ML19256A607
| ML19256A607 | |
| Person / Time | |
|---|---|
| Issue date: | 12/22/1978 |
| From: | Johnston W NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Campe K, Chipman G, Disalvo R Office of Nuclear Reactor Regulation, NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| References | |
| NUDOCS 7901090086 | |
| Download: ML19256A607 (20) | |
Text
.
b,*
UNITED STATES NUCLEAR REGULATORY COMMISSION y
'g VVASHINGTON,0. C. 20666 j
.w s,**v}/
al a
DEC 2 2 579 MEM0hkANDUMFOR: Those on Attached List l
W. V. Johnston, Chief FROM:
Fuel Behavior Research Branch
SUBJECT:
TRANSMITTAL OF MONTHLY REPORT BY NRC RESIDENT ENGINEER AT KERNFORSCHUNGZENTRUM, KARSRUHE, GERMANY Enclosed is the twelfth and final report provided to NRC/RSR by Mr. Walter Murfin, resident scientist from Sandia Laboratories assigned to KfK in Mr. Hurfin's principal responsibility at KfK included partici-Gennany.
pation in Projekt Kernschmelzen (core meltdown research project).
These reports were his vehicle for describing progress in that program.
Mr. Murfin maintained cognizance to the extent possible of other safety research at German facilities of interest to NRC/RSR including:
LWR fuel rod 5ehavior s
experimental techniques for measuring two-phase flow jointly sponsored (US, FRG, Japan) experiments on upper plenum thermal hydraulics.
Your coninents on the contents of the reports are transmitted to Mr. Murfin, or you may contact him directly.
Mr. Murfin has returned to assume his previous duties at Sandia National On behalf of the NRC and myself, I would like to express Laboratory.
our gratitude to Mr. Murfin for an excellent job as NRC representative to KfK.
Future reports in this series will reconcence following assignment of a new NRC representative to KfK.
U AL%
L&
W. V. Johnston, Chief Fuel Behavior Research Branch Division of Reactor Safety Research
Enclosure:
As stated CC w/o encl.:
W. Murfin, Sandia 79010900SG
~
1.
Introduction This is the final report of Walter B. Murfin, USNRC representative at Kernforschungzentrum Karlsruhe, GmbH (KfK).
My duties include modeling of core / concrete interactions, cognizance of German research efforts, and duties assigned in Projekt Nukleare Sicherheit.
The period covered by this report is October, 1978, and the first part of November, 1978.
2.
Core Meltdown Research 2.1 Core / Concrete Mcdeling (PNS 4331) the PAHR A presentation of the WECHSL code was given at Information Exchange Meeting, October 10-12, 1978, at Ispra, l
Italy, in a paper " Calculations for the Decomposition of Concr-ete l
l Structures by a Molten Pool", by W.
B. Murfin and M.
Re ima n n.-
l r
Progress is being made in the documentation of the WECHSL I
code.
A version has been "f rozen" for possible distribution.
I j
i 2.1.1 Influence of a Gas Stream on Heat Transfer Between Two Liquid Layers (PNS 4332) transfer Experiments were conducted to determine the heat between layers in a multilayered system with gas streaming.
Two The water fluid layers - water and silicone oil - were used.
layer was heated by a spiral electrical heating element and the oil was cooled by a spiral heat exchanger.
The container con-sisted of a porous glass filter through which air was blown.
Bubble diameters were controlled by overlaying the bottom plate with steel or glass spheres.
The inter-layer heat transfer could
be inferred from measurements of the temperature difference between the water and the oil.
The experimental setup is shown in Figure 1.
A s'eries of five measurements were made - each with con-stant electrical power.
Stationary temperature distributions Heat were measured as a function of volumetric air flow.
losses were determined by the difference between the electric heating power and the rate of heat withdrawn by the cooling system (mostly less than 10%, max. 20%).
The measu r ed ter..per-ature distributions were corrected in accordance with these If the corrected power and volumetric air flow heat losses.
are divided by the bottom area, one obtains the upward heat flux, j(W/cm ), and the so-called " superficial" gas velocity, 2
Vs(cm/s).
Some of the relevant material properties are givea in Table 2.1.1 (I).
Figure 2 shows temperature profiles as a function of V s*
and a The temperature difference decreases with increasing V s, broad mixing zone is formed.
Even at the highest values of thermocouples TE2 and TE5 and 6 lie outside the mixing V3, these temperatures thus correspond to the layer temper-zone; atures outside the boundary.
influence of gas bubble diameter was first investi-T.. e gated.
Figure 3 shows the results; a variation from 0.1 to 0.4 cm..does not make a great difference.
The following re-sults were for 0.4 cm glass spheres, which give various bubble diameters in the rar;e of 0.1 to 0.4 cm.
Table 2.1.1 (II) summarizes the results.
In the upper part of Figure 4, the experimental values are consistent with the re'lationship 4/3 j = h*(V3)h 7
o The constant h is.006 W/CM g /3 and the value of hh 24 o
was evaluated from the experimental data, with h*(0) = 1.0.
The value of h is in satisfactory agreement with that cal-o culated from material properties (.0045).
The lower part of Figure 4 shows ho'; the heat transfer is increased by V The normalized heat transfer j/h* is shown in the lower left, and h*
is shown at the lower right.
The gas flow causes irregular wave motion and mass trans-port at the interface.
At sufficiently high gas flows, this.
leads to near homogenization of the two layers.
The heat transfer w'as increased by a factor of 630 from V
= 0 to V
= 0.63 CM/s.
The gas flow in a core melt / concrete s
s would be even higher, so that heat transfer should be principally dictated by gas flow.
Increase of the effective boundary area and mass transport are undoubtedly influenced by density difference and surface tension.
A later series of experiments will investigate these effects quantitatively.
4 in 2.1.2 Experiments on the Simulation of Large Core-Melts Interaction with the Reactor Concrete (PNS 4323)(BETA)
T.he experiment plans and schedules are currently being reworked.
The crucible dimensions for 100-300 K melts have g
been fixed and are shown in Figure 5.
2.2 Fission Product Release (PNS 4243)
The expanded SASCHA facility has been accepted and some preliminary tests have been carried out.
Following these tests, the facility was shut down for further alterations.
" Release of Fission and Activation Products During A paper LWR Core Meltdown", by H. Albrecht, V. Matchoss, and H. Wild was presented at the ENS /ANS topical meeting on Nuclear Power Reactor Safety at Brussels, Belgium, October 16-19, 1978.
Copies of the paper are available from:
Walter B. Murfin
,, Organization 4441 Sandia Laboratories Albuquerque, NM 87115 2.3 Other Core Meltdown Research 2.3.1 Experimental Investigation of the Meltdown Phase of UO - Zircaby Fuel Elements under Failure of Emergency Cooling (PNS 4327)
Earliet bundle meltdown experiments have been Previously summarized in this series.
During 1978, 3 x 3 fuel rod simu-lator bundles were tested to investicate the influence of pre-and oxidation the refreezing behavior of tne resulting melt, Preli-the meltdown behavior of rods with ballooned cladding.
on the behavior of minary experiments were also carried out absorber rods.
Preezing Behavior of the Melt in the Bundle The bundles consisted of a central full pellet rod sur-rounde'd by ring pellet rods with 6 mm tungsten heaters.
The bundle was surrounded by a ceramic fiber insulator, consisting fibers and 2 inches of Aluminum-Silicate of a 1 inch Zr02 fibers.
Steam flow (up to 30 ml/ min of evaporated water) was provided by stamm tubes at the lower end between and outside the rods, giving a uniform steam distribution.
In the outer periphery of the bundle a solid oxide layer developed.
However, the sharper temperature rise in the in-terior gave such a thin oxide layer that the oxide skin was flushed away by the melt of Zircaloy and dissolved UO2 The first dropa of the melt froze in the lower colder region be-tween the rods.
Above this frozen material a " clump" of melt-formed at the lower end of the melt region.
The composition of the melt with respect to Zr and U con-tent was investigated by X-Ray fluorescence.
The droplets below the melt lump had a composition of 18.7-20.8 w/o U and 59.2-68.6 w/o Zr.
The droplets are the first to form and freeze and have the lowest mean concentration of uranium.
The samples out of the clump gave a U content of 18.6 - 32.9 w/o and 52.9 - 69.4 Zr w/o.
The analysis of the drops which were frozen on the rods in the middle of the melting area gave a U content of 46.6 - 54.8 w/o and a Zr content of 40.0 - 47.6 w/o.
The U content of this melt originating later in time is thus significantly higher.
Influence of Preoxidation Three fuel rod simulators were preoxidized with a thickness of 20.. No significant difference could be found in the melting or freezing behavior for the rods with oxidized cladding.
Meltdown Behavior of Ballooned Claddino Cladding tubes which had been ballooned and burst in the REBEKA facility were heated with various temperature rates (1/3*C/s to l'C/sec).
The separators were close above the ballooned region.
The rods heated at 1*C/s withstood the heat-up and cooling without damage.
On the other hand, the bundle heated at 1/3
- C/s showed severe enbrittlement.
This was first observable after cooling.
During the test and a half hour after power was shut down no material fell out of the bundle.
Before the start of' '
disassembly on the following day, fine little pieces spilled i
from the surface were observable on the lower electrode.
In ad-dition to these powdery fragments, some large pieces (1 - 3 cm) were lying on the lower plate.
Owing to the mechanical shaking during disassembly large regions of the cladding flaked off to-gether with broken up U02 fragments.
Behavior of Absorber Rods Absorber rods contained Ag-In-Cd alloy in Cr-Ni steel cladding.
The melt temperatures of the silver alloy, the Inconel separators, and the steel cladding are 800', 1300*, and 1450*C[respectively.
Heating was carried out in argon and steam a tebperature rise rate of 0.5'C/s to 1300-1550*C.
at -
The CrNi steel maintained its integrity to 1400*C in inert gas.
The separator was melted together.
A signifi-cant attack on the steel was not demonstrated.
In' steam the cladding also remained as an integral piece; fracture occurred on disassembly.
The appearance of the oxide was much different from the Zircaloy rods.
Instead the steel rods were blistered, of the smooth surface of Zr02, spongy, and deformed.
At 1500*C the CrNi-steel tube was melted away in inert gas.
However, in steam, the tube was converted to a spongy form that was significantly beyond its original diameter.
A pure Cr-Ni tube could be disassembled entire.
Howewer, a tube containing alloy and separator was so brittle t. tat it fell to, pieces.
~
Tubes were also heated slowerL(0.l*C/s) or were held at s
the test temperature for one hour.
No significant attack of the absorber on the tubing was observed due to the longer heating time.
The influence of the tubes and absorber rods on the melt-down behavior of fuel elements will later be investigated in bundle tests.
Photographs show the behavior graphically unfortunately, these photographs do not reproduce well.
It is suggested that readers with special interest in obtaining copies of the photo-graphs apply to:
s Dr.
S. Hagen Kernforschungzentrum Karlsruke (RBT-IT)
Postfach 3640 75 Karlsruke Ped. Rep. Ger many.
Comparison of Ring Pellet and Solid Pellet Simulators Fuel rod simulators with solid pellets are compared with rods containing ring pellets in Table 2.3.l(I).
2.3.2 Constitution and Reaction Behavior of LWR Materials in a Core Meltdown (PNS 4314)
Analysis of Samples from Thermite Tests Table 2.3.2(I) gives chemical composition of the slag, cal-culated from measured elemental values.
The locations from which samples were taken are shown in Figure 6.
Sample E2 was taken near the boundary.
The remaining samples were taken from near' the middle of the slug.
Samples with the same initial letter (e.g., C2 and C3) are taken from the same slug, and would be identical in a perfectly mixed melt.
3.
Fuel Behavior Research at KFK 3.1 Studies of the Influence of Oxide Fuel and Fission Products on the Mechanical Properties of Zrv-Cladding Tubes under Transient Loads Influence of Iodine on the Mechanical Procerties of Zry-Clad Tubes Zry Cladding tubes were internally pressurized with Argon and with mixed Argon and iodine.
Tubes were tested in "as-delivered" condition, preoxidized (oxide layer 25 m) and con-taining'UO2 fuel.
The creep rupture 'results are summarized in Figure 7.
Solid lines were tests with pure argon, dashed lines used 3
It can be seen that, for a given hoop st'ress, the time to rupture is shorter for the specimens with iodine.
The differences are less at lower internal pres-sures and longer times to ruptur e. These results can be con-trasted with earlier transient tests, in which iodine made little difference in the burst temperature or endurance time.
Preoxidized and/or UO -containing tubes showed a reduc-2 tion in burst strain as compared to unoxidized tubes.
The differences are very marked at 700' and 800*C. However, at 900' and 1000'C, the burst strains are unaccountably greater for iodine-containing specimens.
4.
Final Remarks My stay at KFK has been suddenly and unexpectedly termi-nated because of illness in the familj.
I would like to ex-press my gratitude to the Nuclear Regulatory Commission, Fuel Behavior Research Branch, for the opportunity of being allowed to serve in this capacity and to KFK - especially to Projeckt Nukleare Sicherheit and to the Institut fuer Reaktorbauelemente -
for their fine spirit of cooperation and helpfulness.
I have returned to my duties at Sandia Laboratories, Albuquerque, NM, and will be continuing development of the WECHSL code at that location, in cooperation with Dr. M.
Reimann of KFK.
Questions regarding the code can be addressed to either of us.
_9_
TABLE 2.1.1 (I)
Material Properties for the Liquids Used Water Silicon Oil AKS Property (40*C)
(20*C
' Density g/cm3 0.99 0.91 Volumetric Coefficient of Expansion (1/*C) 3.85 x 10-4 10.5 x 10-4 2i Kinematic Viscosity (cm s) 0.66 x 10-2 5.6 x 10-2 2
Thermal Conductivity (W/cm *C 6.3 x 10-2 0.12 x 10-2 Specific Ileat (J/g*C) 4.8 1.42 2
10-3 Thermal Diffusivity (cm /s) 1.5 x 10-3 0.91 x Prandtl Number 4.4 62 i
TABLE 2.1.1 (II)
Experimental AT Values (Glass Spheres)
V 0
.042
.084 l
.17
.32
.47
.63 s
cm/s j
h*(Vs) 1.0 5
11 19 28 220 630 2
W/cm
.16 12.8 3.3 1.4
.38 22.6 6.3 4.0 2.0
.75 37.7 12.5 5.4 4.3 1.17 48.7 19.5 8.6 5.l' l.5 52.1 17.9-10.9 5.9 5.2 1.1 0.5 8.3 7.2 P
s TABLE 2.3.1 (I)
Comparison of Solid Pellet and Ring Pellet Fuel Rod Simulators Phenomenon Differences / Similarities Same Zry attack on UO2 Melt penetration Greater for ring pellets Stability Solid pellet rods remain intact longer Melt breakthrough Always in spacer region 0
's 9
9 O
TABLE 2.3.2 (I)
Chemical Composition of Samples from the Shag Ca0 Fe MgO MnO Cr2 3 SiO2
.HO 0
2 Al O23 Al 4,9 9,9 6,8 (1,5)+
0,4 (75,9)++
0,6 C2 40,2 6,2 11,8 (12,1)+
0,4 (28,4)++
0,9 (28,9)++
0,5 C3 40,2 6,2 11,7 (12,1)+
0,4 D2 48,4 3,9 3,9 (14,5)+
0,1 8,3 (20,4)++
0,5 D3 41,2 3,5 r,d (13,3)+
0,1 8,0 (24,5)++
0,5 j,
El 4,7 9,1 8,9 (1,4)+
0,1 (75,0)++
0,9 E2 46,7 5,5 11,0 (14,0)+
0,4 (21,7)++
0,7 F1 35,9 5,9 9,6 (10,8)+
0,1 (36,8)++
0,9 F2 34,2 6,2 10,6 (10,3)+
0,1 (37,7)++
0,9 15,0 0,6 Il 61,4 1,8 5,6 18,2 0,1 I2 50,1 4,9 5,7 14,9 0,3 23,5 1,8 13 53,7 3,4 6,7 16,6 0,1 19,2 2,2
- Not taken from analysis; calculated under the assumption that MgO content is proportional to the Al 02 3 content, as was measured in samples II, 12, 13.
++ Remainder, assumed to be SiO2-
WT s
s u
r-TE DtAK5 A 6
' mE 7.5
- N Wass er
+3 Kugel-y 2 sgjpg _m 1
.y Ho# 7 !nr1 Pc,,
- ... ;.. : :.. ;- :......;.._. _..,g x, j aq Frit te o
Luft FIGURE 1.
Experimental apparatus.
100 T(z l( 'C )
Wasser AOLAK5 l
s 50
^\\k v (cm/s) s
~
.u -
k,.0
~;g N
~
^
N_
l y
[
}_
J -- - --- i4.65 i
Tu/ -
Heizung Td lE 2
3 4
5 6
20 0
1 2 2(c m )
3 4
5 2
FIGURE 2.
Temperature profiles at j = 1.5 W/cm ;
superficial velocity is shown parametrically.
s 50 Blasen jlW/cm2)
~
6km) 38
.75 Fritta
=.1 I
StoNkugeln p=,3cro
- 4 20 s Glas cugeln e
o= 4cm l,A A T ('C )
10 a
j(W/cm2) o a
- .75 S
A 38 0
2 0-
.05 v (cm/s)
.1 s
FIGURE 3.
Influence of bubble diameter on temperature distribution at various heat transfer rates and superficial velocities.
b e $
2 k7
.32/,
3
/
,/
,/
/
.5 s'
/
[.17 ivF2
.d.2 g=.0cm/s j(W/cm2)
~
2
-/ /
/
5 1
g(cm/s)
.0 o
/
gg2 o
500
'5
'3y j
f
.32 x
J.7 o
.63
=
200
.2
/
100 1
50
.05
. j(g)
J/ri(W/cm2)
= w,
.02 g.y -
20
/
.01
/
/
.05 1
.2 g b m/s).5 l
l I
I
,^
D02
.5 1
2 5
10 20 6T('C) 50 Heat flux as a function of temperature difference (upper).
FIGURE 4'.
i;ormalized heat flux as a function of terrperature difference (1cwer left).
Relative heat transfer ccefficient h* as a function of V3 (lower right). _
i Asbestzementrchr f1174 a
l f830
=
e o i/
4 J
,L 0
0 o
0 O
o o
O
- O o O
O O
o 0 0 0
0 g
O O o O
~
0 0f*
0 O
0 0
O O
O O.
O o
N I
O 0
O o o oe o
o e
o o o
?
O o o
o
- 0 0 o 'n 0 e o
o C
0 0 0
~
I O
O o
o 0
0 0
o O o
0 0
o o
o o o 0
0 0
O o
a C
0 0
0 D
O O
Oo 0
0
- o o
o o 0 e
0 O
o O
O O O
' 380 ---
o O
O o
o o o
o 0,
C 8 0 oo o o 0
O O
0 0 e 0 0 0 N
0 o
0 0 o
Q 0
0 0
0 N
Q b
O Q
O v o a
o o
o y
)0 0
0 e
o 0
0 o
,o e
o o
g 0
0 0
0 0
o o
o*0 0
0
'o o
e o
O 0
o o
o g
Ultra schc!ImeP,k 00 O
O o
o o
O C
0 0
0 /
o e
O t
O o
o i
8 o
o o
o o
0 o N,\\
o e
//
o ye o
o 6
0
- u e
{
/
,\\
o o
0 0
0 0
o e
o
]
t O
o o
o o
o o
e o
o
,o l
t/l o
o e
o o 0
0 0
_f 7
r L el l Y
I e
o a
o O o 7
ti i t
i i
6 il o'.
170,0
'810
=
I
- 1000
~
l FIGURE 5.
Crucible design for 100-200 Kg BETA tests. Dimensions are in millimetars.
- i.
s 7
{
J d'
0
)
I d
~
r n d
g
=
i T
13 k
(
ne ka t
erew s
e
- '0 D
P,l '
e h
lp 8
ma s
C 9
9d c
3 N
m d
s C
Cs 1
C i
u h
eg a
w e
ta sn
~
o
'r i
t i
aco L
6 ER U
l G
E I
F q
lw
'yb' a
A c
~. 3 f
A w
[(;;3, W
Qh.
a
(
s 5
U 9W7
(*'#3 88 Zry -4 (c7s.cnmm)
I"I T aarst
-o w.,.a,.o,a -
e Pgaccest e-weye.angm(by/ coll 200-4,.
,,n 10 0
=.=.vw = ls) y O gg, M' :t== m C 5A 3
.,l A%? h.
30 3 M
Y 10 *C 20 mW u
a E
E I
{ 20 C g,..
%**. }.' A,,g \\.
' i
.o c
ht4 10' tyc -
5 2
14:3*C 50 10 0 500 1000 TrneMupture (s)
(a) Time-to-rupture of Zry-4 cladding tubes in Argen.
Pressure medium was Argon or Argon - Iodine gas mixtures.
Preoxidized and UO containing speciments also shcun for 1000 C.
2
% tsaat P stest s tanst y
tg M Es 24 Pg econst
.go-(
try.4 005 a 172 sun 1 1 : 12-160 s 3
ggfi boneconcectratma faq/cm3 g
2' e
E I,J.4 (11 5 172) 100 p
- 22 a
E
\\
g s
22
\\
.,00 g 4
603
/
7tt *t 3
2 c
/
600 't 30 c E 18 WIC C*t e
t 20
=
l'j' 60-f
\\
j
/
lac *t E
$ 18
[
- CCo*C k
me rl% g\\e s
I/,\\
s00 s0
- 0 ss j
\\\\
x, 20 s
i
\\
- . H
/
12
/
?
20 0
u
/
NN I q
10 20 30 40 5o EO Length of the Zry Tube Cc;suesimmi 10
- 10 20 30 8.0 50 60 Lengthof the Zry TubeCopsues trrrn]
(b)' Deformation of tubes withcut iodine (left) and with fodine (right).
FIGURE 7.
Results of tests on Zry-4 cladding. -
DEC 2 2 m ADDRESSEES FOR LETTER DATED R. DiSalvo, NRC/RES X. Campe, NRC/DSE G. Chipman, NRR/DSE M. Cunningham, NRC/RES R. Frajey, NRC/ACRS s
S. Hanauer, NRC/0E0 Y. Y. Hsu, NRC/RSR
)
J. LaFleur, NRC/ DIP 6
S. Levine, NRC/RES D. MacPherson, NRC/RSR A. Marchese, NRC/NRR R. O. Meyer, NRC/ DSS J. Murphy, NRC/RES J. Norberg, NRC/EMSB J. Read, NRC/DSE L. Rib, NRC/RSR L.Rubenstein,NRf/fjRR M. Silberberg, NRC/RSR L. S. Tong, NRC/RSR R. W. Wright, NRC/RSR A. Malinauskas, ORNL N. Osborne, ORNL J. Gieseke, BCL R. Denning, BCL F. Kulacki, Ohio State L. Baker, Jr., ANL R. Henry, AHL D. Dahlgren, Sandia x
L. Kelman, ANL J. Muir, Sandia D. Powers, Sandia L. Buxton, Sandia L. S. Nelson, Sandia D. Walker, Offshore Power Systems l
A. Millunzi, DOE /RDD J. MacDonald, GE, Sunnyvale H. Morewitz, Atomics International I. Catton, UCLA D. Swanson, Aerospace Corporation R. Ritzman, SAI P. MacDonald, EG&G J. Zane, EG3G T. Pratt, BNL W. S. Fanner, NRC/RES L