSE COOLANT INJECTION CAPABILITY FROM MULTIPLE SOURCES o AFFORDS ASSURANCE OF CONTINUED RPV INJECTION TO PREVENT CORE MELT i

t

't

~

3. APPR0XIMATE TIME SCALES FOR SIGNIFICANT SEQUENCES? E.G.' ,,

TIME TO CORE UNC0VERY, CORE MELT, MELT THROUGH, CONTAINMENT FAILURE. IS THIS GENERIC OR VERY PLANT SPECIFIC 7

RESPONSE

APPR0XIMATE TIME SCALES:

0 TIME VARIATION LARGE FOR IMPORTANT SEQUENCES (E.G., CORE MELT 3.3 - 40 HOURS) 0 TIMING 0F EARLY MELT SEQUENCES APPR0XIMATELY SAME CORE UNC0VERY (1.1 - 2.2 HOURS)

CORE MELT START (1.4 - 3.0 HOURS)

VESSEL FAILURE (1.9 - 3.8 HOURS)

GENERIC OR PLANT SPECIFIC 0 TIMING THROUGH VESSEL FAILURE SIMILAR FOR SAME SEQUENCES 0 TIMING CAN BE PLANT SPECIFIC AND TYPE SPECIFIC E.0. ISOLATION CONDENSER BWR VS. PWR STATION BLACK 0UT

~ . _

12.

4. D0 HIGH PRESSURE MELTS (ADS FAILURE) HAVE A SIGNIFICANT EFFECT ON THE PHYSICAL BEHAVIOR OF THE CORE MELT IN A BWR?

RESPONSE

0 NO SIGNIFICANT EFFECT ON CORE MELT PROGRESSION EXPECTED FOR HIGH PRESSURE SEQUENCES COMPARED TO LOW PRESSURE SEQUENCES.

O DISTRIBUTION OF MATERIAL IN CONTAINMENT AFFECTED BY HIGH PRESSURE VESSEL FAILURE.

I G

h

' ~

-.. - _ _ . ~ , * '

.y._ ,, _ _ . ._,,. . _ _ _ __,,,,__ __._ __,___ -,_,_ ._. . _ _ _ _ ,. . . . . . ,,, . , __ _ _. . ._ , _ _ _ . . . . . _ _

'd

5. D0 CURRENT MODELS INDICATE SUBSTANTIAL DIFFERENCES BETWEEN PWRs AND BWRs IN MELTDOWN TIMES?...IN MELT THROUGH TIMES?

RESPONSE: ,

0 NO. SIGNIFICANT DIFFERENCES DO NOT EXIST. MAT'L TYPES, AMOUNTS AND FUEL DESIGN MAY CAUSE MINOR DIFFERENCES.

0 ' SEQUENCE ASSUMPTIONS CONTRIBUTE MAIN DIFFERENCES IN TIME.

9

,... .-,y+.. ,--y ww = r e , * 'y9.- w. , m - , - -. - s v - - - - - - - - - - - - -

Is

6. ARE THE PHYSICAL PROPERTIES OF THE " CORE-0N-THE-FLOOR" FOR.A BWR EXPECTED TO BE SIGNIFICANTLY DIFFERENT THAN FOR A PWR7 RESPONSE: ,

0 DEBRIS PROPERTIES WILL HAVE DIFFERENCES BUT BEHAVIOR NOT I SIGNIFICANTLY DIFFERENT, I

o

(

J

- '..--..U,[

. - - - . - - - , . , . . , . - . __ . . - - - - - - - _ _ . - , - - , - _ - - , _ . - , - - - - . - - - . _ _ - - . . - - . . - - - _ . . _ - . - - - ._,r__., .

• 85

\

7. IS THE PRESENCE OF A ONE FOOT WATER LAYER ON THE DRYWELL FLOOR BENEFICIAL?

RESPONSE

0 THE PRESENCE OF WATER WILL REDUCE:

J POTENTIAL FOR DRYWELL WALL CONTACT AIRBORN FISSION PRODUCTS THROUGH STEAM CONDENSING 0 ONE FOOT LAYER ALONE NOT SUFFICIENT - MUST BE REPLENISHED.

1 0 IDCOR ANALYSIS SHOWS:

PEAK WALL TEMPERATURE IS WELL BELOW THE STEEL MELT POINT FOLLOWING DEBRIS CONTACT WATER LAYER SUBSTANTIALLY LOWERS THE' WALL TEMPERATURE AND QUENCHES THE DEBRIS.

i a

6 k

+

D w ~ ~ ~~-e..,~. .-~-e-e -e e

n' ll,\ 1 ,

, ~

uO uooz jwI* FmW sOI

_i F

0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 1 1 8 6 4 2 0

_ - - - - 0

- 1 0

' ' 0 R 8 a

E T P K A M '

0 m0 W 1 2 4 0 O 0 1

U S

o N P a 0

6 c P

T M e

H s 1 3 ' ,

P 0 E E M D K P I 0 T S 0 '

0 I 0 4

- H 1 .

a B 2 P

=

f O o M F 5 T 0 0 P '

0 2

- - - ~ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 9 8 7 6 5 4 3 1

xdoOZ jWIM Hm$ HOI tll!l1

~

VARIOUS DEBRIS DEPTHS T Fo:2100 K T So= OOK 900-0.12m i '

1000 a .

800 -

0.09m m

!. m a .

700 -

0.06m -

800 @

z .  : a J ' i w '

600 600 W w i -

F cn i; CO

! F F 500 -

400 m co m

' m i. e

!' F

H  ! .

j F 400 -

~

b F

O 200 O i I I i 300' i 0 200 400 600 800 1000 4

i TIME, sec l .

-J g

O M.485 9 4 pp A gm e g n [ g l l, g g , g , { } } g ,

~-e- .I 17

8. CAN DRYWELL SPRAY INHIBIT CORIUM MOVEMENT TOWARD AND ATTACK OF THE OUTER WALL OF THE DRYWELL? WOULD SUCCESS BE PROPORTIONAL TO WATER FLOW RATE?

~

RESPONSE

0 PRESENCE OF WATER OVER THE DEBRIS FROM ANY INJECTION SOURCE WILL REDUCE CONTAINMENT CHALLENGE.

O A WATER AMOUNT SUFFICIENT TO REMOVE DECAY HEAT OR LARGER WOULD BE ADEQUATE.

0 ALTERNATIVE WATER SOURCES WILL ALSO REMOVE AIRBORN FISSION PRODUCTS.

1

,,-q q + --9y --n ,-%g

-g' ;t IS

9. WITH DRYWELL SPRAY, WOULD A SHORT DIVISION BARRIER WHICH COULD DOUBLE OR TRIPLE THE PATH LENGTH TO THE OUTER WALL SIGNIFICANTLY REDUCE THE LIKEllH00D OF LINER MELT THROUGH?

RESPONSE

0 BARRIERS WOULD CAUSE DEBRIS DEPTH INCREASED HEAT TRANSFER SURFACE REDUCED EXPECT N0 1MPROVEMENT: PERHAPS NEGATIVE EFFECT l

i l

1

2.c

10. WOULD HOLDING CORE DEBRIS IN REACTOR PEDESTAL BE PREFERRED TO SPREADING OVER DRYWELL FLOOR? VESSEL PEDESTAL A CONCERN?

RESPONSE

0 DRYWELL SPRAY AND FISSION PRODUCT REMOVAL EFFECTIVENESS WOULD BE REDUCED 0 HEAT REMOVAL EFFECTIVENESS'IN PEDESTAL IS MINIMlZED 0 POSSIBLE CONCENTRATED ATTACK OF CONCRETE FLOOR WOULD BE UNDESIRABLE

f: ? : ._

2. t

11. IS A RELEASE ATTENUATION EXPECTED FROM THE BIOLOGICAL SHIELD SURROUNDING THE MARK I DRYWELL7...ls IT TREATED IN CURRENT MODELS?

RESPONSE

0 BIOLOGICAL SHIELD:

ATTENUATES DIRECT SHINE ELIMINATES POSSIBLE DRYWELL FAILURE LOCATIONS MAY PROVIDE SMALL AMOUNT OF FISSION PRODUCT REMOVAL IF DRYWELL FAILURE OCCURRED-0 NO CREDIT CURRENTLY TAKEN 0 .N0 QUANTIFICATION AND CREDIT FOR POTENTIAL BENEFIT ARE PLANNED

-r-.,.

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

~

L 7L

12. WHAT EFFECT ON SIGNIFICANT ACCIDENT SEQUENCES CAN BE EXPECTED IF RELIABLE MARK I VENTING IS UTILIZED

RESPONSE

0 VENTING THROUGH THE MARK I CONTAINMENT WETWELL PREVENTS UNCONTROLLABLE' RELEASES FROM CONTAINMENT FAILURE REDUCES RELEASE TO NOBLE GASES ESTABLISHES CONTAINMENT HEAT REMOVAL PATH IN SOME SEQUENCES 0 VENTING SIZED FOR DECAY HEAT REMOVAL QNLY CAN BE USED AS A LAST RES0RT TO PREVENT LOSS OF CONTAINMENT FUNCTION, l.

l

=

2. 3 '  !

13. DO PRESENT-MODELS INDICATE THAT DELIBERATE VENTING 0F NOBLE GAS ,,

ACTIVITY MAY NOT BE JUSTIFIED?

RESPONSE

0 VENTING AS A LAST RESORT CAN REDUCE THE RISK IMPACTS OF SOME SEQUENCES 0 A NUMBER OF NEGATIVE EFFECTS MUST BE CONSIDERED FOR VENTING 0 1DCOR AND OTHER STUDIES HAVE SHOWN THAT SIGNIFICANT VENTING MODIFICATIONS ARE NOT COST BENEFICIAL.

l' l

.-_..L..--._, - _ - . - . . _ . . . . . . - _ . - - . . . - . __ - . , - - . -

24 -

14, T0 WHAT EXTENT COULD RELIABLE CONTAINMENT SPRAY ALONE, WITHOUT VENTING, SUBSTANTIALLY REDUCE CONTAINMENT FAILURE IN THE STATION BLACK 0UT SEQUENCE?

RESPONSE

0 WATER PROVIDED CAN COOL THE DEBRIS AND SUBSTANTIALLY DELAY CONTAINMENT FAILURE.

'O EVENTUALLY CONTAINMENT-HEAT REMOVAL IS NEEDED.

0 FISSION PRODUCT REMOVAL-SUBSTANTIAL.

l 1

-*V: *.: *e *W e h 4+NM '_ y g._ ,

15. IS THERE ANY OTHER PRACTICAL CHANGE TO THE MARK 1 CONTAINMENT SYSTEM WHICH CAN SIGNIFICANTLY IMPROVE ITS PERFORMANCE IN CORE MELT?

~

RESPONSE

0 NUMARC IS CURRENTLY EVALUATING CONTAINMENT PERFORMANCE ISSUES 0 THIS REVIEW INCLUDES:

BASIS FOR CONTAINMENT PERFORMANCE B,$RNERO PROPOSED MODIFICATIONS ALTERNATIVE MODIFICATIONS 0 CURRENTLY NO COST BENEFICIAL MODIFICATIONS IDEilTIFIED BUT STUDY NOT COMPLETE

A A --

g . h e '

4 3L

. 1

~

1 I s-h ~

l .9 l 83.

7 3

'A t ,

Ms '

1 g

~y n

( n' i

I

\:3

,) ,

r ~

i d'

Ef  !

?Ml J

,a 49 ,

m--

l

, , e ,  : s bEe h

l wN r 8 i _-

mg p

• o .

I f

• 9' 9 9 y w, - . , + + . -

.,e- y, .

.,,-,,,-,..,.--c...4.- f w , m,--.--,-y.- -rw.,_+,m-.,%--e-.---,.,-e-,,-----,---..-----..r--.-v.-.e.

e. e 4

13 COO.

l II l I l l l l l l l l I I l I I l l l l l l l l l g g g g ]' '

pp gw r> v . --

2 900. . _ _- --__

0 4-

. i 3 - V.

n -n , . .

j -_.

t l SLUw=zu c 1~a te l~N t.

r kN'c f. 'FcO) u *: 'h 76t'e toO.

e . c h u-ae r* s ac e,ca u w rc 2 i I tr. 0.

• [\ V J

{ p. .- > uA l 81s'

.l I 'I !I l l l l l I l l l l l l l l l l l l l l l l l l l l l l l l g g l g ; g g g

- i

• t 0 0.0 1.0 2.0 3.g 4.0 8.0 e.s t

t 4

y, .g hI=6ed keewA e'.,)gJfgg ,g g gr ,p j c * *, r p ,- +

. ,. __A_ _ _. _ _ _ _ . . _ _ _ . _ . _.. , . _ . .

_.~.

e- *

- 34 1

s  % m,

, , _a Q M

- I U Clk l

\ -

a

)

e 1 e -

. I,

,- J t- \ l 8

( '

-L%l o\

N 5

e

. y I.I 5 ll !! - L ,esa m

. AH44 '

g

~

w *wd I

l o, d i

F j i

W- m-5 O

J @

l 9, -

i 8 I _

l n

! wo l

S k -

k.)

- ~ ~ ---

,.- e .--.d,. ,

b, ,,-.,.,,.,-[, s- _, - , , , _,[.,_,7d...,,._[., ._ ,, ,.,_7,,_- . _ __ .__ ,_,.m,,m.,_,..__,,,,_,,_.,_,,,m,,,,,9w,

15 TMI-2 ,,

DEBRIS COOLABILITY IN THE LOWER PLENUM Mass of Material - 20,000 kg Planar Area ~ 12 m 2 0

w/m 2' Quenching Heat Flux ~ 3.4 x 10 Debris initial Ternperature ~

2OOOK -

mc p (T - T sat}

60 = q/A - A j A O :: 600 secs = 10 min 1 .

i

~.e..

O

____gg -=

4 6-G

.- -- y,- -,v.--.m~.._,_,m.__,,_,. , , , , , , , _, _ _

' ~ '

.: .: i-3<.

CONCLUSION COMPREHENSIVE MARK I EVALUATION

~

0 UNDERWAY BY BWROG 0 TECHNICAL BASIS FOR NUMARC DECISIONS

-0 COST-BENEFIT COMPARISON FOR POTENTIAL MODIFICATIONS O

i i

I

, ,,_.-.L_,-..'--.,,~,a,-,-,.- 0,.~,_,w,l_?l,,. ,,__ . . , , . . , , . . . . , . . _ . . , - . . . , , . , . , ~ _ , , - - - _ - , , , , . _ , , , , , . , . . -

A-n EM C LOC,,C i< E i- .

?;27 mWW~ - -my'Q v

y. Q4.c ~ m .:.2. *~.. n. . r. - ---

Gh'

'( 5;M... :st%'. m m. m.,:.m?

1**? .W*%f?. 7 ]i *b.-t " '* -

'c@;Q;cr .N g w ..y . ;

,,p& ;

. : : 6 yg,W.s . "W..-; ; 4.'5-Q*WNw* vp%q e

y.1 x..... m

.ws..s; , g;e,lv.  :- :3 g . ~ . . , . ., -, ~p . , ,

I *' - -

.. .b * ! h e  ;. . '8' I 7.

k,,e.A .. I k -r

, . A, - ,r. '5*#e210 ,-

4." ' % --

..._a..-

~l'

~

? h ."~ i w &*.4 s : - i:- ' n Y ?. "> .

'~2h & &'.*' W >"Q!}%

- . '"#,'~.".m

w. ;: 5. ;.,y. .

__ I $ b"'$

~.g J'*.N[-h,..y,-. 'p. , 'nt., .

l

'** rv h. .' .'ga I

c v s. .'.. .

q

• • I -

i" ..s yr'*;; .

f.~gy . ..j,y: . . . " M sh.'- i ..: %

!' in i - 7F?.yj I,:? ,- . N.: E%<.,5.fr-Q.g.d g.

sM

! 1

-'., s.. ,

I

w., .. -  :

v

'%jf,i.,;, C, l

c

'- w _ , [.

s e .;r .. . ;.

l

~

F w  :

f ,

' y,g '=:d - [ ;qs - 2- r : : 4.

,a.4f.m&w%5%iWih%ylW . .._M.k.Yk..:. ~

b.

n.. L  ; ,

Mm ., ... . s y. m* .

9.t'f'O e 4 w.

t dsW-

,r! ?

g, iff.7

~

'O%;D ~

4'. .c * -..-

2 b .J. d, , b .g,y!,ih'j%,y.

i.,{ . j

$#~$ " 3 2.,t.q ,, y 3g ..s r,n,yy,gpsg,3,gy,p.g.i.

e em ,,

m..

.c .e e" <  ; .r; _

7 .. ..

M* .. .,*

l I .

dH c1M

1. Cth&

Q'~~ .

y i v 31.-i,pt

~

M d e, w ',.- Mesg.g@s% ' .

,. e. ..

.* g,,,3 w p.

.M~ k iN.

w. .b M . .

E.$dw$ , .$.'

.m F p@.f 9.gpy .- p%

a"' T 1 ~

~

'?

_. .A- ,

2,.

l _

{y_ - - } V -

lx . %f.;. + '?. <.~,l .,h. fy .V $l.

?*

.. p w.

T/

--' - 1 i. Genera (nodalig 1 -4 ' ' ,4.,

l , .

g p _ -: : y r~>g '

. k E..>M,5 bdr. ;.-?-

l Mh y  % M, v ' i'.:r 7-q 7_

_ l T

h = -

%vl.c (hCl,'W W' -iT.N u."ff{$Nh,

e. .

1 -

k 'l S y f. . . &[ &k. . :.,... va..555 y{fs.. 'sf+ ".'$

y'&g

. , , ' .-. y

' yk .9,7,7.:y),q?,:.9 y..n, .

VM. .:iTu;:Mn

.y M ,.:. r,. ,

' ,.?

. f.s ,

. ,p 9.y, ..y .. %g * .,.' 7,,

Joy. ,,$

ip.E!r*

ry' y ,, g

', [, . g,.q&'i;3.

e

.y .

' 7 " *ys:

fi . 4.

.s

'r.lp!&E$'.M:Ch? 9.-ti {a.K %...

i h- ~' i  : re@.' sit . %T,- -

Ml >n h k Qf. A.hj?p'.Y:,b. i r W U ~' .T'*' - -%

w.aw M.s p n M MwWew9.w:M w . .?

xaw MntbN.w v

w.m. n; . . .m M .u y f.if M ' Mi W %.i%%fy. 3 %'MQ k l!MQChpTT'&:~

2.

e ,w..,

% , _ id V

MU..,,,

~

%dR'M.:+-.l.T t.'f, W F ; . *.<.LQ .Q.u.*Q*

..::M. - Q}? .. &r%-2. %*4&.yQft:f;IhQ

$,y?.l.y%pp .s.n. . . 'r. 70$u

.y 6 y m :

5 hf;,.,y.*[4,?A.W.

. t. n . ;..

e qp  ;;p.

~ .: p',5 P 0. ., M6*1..;

.w. u..g z.:

-u., &. 3.8,F 7M. w +;. . . , W. W..,. %w ,. m:n,..Q r r 4%. . .s

.,%.A,, . w.w.

.w,M. .

. rs.4.. rg..._ s .. . a. > .-e.v.: , e , m. . . %.-

m. , . v. c.c. 2r. ..,,........:

. ,7 .. .

.s.^'

1. ,,  :..

. . .' C- - . ;m? ...,p. ..r.m

(; *

. us

• r- . hg.,

*d dir' [s.-;'l @." W . ,

. . . - f.p ,s h .E' '/ .1. ' *

, V *.

(.'l:. , .}, ; y t-

'.**r..ic,1

9%fw.' y a gk & ) y'j..d".- "fra., - - :.r.

y ' 8 .

k l.4-htihl;%.g~"

. 4. _ .

1' ~ 2 ~ uTY' ?g&. .k.s.).ig y . .

~_

, sy .  % ?, gy  ; ,.

y$.* %g:.gyg?2.g'aStY,' .=ms?q:-

~

f~ e  :,n.

. .. ;f4 . , .c3 y* f 4.$. p p: *fg T y ; . .

..&..: [?-y!.&

.e  !- .

,,....c.

^

e:;- w a'

. .J- . .

5 .'-' h, *-

_-- - b

• SQ , $. -  ?,%; '

k

^

m-an>+> a

+ w- ._

g s

N_

e 4,",k' S hh

. .s '

E 5

- [.

c 1

. e&

)-

%- ggg. ggggg ,g - . - - -

m.

-... h;c$.

y --.Y o

h [ it itI " ~ -

.C $ 'e?' $ [ g u .. k J M 2 Ei!!B u E i t GE 4 s a EH s . . . .

i % JDallf

gf.geMi -e S g g + E m m s y s

• g

• sg!M W$5EE* ? 4BHE5P E e ' :. ~ _ @$F _

W . g e'"

N Sx M $ $ W J I @n.B fE S, 7 " ,* 8 M ft6 .,1 m -

.u WF I4'dQ.' .

• • h{

,. hy ,.'. o.

p .

. . s ... t; f+ )4.o.lh '.. . . .

g g. . , .. ' " ll -n -

.,f.:Q; ,c.. .,:.?

r,i;*:e *

,s I

i

p. vr* y',. ; y' q';j '.l'i'.f (o. g. ?%

l ,%y* .p

. . 4 g.y . v4.% ?.

h . .',Y k g , M #bi ,

. . 3 .

ffkN 'k 5. 5f' '

WDt%nw . ..

.. .W. ,, U~. V:-844W. - ~-

g&r ,M.i.f.lj.Q' " g g..' ysf.. .

~

g. c.. . .pg ,. .

pp.x ~& . .

jg

= +e; 3. . n .

c 2 w%.

. w ,

n. a,.y.gg%w.4:.w%yc+k;'{te.

44 .

yyrc .

w+ .

,x '$gg.

r 3 ..

ffp?c..y..:a;a.$

p:

, w ,..

n b .i'r'. V.

,'yf l Y. .q. ,.. w'.hh "y.?.v..v-: ' ,".p&u;y&,1iif55n

• 5 f.f.' f u.8{ h*kY.m$'\bN'b$4&f.._;.;y .>w,., y. s9 a

[

&g r. .c q.g

., .

. . . . x . .2- v. y- a,e ~ &*ir. h

.s .'- . .n : > m .. .s .

1 N. , y. ' ;

I

.

• c ") - . .y . ,.t*.- -' h,;.a.s;e.... X. 3M'i

. . , . . w,:l'p,* --

4 , .t ..e,if;...- ,.w .uy . . g.,d. [ [p M. M.. -b5.Y,,$.i. u N.g:'.i.k , : .3 L .fb. ..,... ....,,-,.,-- ; ,* d * *. '

s( ... .

.k.'.i.

t..,. ..

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

l'

t

~.--e. +--- ---% .-

b .~ .,

r g, -

y , . . .

,,...,.7y7.,, . ... _ ,...,

,. ) * *Y$tW* . ,.. .,&.'- *.*. . , (V5lA H

, ~~

.l. *.lelr);.' -f';i$N.'.~ . '

~ xn ~. .vy ,s .n-

.~. *, ~ &.g'.','N-&.s.,...

f. .!N.;r,%. v.s .*., .ll,X.'yl.*i*%'".Qyt  ; ..K.%*j'.&. .~ n;,G:  ?, , W.?< <.. *-i;.'eT.s$.d.$w, k.

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

,t,?. g..d.

<..,- ' ' N .,Q~ r*

• ps., . f .(*,9f. 4? 't.

n. . *M, , .f T .1,t,W;.y M W,$,

• e s, ., s. s.

'*s *-. -

'f* 2.y M.*. .J 1 2 , .W.

• . . . . \.

jl ~ ' '..' .. ~* ' 3l:; b*N,~ . .?.jifg, .

. .:.; _ , f,h,n:y,&s

'\.yf,,{-j.-Q;&*y ' . * .h'O.

$f;.'.'. . '.* -lQQ}[?. l.N[.g. .; . ,

• ,* V. . .. , , . Q . -x.+= ~ '

= <

. . . . J v:, i ,a,. t. . g~ .r. . .M- * *. .s %

..p. *.f.4 N  ?

e, .,L'-

k, *y J,'-b. (,2

, .s

[1, .[ . ' w.y*.. *1' $*

. fs Q $'*c.* .

, s ,,.

h*( * " . M - ' 'f . pj [.r.~ ~, 4 3. "h

, }'

, k.r* *'e,,'

• ~~ ~

'MJM~

%*ha $ '* .h*./ ** > ' 3* . ' ' -' .2Nfr.M. ; & l'

• 4. $ c* . E i -

' ...i,d.Q.S**d h E.Y; ?.M h,Ir- ,' *-I.;,

i

't"*~h.i.m, ' .-

4*O hS i'e-$ '*.*.. -

4.*hehq;h_k'U- !k. gT. ,.

g M. ENh. .f ' - , cp-Pd,$.(.$[E..'G..  ; ,p .g M,.....h.# '}.S*[$th "

H

~ i. h.; -L I.O' M +Mi%s,IN s. ,A N h.'

Rg C. y; .,l

'~~=" tp

• r u... s **"T * *

~

. ... ~ W*

• Ms =RE = , . . L' + ,,

. ..,c.

x

.s

$se w.a p ,>

- = ~

g - . yst &,

. .: ~~ m.,. - . , . *

+

9  :-

ns.

. w. -

... ., s .. ..

y.[.;?

. f, T +' ' '

W'

• ' . h ;.& . .4

%.c. .aa:% T.$ '.~ d,o. .a'..h. '  % ' ,. ' -%9k *n.,M/

y'h.p. h. *.:"s. ',k'.'. . .!N.Q 4'$ , ,. .

• f.f. W. $OiY ~.n M)d, akf[*,~kNM,f., Efd.?g?*b?* h.fi$.Y,h .w'--

$_"Oh,  !"3.I . d.*e *EI'l r. '((.iZ

. j.

M ; W e.;r 4 .

~

• . - A .. ;

p'M M.4#W9J@lMYI ., [*~,f 'M .;b6 s- Y .,. c- k. . .I.fis rr, -= = b.

W , ..=g: i <,R' f .N' y%' S; ac.,

. . .s ;' -

,? .-Y..

s..<q,".,p a w k M*.cy y.g, >r+..j he t s' r.n v .-!. 3,6'.iW+Qj  :.~ '

.. , /*e-

,"=' N. m' .,L. g~*,

L*48 m M.w .,

gs&

s ., .. m -W . ,. q .

• Nr

% * 'g!

eis:= :~s.h%

f . . :' : . .v. 4 t: s 1 L ,,.~~f.-]: >

N*T , .:. ..

N

,f f r y

• f

=

aweg y - --- - ,

y' w*g . s.  ;,

,n - y ..

gp.,

L 7

.. g,.; y o ^y, 4 ,, gokl;,g,.tp.h.. . v .. O ~

V. , ..

..,3*-'u. '

A .. . .i . ) .7,

~g i

-9 Y'  !,, $ D (a 85 y

s

, 'J 3 . . ,

,..M' -t , 6 .. . + . . a .,. T; '$' <* %s.. y . .,, 1r 3 *h '?. . ta ',t..

, **?

s P. c%a ' . .

,e,. : -p,  ; g" M ,,

, , a > .- - e e- .re . v.; R .

c' r <~ *- ? @.V '

n.yLh.y[gA.,::i. $h.M7 g :. ~

. M I. N ' b -

'SM- .i*, .,,.

E -

• dhs w.#Jh.e $ kNh. ~' #

- s h.. 4 b b g$P7 -? %~ ..

u. ;[m@w 96

! d(,[. Ih ,,,, f.. . . ..

l p,.'& ^*.' .

, .,..$ pj y ,,;* L*.'l-g ,.

k-yg, 3, . *

• 1 *g v

W' we MMMM. A.;

pe a. - w wt A.4%.dM;.'?1%Ihi 1 JimiMid esemme w . .n. .

A$$hMe!W e s

n .,v. . .

, w. w w 'd '% y..,,**; $ c.

I 4 .. . w.k 5'[.s '%c 4h 4

+. u . .c .

. .n , gry.w.4.&.p*$m,e.s.

f'& % & t :. .*

' l 9- . . . 0 -l.W:'l ; ,': .f .' 41*Cf.'""y . WI:-@ " *- '

/ $p ' *

-l,,'.,. -

r..,  ; .3- . p,%,. r. j ,'. A. ~f:

.p.',pt.,c.3,M.M'?.@ef

.r - y , g,<. W - pg a..

. t, .. , , , ,

}.,k. '..'l; f. '.

h;. y 's ;o3

,' *l.1.

.a;.w h,% sh.3 ,w ,k. :..# 'j ' ,,:g

<l . *:~

.-,2;',

i . . , .. ' .

. - .;!iEl*g

. . . .q . .e-t

-nl'.'g.. ;. .u n .. -

., *y , ,

.s

[

a; *4, *

,r y . .+ y : % . ,. * . 's *

t. '. - .'*,.~

'EM C.Lo SU R E 5 t

Key Legend CLASS EXAMPLE DESCRIPTION EXMPLE I - Loss of Inventory TQUX,TQUV

- Station Blackout II Loss of Containment Heat Removal TW III LOCAs AV IV ATWS TC V LOCA Outside Containment A out V

. . .s . . ....

2-T-157-002

BROWNS FERRY ..

IREP - -

~ -

~4 e

. U1 N. .

8 eo ................

I e .. . . . . .............................

\- ... . .. . . . .................................

}

h g ................................................

I l

4, .

,I ........................

i CLASS I (100%)

NOTE: Most of the Class I Events are in fact induced by postulated failure b

[

of the containment Heat Removal system. The analysis had unique scope and boundary limitations that may have limited the problem diversity.

4 i Figure 10 SuharyoftheContributorstoCoreMeltFrequencyfor.BrownsPerry. .

Totpl Core Melt Frequency = 2.0 E-4/yr.

. . SHOREHAM LONG ISLAND LIGHTING COMPANT IPE e=*

v.

N.

8 ro CLASS IV (8.0%)

s ...........

CLASS II 8.0%

................. CLASS III (2.0%)

class I (es.0%) -- -

Figure 9 Susumajry of Contributors to Core Melt Frequency for Shoreham f from LILCO. Total Frequency == 8.5 E-5/yr.

s

J .,; s. . . --

ENCLOSURE 1[s INTERNATIONAL April 10, 1987 Mr. L. G. Hulman, Chief Plant Systems Branch Division of BWR Licensing Nuclear Regulatory Commission Phillips Building Washington, DC 20555

Dear Mr. Holman:

DRAFT

SUMMARY

OF MARCH 27. 1987 MEETING WITH BWR OWNERS GROUP /IDCOR ON MARK I CONTAINMENTS I have reviewed your draft summary of the subject meeting. My comments on your summary are as follows:

1) In the second paragraph of page 2 you indicate that R. Henry presented the responses to questions 3 through 10. You immediately observe that the conclusions were in large measure based upon evaluation of heat transfer in which the containment shall is not postulated to fail. This is misleading. The heat transfer model of the steel shell has no' influence on the answers to questions 4 through 6, 9 and 10 and very little influence on the response to questions 3 and 8.

2) Your note in parentheses in the summary to question 6 is not quite correct. Your note Laplies that the conclusions are based on the containment shall heat transfer model. This is an incorrect implication.

3) The note in parentheses in the summary response to question 7 refers to the IDCOR heat transfer model as an " assumption". This is misleading. The heat transfar evaluation should be referred to as a model. While certain assumptions are made with any modeling of physical processes, the heat transfer model of the debris-containment wall interface is a simple application of heat transfer laws. The principal assumption present in this model is that debris is molten forming a pool in good heat transfer contact with the shell. Such an assumption represents a worse case condition for evaluating the melt through of the containment wall.

4) The summary provided to question 10 should state that a debris barrier to contain debris in the pedestal area would be detrimental. No discussion of usefulness should be made.

Regional O!! ice

$75 Cak Ridge Tumptke . Oak Ridge. Tennessee 37830 615 4013300

. . . . _ . = ~. _ .

'I INTERNATIONAL TECHNOLOGY CORPORATION Page 2 . l

' Mr. L. G. Hulman 1 i Thank you for the opportunity to comment on your summary report. Please call me at (615) 481-3300, if you need to discuss any of my comments.

Sincerely, I James C. Carter IDCOR Project Manager JCC ks cc: A. Buhl A. Deiderich (PECO)

M. Fontana

R. Henry G. Hughes (ERIN)

J. Raulston NO 0487-040 1

l 1

1 l

1 I

I l

l l

. .. . . . _ . . . . j

PHILADELPHIA ELECTRIC COMPANY -

3 2301 MARKET STREET

~

P.O. BOX 8699 .

PHILADELPHIA. PA.19101

\PR 101987 Mr. L. G. Hultran, Chief Plant Systems Branch Olvision of BWR Licensing U.S. Nuclear Regulatory Conmission Washington, DC 20555

SUBJECT:

Oraft Sunnary of March 27, 1987 Meeting on Mark I Containments

Dear Mr. Hulman:

Attached are my'conments on the subject meeting sunnary. These are in the fonn of mark-ups on your letter of March 31, 1987.

I have limited my conments to areas which I presented.

Sincerely, c-A. Richard Olederich

, Supervising Engineer Environmental Branch ARD/cb/04108701 Attachnent i

k

gb8 AI +

1. 'o UNITED STATES

- Ig l' e '

NUCLEAR REGULATORY COMMISSION WASHINGTON,0. C. 20555

,I

,/ MAR S 11967 MEMORANDUM T0: DISTRIBUTION FROM: L. G. Hulman, Chief

, Plant Systems Branch Division of BWR Licensing

SUBJECT:

ORAFT

SUMMARY

OF MARCH 27, 1987 MEETING WITH BWR OWNERS GROUP /IDCOR ON MARK I CONTAINMENTS This draft is being furnished to those participants in the meeting that requested the opportunity to coment on the sumary. Please provide' coments, including any. supplemental material to be incorporated into the final sumary, to reach the undersigned by April 10, 1987.

The meeting was opened by Messrs. Denton and Bernero, who discussed the background. A previous meeting with representatives of the research comunity was referenced. The sumary of that meeting was identified as available through the Public Document Room. A copy of that sumary is enclosed for those meeting attendees that se requested. Enclosure 1 is the attendance list for the meeting. Enclosure 2 contains the proposed meeting schedule and Ifsts the 15 questions.

V. Boyer, Philadelphia Electric Co. (PEco), indicated that the Owners Group /IDCOR were requested to respond to the 15 questions. The responses were coordinated through the NUMARC Containment Issues Working Group of which he is chairman.

He indicated that other NUMARC efforts were being delayed to respond to the request for information on the 15 questions, and that the NUMARC working group draft report to the steering comittee was not expected until mid-May as a result. He indicateo that the IDCOR (Industry Degraded Core Rulemaking) effort was going out of business. He then introduced the responses, sumarized his views on the most critical issues and information available (Enclosure 3, p2-4). The critical issues identified were 1) the progress of core failure,

2) cooling of a core on the floor, and 3) core concrete interaction. f

,R. de rich, PECO, described the industry evaluations (Enclosure 3, p 5-8).

l He indicated that they were evaluating both overall risks (referred to as bottom line), and conditional failures. He indicated their conclusion that

1. 'pNg6

%. ewe sequence and plant dependent 3 He also stated a conclusion that the Chipto Bridge and Iron study is indic ing that the ultimate MK I fcapability is' higher than generally assumed, and hat the torus airspace is the most likely failure location. He compared he IOCOR and NUREG-1150 efforts, including the conclusions from both } at modifications were not justified.

He concluded with a sumary that in 1cated the NUMARC working group is studying MK I containments, that he believed sufficient technical bases exist for NUMARC to make decisions, and that cost /b nefit comparisons will be ma of potential

' modifications. He indicated studi s to date have shown no modific ons to be beneficial. .

! coymnoemL Feat ogg a hl4 Kin 6 er b Ppic.ub7" To

(

i CornPhile DuPFeeeNr PUWT5 l . s u 4 MgANW Wut u;Av.

E. Burns Delian Corp...discusse'd the responses to questions 1 and 2 (Enclosure 3, pg-10). He indicated there were four or five PRA's for MK I plants available that indicate no specific accident type dominate for all MK I's. He, therefore, concluded that the spectrum of potential sequences was important. He also concluded that there was no mechanistic coupling of containment failure to inducing coremelt. (See Enclosure 5)

R. Henry, FAI, discussed the responses to questions 3 through 10 (Enclosure 3, p 11-20); The conclusions presented were in large measure based upon evaluations of heat transfer in which the containment shell was not postulated to fail by perforation (Enclosure 4). This evaluation was noted as significantly different from those of the NRC staff and contractors. The significant points of his analysis are: 1) 12 Cm debris bed depth, 2) water above the debris bed acts as a heat sink with nucleate boiling at the Shell surface, 3) concrete below acts as a heat sink, and 4) debris bed assumed to be near the melt temperature. His other main points were:

(04) high pressure melts have no significant effect on core melt progression, but the distribution of material in the containment is influenced; i

(QS) there are no significant differences between BWRs and PWRs in meltdown or melt through times; (Q6) the debris properties of a " core-on-the-floor" are different, but the behavior is not. BWR's would have more metal with less oxidation. (Note that predicted behavior is in large measure a i functior. of heat transfer modeling - see above);

(07) water on the drywell floor is beneficial, but requires replenish-ment. (Note again that the IDCOR heat transfer assumption results in no prediction of steel containment or downcomer melt through);

(08) drywell spray would reduce containment challenge, sufficient water to remove decay heat would be adequate, and sprays can help remove airborne fission products. Spray rates in the range of 500 - 1500 l

gpm appear adequate. Enclosure 4 was again referred to for, a discussion of heat transfer and related conduction. It was noted that the IDCOR heat transfer methodology was included in submittals to the staff, but little feedback had resulted;

! (09) a debris barrier would not be useful, and could result in negative effects; and l

(Q10) a debris barrier to contain debris in the pedestal area under the i

vessel was not considered useful.

l h dqwd

11. Fe indicated no analysis was made of the gap

/

R. Depfrich discussed between the et:k.~nt and the biological shield. However, if the :: td..,. mat d y d were breached, fission products, r:d ^^* "*^ "'" "" "-- through pene-trations. (Se Enclosure 3, pg 21) l 7 f So Af. hyye TW'PP8D NTil8 gap ew ws 'lWrhM TV W N *

~6ng ,eex seeQ. '

r

.g ., .

9 E. Burns discussed venting (Q 12). He indicated venting was a means of prevent-ing uncontrollec releases and establishing a heat removal path as a last resort.

Further, venting can be used to prevent coremelts in such sequences as TW.

However, he indicated large costs were not justified' generally, but plant specific analyses may indicate differently. (See Enclosure 3, p 22)

. /

R. erich discussed noble gas venting (Q 13). He indicated such venting as a la t resort can reduce the impacts of some sequences, but that negative effects-must be considered. (See Enclosure 3, p 23) He preserged a back.up slide which /

showed substantial reduction in dose if u ntin of nobel gases is delayed.

ReteMd 0

-R. Henry discussed the use of containment sprays for station blackout sequences in response to Q 14. He indicated several benefits (debris cooling, delay of containment failure, and fission product removal), but eventually containment heat removal is required. (See Enclosure 3, pg 24). He also discussed debris coolability referring to pages 25-35 of Enclosure 3 using inferences from TMI, experimental evidence and analytical assessments. Analogies were also made to debris coolability in coal fired power plants and experience in the steel industry with electric furnaces by several participants.

6 R. Okt'erich discussed Q 15 (See Enclosure 3, p 25). He indicated that the NUMARC <

evaluation is not complete, but that to date no cost beneficial modifications has been identified.

R. Bernero asked whether modifications such as a more reliable ADS system cculd help. R. Henry indicated he did not consider such modifications cost bene ficit.1.

The issue of steel shell perforation was again raised. R. Henry egain sumarized the IDCOP. view that the carbon steel and heat transfer capabilities precluded such as occurrence.

4 V. Boyer concluded by indicating the NUMARC working group report was expected in mid-May, followed by a review by a supervising technical committee. He indicated no firm dates had been established for briefing the Commission or the staff.

~

ef Plant Systems Branch Division of BWR Licensing

Enclosures:

As stated

cc w/ enclosures:

H. Denton T. Murley -

E. Beckjord T. Speis D. Ross R. Bernero .

- . .._n, .,, -... . , . , , , . . - - . , , - - - , ,--. , . ..-.we - . . - - -, ,

- (p *.n s . '.

w:n

,p3j

• g j .3j.a BROOKHAVEN NATIONAL LABORATORY.

I?l 9 [l El A A A a ::A ASSOCIATED UNIVERSITIES, INC.

. Upton. Long Isfond. New York 11973

. (516) 282s Department of Nuclect Energy . FTS 666 ' 2296 April 13,1987 P

< l_ t A .Mr. L. G. Hulman N-007 Severe Accident Issues Branch U.S. Nuclear Regulatory Commission Washington, DC 20555

Dear Jerry:

Here is the assessment we did concerning the IDCOR-IPE methodology for BWR's as reported in FAI-86/1. Specifically, this part of the assessment examined the Mark I steel liner response to contact with core debris. Let me remind you that this is somewhat " anonymous" due to the administrative turf battles I alluded to over the phone.

Should you need any follow-up action or wish to discuss this, please give ma a call (FTS-666-2296).

4 Sincerely, t* ,

G. A. Greene Experimental Modeling Group

! kb Enc.

i

/.

i,i

' ' ~ " ' ' '

s l

oD s *, '

,~ ( ,

P Thermal Res >onse of Mark I BWR Steel Containment Shell When Contacted )y Core Debris During Severe Accident Condit!i'ons i

. i r

) .

It is' the stated objective of. the IDCOR-IPE , program and the NRC ' Severe Accident Policy Statement to ascertain if there are 'any potential risk ,out-liers with respect to core-melt frequency or unusual containment vulnerabili-ties.- ' One such containment vulnerability has been identified for the- Mark I -

BWR~ containment steel liner, and an analysis of the potential for liner melt-through has been published [1]. Primarily on the basis of Reference 1, the failure of the Mark. I liner when contacted by . core debris following vessel failure was included by the SARRP program in the NUREG-1150 source term anal-yses [2]. _ An average of the eight SARRP analysts' estimates of liner failure probability upon contact with core debris is shown below.

Postulated Accident Conditions Probability of Liner Failure 4

High pressure vessel failure 83%

Low preisure vessel failure', dry floor 76%

Low pressure vessel failure, wet floor 61%

+

.Trie IDCOR analysis in the' draft report " Approximate Source Term Method-ology . for . Boiling Water Reactors (FAI/86-1)" [3] recognized this potential

containment failure mode and ~ reexamined the liner vulnerability or survivabil-

. . ity in a separate a'nalysis. In what was. characterized to be' a " conservative" analysis, the report indicated ' that the- steel containment . liner would not fail under any of the postulated conditions. This conclusion is in disagree-ment with the analyses presented in Reference 1, as well as with the contain-i ment event tree issues in _SARRP for the Mark I containment analyses. As such, the models and assumptions inherent in the 10COR analyses will be assessed.

The IDCOR analysis of the behavior of the Mark I containment shell was bised, upon numerous assumptions and judgements. It is on the basis of these l' assumptions and judgements that the initial and boundary conditions, physical

~ -properties, and phenomenological models were developed. Those assumptions i

' that could be-identified from the text in Reference 3 are discussed below:

. 10COR Model Assumptions

, -(a) The core debris that escapes the pedestal region of the drywell is as-

! sumed to be in a thin layer 6-12 cm deep and to be,'by definition, solid-l- -

,<ifled [4]. This debris, for the purpose of the analysis in Reference 3, i y sis assumed to consist only of-uranium. oxide fuel.

l-l' r (b)- Heat transfer within the core debris is assumed to be by conduction l<' '

a v

,nnly. There .is sno allowance for internal convective processes.

., L

$^. (c) Heat generation within the core'debrfy is by decay power heating. There

[ are no provisions for the chemicil erergy source resulti'ng from metal-gas

! -phase reactions between concrett decomposition gases and metallic core debris.

L

/

^

L ',_f c L_ _ _ _ . _ . _ .,. % . _ ..._ _ _ .

^ ^ ^ ^^ - ~

~ ' ~ '

w .w.  :.. - . .-

o

• h '.

- 2- D (d) A pool of water overlying the core debris is assumed to boil at the cri-tical heat flux. The film boiling regime is not modeled.

(e) The steel liner is modele'd to transfer heat from its . outer surface by

~

thermal radiation to the surrounding concrete shield wall as well as by convection to the gas in the gap. Both the concrete shield wall and the

,' gas in the gap appear to be heat sinks at a constant low temperature.

All emissivities are apparently equal to 1.

( f) The area of the steel liner that is in contact with the overlying water pool is assumed to transfer heat to the water at a rate specified by an arbitrary heat transfer coefficient, h,.

(g) The core debris, consisting of UO2 , is assumed to be at a temperature of only 1800 C and only 12 cm deep. An unspecified " protective layer on the inner steel shell surface" is postulated.

(h) The core debris transfers heat to underlying concrete by conduction.

' However, the basemat concrete is not allowed to outgas (i.e., dehydrate

'and decarboxylate) or to ablate. This prevents concrete decomposition gases from entering the debris from below and rules out convective heat transfer and exothermic chemical reactions from occurring in the melt.

,' There may be other fundamental _ assumptions inherent in the model - for

~

liner response when contacted by core debris. However, assumptions (a) - (h) were those that could be readily ' identified from Reference 3. Nevertheless, these eight categories of assumptions appear to form the basis for the IDCOR approach to the problem; each will be addressed in the following discussion and compared to representative NRC positions or assumptions.

Discussion of 10COR Assumptions 1

10COR assumption (a) assumes that the debris is solidified, and consists of UO 2 fuel only. Since the -debris is assumed to be pure U0 2 , its thermal conductivity is only 3 W/mK. However, IDCOR's own core-concrete interaction model, DECOMP, does not agree with these conditions. DECOMP assumes that the ex-vessel debris is a homogeneous mixture of oxide and metallic core debris

' phases, not just oxide fuel. This results in a debris pool with a lower melt-ing temperature that can sometimes be molten, a more fluid pool of debris, and a higher debris thermal conductivity, 'in the range of 10-20 W/mK. NRC anal-yses rely upon the CORCON code. These analyses allow the debris to be molten or solid, depending upon the calculated conditions, not only assumption. The molten oxide and metallic phases solidify in a mechanistic framework in a man-ner consistent with prevailing thermal hydraulic conditions in the melt and

!' the boundary conditions experienced by-the melt. These analyses show that the

, liner may be contacted by a deeper pool of core debris (> 25 cm) than assuned by Reference 3. Also, this pool can be molten and have e~ considerable quanti.

ty of molten metal phase present, with a thermal conductivity as great as 47 W/mK.

+

, wr w m .

.,43 e ya> ,- - *A* - * * ~ * * * * ^

>m , n-- , . -

+~~,.,.em,,~<+-s

,'~^w~,n~--e ,n-. w w ,

. 1 ..

o. . 1, ,

z IDCOR assumption (b) assumes categorically that the U0 2 core debris is a solidified mass. This precludes internal convective processes from transfer-ring heat to boundaries, especially to the basemat concrete and the steel liner. I'n deeper pools, this has been shown not to be the case, and both NRC and EPRI presently have reactor materials experimental programs in progress to examine the molten stage of debris-concrete interactions.

IDCOR assumption (c) allows for internal heat generation in the solidi-fled fuel by decay heating only. However, reactor materials experiments and -

code analyses have shown that, especially for BWR cases which may have a large inventory of unoxidized Zr in the melt, the internal heat source due to metal-gas phase chemical reactions will in general exceed the decay heat generation by a large margin-, in most cases representing the driving heat source 'for the aggressive melt-concrete interaction stage.

IDCOR assumption (d) considers a pool of water over the debris, boiling at the critical heat flux. At the temperature specified for the debris, 2100 K, clearly this boiling regime would most appropriately be represented by film boiling. For most cases of interest in the NUREG-1150 analyses there would be no water present since containment sprays are assumed to not be available.

The availability of fire sprays must be evaluated on a plant-specific basis.

IDCOR assumption (e) models heat transfer from the outer surface of the .

liner by radiation to the concrete shield wall and by convection to the gas in the narrow gap. The concrete and gas appear to be isothermal ~ heat sinks at 350-400 K and the emissivities representative of blackbody radiation. How-ever, the gap between the liner and concrete shield wall, at least for the Browns Ferry Nuclear power Station analyses reported in Refetence 1, is not -

empty but full of fibreglass and polyester foam. Over the time intervals re-ported in Reference 1 for liner failure, this would be sufficient to insure an adiabatic boundary condition on the outside surface of the liner, not a radia-tion-convection boundary condition.

IDCOR assumption (f) assumes that an overlying pool of water exists over the core debris and that it cools the exposed surface of the liner with an ef-fective heat transfer coefficient hw. In most Mark I BWR drywells, the down-comer vents to the torus are only one foot above the drywell floor. If core debris were to at. cumulate to this depth, the overlying water pool would simply overflow into the suppression pool. This would prevent the water heat rejec-tion mechanisms proposed, both for the liner and melt (debris) surface, and i expose the liner to direct radiant heat transfer from the high tenperature debris.

IDCOR assunption (g) proposes a debris tenperature of I'OO d C and a debris

depth of, at most, 12 cm. For similar low tenperature cases studied in Refer-ence 1, the steel liner was sonetimes calculated to survive nelt-through.

However, the steel was calculated to be at a high enough temperature so as to have greatly reduced rechanical strength, and failure by mechanical deforca-I tion would be likely. Furthernore, a simple examination of the ex-vessel de.

bris inventories calculated in recent studies such as BMI-2104, NUREG-1079, NUREG-0956, and NUREG-1150 indicate that debris depths (assuming uniform I

spreading over the entire drywell floor to ninimize the depth) may exceed one

! foot.

i e

w.--w. . . . - * ------ e 4 * .*

- 3- .. .

c . 5 s

. '. J Finally, IDCOR assumption (h) allows for heat transfer to underlying dry-well concrete from the core debris by conduction only. By assumption, the concrete is not allowed to decompose or ablate. This is in spite of the fact that concrete needs only tc be heated to 100 C to start boiling the free water in the aggregate matrix. By not accounting for debris-concrete interactions, the gases (H2 0, CO2 ) which would bubble up through the debris and react with metallic species (if there were any) are eliminated, thus precluding the pos-sibility of exothermic chemical reactions in the melt.

Other issues that may be imbedded in the IDCOR assumptions in Reference 3 but were not apparent to this assessment are the concepts that (1) water over-lying molten core debris quenches that debris and (2) water on the floor pre-sents an ebstacle to the migration of high temperature melts across the floor. Data from ongoing experimental programs at SNL and BNL exist which contradict these concepts. Instead it is found that water overlying melts en-gages in film boiling and that melts flow through or under water obstacles as long as the debris is molten. Neither of these two concepts presents a con-vincing case to argue that core debris cannot flow to the containment liner and still be molten.

It is clear that there are major differences between the assumptions in the IDCOR analyses [3] and the NRC analyses [1] for the Mark I BWR containment liner response to contact with core debris. The IDCOR analyses pertain only to a limited, optimistic set of assumed accident conditions and are not gener-ally applicable to a wide range of accident conditions such as those addressed by NRC in Reference 2. The IDCOR analyses specifically are not applicable un-der the conditions that (1) the debris pool is hot, molten, and deep, (2) the debris has a significant metallic component, (3) the debris is attacking the drywell basemat concrete, and (4) there are exothermic chemical reactions in the melt. In addition, some of the IDCOR models are suspect and should be re-evaluated. In particular, (5) the heat transfer from the outer surface of the steel liner, (6) the existence of an overlying pool of water over the debris when containment sprays are not available, and (7) the mode of boiling of an overlying pool of water when water is available. Finally, some of IDCOR as-sumptions with respect to physical properties should be assessed, specifically (8) radiative emissivities of steel, core debris, and concrete, and (9) the debris thermal conductivity.

References

1. Greene, G.A., K.R. Perkins, and S.A. Hodge, " Mark I Containment Drywell:

Inpact of Core-Concrete Interactions on Containment Integrity and Failure of the Drywell Liner," Proceedings of the International Symposium on Sourt.e Term Evaluation for Accident Conditions, IAEA (October 1985).

2. Reactor Risk Reference Docunent, fiUREG-1150, Draft for Conment (February 1987).

3. Approxinate Source Term l'.ethedology for Boiling Water Reactors, FAI 86-1 (Decerber 1986).

4 Plys, M.G., J.R. Gabor, and R.E. Henry, "Ex-Vessel Source Term Contribu-tion for a BWR Mark I," Proceedings of the International AriS/ ENS Topical Meeting on Thermal Reactor Safety, San Diego, CA (February 1986).