•
!
.
.
wy
.
-
-
..
.
.
I OF I
ORNL:P
1287
ſ
.
.
.
. .
.
L'
.
SO
13.2
0436
t.
T 4 O
EEEEE
r
.
TI
?
.
:
So..
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS - 1963
LEGAL NOTICE
This report was prepared as an account of
Government sponsored work. Neither the
United States, nor the Commission, nor any
person acting on behalf of the Commission:
A. Makes any warranty or representa-
tion, expressed or implied, with respect to
the accuracy, completeness, or usefulness
of the information contained in this report,
or that the use of any infurmation, appa-
ratus, method, or process disclosed in this
report may not infringe privately owned
rights; or
B. Assumes any liabilities with respect
to the use of, or for damages resulting from
the use of any information, apparatus,
method, or process disclosed in this report.
As used in the above, “person acting on
behalf of the Commission” includes any em-
ployee or contractor of the Commission, or
employee of such contractor, to the extent
that such employee or contractor of the
Commission, or employee of such contractor
prepares, disseminates, or provides access
to, any information pursuant to his employ-
ment or contract with the Commission, or
his employment with such contractor.
olyp -/287
CONF-65 otor-8
RELEASE OF FISSION PRODUCTS FROM REACTOR FUELS
TO
PO DI
AJUN< 4 we
Ým
medicine
whether in law released the short warum:n.
c
p window
DURING TRANSIENT ACCIDENTS SIMULATED IN TREAT
G. W. Parker, R. A. Lorenz, and J. G. Wilhelm
Oak Ridge National Laboratory
The purpose of the ORNL fission-product-release experiments
conducted in the TREAT reactor is to study the release of radio-
active material from fuel during a reactor transient in which
the fuel cladding fails or the fuel melts or vaporizes rapidly
as a result of a nuclear accident. The objectives of this
program are to measure and interpret the release of fission
products as a function of fuel type, cladding, atmosphere, burn-
up, and transient characteristics. The extent of reaction of
water (steam) with metal cladding and with VO, fuel is determined
in experiments having a water or steam environment.
Experimental Method
The experiments are performed so that metal-clad UO, pellets
ure enclosed in a high pressure autoclave and the specimen and
autoclave are electrically preheated to the selected initial.
conditions. Immediately after the reactor transient, a valve
is opened to allow gas to flow from around the fuel specimen at
a controlled rate into a second autoclave which is initially
evacuated. A diffusion tube and membrane filters collect samples
Research sponsored by the U.S. Atomic Energy Commission.
under contract with the Union Carbide Corporatio
*Visiting scientist from Karlsruhe Center for Nuclear Research
and Development, Karlsruhe, West Germany.
PATENT CLEARANCE OPTAINED. RELEASE TO
THE PUBLIC IS APPHOVE, "ROCEDURES
ARE ON FILE IN 15.1. PNG SECTION,
.
2
of suspended particles. The various components are then analyzed
radiochemically for fission products.
Those experiments which
use large amounts of water contain a condenser and water-collection
traps in place of the diffusion tube.
Six experiments have been performed to date; four used low
pressure argon or steam-air mixtures around the fuel specimen
and two used 1000 psia steam atmosphere. The first four experi-
ments have been reported in detail in progress reports. -,-
Figure 1 shows the flow diagram for the low pressure experiments.
In these experiments the fuel specimens were preheated electrically
to 800°c resulting in approximately 45 psia gas pressure in the
fuel autoclave. The reactor transient was performed and 12
seconds later the valves were opened allowing the transient-
generated aerosol to flow slowly (47 cc/min) through the 0.195 in.
inside diameter diffusion tube, through a series of membrane
filters, and into the gas-collection autoclave. Figure 2 shows
the arrangement of the fuel specimen and heater within the fuel
autoclave.
The flow diagram for the high pressure steam experiments
is shown in Fig. 3. In these experiments 25 ml of water was
placed in the fuel autoclave below the fuel specimen. The fuel
autoclave was heated to 285°C (1000 psia steam pressure) and the
fuel specimen to approximately 330°C. At the peak of the reactor
transient the first valve was opened allowing steam to flow
rapidly into the condenser. Noncondensbble gases (hydrogen,
xenon, etc.) passed through the water traps and filters into the
.
.
.
...
.
.
.
..
.
.
..
.
-
.
gas-collection autoclave. Suspended particles were collected
on membrane filters and iodine compounds were adsorbed on a
series of charcoal-loaded fiberglass papers. Twenty minutes
later the second valve was opened allowing argon gas to sweep
slowly through the fuel autoclave to flush remaining gases and
volatile material into the gas-collection autoclave.
..........
The first two experiments were performed in argon atmos-
.....
phere for simplicity of operation and interpretation, as well
--.
..
.
as permitting comparison of results with the many other in-pile
and out-of-pile fission-product-release experiments being con-
ducted by others using inert atmospheres, Succeeding experi-
ments have been designed to simulate various conditions which
might be encountered with pressurized water or boiling water
-
-
.......
-
-
-
-
Ilong.
-
.
.
reactors. Future experiments will be conducted with fuel specimens
submerged in water, and burnup will be increased to thousands of
.
.
...
.
Mwd T.
-
-
-.
..-
Experiment Operating Conditions
.
..
..
Table 1 presents a summary of the six experiments.
TI
.
.
.-',
..
2
-
,
... . -
.
"Volatile fission product" values in the table represent an
average of 1, Te, and Cs release results for the four experi-
ments using preirradiated fuel and it is an average of mixed
I, Br, Xe, and Kr for the two experiments which receivad no
preirradiation. This is a rough comparison of fission product
release and serves only to illustrate the wide range of release
observed. The specimen cladding was type 347 stainless steel or
Zircaloy-2 and was 0.035 in. thick in all experiments. The Voz
---
.
*
*
4
pellets were 0.4 in. dia and their density was 95% of the
theoretical value. The weight of UO, in each experiment was
approximately 31 grams and the uranium was 10% enriched. The
first four experiments used a batch of pellets which contained
2% of metal contaminants, most of which was tungsten. The second
batch, employed in the last two experiments, contained only
approximately 300 rpm of impurities. The irradiations for fission
product buildup were made in the ORR and approximately 3 weeks
decay was allowed before the TREAT reactor transient was performed.
The rate of specimen heating during the transient is shown
in Fig. 4. The stainless steel temperatures were measured with
thermocouples and the UO, temperatures were calculated assuming
no heat loss. Significant heat loss does occur from the UO2
during the time required for the transient. For example, the
heat input was greater than 300 cal/g UO, ior all experiments
and only 280 cal/g should result in complete melting. In three
experiments no signs of fuel melting were observed and this is
attributed to greater heat loss than anticipated, resulting from
slower heat input (Exp. 1) or improved heat transfer in high
pressure steam (Exps. 5 and 6).
Experiments in Which the Fuel Melted
Meltin of UOz occurred in three experiments (2, 3, and 4).
Figures 5, 6, and 7 are photographs of the fuel specimens from
these experiments. Melted stainless steel can be seen in the
bottom of the alumina crucibles. The loose pieces of UO, broke
of£ juring disassembly of the equipment. A thin shell of unmelted
UO, remained at the relatively cool pellet surface. Ten
percent (Exp. 2) to 50 percent (Exp. 3) of the UC, flowed out
of the pellet shell in a foamy or porous form which had a bulk
density of approximately 1/2 of the original UO, density.
Figure 8 shows sections of the top pellet of Exp. 2. Swelling
occurred in this pellet and in others that melted.
Stainless-steel-clad to, specimens preirradiated to
~10 Mwd/T were used in experiments 3 and 4 with low pressure
steam-air atmospheres. The results were essentially identical
in regard to extent of fuel melting and change of atmosphere.
Approximately 75% of the o: originally present in the fuel
autoclaves was consumed (38 cza3 , STP) and this suggests that
more oxidation night have occurred if more oxygen had been
available to the heated fuel specimen. The melted stainless
steel cladding did not appear to be greatly oxidized, for
example. The distribution of fission products and Voz was also
essentially identical for the two experiments and the results
72
were averaged for presentation in Table 2. All fission products
and UO, were determined individually and where the distribution
of two or more was similar the results were averaged for this
table and for other tables in this paper.
Because of the short preliminary irradiation for fission
product buildup in Experiments 3 and 4, some of the short half-
life fission products formed during the transient interfered ·
with the determination of the behavior of the isotopes of interest.
The contribution of these transient-formed isotopes was corrected
for, based on their behavior and distribution observed in
Experiment 2 and the correction was greatest for ysr and 14 °Ba
because of the volatile precursors of these isotopes. In
general, results from Experiments 3 and 4 show that the release
of fission products from transient-melted fuel is considerably
less than that from fuel in experiments simulating loss-of-
coolant accidents with longer heating periods. The large amount
of +49Te in the stainless steel is believed to be the result of
alloying. A large amount of the three volatile fission products,
I, Te, and Cs appears to have deposited in the alumina heater.
As in all low pressure experiments, only 50 to 60% of the
aerosol in the fuel autoclave entered the gas transport zone
and was sampled by the gas-collection system. All the low
pressure experiments showed a mean primary particle size (not
agglomerate size) of approximately 0.054, which is similar to
that observed in other vo: melting experiments.
The distribution of material in Exp. 2 is summarized in
Table 3. This experiment employed stainless-steel-clad UO, in
low pressure argon and there was no irradiation previous to the
reactor transient. The release and transporċ of fission products
occurred during and immediately after the reactor transient when
the fuel was heated. At that time the elements of each fission
product mass chain were a mixture of recently formed short half-
life isotopes. Radiochemical analysis was performed by analyzing
for the radioactive isotope of each mass chain which dominated
after several weeks of decay. For example, the most abundant
elements of mass 89 were 89Br and 89Kr at the time of the transient
.!
.
.
...
.
.
.
.
Y-1.
-
.
. -
-
-
-
-
-
-
-
-
-
---
-
-
--
-
-
-
heating, but three weeks later these had decayed to Ysr,
which was the isotope selected for radiochemical analysis.
Results for the various mass chains were found to be
divisible into three groups depending on the chemical and
physical characteristics of the elemental forms existing at
the time of the transient. These groups are called "volatile
isotopes," "alloying isotopes," and non-volatile material,"
and the results within each group were similar and were
averaged for this table. The "volatile isotopes" are mostly
halogens and rare gases and their release was somewhat greater
than 1311, 129Te, or 137cs observed in Experiments 3 and 4. The
"alloying isotope" group was found in high concentration in the
cladding probably because of alloying between the Sn, sb, and
stainless steel. There was very little release of the
"non-volatiles."
-
-
..
.
Upmelted, Intact-Fuel Experiments
Experiment 1 (Fig. 9) was identical to the preceding experi-
ment (Exp. 2), except that the reactor period was longer. The
slower heating permitted greater heat loss and the UO, therefore
did not reach the melting point. Based upon melting of a
molybdenum wire placed between the two pellets, the maximum
UO, temperature was estimated to be 2700 + 100°c. The pellets
were essentially unchanged except for extensive surface cracking.
Figure 10 is a section of the lower pellet and grain growth was
less than 20% in the interior.
The distribution of fission products and UO2 in Exp. 1 is
shown in Table 4. The same grouping of isotopes is used as for
ies*
.
8
8
Exp. 2. In the "non-volatile material" group, UO, had the lowest
release. The "volatile isotope" release was surprisingly large
considering the small change in the U02. Distribution of the
"alloying isotopes" was nearly the same as for Exp. 2 in which
65% of the UO, melted.
Experiment 5 (Fig. 11) had stainless-steel-clad itofuel
with a preirradiation burnup of 7 Mwd/ton exposed to 1000 psia
tiiose which had previously resulted in melting of specimens in
low pressure atmospheres. Melting of the UO, did not occur
probably because of the combined effects of better conductive
cooling to the cooler cladding held in firm contact with VOZ
by high external pressure, better convective cooling by the
high pressure steam, and greater radiation cooling to the cooler
alumina crucible. The stainless steel cladding melted, oxidized
heavily, and slumped down into a cauliflower-like mass. The
oxidized cladding completely covered and adhered to the UO2
pellets which appeared to be unchanged from their original
condition. The melted and oxidized cladding was very similar
in appearance to stainless steel cladding heated slowly to its
melting point in low pressure steam experiments previously con-
ducted out-of-pile. 3
A metallographic section of the upper pellet is shown in
Fig. 12. The hottest portion of the UO, was at the top below
the end-cap, as evidenced by grain growth and plastic flow which
occurred there. The original end clearance between pellets and
end caps was approximately 0.090 in. so the helium-filled gap
provided insulation in this area. Cladding in contact with the
pellet sides provided better cooling. Both stainless steel and
stainless steel oxide were in contact with UO2, but there was no
evidence of reaction with the UOz. Based on comparisons of grain
growth and pellet distortion, it was concluded that the UO, in
this experiment reached a higher maximum temperature than the
fuel in Experiment 1 which also did not melt.
The extent of reaction between stainless steel cladding and
water during the transient was determined by two methods: col-
lecting and measuring the hydrogen produced during the transient
oxidation, and analysis of a sample of the cladding to determine
the ratio of unoxidized to oxidized stainless steel. Previous
metal-water reaction studies utilizing transient heating methods
.
.
".
have assumed that the only reaction which occurred was Fe + H2O
- FeO + H2. Recent studies at ORNL and ANL show that oxidation
of other components of stainless steel occurs. For example,
examination of stainless steel exposed to steam at 1500°c for
several hours showed no unoxidized Fe, Cr, or Ni and that the Fe
formed mostly Fe304. Therefore the percent of oxidation was
calculated for both limited reaction (Fe - Feo) and complete
reaction (Fe, Cr, Ni – Fe304, Cr2O3 , Nio). The method of
chemical analysis involved dissolving the partially oxidized
-
cladding from the lower pellet in hydrochloric acid and measuring
the hydrogen release for comparison with that from other samples
of unoxidized stainless steel. From this analysis the metal-
water reaction was found to be 34% complete assuming that the
oxygen-consuming reaction was (Fe - Feo) or 22% complete assuming
: 10
that the higher state of oxidation represented complete reaction.
Measurement of the hydrogen produced during the experiment itself
verified this extent of reaction. The amount of stainless steel
oxidation found in Experiment 5 was approximately three times
that observed in transient metal-water reaction studies with
specimens immersed in water conducted by ANL, and is probably a
result of slower cooling of the metal exposed to the gas phase.
The distribution of fission products and UO, in Experiment 5
is given in Table 5. The release of fission products was ex-
ceptionally small, especially when compared with Experiment i
(the other experiment that gave unmelted intact UO2).
The
lower release may be attributed to the cooling and sealing effect
of the oxidized, adherent cladding produced in the high pressure
steam atmosphere. Of the material carried out of the fuel auto-
clave during steam release and argon purging, most of the 15tI
and Cs were found in the water in the main water trap, and
most of the 129Te was found on the walls of the condenser tubing
and main water trap.
The distribution of 1511 removed from the fuel autoclave
by the steam release and argon purge is shown graphically in
Fig. 13 where the components are shown according to the sequence
of their exposure to the flowing gas stream. Of this 1341, 18%
was found on the walls of the condenser and water traps, 79% was
in the condensate in the main water trap (the second water trap
was dry), 0.06% was on the membrane filters, and 2.5% in the
charcoal-loaded papers. In contrast to the low pressure experi-
ments in which 10 condensation occurred, there were very few
.
.
.
..
i
i
11
n
,
m erineisiin
: tri
.,.-'
uc-.-
icann... L
w
.
.-.
.-.-.-.
particles on the first membrane filter. The first membrane
filter (1.24 pore size) was apparently 100% efficient for the
particles present, since filters 2 through 5 (0.454 pore size)
did not show a decrease in activity. The charcoal papers were
1/32 in. thick fiberglass mats loaded with charcoal. Twenty-
seven charcoal papers were used in series and these showed an
orderly decrease in activity in the first 16 papers, indicating
equal fractional adsorption of iodine-containing material on
each charcoal paper. The calculated iodine penetration through
each filter was 70% for the linear part of the graph.
From the experience of other experimenters, the 15 LI
released from fuel occurs in the elemental 12 (vapor) form, in
forms associated with filterable particles, and as filter-
penetrating forms including organic compounds such as CHz I.
Based on experience in other experiments, all the elemental In
(easily collected in condensate or deposited on cool metal
.
.
.
.
.
surfaces) was collected in the condenser and water traps. There
-...
was very little particulate -I activity and all of it was
--..
N
e verwinnin' .... ..-:--...
collected on the first membrane filter. No elemental I2 reached
the charcoal papers. Instead, a penetrating iodine compound
was found to have been adsorbed on the charcoal papers.
To help verify these conclusions, the filter pack (membrane
filters and charcoal papers) was tested with a high purity
elemental iodine source under conditions approximating the
humidity, flow rate, and total gas flow employed in Experiment 5.
The iodine contained 1341 tracer with added carrier, and volatile
i
susivieniniwerweni-internatinait
compounds were removed by evacuation while the source was cooled.
Figure 14 displays the results obtained and it shows that two
different iodine forms were adsorbed on the charcoal papers.
The penetration of each charcoal paper by elemental I, was only
0.01%. Only a very small amount of iodine penetrated the first
2 charcoal papers, and this iodine was apparently in a compound
form of which 93% penetrated each charcoal paper.
Fragmented, Unmelted Fuel Experiment
Experiment 6 was performed in 1000 psia steam and it was
the only experiment in which a Zircaloy-2 clad fuel specimen
was used. The operating conditions and reactor transient were
identical to those used in Experiment 5, but the condenser
tubing plugged so that steam was not released from the fuel
autoclave and the argon purge did not occur. Pressure in the
fuel autoclave remained above 1000 psia for 23 minutes following
the transient. The uo, and Zircaloy cladding were found
fragmented and dispersed within the alumina heater after the
. fuel autoclave was disassembled. Figure 15 is a photograph of
the fragmented specimen and Fig. 16 is a section of the upper
end cap and an intact portion of the UO, pellet. The Voz appears.
to have reached approximately the same maximum temperature as
in Experiment 5, based on grain growth and upward swelling or
flowing of the VO2. Extensive oxidation of the Zircaloy-2 is
evident. Material that had flowed can be observed at the edge
of the pellet. Figure 17 is a photomicrograph of this area
and the fluid material appears to be a mixture of Zircaloy-2
zirconium oxide, and UO2. This type of reaction or mixing has
13
4.5
been observed previously in two different types of out-of-pile
melting experiments. 4,5 Metallurgists reported that the cladding
was weakened and that loss of UO, grain boundary integrity
occurred in this experiment but not in Experiment 5. The cause
of the fragmentation is not known, but the above-mentioned
weaknesses probably contributed to the breakup. Oxidation of
the Zircaloy-2 cladding, excluding the end caps, was estimated
to be more than 30% complete based on metallographic examination.
Hydrogen formed during the reaction was lost through a leak in
the equipment.
Fission product and UO, distribution in Experiment 6 is
shown in Table 6. No material was released from the fuel auto-
clave because of the plugged condenser tubing. The release of
fission products from fuel and cladding was considerably greater
than from the unfragmented fuel in Experiment 5. The extent of
release in Experiment 6 may have been influenced by the longer
exposure to high pressure steam-hydrogen mixture (hydrogen
released during metal-water reaction) but it was probably higher
because of the fragmented fuel and cladding.
-
-
-
-
-.-.-
.
.. ..
.
..
.
...
-
--
i
-
..
.
-
..-
-
.
-.
....5-.
...
...
..
.
.
.
..
Conclusions
.
.
A
.
L
Approximately one-third of the 1311, 129Te, and 137cs were
released from UO, and stainless steel cladding in transient-
heated experiments which used 45 psia steam-air atmospheres where
75% of the UO, melted (Exps. 3 and 4). In-pile loss-of-coolant
type experiments performed by other ORNL experimenters have shown
considerably greater release of these isotopes from the fuel zone.
twistina LXmas
14
Release values averaged 95% for a variety of low-pressure
atmospheres when the fuel was held molten for 5 min.
In three of the transient experiments, the UO, fuel
specimen was heated close to the melting point, but did not
melt. The release of fission products from fuel and cladding
in these experiments were: 20% of mixed halogens and rare gases
in low pressure argon atmosphere (Exp. 1); 0.5% of the I, Te,
and Cs in high pressure steam where the melted and oxidized
cladding adhered to the UO, pellets (Exp. 5); and 10% of the I,
Te, and Cs in high pressure steam where the Zircaloy-2 cladding
and UO, fragmented (Exp. 6).
Release of high pressure steam and collection of con-
densate was performed in Exp. 5. Of the 1511 carried in the
steam release, 79% was collected in the water and 2.5% was in
a penetrating form that was collected in a filter pack (a series
131
of membrane filters and charcoal-loaded papers).
In the 1000 psia steam experiments, oxidation of stainless
'steel cladding was approximately three times greater than for
specimens exposed under water by ANL. The partially oxidized
stainless steel cladding adhered to the UO, pellets, but there
was no evidence of reaction between UO, and the stainless steel
or its oxides. With Zircaloy-2 cladding in high pressure steam
there was definite mixing or reaction between Zircaloy, zirconium
oxide, and UO2.
15
Experiments in Progress
Experiments are being constructed which will simulate
startup transient reactor accidents. This type of accident is
considered to be one of the most probable and serious of transient
accidents. In the type of accident to be studied, metal-clad UO2
under water is transient-heated into the UO, vaporization range
resulting in rapid expulsion of steam from the reactor tank
(fuel autoclave) followed by a long after-heat period in which
steam is slowly evolved. The use of after-heat with slow steam
evolution should be a realistic simulation for maximum fission
product release.
:16
References
1. G. W. Parker, R. A. Lorenz, and C. E. Miller, Jr.,
Nucl. Safety Semiann. Program Progr. Rept. Dec. 31, 1963,
ORNL-3547, pp. 25-42.
2. G. W. Parker, et al., Nucl. Safety Program Semiann. Progr.
Rept. June 30, 1964, ORNL-3691, pp. 16-28.
3. G. W. Parker and J. G. Wilhelm, Nucl. Safety Program Semiann.
Progr. Rept. Dec. 31, 1964, ORNL-3776, pp. 37-40.
4. G. W. Parker, et al., Nucl Safety Program Semiann. Progr.
Rept. June 30, 1962, ORNL-3319, pp. 27, 28.
5. G. W. Parker, et al., Nucl. Safety Program Semiann. Progr.
Rept. Dec. 31, 1962, ORNL-3401, pp. 18-21.
6. W. E. Browning, Jr., et al., Nucl. Safety Program Semiann.
Progr. Rept. Dec. 31, 1964, ORNL-3776, pp. 109-115.
TABLE 1. SUMMARY OF TREAT FISSION PRODUCT RELEASE EXPERIMENTS
RELEASE OF
VOLATILE F. P.
FROM UO, AND CLAD
FUEL MELTED
FUEL AND CLADDING
FRAGMENTED
TOTAL FISSION HEAT
AND PREHEAT TO
SPECIMEN, Cal/g VO2
REACTOR PERIOD,
msec.
INITIAL PRESSURE,
psia
AWN EXPERIMENT
ATMOSPHERE
ELECTRICAL PREHEAT
OF SPECIMEN, °C
CLADDING
VO, BATCH NO.
BURNUP, MWD/T
BURNUP, UND/T
50%
65%
315
ARGON
S.S.
008
30%
75%
346
74
S.S.
008
0.2
30%
75%
335
77
53
3% Steam-
97% Air
30% Steam-
70% Air
S.S.
008
0,2
17

20%
309
108
39
ARGON
S.s.
00
10%
on a W
327
78
1000
STEAM
Zr-2
1350
NNE N
| 13
0.5%
330
1000
STEAM
S.S.
310
7
TABLE 2. DISTRIBUTION OF MATERIAL IN TREAT EXPERIMENTS 3 AND 4:
LOW PRESSURE STEAM-AIR, TRACE PREIRRADIATION, 75% MELTED
LOCATION
131,
129Te
PERCENTAGE IN EACH LOCATION
137CS 106Ru 89sr
140Ba
95zr
144ce
UO2
UO, FUEL + S.S. CLADDING
60
79
66
95
98
(0.25)
98.3
(0.24)
(S.S. CLADDING)
9
(0.8)
(0.2)
ALUMINA HEATER (1
25
13
28
2.3
0.4
0.43
AUTOCLAVE LINER
6.8
18
5.6
5.1
1.5
1.1
0.8
FUEL AUTOCLAVE WALLS (1
6.3
0.6
0,7
-
-
0.7
- - - -
0.001
-
-
-
-
0.46
- - -
0.001
- -
0.5
-
1.1
0.15
0.5
GAS TRANSPORT ZONE
FILTERS (MEMBRANE)
GAS-COLLECTION AUTOCLAVE
0.02
0.15
0.01
0.001
0
0.2
0.01
0.001
0.0001
0.00008
0.05
A
irA
T
E
V
+
---
- SAP .-
.-
'
-
.
--
-
.
-
.
-
.
-.-.
.-
.
'
'
.
....
.
.
.
..
..
.
..
.
.
.
-
-
-
-
-
.
- -
-
-
-
-
TABLE 3. DISTRIBUTION OF MATERIAL IN TREAT EXPERIMENT 2:
LOW PRESSURE ARGON, NOT PREIRRADIATED, 65% MELTED
PERCENTAGE IN EACH LOCATION
VOLATILE ISOTOPES ALLOYING ISOTOPES NON-VOLATILE MATERIAL
MASS 89 (Br + kr) MASS 131 (Sn + Sb) MASS 95 (Rb + Br)
MASS 137 (I + Xe) MASS 129 (Sn + Sb) MASS 103 (MO + Tc)
LOCATION
UO2
46
92
99.5
UO, FUEL + S.S. CLADDING
(S.S. CLADDING)
(1.7)
(18)
(0.5)
6.7
19
ALUMINA HEATER (1000°c)
METAL HEAT REFLECTORS
0.3
5.5
0.45
0.04
FUEL AUTOCLAVE WALLS (170°c)
11.0
1.0
0.06
:
GAS TRANSPORT ZONE
13.8
0.04
0.02
FILTERS (MEMBRANE)
2.2
0.003
0.12
0.02
GAS-COLLECTION AUTOCLAVE
14.8
0.0005
TABLE 4. DISTRIBUTION OF MATERIAL IN TREAT EXPERIMENT 1:
LOW PRESSURE ARGON, NOT PREIRRADIATED, NOT MELTED
PERCENTAGE IN EACH LOCATION
LOCATION
VOLATILE ISOTOPES
MASS 89 (Br + Kr)
MASS 137 (I + Xe)
ALLOYING ISOTOPES
MASS 131 (Sn + Sb)
MASS 129 (Sn + Sb)
NON-VOLATILE MATERIAL
MASS 95 (Rb + Sr)
MASS 103 (Mo + Tc)
UO,
77
92
UO2 FUEL + S.S. CLADDING
(S.S. CLADDING)
99.9
(0.03)
(1.0)
(25)
ALUMINA HEATER (1000°c)
3.7
6.6
0.06
20
METAL HEAT REFLECTORS
2.1
0.6
.
0.0003
FUEL AUTOCLAVE WALLS (2
3.1
1.0
0.005-0,00001.
5.5
0.06
0.004-0.0001
GAS TRANSPORT ZONE
FILTERS (MEMBRANE)
0.35
0.1
0.004-0.00001
GAS-COLLECTION AUTOCLAVE
8.1
0.02-0.00005
-
nevromasimmerein
m'a
-------
---
-
-
-
.
.
-
.
.
. .
.
.
.
-..-
.
-.
.
Y
.
-
-
-
.
.
.
-
-
...
.
.
...
:-...
4.
..
....
/....,-,
TABLE 5. DISTRIBUTION OF MATERIAL IN TREAT EXPERIMENT 5:
1000 psia STEAM (RELEASED), TRACE PREIRRADIATION, NOT MELTED
PERCENTAGE IN EACH LOCATION
129Te 1370s 89sr
LOCATION
131,
95zr VOZ
. 14oBa
144ce Ru
UO, FUEL + S.S. CLADDING
(S.S. CLADDING)
99
(5.3)
99.6
(1.1)
99.9
(2.1)
99.99
(0.12)
99.998
(0.1)
0.5
0.3
0.06
ALUMINA HEATER (400°C)
FUEL AUTOCLAVE WALLS (300°c)
0.0006
0.046
0.004
0.006
0.0014
0.002
0.0009
0.0004
21
MAIN WATER TRAP WALLS
0.02
0.09
0.0005
0.00004
WATER IN WATER TRAP
0.12
0.0007
0.01
0.0009
SECOND WATER TRAP
0.008
0.002
0.0005
0.0008
0.0001
0.00002
MEMBRANE FILTERS
0.0001
0.0001
0.00005
CHARCOAL FILTERS
0.0001
0.0036
0.0001
GAS-COLLECTION AUTOCLAVE
0.0001
0.0002
0.0003
0.00001
TABLE 6. DISTRIBUTION OF MATERIAL IN TREAT EXPERIMENT 6:
1000 psia STEAM (RETAINED), TRACE PREIRRADI ATION, FUEL AND CLADDING FRAGMENTED
PERCENTAGE IN EACH LOCATION
131,
129Te
137cs
LOCATION
103Ru 89sr
106Ru 140Ba
95zr UO,
144ce
97.7
98.9
VO2 + 2r-2 CLADDING
(Zr-2 CLADDING)
90
(2.5)
95
(11)
77
(5)
(10)
(7)
N
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
ALUMINÁ HEATER (400°c)
2.6
4.4
1.7
21
2.1
0.11
4.8
METAL HEAT REFLECTORS
0.06
0.5
0.11
0.2
AUTOCLAVE WALLS AND
SPACER (300°C)
0.7
0.2
0.5
0.07
0.05
0.0002
WATER IN AUTOCLAVE
0.01
0.004
0.2
0.07
23
FIGURE CAPTIONS
Fig. 1.
Flow Diagram for Fission Product Release Under
Transient Reactor Conditions (For Low Pressure
Gas Atmospheres).
Fuel Autoclave Assembly for Fission Product Release
Fig. 2.
Experiments Under Transient Reactor Conditions.
Fig. 3.
Flow Diagram for TREAT Fuel Melting Experiments
(For High Pressure Steam Atmospheres).
Fig. 4.
Comparison of Transient Heating Rates for Fission
Fig. 5.
Product Release Experiments.
Fuel Specimen From TREAT Experiment 2; 65% of the
UO, Melted in Argon Atmosphere.
Fig. 6.
Fuel Specimen and Alumina Heater From TREAT .
Experiment 3 in Which 75% of the UO, Melted in
3% Steam - 97% Air Atmosphere.
Fig. 7.
Fuel Specimen and Alumina Heater From TREAT
Fig. 8.
Experiment 4 in which 75% of the UO, Melted in
30% Steam-70%'Air Atmosphere.
Metallographi
Metallographic Sections of Upper Pellet From
Experiment 2 Showing Porous Melted UO, in the
Interior.
Fuel Specimen From TREAT Experiment l in Which the
UO, was Heated to 2700°c £ 100°C in Argon Atmosphere.
Metallographic Sections of Lower Pellet From
Experiment 1 Showing Little Change in UOStructure.
Fig. 9.
Fig. 10.
24
Figure Captions (continued)
Fig. 11.
Fuel Specimen From TREAT Experiment 5 in Which
the Stainless Steel Cladding was Oxidized and
Adhered to the Unmelted UO, Pellets in 1000 psia
Steam.
Fig. 12.
13.
Metallographic Section of the Upper Pellet From
Experiment 5 (1000 psia Steam) Showing Grain
Growth and Plastic Flow of the UO2.
Distribution of 1311 in Steam Released in
Experiment 5.
Distribution of 1341 in the Elemental I, Test
of the TREAT Filter Pack.
Fuel Specimen From TREAT Experiment 6 Showing ·
Fragmented Zircaloy-2 Cladding and UO, Resulting
From Transient Heating in 1000 psia Steam.
Fig. 14.
Fig. 15.
Fig. 16.
Metallographic Section of the Upper Pellet From
Experiment 6 (1000 psia Steam) Showing Grain
Growth and Plastic Flow of the UO, and Oxidation
of the Zircaloy-2 Cladding.
Photomicrograph of the UO, Pellet Edge From
Experiment 6 (1000 psia Steam) Showing Mixing
of Zircaloy-2, Zirconium Oxide, and UOz.
Fig. 17.
UNCLASSIFIED
ORNL-DWG 63-1445R

-
GAS COLLECTION AUTOCLAVE
(FULL VACUUM ORIGINALLY)
FILTERS
VACUUM TESTING VALVE
_GAS COLLECTION
SYSTEM
EVACUATION VALVE
VACUUM TESTING GAGE
DE
EXPLOSIVE VALVE
-FLOW CONTROLLER
-FLOW RESTRICTOR
PRESSURE CELL
Y
AF
DIFFUSION SYSTEM
FILL VALVE (FILL WITH
47 psia Ar AT 25°C)
--AEROSOL TRANSPORT ZONE
DIFFUSION SYSTEM -
10
FUEL AUTOCLAVE
FILL VALVE
- DIFFUSION TUBE
-
THERMOCOUPLE AND HEATER ·
WIRE SEALING GLANDS
.
.
.
.
- EXPLOSIVE VALVE
de
.
.. .
..
-GAS OUTLET TUBE
.
....
..
en
.
. ..
FUEL AUTOCLAVE
(ORIGINAL FILL 25 psia Ar AT 25°C)
FUEL AUTOCLAVE
SYSTEM
---• -,.
-
-'.
.
,.,
..
.
vvv
Flow Diagram for Fission Product Release Under Transient Reactor Conditions.
n
Figol.
(For Low Pressure Gas Atmospheres)
A
n.
- -
-
::.-
UNCLASSIFIED
ORNL-DWG 64-650

- FUEL AUTOCLAVE
STAINLESS STEEL CLADDING
-CLADDING THERMOCOUPLES
- HEATER THERMOCOUPLE
- AUTOCLAVE THERMOCOUPLE
1
- UO2 FUEL
METAL HEAT REFLECTORS
- ALUMINA HEATER
SUPPORT
tag.2. Fuel Auto
Fuel Autoclave Assembly For Fission Product Release Experiments
Under Transient Reactor Conditions.
ORNL-DWG 65-3358

GAS-COLLECTION AUTOCLAVE
|
B
101
CHARCOAL-FILLED
FILTERS
illil
MEMBRANE FILTERS
achimb
>
FLOW RESTRICTOR
PRESSURE CELL-
SECOND WATER
FIRST WATER TRAP
ARGON TANK-
- EXPLOSIVE VALVE NO. 1
FLOW
RESTRICTOR
DIDII
CONDENSER
.
.
.
.
EXPLOSIVE
VALVE NO. 2
.
.
.
-
-
.
-
.
.
.
-
G
-
-
IN
.
-
---
-
.-
-
-
.*...
-
--
.
--
-
-
-
'
M
-
+
w
-
.
-
FUEL AUTOCLAVE-
-
+
'
L
atura mai are un
*
-
4FUEL SPECIMEN
Inklim
on or restriction
.
.-
.
Flow Diagram for TREAT Fuel Melting Experiments.
-
**
(For High Pressure Steam Atmospheres)
-
"
---Cicati
-
ORNL-DWG 64-7116R
2800

APPROXIMATE UO2 FUEL
TEMPERATURE, CALCULATED
ASSUMING NO HEAT LOSS
2400
ZOPEN THERMOCOUPLES
M
TEMPERATURE (°C)
OPEN AT 6.7 sec
MEASURED STAINLESS STEEL
CLADDING TEMPERATURE
------ EXPERIMENT 1
----EXPERIMENT 3
EXPERIMENT 4
TIME.(sec)
Comparison of Transient Heating Rates for Fission Product .
Release Experiments.
. .
.
. .
•
•
Fig. 5
ا
... .
....
. . .
.-
.
-.
منم . مه سهمممممممه
مساله من
نی مهنده
. .. .. منها
1
-
-
-
- -
-
-
-
-
-
-
1 .
-
-
Fig. 6
:: ...
۰۰:::
:: :: ۱۰
نی
،مد، ۰
۰۰
.
۱۱:.. از مدینه وند
.
-
-
-
ملتاعسم
... مه مه ...
--
15 دوره ۰ة
ج معه
|
..
و
.
***..; op
.
..
-":-"*****
-
.
.
Fig. 7
mer
..
Ś
han
di convient
:
.
.
-
.
. .
.
-
-
.
.-
-
.

.
.
.
92
ON
:
.
.
V
Join
TE
.
.

TO
..
- VW VALVL
Fig. 8
1
.
.
..
Y
.
Fig. 9
....
...
..
maitemente
.