ML20207H353

Summary of 990317 Meeting with SCE Re Issues Staff Identified in Review of 980910 License Amend Application & Exemption Request to Remove Hydrogen Monitoring & Control Sys from SONGS Licensing Basis.Meeting Attendee List Encl

ML20207H353
Person / Time
Site:	San Onofre
Issue date:	06/14/1999
From:	Mel Gray NRC (Affiliation Not Assigned)
To:	NRC (Affiliation Not Assigned)
References
TAC-MA3543, TAC-MA3544, TAC-MA3545, TAC-MA3546, NUDOCS 9906160173
Download: ML20207H353 (33)

Text

k 3 g

'o p -

44 UNITED STATES -

j NUCLEAR REGULATORY COMMISSION L

4 WASHINGTON, D.C. 20066 4001

• • • • • June 14,1999 -

L . LICENSEE: Southem Califomia Edison Company FACILITY:. San Onofre Nuclear Generating Station (SONGS)

SUBJECT:

MEETING WITH SOUTHERN CALIFORNIA EDISON (SCE) TO DISCUSS

- THE PROPOSED LICENSE AMENDMENT AND EXEMPTION REQUEST TO REMOVE HYDROGEN CONTROL SYSTEMS FROM THE SONGS LICENSING BASIS (TAC NOS. MA3543, MA3544, MA3545, MA3546)

REFERENCES:

(1) SCE letter from D. E. Nunn to NRC dated September 10,1998 (2) NRC letter from J. W. Clifford to H. B. Ray dated February 9,1999 (3) " Generation of Hydrogen and Oxygen by Radiolytic Decomposition of Water in Some BWR's," K. Parczewski and V. Benaroya, presented at the Joint ASME/ANS Nuclear Engineering Conference, Portland, Oregon, August 5-8,1984 In a letter dated September 10,1998, the Southern California Edison Company (SCE)

. submitted a license amendment application and exemption request to remove the hydrogen monitoring and control systems from the SONGS licensing basis (Ref.1). On March 17,1999, the staff met with SCE personnel to discuss issues the staff identified in their review of the l submittal. Personnelin attendance are listed in Attachment 1. Attachment 2 provides the NRC i meeting presentation material.

SCE submitted the amendment application and exemption request based on its conclusion that the hydrogen control system is not a risk-significant system during design-basis accidents or j severe accident sequences. Approval of the subject amendment application and exemption request would remove the hydrogen control system from the SONGS licensing basis and allow l SCE to remove or secure hydrogen control system in place. j  !

In reviewing the GCE submittal, the staff performed analyses to investigate the effects of Y f

removing the hydrogen control function from SONGS during severe accident conditions. The l' preliminary results were provided to SCE prior to the meeting (Ref. 2). During the meeting the staff indicated that, based on these preliminary results, during severe accidents, the SONGS hydrogen control system design capacity would not preclude an initial hydrogen burn in k l i

containment due to buildup of hydrogen gas from the zirconium fuel cladding metal-water reaction. This is consistent with conclusions made in the SCE submittal.

/  !

However, the staff's preliminary evaluation identified that the SONGS hydrogen control system  ;

could function to limit the long-term buildup of hydrogen gas in containment due to radiolytic l decomposition of water and prevent subsequent repetitive hydrogen gas burns in containment.  ;

While it is unlikely the subsequent hydrogen burns would challenge the SONGS containment, j the staff was concerned about the potential increased risk of damage to available '

instrumentation and equipment in containment. Therefore the staff concluded that the  :

I e ~ ~k k kw C N Btcopyj x  ;

v.1., 2' June 14, 1999 hydiogin contral system hIs s:me s;f:ty vclue. The staff cgreed th t, b:s:d cn risk insights prese.nted in the SCE submittal, the current level of control (technical specifications) for this system may be considered for relaxation.

The staff also discussed assumptions that were modified in the SCE submittal for calculating hydrogen generation. With regard to hydrogen generation from radiolytic decomposition in the containment sump, the licensee's submittal referenced an Oak Ridge Report (ORNL-NSIC-23) as the basis for modifying the calculated hydrogen yield. The staff indicated that the methodology in this report has not been approved by the staff. The staff identified Reference 3 as an acceptable method for considering modifications to the calculation of hydrogen generation from radiolysis. A copy of this reference is provided as Attachment 3.

The meeting concluded with discussions regarding the hydrogen purge system and hydrogen monitoring system. The SCE submittal also requested removal of these functions from the SONGS licensing basis. The capabilities of the SONGS hydrogen purge system and separate mini-purge system were discussed. The usefulness of hydrogen monitoring information to SCE and NRC emergency response organizations was also discussed.

The staff review of the SCE submittal will be addressed in future correspondence to the licensee.

ORIG. SIGNED BY Mel Gray, Project Manager, Section 2 Project Directorate IV & Decommissioning Division of Licensing Project Management Docket Nos. 50-361 i and 50-362 1

Attachments: 1. List of Meeting Attendees 2.- NRC Meeting Presentation Material

3. " Generation of Hydrogen and Oxygen by Radiolytic Decomposition of Water 4 in Some BWR's," K. Parczewski and V. Benaroya, presented at the Joint )

ASME/ANS Nuclear Engineering Conference, Portland, Oregon, August 5-8,

~

1984. l

)

cc w/atts: See next page DISTRIBUTION - Hard Coov - DISTRIBUTION - E-Mail g'Qechsta SCollins/RZimmerman CBerlinger JClifford PUBLIC JZ4dnski/SBlack MRubin BSheron PDIV-2 Reading CJamerson RBarrett DLange  :

LRaghaven SRichards MCunningham MGray- GHolahan PBoehnert MSnodderly ANotafrancesco CTinkler OGC :C]a33 AMalliakos - GHubbard ACRS RHasselburg JO' Brian

• Previously Concurred To receive a copy of this document, indicate c in the box OFFICE PDIV-2/PM C PDIV-D/LA DSSA/SPSB' PDIV-2/SC NAME lbr[ CJamerson k MRubin SDembekM DATE 6 / / Y /99 4 / li /99 N 06/09/99 [o //Y /99 DOCUMENT NAME: G:\PDIV-2\ SONGS \MTSA3543.WPD OFFICIAL RECORD COPY

^

• 3 ,

g . g.-

. hydrog

n control sy:t:im his soms giftty VIlue. Whiledha stiff agre:d that, bis d on risk

' E insights presented in the SCE submittal, the current level of control (technical specifications) for

, .this system may be considered for relaxation, oc=pl:t: :: mere! d te hy+~p reatre! eyrMm f, neR we; net wWeWthe-staff'c p;fah=y evaluation, The staff also discussed assumptions that were modified in the SCE submittal for calculating hydrogen generation.' With regard to hydrogen generation from radiolytic decomposition in the containment sump, the licensee's submittal referenced an Oak Ridge Report (ORNL-NSIO.23) as the basis for modifying the calculated hydrogen yield. The staff in-ficated that the methodology in this report has not been approved by the staff. The staff identified Reference 3 as an acceptable method for considering modifications to the calculation of hydrogen

. generation from radiolysis.: A copy of this reference is provided as Attachment 3.

The meeting concluded with discussions regarding the hydrogen purge system and, hydrogen monitoring system. The SCE submittal also requested removal of these functions from the

, _ SONGS licensing basis. The capabilities of the SONGS hydrogen purge system a'nd separate mini-purge system were discussed. The usefulness of hydrogen monitoring information to SCE .

and NRC emergency response organizations was also discussed. f

/ _

-The staff review of the SCE submittal will be addressed in future correspondence to the licensee.' /

/

I Mel Gray, Project Manager, Sectibn 2 Project Directorate IV & Decomrnissioning Division of Licensing Project Management Office of Nuclear Reactor Regulation Docket Nos. 50-361 /

and 50-362 j

-- Attachmentsi 1. List of Meeting Attendees /

2. NRC Meeting Presentation Material /

3. " Generation of Hydrogen and Oxygen b Radiolytic Decomposition of Water in Some BWR's," K. Parczewski and V! Benaroya, presented at the Joint ASME/ANS Nuclear Engineering Conference, Portland, Oregon, August 5-8, g 1984. /

l '

cc w/atts: See next page / j DISTRIBUTION - Hard Coov DISTRIBUTION - E-Mail ,  ;

' Docket - SCollins/RZimmerman CBerlinger JClifford i PUBLIC JZwolinski/SBlack MRubin BSheron j PDIV 2 Reading CJamerson' RBarrett DLange  !

LRaghaven SRichards' MCunningham 1 MGray. GHolahari PBoehnert  !

MSnodderly ANotafrancesco CTinkler '

l OGC AMalliakos GHubbard  !

ACRS RHasselburg ; JO' Brian DOCUMENT NAME: 'J:\ SONGS \MTSA3543.WPD <

OFFICE PM/PDIV-2 y LA/PDIV e m:l DSSA/SPSW SC/PDIV-2 )

NAME MGray/t 1b/. CJamerson- (y\ / MRubinC7 SDembek l DATE i /- 3 /99 I - F / 3 /99 7 [o / 4 /99 / /99 OFFICIAL ITECORD COPY-u 4

{

L

- .j

y, ,

' San Onofre Nuclear Generating Station,. Units 2 and 3 cc:

- Mr.' R. W. Krieger, Vice President ' Resident inspector / San Onofre NPS Southem Califomia Edison Company c/o U.S. Nuclear Regulatory Commission San Onofre Nuclear Generating Station Post Off~m e Box 4329 P. O. Box 128 San Clemente, Califomia 92674 1 San Clemente, Califomia . 92674-0128 Mayor

)

Chairman, Board of Supervisors City of San Clemente County of San Diego 100 Avenida Presidio 1600 Pacific Highway, Room 335 San Clemente, Califomia 92672 >

San Diego, Califomia 92101 Mr. Dwight E. Nunn, Vice President Alan R. Watts, Esq. Southem Califomia Edison Company Woodruff, Spradlin & Smart San Onofre Nuclear Generating Station 701 S. Parker St. No. 7000 P.O. Box 128 Orange, Califomia 92668-4702 San Clemente, Califomia 92674-0128 Mr. Shenvin Harris Resource Project Manager Public Utilities Department City of Riverside 3900 Main Street Riverside, Califomia 92522 Regional Administrator, Region IV l

U.S. Nuclear Regulatory Commission i Harris Tower & Pavilion

' 611 Ryan Plaza Drive, Suite 400 Arlington, Texas 76011-8064 Mr. Michael Olson San Onofre Liaison San Diego Gas & Electric Company P.O. Box 1831 San Diego, Califomia 92112-4150 Mr. Steve Hsu Radiologic Health Branch State Depadment of Health Services Post Office Box 942732 -

Sacramento, Califomia 94234 -

Mr. Harold B. Ray Executive Vice President Southem Califomia Edison Company San Onofre Nuclear Generating Station P.O. Box 128 San Clemente, Califomia 92674-0128 A

1 8 2

hydrogen control system has some safety value. The staff agreed that, based on risk insights presented in the SCE submittal,- the current level of control (technical specifications) for this system may be considered for relaxation.

The staff also discussed assumptions that were modified in the SCE submittal for calculating

~

hydrogen generation. With regard to hydrogen generation from radiolytic decomposition in the containment sump, the licensee's submittal referenced an Oak Ridge Report (G !NL-NSIC-23) as the basis for modifying the calculated hydrogen yield. The staff indicated that the methodology in this report has not been approved by the staff. The staff identified Reference 3 as an acceptable method for considering modifications to the calculation of hydrogen generation from radiolysis. A copy of this reference is provided as Attachment 3.

~ The meeting concluded with discussions regarding the hydrogen purge system and hydrogen monitoring system. The SCE submittal also requested removal of these functions from the SONGS licensing basis. The capabilities of the SONGS hydrogen purge system and separate

- mini-purge system were discussed. The usefulness of hydrogen monitoring information to SCE and NRC emergency response organizations was also discussed.

The staff review of the SCE submittal will be addressed in future correspondence to the J licensee.

HJ Mel Gray, Project Manager, Section 2 Project Directorate IV & Decommissioning Division of Licensing Project Management Office of Nuclear Reactor Regulation Docket Nos. 50-361 and 50-362 Attachments: 1. List of Meeting Attendees 1

2. NRC Meeting Presentation Material . ,

3. " Generation of Hydrogen and Oxygen by Radiolytic Decomposition of Water ]

in Some BWR's," K. Parczewski and V. Benaroya, presented at the Joint  !

ASME/ANS Nuclear Engineering Conference, Portland, Oregon, August 5-8, 1984.

cc w/atts: See next page i

a

< i i

t 4

n-

-]

l MEETING WITH SOUTHERN CALIFORNIA EDISON COMPANY LIST OF MEETING ATTENDEES i March 17,1999

, Southern California Edison (SCE) )

l J. Rainsbury T. Hook O. Thompson Performance Technoloav B. Christie McGraw-Hill D. Stelfox MB_Q W. Bateman C. Nilinger M. Snodderly A. Notafrancesco A. Malliakos R. Hasselburg M. Cunningham P. Boehnert C. Tinkler G. Hubbard l J. O' Brian J. Clifford G. Holahan M. Gray l

ATTACHMENT 1

l l

1 l

l NRC Meeting Presentation Slides l

1 ATTACHMENT 2

S T

N e t n

E b e M n m E i o p t n i R

I a

r oi u

q et e U np Q em ge g

n E n x i

R e w _

e ga l o _

L l o d oi r v O f n a

- R dyi s e -

s h T

h a t s r -

N f o b e v o -

O n n o i t

n _

C oi g

m o g 2

i t s e m H ae rd r i n -

n e d g 1 M d i r e l

u e g

r u -

O sf o o p R n o o n w s d

r d

F cO s r y e n: e h l

l o

an t

T os n e r S h a t S ii t s pa i

b n t n

E d e mb m i l

o -

U eh eg o n c

t x c o r Q st ei n e s o E em u o d n s r n u f

y R qr e ec e o u

i l

t ef s g i N rd ol i o n b a

O se p r i

t I at h a oe rh d

y n

o p

a T n pt H C C P Ei m em M Ci l h o E Se Tf r X

E =

• l l l

y .

l a3 u2 d

? s am o sy r o i

f S gr R pl o yt f

E ui b n N d da l

2- e I

i u r m B b ot I

Mi n M me Ta t O r u e d t n ao C t

- t c E gly n R ni r ee did oa i c sn N l ei m E ht p r cis a ne G gs eeu O h gl R ih ni c c t

gr o vra D dh ui i nd ye 2 Y ew r

uh w H r t df ol o n n u onto E e e e i

R umn l ai A l ai t 1 valri E va t f

otnp r L f n B o ec o eeA r c A en ,

U r

a he woM cP L st r s r ee A ei n eht V i n n i nt a l b e b g ,

W mg o or min o

nt O

H cd ey c ce ucr Rh edp e Rr e

S -

T N d e

E t M l a

E t u

R s I o p

U a Q g?

E nde R i r

ug L d an O na R em g

T oe N r O dye b _

C h rf _

f o 2

on H pO 3 M u d na O l i

uS R b t F a mt T

r n _

S t e e .

E gi d _

U n c Q l o ca.

E ee .

R h r t o N d c l

O ud oe wdar I

T P wg M

Hd oe E

X E

• e k

a n t

di o

? e t o ec t

Y v l n

L i

t n l ou c

- T e d l

t rf ny N

- f f u ot e

- E e o cf I

o w na C t r n us I

F i o i o f f oo .

r t F p r: a ss _

E n es t e i co D bt uh b nl t E

i

,e m o e u u o

T A ,

c qh N Cm e r et A Oe t n sw n

i R L s e on .

G ays g co gl y i x et E nor o ht n a 4 B

i wtn n db i

N l ooc e nm O

l o g o t a o I f n r sc T t e ag d y

ht r e

P h or wne i

t h M sd d d g E ey e l uy X rh l E

i u e l

o r

ox co qh t -

N et n r t

n e n A )

r f o o c o a g n

ue l

C 1

( n po

)

ci o c ed r

W ( t 4a r Ap na l h y O Th H 4.

0p e

5o

~

- 1

- .  ; l 7

S t uA d T oC e  %

N bO w 0 E aL o 2 t h y M as s: s e b i

s sy h E n%a na t

p R e4b od s m I g i e u U o r o g f

n t aty cr m s Q

i t

di t i e

yme s i i E f h i

e h hid dt v t R l eya o o r i f

n L ht t

i g

mf n a

i s

O g d l

i i n de s t h a b n R ai t y we i

h m w r

o l

T t o oe e N

i sml o l l

l c ht a

3 d a

O w af of l ox f e r o r C h r ys et h3

)

t y 2 sswdea h o t (

p 5 b H i so h t n s e )b d e

M 0 t i d (6 m o t a

5 bal ht i

e3 wi t n4 r r O nt u Em o0 f d

e R

i g

i ta 5 n F ih o Cl y r r l

e e scb St tefo i y g T ea a d et yi i

l c n S r a b ba nd e E

t e pt e i

t y g U nd%

el r u3 dme gl a l

o o r

Q r o m r m a nu i d d y

E cu w5 of l ic dl a r h d f d a R s n n r n e a c d d

'e oi a p eg t e e l

N r t forasy cn c c O so e r d u eo d u

d u

I T

n na t

sd y y sm e e P

Oed Ua R R n nc5 a h l

M a o3 ne E Sc1 Ath X

E .

• S T t o

N se r

E e r u u M i d cd E u qan cg R e rt on i I

n ed U ) sd Q

4e

(

) m a el a E ( n i l e c

.i r

R a. ta t m -

n c u G 2. o I un i N Vc t n d o I

n e e ocr r

R oh t t

,m pi z -

O i t

niS r o r n n e T c e p ei) oh I

SnD p t

a6 nt n9 i

st N ,iR o u e sf o

O t s CcM o- i f

0aE 5r o wn M T oi o t

t a t r t 6

t ni o pe nR ae v r t M d e s p v (r h an O Pct n n sd o d

e uoe sixt l i R oo e cm e vSt n t

n o oc F

t e el de a En er n i ee T xeu s L ae nins cec l t

ah r u t f d gs i

S i g o n n cgo E nroae n ih t e c clbi o l

aia nt U ed pym i

r o Aeia t cwe Q ph e i t

y Ts s e r t oh nl l E Aeh n o c sop eoi c _

R dht n t g m n ee 'e p s O d iyr cl f

m N agn n gs ss ceeh _

O ni tt e r n ef '

atat

)i eesa f _

b ri g I

T (

um o mcst eio rdx e ee i

P 4 s s r ELAS M

4 a n 0 e ar d y vm e e 5mt H Sitmh E

X E * *

• S g T i n ,

N r u 1 f ot n E s . n a o ne a M e N om l p

E m d i

tain y R

I d i

A r et a c nb e

U n pn es a

l Q

e n o oc ge r

E m o e n m i d e ei?

mleg R o t

a uht edn c l u

l ct G

i i e nn ,ur r c sgo N s l

a ie sv e t i

unn r t I

C e nt o R e n s t d e o O i l

ee r

o i

l y em m T et c I

N diu t p p dl i

i i ogeen r

un u ub ga pag O gi m s p t no M

t o t na n ar n5 e 1 t e c emdy 7 M my d e me mt s n s h

O er ge sut "y eh g t se u ed R av i ad sio n cu an l

F ne ei rb a sa cain T an ai t S mo i t t snbu s msr e gr e o n t

E t n ar em t e nn aec U dn et mo ei d b mvt eu Q ie c im a so c c E eC c d a n r

ut co ace d h R

cc o

asn e cr eni r

o aw t

N i r n emn i r n c ,d e

O e e mia e e st e l

n g n g l wnr i

I T

P e o gr et sn e

gr o wp o op Ey d eo Eyd o esp M

E Ch hC T" Ch Hr s u X

E * * *

• t o

S e t y f i

T b i

l d

N i

l b e E

l a a p n h

a i

M s a c t E g n R n e i a

h m?ss i

I g

U r u

t e

i Q e p d u b ba E d l t .

R c nn l

e n eg o

l G

i r s mi ne s _

N t

n e id I

o n a _

G c i

l e tn e R r d ot h U o i u c e m f

P t y g h o t r M

i l

i n t f 8 b e t nd O a e R

F p

a m

e g

ve vo c a t om T e n ye S ht a t r i

E t m l

i s a t b a U ht tn n aw p

Q e e ag E s e dm in cn R

i r ci eg r i

u ca h N q at n t u p

O e r i co r c d

l I

e ude T )d ee n e ol P (d 4 eh t wlo r M gt i

4 v Ene wtn.

E 0r o oo X 5p Cv Hc E

\

t'

F 1

" Designing for Hydrogen in Nudear Power Plants," Parczewski, and V. Benaroya, presented at The Joint ASME/ANS Nudear Engineering Conference, Portland, Oregon, August 5-8,1984 ATTACHMENT 3

-- -~ .

}gfMfQff'Q?WY&???S5 1

TRANSAC ICJNS OF THE AMERICAN NUCLEAR SOCIETY JOINT ANS/ASME CONFERENCE ON DESIGN, CONSTRUCTION, AND OPERATION OF NUCLEAR POWER m

August 5-8, 1984 Supplement Number I to Volume 46 Portland, Oregon TANSAO 46 (Suppl.1) 1-151 (1984)

ISSN: 0003-018X General Chairman Donald J. Brochi(PGE)

Technical Program Cochairmen Jack W. Lentsch (PGE)

. Soung M. Cho (Foster Wheeler)

Transactions Editor Lorretta Palagi(ANS)

Transactions Coordinator Joann flollensteiner (ANS)

TECliNICAL PROGRAM COMMITTEE ANS J. F. Firlit (Consumers Power)

John C. Guibert (Impell Corp, Walnut Creek) Charles K. Soppett (Bechtel)

Wes llartley (MAC) Gerald C. Sorensen (Washington Pub Pwr Robert W. Hess (Cygna) Supply Syst)

Karl Hornyik (Oregon State Univ) Frank Spangenberg (Northwest Energy Sve)

Michael J. Kolar (EPRI) R. Jon Stouky (PCI)

James J. Rocca (PG&E) Kathryn M. Tominey (RilO)

Frank Rogan (PGE) Howard T. Watanabe (GE, San Jose)

Alan M. Ross (Consultant) George L. Wessman (Torrey Pines Tech)

W. Morris Sample (Duke Power) Paul Williams (Stock Equipment)

11. J. Worsham, Jr. (MAC)

Jay C. Young (NUTECH)

ASME Kenneth L. Adler (Energy Tech Engrg Ctr)

Richard Baker (NUS) Kalyan K. Niyogi(UE&C)

Martin D. Bernstein (Foster Wheeler) Scott R. Penfield (GCRA)

Donald F. Casey (GE) George Plimi (CECO)

David Elias (CECO) Alfred J. Rino (DOE)

Jack G. Evans (Toledo Edison) Brian C. Ryder (Teledyne)

Douglas K. Warinner (ANL)

William C. Jones (Omaha Public Power)

This Meeting is Cosponsored by the American Nuclear Society's Power Division Reactor Operatiom Division, Fuel Cycle and Waste Management on, ,

Disision, the Oregon Loca and the American Society of Mechanical Engineers' Nuclear Engineering Division COPYRIGHT C 1984 AMERIC AN NUCLE AR SOCIETY, INCORPORATED , L A GRANGE PA RK lillum *a'*'

- l

,o .2 ,e *

.. 128 o'

l l

DESIGNING FOR HYDROGEN IN NUCLEAR POWER PLANTS-Il

1. Radiolysis of Wate< in Some BWRs, K, I. between the decomposition products are very complex.

There are Several computer codes that take into considera-Parczew. ski, V. Benaroya (NRC) tion these reactions and compute the net generation of radiolytic gases.

Water cooled reactor plants under certain accident condi- These codes are very complicated; however, where high tions can generate significant amounts of hydrogen and accuracy may not be wananted, a simple method suggested oxygen by radiolytic decomposition of water. These gases by A. O. Allen could be more appropnate. In addition to .

when accumulated in sufficiently high concentrations can being simpler, it has the further advantage of being able j threaten the integrity of the containment building. It is to account for the effect of impurities such as iodme. The important to calculate the rates at which these gases can method hmges upon an assumption that during postaccident accumulate m boiling water reactors (BWRs) with Mark I conditions, the reactions between OH free radicals and and Mark !! contamments where they can deinert contain- rnolecules of hydrogen and iodine have a controllmg effect ment atmosphere, on recombination of gases. Based on this assumption, the When energy from radioactive sources is absorbed by follow ng equations for hydrogen and oxygen generation are water, in addition to molecules of hydrogen and oxygen, a obtained:

number of chemical spec'es, called " free radicals," are GOH III abo formed. These species react in several secondary reac- G(H3 ) = GHi - kdt-)

tions and recombine some of the onginally formed hydrogen and oxygen back into water. The mechanisms of interaction 1 + kHlH3 )

0.5 - c(H,)

! 0.4 '

i 0.3 -

0.2- -

lodine Concentration 0.1 -

0.05 ppm 0.5 ppm 1.0 ppm 2.0 ppm 3.0 ppm i ' ' ' '

0 8 35 O 5 10 15 20 25 30 Hydrogen Concentration, cc H2 /kg H2 O Fig.1. Effect of dissobed hydrogen and iodine M G(H 21-

.. . , 3

1. l 1 4 .

Designing for Hydrogen in Nuclear Power Plants-ll 129 GlH2I l

j

'5% 10 %  ! 25 %

l50%

0.4 l l

Percentage Omidationl l of Fuel Cladding _ 75 %

0.3 - l l l l l Percentage  !

l l l Release of l l Total lodine l Inventory 0.2 - l l l l l l l

1%l 10 % 20% l 40 %

60 %

l l l

0 i f I ' ' ' d 0 0.2 0.4 0.6 0.'8 1.0 1.2 1.4 1.'6 1.8 2.0 2.2 Partial Pressure of Hydrogen, atm.

Fig. 2. G(H:) salues in BWR with Mark I containment, and along the constant iodine release curve until it reaches G(0 ) =;G 3

1 (H ) , (2)

G(H 2) = 0; at that instant, radiolysis stops.

in addition to llecreasing G(H2 ) and G(02), rates of when radiolysis will also be affected by the decay of radiant energy. To calculate generation of radiolytic gases in the {

GtH3),GtO )= net generation of hydrogen and oxygen plant, both of these effects would have to be considered.

(molecule /100 eV radiant energy ab- The methodology described in this paper provided the sorbed) means for determining variations of G(H2 ) and G(02 ). The decay of radiant energy sources can be evaluated using GH,, GOH = initial generation of hydrogen and OH methods available in nuclear engmeering.

free radical (molecule /100 eV radiant en-ergy absorbed) kH,ks a velocity constant for the reactions be- .

arameMrs W Gntainment @ ogen tween OH and hydrogen and iodine, Burn Analysis, R. G. Gido (LANL), A. Koestel respectnely (t/g mol-s) (ConSultanf)

[Hal,(l*] = molar concentration of molecular hydro- '

gen and iodine (g mollt) INTRODUCTION These equations indicate that the generation of radiolytic Some nuclear containments now can burn degraded core )

gases decreases with increasing concentration of dissolved hydrogen under control.' This paper desenbes experimentally I hydrogen and increases with increasing amounts of iodine. verified models for the parameters used by codes ** to estimate the pressure-temperature response to the burning. {

For certa;n concentrations of dissolved hydrogen and iodine.  !

The parameters are the mean hydrogen concentration at dissociation and recombination rates become equal and the net generation of radiolytic gases becomes zero, even if the ignition,8, the flame speed, FS, and the fraction burned, F.

water remains exposed to the radiation environment. This is Because of the sprays, fans, etc., the models are for a highly illustrated in Fig. I. turbulent atmosphere, which causes burning to be initiated at relatively low mean concentrations.

In nuclear plants under accident conditions, radiolysis of water is one of the several sources of hydrogen. By far TURBULENCE the most important is oxidation of fuci cladding by steam.

This reaction yields a large amount of hydrogen at the Angaccounting for turbulent kinetic ,en,ergy (TKE) is beginning of the accident. This hydrogen, when released to p  ;

g P

the containment building, may saturate the containment greatly simphfied if applied on a global basis. Neglecting water and impede radiolysis. The curves in Fig. 2 illustrate TKE gradients produces how the hydrogen and iodine present in the containment building control radiolysis of water in a typical BWR Mark I E 8t=8+dK containment. Intersection of the constant iodine release Ti . (1) curve with the vertical fuel cladding oxidation line gives the where 8 8: sums the TKE sources, such as sprays and fans; initial value of G(H ).2 As the radiolysis proceeds, more e is the rate of TKE dissipation to heat; and dK/dt is the hydrogen is generated and the operating point moves down rate of change of TKE.

l

'l l- j pesigning for Hydrogen .l

, n Nuclear Power Plants-  :

1 ra

, .y presentedet .

THE JOINT ASME/ANS NUCLEAR ENGINEERING CONFERENCE '

PORTLAND, OREGON

• I d AUGUST 5-8,1984 'dd w

sponsoredby .

' ~

,' .,.'h.R 47/

r THE PRESSURE VESSELS AND PIPING COMMITTEE,' ' "* ' ' M.@n NUCLEAR ENGINEERING DIVISION, ASME 'C... 4: dA '

,Ad. y.:y;

. .v ' ,, .q, editedby  : -< unt+ Y'~FL&e  : i

. c. . .... % r &  ;

K. K. NIYOGI

. * * ~, ~. N ~ ". M...t. .. . g.qJ UNITED ENGINEERS AND CONSTRUCTORS,INC..y . .d.qqOy

. . . . , . v.. . e.o.. ..

M.D. BE R NSTEIN c /m 4.,

• 6'

' --I<, ..

~'

FOSTER WHEELER ENERGY CORPORATION  :

P. ci. m..* .,e

. 1:

4 p.. .. * /* .', -

j. r; ;, <,

. '~

':; "!U

....n,:y..T.O$

, . . + . .;.cs Y

, r,n+s.1 s ;.r/;, .: ff. . . .

lj3ssi.i v '. .p a l4, w. . , . .,-t* ....-s J.;. . .

.m.,__ m.  :. , g. . ,qq 'p u y.

, . ,$es ... -

m.y 1 N $; %. . . .h. ., . .

. n ; . < cp :: );;.g

n.: n .' ' t.ny THE AMERICAN SOCIETY OF MECHANICAL ENGINEERSp' id . . .

. e ; . ..

United Engineering Center 345 East 47th Street New York, N.Y.10017 p-

....~. u 1

p.

4 J.:

' d' .N.1;o@k U. S. NUCLEAR REGULATORY COMMISSI '8)w..ps/,y.,4.g ltM'#Tpi;ik;t LIBRARY WASHINGTON,D.C. 20555 STOPNif s .

LNd98Misti.$II

• w ,O,bMNW.W! s 4..A .

,,:ypm

<.-. rlp Qy.a 1 we<. . k' .; o ...; a;n. . ....

p,.,..y,,.;Q........:.;s'?:$5;.;.%ggQQ h'

o

3 . ,

..,0...

/ GENERATION OF HYDROGEN AND OXYGEN BY RADIOLYTIC DECOMPOSITION OF l r WATER IN SOME BWR'S i

K. l. Parczewski and V. Beneroya U.S. Nuclear Regulatory Commission Washington, D.CV. 20555 i

ABSTRACT Water cooled reactor plants under certain accident conditions can generate significant amounts of hydro-gen and oxygen by radiolytic decomposition of water.

Radiolytically produced oxygen can pose problems in the BWR plants with Mar.k I and Mark 11 containments which rely on a nitrogen inerted containment atmosphere for preventing deflagration or detonat' ion of the hydrogen i accumulated.during an accident. Fortunately, there l are several factors which tend to limit the extent of radiolysis. Although initial decomposition of water ,

is proportional to the amount of radiant energy ab-sorbed, due to the several secondary recombination  !

reactions, not rates of rad'iolytic gas generatign tre l clependent of the presence of hydrogen and iodine dis-

, solved in water. For certain hydrogen concentrations net generation of radiolytic gases will stop even if the containment water remained exposed to the radiation  !

environment. It is important to be able to predict .

these concentrations.

' i l NOMENCLATURE 8 A- Avogadro's Number i

[H2 ], [0H], [R], [S] u molar concentration of molecular i l hydrogen, OH free radicals and .

impurities,-respectively, 9 mole / liter kg, kg , kg, - velocity constants for reactions between OH and hydrogen, other free radicals and impurities, respectively liter /g mole-sec l 2 - rate of radiant energy absorp- -

l tion per unit volume of water, 100 ev/ liter-sec i

e

. l'

~

, [ ,0 , , .

INTRODUCTION In water cooled nuclear plants, large amounts of hydro-

, gen could be produced during certain accident condi-tions. This hydrogen, when mixed with containment air, can form flamable or even detonable nixtures, which when ignited, maf post significant threat to contain-ment integrity and to operability of the equipment located in the containment. Such a hydrogen burn was experienced in the Three Mile Island Plant, Unit 2 resulting in a 28 psi pressure spike. Subsequent examinations have indicated that several pieces of equipment showed signs of thermal or mechanical damage. -

Hydrogen can be generated during an accident involving

• ioss of cooiant in three different ways; by oxidation i

.of the overheated fuel cladding, by radiolysis of water '

and by corrosion of non zircaioy metallic surfaces present in the containment, By far the most imnortant source for hydrogen production is oxidation of fuel cladding. Oxidation usually occurs at the beginning of the accident and results in generation of hydrogen at very high rates. Corrosion of metallic surfaces also produces hydrogen, but in much smaller, quantities and at considerably lower rates. Radiolysis of water differs from these mechanisms in that in addition to hydroge,n, it also produces stoichiometric amounts of oxygen. The rates of production of hydrogen by this mechanism are much slower than by metal-water reaction; however, the radiolytic gases can be generated as long as there is radioactivity present and the concentration of hydrogen dissolved in water is kept below certain threshold values.

We will concentrate on the radiolysis as'pects of gas generation. Radiolysis, although not necessarily a -

principal source of combustible gases, can play an -

important role during certain accidents. This is .

especially true in the BWR's with Mark I and Mark II containments. These plants use nitrogen atmosphere .

1, _

2-t l

....a,,. '

.a .

.j in the containment as a means for preventing ignition -

l- of accumulated hydrogen. When a sufficiently large -

l quantity of radiolytic oxygen is produced, these con-tainments may become deinerted,.and a hydrogen defla-gration or even detonation may occur.

The amount of radiolysis that may contribute to deinert-

. ing of the containment depends on the particular design of the plant. Therefore, plant specific analyses have -

to be performed taking into account the plant design features as well as different accidents scenarios. -

These analyses have to rely on the predicted rates for ,

radiolytic decomposition of water. l Methods for predicting these rates' are described in this paper with an illustrative example of radiolytic gas generation in a typical BWR. It is recognized that in actual accident analyses hydrogen generated from other sources should also be considered. However, discussion of this subject lies beyond,the scope of this presenta-tion.

MECHANISM OF RADIOLYSIS When energy from radioactive sources is absorbed by water, the water molecules decompose, giving rise not only to molecules of hydrogen and oxygen, but also to l a number of chemical species, called free radicals. ,

These, free radicals react in several secondary reactions ,

and recombine some of the originally formed hydrogen i ar.d crygen =lecule:: back into molecules of water. The initial radiolytic decomposition of water is proportion- 1 al to the quantity of radiant energy absorbed. It is customary to express the amounts of chemical species produced in radiolysis by G values, which are defined as the number of molecules of a given chemical species j produced per 100 ev of radiant ehergy absorbed. It is -

i usual to denote initially generated chemical species by a subscript to G, e.g., GX for X chemical s and net generation (after secondary reactions)pecies, by a bracket following G symbol, e.g. G(X).

The mechanism of interactions between the decomposition products of water is very complex. There is a number i of secondary reactions between free radicals themselves, between free radicals and water molecules, hydrogen and oxygen molecules and finally between free radicals and different impurities present in the water. To determine the extent of recombination of initially produced hy-drogen and oxygen molecules, a very complex system of differential equations representing kinetics of indivi- ,

dual reactions has to be solved.

n ,

• 4 i 40 .U,g .

.a .

This task can be accomplished by several computer codes

such as PAKSIPA-CHEMIST developed by the Atomic Energy I

of Canada Limited (1) or FACSIMILE developed by the a 2 These codes United

,are very Kis dom Atomic Energy Authority (Riolytic gasFor calculating r ?.

complicated.

' generation in nuclear plants, where high accuracy may not be warranted, a simple method suggested by Dr. A.O.

Allen 3

(Tm),could be more appropriate. In addition to being s pler, it has a further advantage to being able to account for the effect of impurities, which in the case of post-accident conditions, is an important con-

. sideration. In this method, only the reactions contri-buting most significantly to radiolyt,1c decomposition of water are considered. Since the OH free radical is .

one of the most reactive species, it is assumed that at temperatures belog 300'F, the reactions involving this  :

free radical have a controlling effect on radiolysis. '

The main reaction entsiling OH is oxidation of molecular l hydrogen to water. During this reaction OH is dee .

stroyed. OH is also destroyed in reactions with other free radicals and with certain impurities present in i the water. It is however continuously regenerated by j radiolytic dissociation of molecules of water. These i

j processes can be represented by the following chemical l reactions:  !

)

t Oxidation of molecular hydrogen: i H2 + OH --> H + H O 2 j l

I Reactions with other free radicals

i OH+H HO 2  !

OH + OH --+ H O 2

+ O2 l

Reactions with impurities:

S + OH  ; S + OH~ '

Since each of these reactions will proceed at a certain -

rate, it is possible to make the following mass bal- t ances for molecular hydrogen and for the OH free radi-cal:  :

Z.G X (H2) = Z.

(1) , '

I hj - k 'EH H 2].[0H] __

Z.G 0H - kH .[H2 ].[0H] - k R.[R],[QH] '

I (OH) = XZ.G

- k 3dS],[0Hj (2) 4

p ,

. 9 ,0 . -

If it is assumed that at equilibrium conditions the concentrftion of OH free radical remains unchariged, )

1.e.,G(OH)=0, hydrogen -

from Eq. (1) and Eq. (2) generation  ; can be obtained '

. (3)

G(H2 ) " GH2 --

, 1 + kR*[R) + kg.[g] -

l kH.[H23

~ When the rate at which OH free radicals react sith impurities is high, i.e. , k [5] >> k R[R], Eq. (3) can

,be further simplified: 3

. I l

I G(H2 ) = GH2 - H' (4) k3.[ 5.] ,

1+k H*[Hg] l o -

l r ,

The corresponding generation of radiolytic oxygen can i be. assumed to be equal approximately to half of that

'of hydrogen: -

G (02 ) = 1/2,G (H2 ) (5) ,

The last two equations indicate that the generation of

'radiolytic, gases decreases with increasing concentra-

. tion of dissolved hydrogen and increases with increas-ing amounts of certain impurities which tend to destroy OH radicals.

' ~ ~

~

The majority of impuritiis or addittyes p"r'e ent in re .. 1

. actor coolant do not react with OH free radicals and' T consequently they do not affect rates of radiolysis.

For exam le, it was demonstrated experimentally by ,y( -

Zittel ( ), that the rates of radiolysis of a boric ,j '

acid sol tion containing 3000 ppm of boron are the same  ;

as those of pure water. However, there are few chemi-  !;

cal substances which react strongly with OH radicals. ' '

In nuclear reactors iodine is the most important of them. Iodine is formed in fuel by fission. Under normal conditions, it remains confined to fuel rods L

l with only very small releases to the coolant.during sudden changes in operating conditions, cau'cing the

'. 1 s P so-called, iodine spikes. During accidents involvina d l

I

.s.a...

a l

damaged fuel, large amounts of iodine can be released to the coolant system and to the containment. The released iodine can exist in several different chemical forms. It can exist as elemental iodine, as iodides, as iodates or as organic iodine compounds. Only the iodine in iodide form is effective in reaction with OH '

free radicals. It reacts as follows: .

21 ~ + 20H ----* 12 + 20H" ,

21~ + 2H+

12 + 2H ,

The fraction of iodine in the I- form depends on the l plant configuration and the accident scenarios. Since it is impossible to determine it exactly, in licensing l calculations it is conservatively assumed that all '

released iodine is in I~ form. Curves representing hydrogen generation in terms of G values as a function of hydrogen and iodine concentrations are shown in .

Ficure 1. These curves indicate that for certain l hydrogen and iodine concentrations, radiolysis will stop even though radiant energy may still be absorbed by the water. ...

e 05 - p$i ,

04 ,

3e OJ .

02 -

tration

51. =

DD5 ppm 0.5 pom 12 pspn 2Sepm 3.0 ppm 0 10 16 b M Hyakopen Concentrate ac VC H3 0 FIG. 1 EFFECT OF DISSOLVED HYDROGEN AND IODINE ON G(H )

2

.e.

+=

- 4

• (

1

~

,, RADIOLYSIS OF WATER IN BWR'S UNDER ACCIDENT CONDITIONS Although in BWR plants, water exposed to high radiation' levels undergoes continuous radiolytic decomposition, under normal operating conditions this radiolysis does -

not pose any significant problems because radiolytic ,

gases are generated inside the coolant system and are ;

usually recombined to form water in the off-gas system.

During accident conditions, because some radioactive '

. . . material is released to the containment's suppression .

pool, .radiolysis takes place not only in the reactor vessel, but also in the containment buildina. In addi-tion, some of the pases produced inside the vessel may be vented into the containment. Since a hydrogen-oxynen mixture produced by radiolysis may become flammable and

' threaten the integrity of t6e containment, different In hydrogen with Mark control schemes have III containments been (about 1.5 devised x 10 $ cu. ft.)BWR's, which are relatively large, hydrogen is deliberately '

ignited and burned before it reaches dannerously high concentrations. In BWR's with smaller egntainments, such as Mark I and Mark II (about 3 x W cu. ft.), -'

hydrogen innition is prevented by maintaining a nitro-  !

pen inerted atmosphere and assuring that the concentra-tion of oxynen does not exceed the combustible limit of I SV /o. This last method is very effective as long as, j there are no potential sources of oxygen in the con-tainment. This, however, is not the case, since during' some accidents sinnificant amounts of oxygen may be produced by radiolysis. This oxygen is not produced at  : ,

a constant rate, but its production rate keeps decreas- '

inn as the radiolysis slows down due to build up of higher hydronen concentrations in the free snace of the containment building.

Eventually a hydrogen concentration is reached when radiolysis stops and no more oxygen is generated.

Whether this will happen before or.after the SV/o. limit

{ is reached depends on several plant. characteristics and on the type of accident considered. Each plant, there-I

fore, has to be individually examined to find if its

containment atmosnhere will remain inerted throughout the duration of an accident. -

i

.7

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

i. .

.s.*... .

.a 8eaca tha emunt of hydrocen dissolved in subcooled . .

! water is proportional to its equilibriu. r,ressure, tt.e L rate of radiolysis will be a function of the concentra-l ,

tion of hydronen in the containment. free volume. In piants with inerted containments, this hydrogen plays ,

! a dual role: it acts as a diluent for oxygen in the

! containment free volume,,and it controls the rate at which radiolytic oxygen is produced. The hydrogen pro- -

duced by cladding oxidation or coming from other sources )

l in the containment will help in reducing the buildup of l oxygen concentrations, althounh higher degree of clad-ding oxidation usually will result in higher damane to , 1

.the fuel, which in turn will cause a higher iodine '

l rel ea se. The ben;^1t accorded by this source of hydro- i gen will be thereiere partially offset by higher iodine i  !

concentrations. )

The exact relationship between degree of cladding oxida- l l tion and iodine release is not' presently well known. l Therefore, in accident analyses, a conservative anproach "

has to be taken by assuming conservative iodine releases. ,

During loss of. coolant accident in BWR plants, radioly-sis of water occurs in two locations: in the reactor yessel and in the suppression pool. Water in the reac- l tor vessel radiolyzes by the absorntion of radiant '

energy from the core and the radioactivity released to the coolant. In the suppression pool, the activity re-leased from the damaged fuel is the only source of

' energy. At the beginning of a loss of coolant accident, the water in the core region boils and hydrogen is stripped from it by the steam bubbles that are released.

The rates of radiolysis' would correspond to the maximum value of G (H2 ) = 0.45, regardless of enuilibrium nar-tial pressures of hydrogen. However, later in the acci, dent, when boiling stops, the rates of radiolysis become -

a function of the hydrogen partial pressure, In the suppression pool, water remains in the subcooled state

~

throughout the accident and the partial pressure of

. hydrogen controls the rate of radiolysis at all times.

4 -

8- ,

n .

g_

.Jt u,.**.

GtHy, ,

1 5% 110 % 25 % l50%

g,4 l l

l\ g l

Percentage Oxidation l of Fuel CladdinD  : l75% -

s l , l ,

o.3 i

l l l Percentage l l l Release of l i

l ll Total lodine l Inventory i

~

l 00 %

l l l

i y.. l 60 %

I I l l l  ;

o o 0.2 I t o.4 0.6

\i o.8 e i 1 e i s 1.0 1.2 1A 1.6 1.8 2.0 2.2 .

Partial Pressure of Hydrogen, atm.

FIG. 2 G'(H ) 2VALUES IN BWR HITH MARK I CONTAINtiE T The curves 'in Figure 2 illustrate how the hydrogen and iodine is expected to control the rate of radiolysis  ;

, G (H2 ) value) in a typical BWR plant with Mark I of '

tiark II containment. In this figure, G (H 2) values are l plotted for different partial pressures of hydrogen in the containment atmosphere, and for different percentage releases of total iodine inventory, which is of the order of 12-15 kg. In addition, vertical lines mark the partial pressure of hydrogen produced by the oxida-1 tion of different fractions (expressed in percentages) of Zircaloy fuel cladding. In this example cladding i

oxidation is considered to be the only other source of hydrogen besides radiolysis. The other sources of L hydrogen as well as the reactions contributinq to the l

consumption of hydrogen are not considered because in .

most cases they are of a considerably smaller magnitude and will rr>t have siqnificent effects on the mechanisms of radiolysis.

Because oxidation of cladding occurs at the beginning -

of the accident and is completed in a considerably shorter time than it takes for. water to radiolyze, it can be assumed that the hydrogen Droduced b

. of cladding will affect radiolysis from its,vinception. oxidition

. The initial value of G (H 2) is given by the intersec-tion of a a for a give'iven n iodinemetal-water release. reaction line with the lineAs the radiolys more hydrogen is produced, thus the operating point moves to the ri*ght along a constant iodine release curve. Since these curves are sloping down, the value -,

of G(Hp ) will be decreasing until G (H2 ) = 0. At this ,

point radiolysis will stop.

I. g .

i. . .

• The total amount of hydrogen which needs to be gener-

• ated before radiolysis stops is shown in Figure 3. As expected, this amount will increase.with increasing iodine releases. This total amount also includes the hydrogen generated by cladding oxidation which is indi-

, cated by horizontal lines in the figure.

It can be seen that for a certain fraction of cladding oxidized, enough hydrogen is generated to prevent radi ,

olysis from taking place in subcooled water. In this '

case. the only place radio 1.ytic oxygen can be formed is in the core region during the boiling regime when boil-ing will remove dissolv M hydrogen from the water.

o 2000 Mass of Hydrogen in Containment kg.

1500 -

--- 100 %

Percentage

--- 75 % (sja

---m j

e y,,

• ___s%

--- 10 %

0 l I ' I '

O 20 40 60 - 80 100 iodine meisese. % i

?!G. 3 #10tlNT OF HYDR 0rEN IN MARK I CONTAIN.NENT REDUIRED FOR STOPPING RADIOLYSIS OF WATER

~

There is also another reason for the reduction'6f ~ '

radiolytic gas generation with time which has to be considered. This reduction is due to the decay of the energy source required for radiolytic decomposition of water. As 1t was mentioned previously, the number of molecules of water decomposed is directiv proportional to the quantity of energy absorbed. As the strength of

'the radiation source decays #with timer less energy is

.. . . . . . ...... .. . ... . .,...w , . . ... .

f - .

,,se,o*

• availathe to be absorbed by water and, as a result,

, hydrogen and oxygen are produced at lower rates. To express the actual rates of generation of these gases, '

both the change in G value and the decay of energy l

source should be included. .

.This is illustrated in Figure 4, where the actual rates l of hydrogen generation for different, periods after l t

.beginning of radiolysis are presented. These rates are 0.050 I Hydrogen Generation Rate kgImin. ,'

0.045 - - e

{

0.040

} Release of Fuel Total lodine Cladding

{ inventory, % Oxidation, %

0.035 j Curve: 1 100 75 i

2 10 10 j g

3 1 1 0.030 J\

\

\ .

\

0.025 -\\

\\

• \

O.020 - \ '\g l

N s% %

91 0.015 -

\ .%.

I 2

• 0.010 -

l '

~~_____1 0.005 -

, 2

' I . . i l3 i

~

i i i i i *r i i o l l 0 2 4 6 8 10 12 14 16 18 20 22 24 , j Time, hrs F15 4 HYDROGEN GEtiERATION RATES IN BWR WITH PiARK ! CONTAINtfENT I

s e e ..* *a , e p e

t det& mined for a typical BWR plant with Mark I contain-ment assuming activity release to the suppression cool .

water in accordance with ANS radioactive decay corre-lation (5) and the G (H 2) values given by Figure 2. ,

The marked jump in the curves in Figure 4 is due to the assumption that boiling in the core region will

• " cease after twelve hours and G (H2 ) will decrease due to build'up at the dissolved hydrogen in water.

Although in.this example'it is assumed that this will occur instantaneously, in reality it will take some -

finite time for the hydrogen concentration to increase and hence the hydrogen generation rates will gradually approach their lower values.

SUMMARY

AND CONCLUSIONS

' In water cooled reactors, water can undergo radiolytic decomposition with the resulting generation of hydrogen -

and oxygen gases. Under normal operating conditions, there are well developed procedures for handling this situation. Under accident gonditions, the problem may become considerably more complex because radiolytic gases could be either produced in the containment or vented to it from the damaged primary cooling system. ,

BWR plants with Mark I and Mark 11 gentainments pre-

_ sent a special case. In these p1&nts., hydrogen ijni-tion is prevented by naintaininn the inerted contain-ment atmosphere. During an accident, the atmosnhere can be deinerted by oxygen generated from radiolysis.

Because radiolytic oxygen production is dependent on plant oarameters and accident scenarios, each cas.e has ~

to be examined individually to ascertain whether the containment atmosphere could be deinerted.

A methodology has been developed, based on a simplified

• mechanism of radioly' sis, which permits estimating rates of peneration of radiolytic gases under different acci-l dent conditions. These rates decrease with increasing  !

concentrations of hydrogen and decreasing concentration of iodine dissolved.in the water. Radiolysis does not continue indefinitely, but stops even before the radia- ,

i tion level is reduced to insignificant 1y low value when '

certain limiting concentrations of dissolved hydrogen  :

are reached. Since during accidents involvinn loss of coolant, the main source of hydrogen is oxidation of fuel cladding, production of radiolytic gases will be -

reduced. However, high degree of cladding oxidation is  !

usually accompanied by high fuel damage, which in turn --

results in a.high release of iodine. Torsatisfactorily '

I predict the amounts of oxypen produced in BWR's with  !

Mark I and Mark II containments, one should be able to i realistically postulate the relationship between clad- j ding oxidation and iodine release; .'

l 12- '

. _ _ . _ . _ __ _ . . . . . = _.. _..... - - - .

" f" . -

l

• 2 ~ j,o *

, l

[ . e4 l

I l

l . REFERENCES ,

. 1 (1) M. B. Carver, D. V. Hanley and K. R. Chanlin, l Maksina, Chemist, a Program for Mass Action Kine- i tics Simulation by Automatic Chemical Ecuation l Manipulation and Integration Using Stiff Techni- ,

ques, AECL 6413, February 1.079. i i

(2) E. H. Chance, A. R. Curtis, I. P. Jones and C. R.

, Kirby, FACSIMILE: A Comouter Progran for Flow 1 and Chemistry Simulation, and General Initial i Value Problens, AEDE D.8775, Decerber 1077. i l

' 1 l

(3) Dr. A. O. Allen nrivate ,corrunication. .

(4) H. E. Zittel and T. J. Aow, Radiation and Thermal

! Stability of Spray Solutions, Nuclear Technology, .

l l Vol .10, April 1971, p. 436,4a3. "

4 l  ! (5) U. S. Nuclear Regulatory Conmission, Fenulatory .

l l

1 Guide 1.7, Revision 2, Control of Combustible Gas ,

i Concentrations in Containment following a Loss-of-Coolant Accident, November 1978.

t =

I l

i - --

e age a.

f i

l .

.