ML20005A892

Forwards Guidance Re Concentration of Boric Acid in Reactor Vessel During long-term Cooling.Encl Provides Method for Reviewing App K Submittals

ML20005A892
Person / Time
Issue date:	01/21/1976
From:	Parczewski K Office of Nuclear Reactor Regulation
To:	Novak T Office of Nuclear Reactor Regulation
References
NUDOCS 8107020118
Download: ML20005A892 (12)

Text

l

_ p

,n..--

-~

/V

/+N OIT,T c, ni M

'o UNITED STATES hO d2dO y*w 1

j f[H1, ', I NUCI. EAR REGULATCRY CCMMissICN

• r.

'E WAL4tNGTON. D. C. 20555 s % ~.-.

a gv j-

+....

JAN 21575 Thomas M. Novak, Chief, Reactor Systems Branch, SS t

TliRU:

Robert L. Baer, Chief, Reac:cr Safety Branch, OR /0",,,'

CONCENTRATION OF BORIC ACID IN REACTOR VESSEL DURING LONG TER'4 COOLING - METHOD FO?, REVIE; LING A??E.'; DIX K SL'5MITTALS A:: ached is a me.norandum entitled: "Concentra ica of Scric Acid in Reactor Vessel during Long Term Cooling - Method for Reviewing Ap;:endix < Submittals".

of Accendix K submittals. The memorandum is intenced for One reviewers It describes the methces used in reviewing the calculatiens cf beric acid buildup during a post-LOCA 1cng term cooling.

m

'\\. b. Y'., o...... k K. I. Pare:ewski..

' Reactor Safety Branch Divisien of Ocerating Reactors

Attachment:

As stated I

cc:

D. Ross D. Eisenhut

{

Reactor Systems Branch Members g y:I t,

Reactor Safety Branch Members gs f

V

.j Q'o ['f[,(

s 1

lr}.4 y

.OM,

./h

,jl]*.I l,a.egw '~ )\\

\\6-u.qgua..,,,

o

~,

j u

-wu e,.,

NG

,,N j j

"%;mgQY s

1 t

i I

t E

ti k.~

8107020llTen; f

i

\\

)

gu.

,,.. v. r-v a v

..?...,.

E

~-

r~

.A t

5 l

Encicsure 1 y

u 1.

CONCENTRATION OF BORIC ACID IN REACTOR VESSEL DURING LONG TERM COOLING - METHOD FOR REVIEWING APPENDIX X SUSMITTALS I

t I.

Descrioticn of the Problem l

i Following a LOCA, boric acid solution is introduced into the In the initial reactor vessel by two modes of. injection.

injectiga mode, borated water is provided frem the accumulators from the refueling water storage tank and frem the boron injection tank (Westinghcuse plants only). After this initial period, which may last somewhere between 20-50 minutes, the ECC s

o In this acde system is realigned for tne recirculation mode.

It is borated water is provided frem the containment sumo.

recirculated from the sumo to the reactor vessel and back to l

the sump througn the break. A portion of the water introduced

)

into the reactor vessel is converted into steam oy the decay Since the steam contains,'rtually heat generated in the core.

no impurities, the beric acid centent in the water that was vacorized remains in the vessel.

The concentration of boric acid in the core region'will therefore continuously increase, Withou:

unless a dilution ficw is provided througn the core.

the dilutien ficw the concentration of boric acid will eventually l

reach the saturation limit and any further increase in boric Soric acid descsited acid inventory will cause its precipitation.

in the core may clog ficw passaces and seriously comprer..ise the To prevent this frcm happening, performance of the ECC system.

the ECCS should be designed and ocerated in such a manner that a sufficient throughflow is provided before the concentration of The general performance boric acid will reach its saturation limit.

criteria for the baron dilution systems are given in Appendix I to this memo.

There are two situations when ECC system cannot provide sufficient j

diluting flow through the core.

It occurs during:

(a) cold leg injection with cold leg break l

(b) hot leg injection with hot leg break In both these c4ses the injected fluid does not have enough head Since it is not possible for an ocerat:r to pass through the core.

to distinguish between cold and hot leg breaks, the only way of assuring dilution ficw througn the core is to orovide ene of the folicwing medes of Operation for the ECC system:

l' l? ? f o

r

._3 my r

,pp-(

z } yyw W& ?3$T--q[_-._ - -

-Q-_

-- Qpa y _ y

. ~ -

t1 e

o e

c o

9

.O_

(a) alternate injectiorr to cold and. hot legs.

14 w

N to cold and hot legs.

(b) simultaneous injection cu simultaneous cold leg injection and hot leg suction.

jg IN (c)

,g Methods Proces'ed bv the Vendors for Preventino B

• 9 II.

-i Concentrat,cn

~ Babcock and Wilcox 1.

The B&W plants are unique in that they cossess internal v

-1 The existence of

'J between the upper plenum and the downcemer.

l in the

((

these vent valves allows natural circulation to t is y

S&W has hign enougn to reach the vent valves and keep them open.

id perfor ed an analysis indicating that regardless of hot er ::

1 leg break, tne natural circulation flow througn :ne : re ::ul:

I ce i

is j

maintained for over 30 days after a LOCA and no ocerator act cn 3). After needed during that period of time (Reference l

They are as follcws:

4 e

- - _I __

~~~ Cestablishing diluting flow through the.cor.

I injection into de downcemer and sucticn or drainage 1

(a) frem the hot leg thr ugn the cecay heat line.

.)

into the dcwnc:mer nrough I

simultaneous injectionthe injection line and into the hot le 4

(b)

[

decay heat line.

into the downcemer tnrougn the i

simultaneous injectioninjection line and into the hot leg i

(c) i 5

pressuri:er spray line.

!t B&W claims that any of these methods, if initiated within 30

(

ation will prevent beric acid concentration frca reaching the satur 1

l - )

limit.

d The NRC staff has reviewed the claims m The dilution flow through the core well in excess of 30 days.

i flow medes of cceration, proposed by B&W for l')

because their performance decends on the parameters of eac l

t I

j individual plant.

l l

1 l

I I

t I

i

~

3

)

hdh

, %---e%.*.QTQ 'mm%.

Q C ~k % ;.y ;

Q4. i+%ggp" ~.~.

..u.. 2..g.9. e,,3,.c. _% x

.,,y.__97M y p1y2 T ' _ =rt %.;;

w m.-.,%.;:::-.m. g.

Qy2.,.

% %W JSyp3

~

2.

Ccmbustion Engineering L

In the Ccmbustien Engineering plants the initial injection of borated water is through the cold legs.

After some specified time, cold leg injection is reolaced either by simultaneous hot and cold leg injecticn, or by simultaneous he leg sucticn and cold leg injection.

The time wnen the switchover mus: take i l place is ce: ermined by tne concentration of boric acid in the reactor vessel.

i l

In the analysis (Reference a) Cembustion Engineering makes two assumations.

It assumes ideal mixing of beric acid solu:icn in t

the ' reactor vessel which produces uniform concentration builduo in the whole vessel and it assumes that in all cases there is a -

i residual cressure of at least 20 psia in the reac:Or vessel.

.3 This pressure causes higner boiling licuid temoeratures witn the resulting increase of 41/2 weight percent in beric acid sciubility, Both these assumations are ncn-conservative and Ccm:ustion c

Engineering does not provide satisfactory justification f:r i

including them in :te analysis.

t

)

In the 'preceduce provided by Combustien Engineering, het l'ag i

suction is ac::mplished by use of the shut:cwn c: cling suc:icn s

line and ne no: leg injection can be mace either enrcucn :he

,j shutdown 131ing suction line or tnrcugh the auxiliary ;ressuri:er i

spray line.

In most plants these three alternatives provide a i

system that can withstand a single active failure.

}

}

The procedure proposed by Combustion Engineering nas certain

-i limitations.

If the steam or:duced in the vessel cannot be f-freely vented through the hot leg, it can decress ne level of the licuid in the upper plenum belcw the bott:m of het leg j '

nozzles and impede the suction througn t.e shutdown c: cling

?

line. This case will occur when all 1 cop seals, formed by the U bend in cold legs, are filled with water.

On the other hand, if the steam leaving tne vessel reaches velocities tec high it s

may carry the injected fluid into the steam generator and prevent k

it frca reaching the care.

V.

E Both these problems have beer addressed by Combustion Engineering Z

(References 5 and 5) and it was demonstrated that, for :ne :lants h

censidered, these effects did no: seriously affec: :ne performance 1.

of ECC system.

m

.C'

s s4

$W nU ii*

"=

w W

1 m

6 '

3.

Westinghouse After an initial period of. cold leg infection, Westinghouse recommends hot leg injection as means for providing dilution flow through the core.

The ceneric analysis performed by

. Westinghouse (References 7 and 8) assumes that the fluid I

injected through the hot leg to the upper plenum mixes with the fluid in the reactor vessel.

Hot leg injection, therefore, provides a continuous dilution flow througn :ne core for both hot and cold leg breaks.

The Westinghouse analysis, whicn forms tne basis for these assumptions, is not comolete. Until more information is obtained, it is recommended that the initial cold -leg injection should be reciaceo either with simultaneous 4

i hot and cold leg injection, or alternate periods of hot and cold i

leg injection, so that aceouate dilution flow is provided fer sitner a hot leg or cold leg break.

III. MRC Review 1.

General Description The NRC review of the submitted procedures includes ince:enden~

evaluation of basic paramecers (switchover time, m aicum ficw rates,etc.).

The indeoendent evaluations have been performed for the follcwing reasons:

(a)

In some cases the information provided by the applicants were incomplete and it was not possible to check the soundness of their initial assumptions or their methods of analysis.

(b)

In order te assure that under all operating conditions the procedures provided by the applicant will guarantee the maintenance of safe boric acid concent.ations, the NRC staff proposed a series of conservative assumptions.

It is necessary to determine how this additional f

conservatism'would affect the procedures proposed by l

the applicants.

One 'of the most imcortant assumotions introduced by the staff, was tne four weign percent margin in tne saturaticn concen:raticn I

of boric acid in the core region.

This safety margin was introduced to take care of inaccurhcies in prediccing saturation concentrations in the core.

The NTC review censists of two parts:

(a) analysis (calculation) of boric acid concentrations (b) hardware review sup

)

l 2.

Analytical Methods The folicwing methods were used in calculating different performance paramaters for the systems reviewed:

(a).Switchover time from cold leg to hot leg (or

~

simultaneous not and cold leg) injection.

In order to determine the switchover time the following conservative assumptions were made:

(1). During the initial cold leg injectien, boric acid 3,,

-dets not leave the core.

(2)

The maximum concentration of boric acid in tne core region (C-) should not exceed 23.5 w/o (4 w/o below 3.

\\f;

, tne saturition concentration at 212*F).

(3["'The initial injection period lasts for 20 minutes after a LOCA.

At this time the concentration of boric acid'(C )

3 is uniform in the wnole plant.

(4) The effective volume in which boric acid builcuo occurs consists of a volume of licuid in :ne core and in the upper plenum up to the height of the botten of the cold

. leg no::le.

The rate of boric acid buildup is expressed by the following differential equation:

dC

= ;. t ". (C-C)

~

dt Where:

C - boric acid concentration in core region C,,- maximum boric acid co.1 centration if all boric acid were concentrated in core region.

t - time A - group depending en several plant parametar n - constant O

e e

e -,

a-ne

,,w.

-e.-,--n

,y- - - - - -.

r,

o Selving the equation for the following boundary conditions:

t=to, C = C, t=ts, C = Cc and rearrangiag the terms, the following expression for the switchover time (t ) is obtained:

3 C - C, y 1/(1-n) t

=t 1

,. 1-n In s

p A. t l-n C,- C }

g g

Where:

A=K.P (anf +anfg). Ms a

for: t <A x 10" see (M days):

K = 0.156 and n = 0.233 s

M - mass of water in the containment sump, lb 3

P - reacter cper

!? Laer, 3:.uh e:

C - 0.3077 boric acid / water o

In the case of alternate hot and cold leg injections, the i

subsequent switchover times could be oo:ained by substi:u:ing t

e: frt in the above equation.

sl, Es2, ts3 o

(b) Mininum cold leg infection l

l The minimum cold leg injection recuired during recirculation (prior to switchover to dilution mode) is determined by the i

1 l

rate of boil-off in the core at the very beginning of the long

~

tenn cooling mode.

I It is determined from the following expression:

j t

.Q,, = 449 H_

gpm i

fin *(ah,+ah,.,)

I 3

l l

l

~~

Where: H - decay heat generated at the beginning of recirculatien 0

period, Stu/sec fin,- density of injected fluid, lb/ft ahf subccoling enthalpy, Stu/lb ah,g - vaporization enthalpy, Stu/lb a

(c) Minimum hot leg injr.ction The minimum injection rate, after the switchover from cold to hot leg or to simultar.eous hot and cold leg injections, is determined by making conservative assumotions that the m.nimum injection should previce enough ficw to reolace the boiled-off lituid and to maintain sufficiently high dilution ficw througn the core.

The boiloff rate is calculated using the decay heat generation rate at tne switchover time from cold to hot leg injection.

The dilutien flew is determined for the maximum allcwable cencentratien of beric acid.

Expression for the minimum injection rate is given below:

Oh=

44g H s

- (1 - 3.25 C ) an j finLanf o

fg Where: H - decay heat generated at switchover time. Stu/sec.

Thedeca)heatiscalculatedusingtheme:bodsgiveninReferenceg.

(d) Steam binding i

'ihen the switchover from cold to hot leg injection occurs

. shortly after a LCCA, large volume of steam leaving the reactor vessel may incuce high steam velocities in outlet nozzles. These velocities impede the injected boric acid solution frem reaching the core.

In order to assure that hot leg injection is not comprcmised, steam velocities have to be estimated and the f

rate of entrainment of the injected liquid calculated.

The rate of entrainment can be calculated using the methods given in Reference 10.

j I

m-v.,.

n n,

3.

Procedure In reviewing the boric acid submittals, the following procedure was used:

Steo 1 Review of the submittal and identification of any special features the applicant may be taking credit for.

In most cases the apolicants follow the generic analyses devel.oped by the. vendors. Hcwever, occasionally tne applicant may take credit for the phencmena which are not considered in the generic 4tudies.

For examole, no consiceration was given in the generic studies to the possible increase in the solubility of boric acid when sodium hydroxide is present.

However, some acclicant may use this phenomenon to increase the allcwable solubility limi:s.

In such cases the reviewer should identify and individually evaiua:e such features.

Steo 2_

Cetermination of the basis parameters identified in Section III-2 of this memo.

The information needed for the determination of :nese parameters are.either provided in the original submittal or can be found in FSAR or PSAR.

Steo 3 Hardware review Using the P&I diagrams or soecial schematic diagrams provided by the applicant, the systems are reviewed to assure nat they meet the requirements identified in the analytical part of :ne review, without violating the single failure critaria.

IV.

Conclusions l

The procedure for reviewing nuclear plant for beric acid i

precipitation is outlined in this memorandum.

It should be mentioned that although this procedure may apply to the

- majority of plants reviewed, there may be some cases where the reviewer may be forced to use comoletely different approach. He should be, therefore, flexible in choosing his prccedures, guided by the considerations described in Appendix I.

Note the Accendix I is intended to apply fully only to C? applications.

I I 7 1

7.-

I i 3

V.

References 1.

Babcock and Wilcox, Topical Repor: 3AW-10102, ECCS Evalua-tion of B&W's 205-TA NSS, June 1975.

2.

Babcock and Wilcox, Topical Report BAW-10103, ECCS Analysis of B&W's 177-TA Lcwered-Loop NSS, June 19.75.

3.

Babcock and Wilcox, Topical Reporr BN4-10105, ECCS Evalua:icn of B&W's 177-FA Raised-Loop NSS, June 1975.

4.

Switzer, D.

C., (NNICO) letter to 0. D. Parr CIRC), dated June 27, 1975, a::ach=ent A.

5.

Swi::er, D. C. (NNICO) letter to 0. D. Parr (NRC), dated Septa =her 25, 1975, a::ach=ent 1.

6.

Telecon, J. Lengo (CI) and RS3 personnel CIRC), oc:ober 30,'1975.

~

t 7.

Case, C. L. (Wes:inghcuse) letter CIC-NS-309 to T. M. Nevak l

(NRC), dated April 1, 1975.

8.

Ca. ::ak, J. O. (Westinghousa) -let:er JOC-NS-369 to T. M. Novak (NRC), dated August.15, 1975.

9.

ANS, Decay Energv Release Rates Foll:ving Shutdown of Uraniu= - Fueled Ther=al Reactors, (Propcsed ANS S:andard),

October 1971.

10.

Wallis, G.

3., One-dirensional Two-phase Flev, McGraw-Hill Book Cc :pany,1969, Section 12.10.

I t

l l

l t

i i

l

.. -, _. ~ -,

Accendix I W

PWR BORON DILUTION SYSTEMS FOR'CP APPLICATIONS WHICH MUST MEET 10 CFR 50.46 CRITERIA FOR ACCEPTANCE 1.. The borep dilution function shall not be vulnerable to a single failure. A singie active failure postulated to occur during the long term cooling period can be assumed failure would then be in lieu of a. single active failure during the short term c: cling period.

2.

The spurious operation of any motor operated valve (coen or closed) i shall not c:mpromise the boron dilution function nor shall it jeopardize the ability to remove decay heat frem the primary system.

si 3.

All ccmconents of the system which are within containment shall be designed Oc seismic Category l recuirements and classified

~

.l Quality Group 3.

4.

The primary.mede for maintaining acceotable levels of baron in the vessel should be established.

Shoulc a sinale failure disable :ne primary mode, certain manual actions outside the centr:1 reem wculc be allcwed, depending on the nature of the action and the time j

available to establish back-u: moce.

j 5.

The average beric acid concentration in any region of the react:r e

vessel should not exceed the level of a weignt :ercent belcw the

~ l

'j solubility limits a: the tamperature of the solution.

r:.

L 6.

During the post-LOCA long term cooling, :he ECC system normally lg operates in two modes:

the initial cold leg injection mode, l

folicwed tiy the dilution mode.

The actual operating time in the t

cold leg injection mcde will depend on plant design and steam

^

i l-t binding considerations, but, in general, the switchever to the H

dilution mode should be made between 12 and 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after LOCA.

S

!?

7.

The dilution mode can be accomplished by any of the following means:

t t

(a) Simultaneous cold leg injection and hot leg suction (b) Simultaneous hot and cold leg injections i.

(c) Alternate hot and cold leg injections.

! ?

8.

In the alternate hot and cold leg injection mcde, the operating j

time at hot and cold leg injecticn shculc be sufficiently sner:

3 to prevent excessive boric acid builcu:.

l l

l.i!

l l T.

e

!{

t.

-.2 4

L

. l

i. The mini :um ECCS flow rate delivered to the vessel during the dilution :.cde shall be sufficient to accormiodate the boil-off due to fission product decay heat and possible liquid entrainment in the steam discharged to the containment and still provice sufficient licuid flow througn the core to prevent further increases in.beric acid concentration.

10.

All dilution modes shall maintain testability comoarable to other. ECCS modes of operation (HPI-short tenn, LPI-short term, etc).

The current criteria for levels of ECCS testability shall be used as guidelines (i.e., Regulatory Guices 1.58, 1.79, GDC37).

l

,s

~. -

l I

l i

i l

l l

.~ ~.....' L=i=r_' _ =x==

- -fT

= _ - _

w =" m = = re-m M

.y

~ ' ~

....