ML20147C111

Forwards Sixth Monthly Rept by NRC Resident Engr Re Progress of Core Meltdown Research Project in Germany for February 1978

ML20147C111
Person / Time
Issue date:	04/12/1978
From:	Johnston W NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:	NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
Shared Package
ML20147C115	List: ML20147C111
References
NUDOCS 7810110278
Download: ML20147C111 (1)

Text

- _ - _ _ _ _ - _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .

y.d Mo vy*o UNITED STATES e 8 '

o NUCLEAR REGULATORY COMMisslON

$ .E WASHING TON, D. C. 20555 IIdqtts. PUR

.E

%S ,,,,*# APR 121978 MEMORANDUM FOR: Those on Attached List FROM: W. V. Johnston, Chief Fuel Behavior Research Branch

SUBJECT:

TRANSMITTAL OF MONTHLY REPORT BY NRC RESIDENT ENGINEER A7 KERNFORSCHUNGZENTRUM, KARSRUHE, GERMANY Enclosed is the sixth of a series of monthly reports to be provided to NRC/RSR by Mr. Walter Murfin, resident scientist from Sandia Laboratories assigned to KfK in Germany. Mr. Murfin's principal responsibility at KfK is active participation in Projekt Kernschmelzen (core meltdown research project). The monthly report is his vehicle for describing progress in that program.

Mr. Murfin also maintains cognizance to the extent possible of other safety research at German facilities of interest to NRC/RSR including:

- LWR fuel rod behavior

- experimental techniques for measuring two-phase flow

- jointly sponsored (US, FRG, Japan) experiments on upper plenum thermal hydraulics.

Your comments on the contents of the monthly reports are transmitted to Mr. Murfin, or you may contact him directly.

l bt/,L] lw W. V. Johnston, Chief Fuel Behavior Research Branch Division of Reactor Safety Research

Enclosure:

February 1978 Monthly Report cc w/o encl.:

W. Murfin , ~KfK 7 P l # 1 1 e .1 7 f

Kornforschungszcntrum Karlsruho i Gesefschaft frdt beschrankter Haftung i

1332

~

K.,nvorsce.nirum Kae o,e Poet,ach sc4o D 7soo xa,t.ruhe s '

Institut for Dr. William V. Johnston Reaktorbauelemente [

United States Nuclear Regulatory Com.

Fuel Behavior Research Branch Washington, DC 20555 o,,um March 22, 1978/de searDener. W. B. Murfin U.S.A. @ y Towton:Onm 82 - 3462 Ihre Mitteilung:

m

.e*V 2

~~% a

- - en -.

~

cra

.v 5 9

4 9 ... l i

i s t j

Dear Dr. Johnston,

-. enclosed is my sixth monthly report, covering activities during l February, 1978. i i

Sincerely,  ;

l a

W.B. Murfind' 4i T

- Dr. R. DiSalvo, USNRC Dr. D.J. McCloskey, Sandia Prof. Dr. U. Muller, IRB - KfK Dr. M. Fischer, PNS - KfK s

i Komforschungszentrure Kartsruhe, Wissenschaftbch-Techrusche Ebnchtungen und Verwarung 7514 Eggensten-Leoceedshafen; Tel 407247) 821. Telea. 7826484. Drshiwort. Reaktor Kartstuhe; Stadtburo: 7500 Karlsruhe. Weberstra8e 5 l  ?

' Vorsitzender des Aufsschtsrats Staatssekreter Hans Hager Haunschtd;

• Vorstand. Dr. Rudolf Harde. Vorsitzender, Dr. Henmul Wagner, Stellv. Vorsazender, Prof. Dr. Horst Bohm, Dr. Hans Henning Henrues, Prof. Dr Handelsregister Amtsgercht Karlsruhe HR8 302. Baden-Wurttembergische Bank AO KarlanAe, Corimertbank AG Kartsruhe, De b:

f' I  ?

Y' ,

- 1v Introdu, tion his is the sixth report covering the activities of W. B. Mtrfin, USNBC representative 'at Esmforschungsentrum Karismhe GmbH.. he report covers

- activities during the month of February,1978 he. principal duties of 7' the representative are tasks assigned in Projekt Kernschmelsen, information -

> gathering, and monitoring German research results.

2 Projekt Kernseheelsen_

2.1 . Core / Concrete Modeline (PNS4331) 2.1.1- Decomposition Enthalpy o_f, f Concrete

- hree methods of detamining the decomposition enthalpy of concrete have been investigated: the theoretical model of M. Reimann /1/, integration of the specific heat /2/, and total heat of decomposition measured in plas-ma tests /2,3/;. P

- he theoretical model considers the following reactions s ,

(1) H 2O(1) +H 20(g) 39.4kJ/mol (T=400K)

(2). Ca'(OH)2(s) +Ca0(s)+g0(g)-99 5 kJ/mol (T=796K)

) (3) CACO)(s)-+Ca0(s)+CO 2 (g) -165 5 kJ/mol (T=11671)

(4) Ca0(s)+SiO 2 (s)-+CaS10)(s) 488.5'kJ/mol (T=1573K)

(5) 81 02(8 ) - 8 10 2 (1 ) -8.53 kJ/mol (T=1573K)

(6) CaSiO 3 (s)-+CaSio)(1) .46.5 kJ/mol (T=15731)

Se temperature for melting has been taken to be 1573K (1300' C). Bis is very close to the reported temperature for complete melt for limestone

' and basaltic concretes, but is somewhat higher than the temperature for the onset of melt /4 ./ Se heat of decomposition has been calculated both  :

't with and witheut consideration of reactions (4) and (6).

4 -

For the theoretical model, the initial composition of concrete was simpli-4 fled. Only CaC09 Ca(OH)2, SiO2 *nd H2 0 were considered to be initially preser.t.' his it only a fraction of the many substances in concreta, just

. as the reactions listed above represent only a part of the possible reac-6 tions /5/. One purpose of the analysis conducted here was to determine

_f_

_w hether the simplified model was adequate: 1. e., whether the substances and reactions considered represented the most significant part. ,

Figure 1 shows comparisons of results for a silicate concrete, without

.< consideration of reactions (4) and (6).

All methods show surprisingly good agreement. Se measured value of spe- ,

cific heat does not go to a sufficiently high temperature to determine the

~ decomposition enthalpy at the_ point of complete melting (1573K). However, the-integral of specific heat follows the theoretical curves quite closely.

I t #

6

- , ~ . . , , , , -r,-r, r-,..- ,,+, -wn.,a , n r s,w gn, w

3 ,  ;

,. ,'A c, * '

Figure 2 shows the results fcr lim 2stens concrete. *E th:rstisc1 caddl has been used both with and without consideration of reactions (4) and (6).

he agreement of the- horetical curves with integration of measured specific it is' not possible to determine frem W

curves heat is .whether reasonably good. - However,6) reactionst(4) and ( sh ould be included: the s intsgral could be reasonably extrapolated to either point (B) pecific heator p int (C).

. It would be difficult to $stify extrapolation to either of points (D) or (E).

b experimental detamination of heat of decomposition from high heating rate testa requires a rather complex analysis involving uncertain quanti-ties' /2,3,6/.c Because -of the exper$menta1 difficulties involved, point (B) of Figure i and points (D) and '(E) of Figure 2 are presently assigned 6 lower order of reliability. The agreement of point (B) of Figure 1 with other values may be fortuitous.

Until better experimental values of decomposition .enthalpy can be obtained, the theoretical model is being considered adequata - in reasonable agree-ment with measured e b discrepancy between theoretical and experimental values needs to be e$e.ered up. However, in W present state of uncertainty, it does not appear reasonable or desirable to include a more complex initial composition or more reactions in the model. It does appear that a simple, unequivocal test for decomposition enthalpy is required.

References

/1/ M. Reimann, et al., 'Ein Modell zur Beschreibung der Wechselwirkung

- einer Kernschmelzen mit Beton", UX2395 (Oct.1W7). ,

/2/ M. hehs, et al., k. Techn. Bericht, IEFT RSiy4 (aug.,1977). l

/3/ T. T. Chu,* Radiant Beat Evaluation of Concrete - a Study of the Erosion i of Concrete Due to Surface Beating *, SAND 77-0922 -(Jan.,1978). l

/4/ Light Water Reactor Safety Research. Prp 7-

- June 1976, p. 65, SAND 76-0677 (WF10766519) (Feb. ,1977).gram quarterly Report - April-

/5/ Light Water Reactor Safety Research Program Q2arterly Report - January- j March,1976, pp.115-124, SAND 76-0369 (EREG766506) (sept.,1976).

- . )

. /6/ J.' F. Lir,'

• Response of Conerate . Exposed to a High Boat Flux on One Surface", SANIf/7-1467 (Nov.,1977) 2.1.2 Oxidation g Metallic Components g the Melt A diffusion model for steel oxidation /1/ has been utilized to show that the diffusion time is very shorts equilibrium will be reaebed rapidly. It has also been shown that the reaction -

Fe + H O Teo + H2 is of primary importance.2The ex.idation of Feo to Fe30 g can play only a

( ,

f s

9

.h.

4 t - . ,, . ~ . . - .-,..,u ., y_c., a

i

/ ,

r:10tiv1y tin:r rolo. F r tha oxidstion cf Er end Cr, th3 cquilibrium partial pressure ratio p(R2)/P(B,0) lies very near to pure hydrogen. It em be assumed that water vapor passing through the melt is completely re- .

duced if any Er or Cr is present. Similarly, CO2 passing through the melt '

will be completely reduced to CO by Er and Cr, and partially reduced by

• Fe.

Oxidation has been modeled in the WECESL code by assuming complete redue-tion of H 2O and CO 2 by Er, as long as any metallic Er is present. After complete oxidation of Er, gasee can be reduced by Cr, as long as any is present.

- Af ter complete oxidation of all Er and Cr, Fe is oxidized to Fe0 he equilibrium constants for the reactions of H O2and CO2 with Fe have been approximated by legio =0.04301.5/T

)

loggp p(CO) = 1.25-89 3/T, p(CO2 )

where .T is the melt temperature.

%e oxidation of Ni is not considered: the partial pressure ratios for Nio lean strongly toward pure H O 2 and CO2

• ne oxidation of Er and Cr is strongly exothemie. Calculation of testa l with sustained heating show that the exothermie oxidation of Cr can con-stitute a significant addition to the heat of a stainless steel melt. ,

he oxidation of Fe by H 2O has only a small heat of reaction, and the oxidation of Fe by CO2 is weakly endothemie.

Tne space above the melt contains H2 , H2 0. CO, and CO2 . Other gases could

- be present in small quantities. It is assumed that the vigorous gas flow )

from the melt drives off all air. Under these conditions the hongeneous i water-gas reaction is to be expected:

H2O + CO. *e__.H2 + CO2 )

Se equilibrium ecinstant for this reaction has been estimated to bes l

logio K = 2010./T,-1.778, where T is the gas temperature. The homogeneous water-gas reaction could also taEe place within +.he melt; however this is of reduced importance.

he reaction is important for the containment, because it alters the liber-ation of R2'

- Calculations have shown that the assumption that all gases are completely l J

reduced in their transit through the melt is very conservative. Even when Zr and Cr are present in the melt, a considerable fraction of H O2 and C02 will be delivered to the containment atmosphere, because gases liberated

• I s l T

i @)

r - . .

en the sides of the melt cannot be completely reacted.

References:

/1/ ' . Reimann, et al., *Ein Modell zur Beschreibung der Wechselwirkung eirer. Kernsehmelten mit Bete #, ifK 2395, (Det. 1977) 2.1.3 Delays in, Arriving af Steady-State Melt Conditions he IFK film model assumes a quasi-stationary melt front velocity. However, under transient conditions, the melt front velocity could be changing rapidly.

he delay in acheiving a steady velocity has been calculated.

Another secree of delay is given by the kinetics of concrete decomposition.

A kinetic model for concrete decomposition has been formulated by Powers

/1/. Only chemically bcnnd water release is well fitted by this simple.

model, he delay due to chemically bound water release is of the same order of magnitude as those for release of. physically bound water and calcite decomposition, and can be considered a reasonable praxy for all delays due to decomposition kineties.

2e table below gives approximations for both types of delay for a typical k stainless steel melt atteking a typical silicaceous concrete, b two types of delay are actually ecupled however, a rough estimate of their magnitude can be obtained by calculating them separately.

Melt Delay in Steady- Delay Due to b p. State Ercsion Vol. Decomp. Kineties*

(OC) (Tg - s ec .) (T2 sec.)

1500 195. 23.

2000 18, 3.4 2500 3.7 0.85 4

3000 2.0 05 j

-9 .

.

• Time required to' release 90% of bound water, when temperature is increased at a uniform rate from 200 C to 1300 C ever the time Tg.

b The delay due to decomposition kinetics is always small compared to the delay in achieving a steady melt velocity. Both types of delay are neg-

. ligible, except for low temperature melts or for simulation of short dur-ation tests. In view of this, the model ignoring the complicatiras of non-stationary melt front velocity has been retained in the WECFSL code.

References

/1/ Light Water naastor Safety Research Program Quarterly Beport - January

. March,1976, SAND 76-0369(NURF4766506) (Sept.,1976).

+

h i.

e. ,

d4

$_: - ~- -

.s-

7 < . .

,- t

' 2.2 - Pi-sion Produnt Reinsm (PNS4315) 1bere is as yet litk.le progress to report on the improved SASCHA facility.

Time has been~ mostly spent on technical inspection, improvement of comp. iter codesg etc.

3. Fue: Rod Behavior Research No report in this period.

4 Wo-Phase Flow Instmrentation Evaluation g Infra-ref Absorption basurements_ i_n Blowdown hsts The infra-red measuring system for two-phase flow was developed jointly by KFK (IRE) and the University of Karlsruhe. A sketch of the system is shown in Figure 3. Tbe following assumptions are made for analysis of the measurements :

-Air and water vapor behave as perfect gases

-Steam and water exist in thermodynamic equilibrium

-The liquid phase is uniformly distributsd over the flow cross-section l

-Water volume is negligible compared to steam volume  !

-Liquid droplets are not influenced by neighboring droplets in light scattering (no multiple scattering)

-Droplet size distribution is monodispersive.

Subject to these assumptions, it is possible to specify the steam density, water concentration, droplet site, and flow velocity of the liquid phase.

Tbe instmmentation system was employed on blowdown tests at the krviken facility in Sweden. Data were reduced by the University of Karlsmho and by the hrviken Project as well as by KFK. Typical data are shown in Figures .

4-6 Figure 7 shows a comparison of liquid phase velocity evaluated by the University of Karlsruhe and brviken.

Further analysis is required for comparison with other measurement techniques.

s e e e

4 8

t i

e K _

-m - -

d

)

if ) **' g77- -

l t'" ? r : s 2$

~

$ha  !

2500 $

, r 3: . ,. j pi ,'

2000 .- () (C)

/

/ .

.n /

.s

• /

/

4 /

M

/

/

g_

1500 -

/

(i I Integral of cp/2/

d /

U / j

/ /

/ *

/

1000 - f ,

/

/ */

. ('

/

/

/

500 -

/

]

/

/ /

g,

, j FIGURE 1. Decomposition enthalpy of silicaceous.  ;

,' concrete. (A)- thoretical, 4% espillary water.

e < (B)- measytred value /2/, (C)- theoretical, dry. -

?.A i 1 l 500- 1000 1500-Temperature, K

g g (D) 3000 -

n.

e.i *p e m e. m

}* *s .. -

it E ,i ,

0 @ (A) .;. . >

2500 -

, (B)

/

/

/

/

/ )

/ 1

' )

, , @ (C) s' .l 2000 -

4., i

- li '

/

5

- /

1 /

s j 1500 -

p'

, Q (E) t ./

A /

/

j/

/

/

,/ ,

Integral of cp /2/ '

1000 -

,- / ')

• /

/ o

(

,I

/

./

/

/

/ o

~

/ -

A

'" f / FIGURE 2. Deccaposition enthalpy of 1$estone c oncrete. (A)- theoretical, 4% capi 114ry b. ,

,f ',/ -

water. (B)- theoretical, dry, without wolla,,s ,

• /. tonite. (C)- theoretical, dry, with wollas (D)- measyred q/. (E). measured /2/.* je , ,tsitte 1500 500_ _1000 bzparature, I

w. . . -

, t 4

1 i.

, l

! PbS Test Section~.

s 4

1 ML1f. transparent {

' 1 MiFFOPS .

- i j s N

i L L L .

Pb5 La g' jh l (h i

,\'d

% l-(T(C{! l j $.. z ~\ //

l / A2  !

Motor /

h

/

i Interferen.tfitter \

/

A3 k Pb5 i

2

.4 _ , . - . .

g I 1

-s Y DB i'  !

. Figure 3. Opties of the IR measurement system.

.s i

< i s

e

m . .-

e R ,

.;a ;;. ,

h- -

. i-f.

w; -

-J--

f'Ipff 0 hI'y!f f L-4 .

1 g.

fbefff $M t,W ,&.\

n_ -].

<c '; '

i-i , i i, i, it i

i * * *

• . , 3.c ', st o .

i 6

e..

.u, .Pigure 4. IR signals during the start phase of blowdown 18 c

s V

,j '

,i

.t-

o > '

-f J .

e .V ,

~

, - - - -.a.a n , , . , , . , ., , _ _ _ _ , ,

~ ~~ ~ '

~~

nc >

(9) r- .. . ,-

t '

. ~1 ,

f

~l f

./,

~. p-b -

..A i.! g r

1.

r .

I 1 r

z.i

. :4 -

.I r Ea s

.h:'

' a

_o v c  :.

c. . -

r..:

l a .

y  ;  ;  ; . i s

• r . 9:

f:

, -]

l

, 1 gm e~N  % . ,,

e a.g. .. s.

. .l

- EJ I

I o ..l ame

, =

ra 4 a~

F.- . .J 4 . q 4 C t N s; *

., i -

r i r -

/

o  :.

9 1 _ , _ .

4

, a g ,.

• 4 ting istc i Figure'5. Pressure (above) and.temperatur~e Abelow) during

~ '

the starting phase of blow own -18

• d .

2 3 .

2%.=

ryc _

. m p . .

p ...

. r

\

e 80 4 4 i e i

, l O

• O 1

. s-- .

O E

o p -- -- O - .o . , ~ ~ 0

^

s _ ,

+

1 O / p # p #o l %y%

2 to -- / o t ,

O '

' O I Op o

{0 0

/

.' . .': is - / -

O '

so /

E /

3 >

E 0l0 -

,o -- 10 -

O '

I

. s_-

[ _

c 4

'g Blowdown 18 r i , , , i o

, e i 4 i .

p, o 25 to 15 100 tal isc o

Zeit / sec

. l 1

l 1

1 7 SD 3 i ' ' ' j O

.,9 . O o O

• s%g ,,,O, , , s u n

' ,0 s' # *tY t

, . O o

"~~

. 090

• A.. w.,~- ~O. m. J

• O. *""

) -.

w 10 , ,g,. +# , ,, p . -- . % - .y.  ;

4 # +

>g/m i o-- -

1

. s.

. -l.

.. 3e _- 9 ----- -

y *

/

6

+ I 2c - J, '

,, i l

.e) s,. .

l l

i, _ _@ ,

l -

I e Blowdown 18 e

. o f i f i f

, i .j

, J,: , . . e a so is too $25 '5:

j

<1-Zeit I sec

. l a

~

Figure 6. Mean drop radius'.(sbeve)'and, water' concentration

""d '

l (below) on blowdown 18'.- ...- -

• 1 1

4 l

(11) 1 E

1

.i f, f Y

.. ,*( ,

1.p s.

I t

- -- r ,'

• 1I .

5 .

i

? I

' I d

- \

e

=

I %I -

g ,,

s)* H k

E  ; :Y fin E

0-8'

'. i h\)

Il vi i l ,

v

r. L

(

, a m -

- .c' c,  ;, u ,., 2 ies -

i.: in

c istc .

. - . i  ;

o. .

, l i i 6

I

. w' g _ .____ . _ -

n .

3:

g 'MV -

i

\h\ ' '

c 4

-3 ____ __. ,

\

S (3

= i i i

i x -

VH %l

. .(

_ _ a._.- ,,,

200 o<

• 140 150 183 g g 30 100 120 O 20

- Tine (SEC)

I . Figure 7. Droplet velocity on . blowdown 24. Evaluation by ,

! University of Karlsruhe (Above) and hsiken Project '(below).

a

__