Note: Descriptions are shown in the official language in which they were submitted.
~ AC~GROUND OF THE INVENTION
Field of the Invention:
This invention provides a device to remove and
store tritium from a gaseous medium, and a method for manu-
facturing the device. It specifically provides a device
which may be incorporated in a fuel rod of a nuclear reactor
to minimize release of tritium to the reactor coolant.
Description of the Prior Art:
The operation of a nuclear reactor necessarily
forms tritium. As a product of ternary fission3 which ~s
typically the largest source of tritium~, tritium is formed
within the solid matrix of uranium containing pellets and
other fuels, typically encased in metal tubes or cladding.
Most water reactors utilize fuel cladding of zirconium
alloy, known more commonly as Zircaloy, and a typical ccm-
mercial reactor includes thousands of such rods. The pro-
perties of typical zirconium alloys are defined in ASTM
Standard B 353-71, "Wrought Zirconium and Zirconium Alloy
Seamless and Welded Tubes for Nuclear Service." After
formation of tritium in the solid fuel pellet matrix, the
gaseous tritium may diffuse through the pellet matrix and
into the vold volume between the fuel pellets and fuel
--1--
A.
`~?- ~1
'
'.
''' '
., .
.. ~ , ~ . .
, , :
'. ~ , ~ '
.
:. . ' ' : .
45,970
1081673
cladding, as do a variety of other fission product gases.
These fisslon product gases are then free to migrate through-
out the fuel rod, and contrlbute to a pressure buildup
withln the cladding. The tritium, and other fission product
gases, circulate inside the fuel rod due to convection. A
typical fuel rod includes a plenum area at top of the rod,
where, due to the free volume, these gases tend to collect.
Although the radioactivity emitted by tritium is a
weak beta emission, and although it has a relatively short
biological half-life (ten days), tritium has a relatively
long radioactive half-life (twelve years). Also, tritium
will readily diffuse through most materials, including
materials such as zirconium, alloys of zirconium, and stain-
less steel, whlch are typically used as fuel rod cladding.
Because pressurized water reactors in operation today uti-
lize boric acid in the coolant for power level control,
; tritlum ls also formed wlthln the reactor coolant ltself.
Once trltlum reacts wlth water to form HTO, it is technl-
cally difficult and very costly to separate.
Regulatory authorities have therefore placed
strlngent restrlctions on allowable releases of tritium to
the environment. One way to lower the tritium inventory in
the reactor coolant, and hence the amount of tritium which
may be discharged to the environment, is to provide a means
within each fuel rod to speclfically collect and store the
tritlum produced wlthln the fuel pellets which diffuses into
the void volume. This invention provides such means, which
further are easily removable from the fuel rod during re-
, processing, As tritium is widely used as a tracer element
and in the medical and other fields, being able to simply
.
.
45,970
lO~t73
and less expensively recover the tritium, as compared to
recovery from an aqueous solution, is a further benefit
provlded by thls invention. The device disclosed herein
also may be utilized in other functions wlthin a nuclear
plant, as well as in other applications where it is desir-
able to remove tritlum from a gaseous medlum.
Although many systems and modes of operation have
been used and proposed to control trltlum subsequent to its
enterlng the reactor coolant, in accordance with this inven-
tion tritium is specifically collected and controlled within
the fuel rod itself. Thls lnvention, in the preferred
embodiment, does so by means of a device consisting of an
inner core of zirconium or alloys of zlrconlum, covered on
all surfaces wlth an adherent layer of nlckel, whlch nickel
layer acts as a selectlve and protectlve wlndow for the
passage of trltlum. At reactor operating temperatures, the
layer of nlckel is generally unreactlve to species in the
fuel rod environment, including any high temperature mols-
ture present. The nlckel layer, however, ls selectively
permeable to tritium, also allowing passage of such atomi-
cally small and avallable lsotopes as hydrogen and deuterium.
Once through the adherent nlckel layer, the trltlum reacts
with the inner core of zirconium alloy to form a solid
solution or hydrlde, and is flxed wlthin the zlrconium alloy
matrix until such tlme as lt is desirable to remove the
; trltlum.
~, .
Other devlces have been disclosed which may per-
form a somewhat slmllar functlon, although of different
deslgn and wlthout the trltium selectlvity provided by the
device of the instant invention. A United States patent
-3-
s''
~"
45,970
108~73
issued to L. N. Grossman, No. 3,742,367, June 1973, disclosesa non-destructive detection process for nuclear fuel rods.
The Grossman patent provides, in part, a device consisting
of a homogeneous alloy of titanium, zirconlum, and nickel,
as differentiated from the layered window of nickel over a
zirconium alloy core of this invention. An amount of the
alloy of the Grossman patent is placed in the fuel rod
during assembly. The assembled rod is then heated prior to
installation ln the reactor, to vaporize moisture, and free
from the fuel pellet matrix gases such as hydrogen, oxygen,
nitrogen, carbon monoxide, and carbon dioxide which react
with the alloy. The alloy within the rod is then examined
by neutron radiography to detect metallic hydrides prior to
puttlng the fuel rod into operation in a reactor. Detection
of moisture provides an indication that sufficient heating
of the fuel has occurred to remove moisture from the fuel
pellets. This reaction of the named elements and compounds
with the homogeneous alloy is designed to occur to eliminate
subsequent embrittlement and induced stresses in the clad-
ding durlng reactor operation. Since the homogeneous alloyof the arossman patent is reactive with hydrogen, it should
also be reactive with tritium released during reactor
operation.
However, it is seen that significant differences
exist between this invention and the teachings of the Gross-
- man patent. Most notably, these distinctions include dif-
ferences in elemental composition and in the methods of
~oining the elements. The prior art device consists of
titanium, zirconium, and nickel, compared to a nickel coated
3 zirconium alloy of this invention. More important, the
-4-
~;
,. '
. .
45, 970
~08~6~73
prior art rorms these elements into a homogeneous alloy,
with the reactions taking place on the surface and within
the alloy. This lnvention, on the other hand, provides a
two-layered composite device, containing zlrconium or zir~
conium alloy as an internal core and a layer of nickel on
the exterior. The device disclosed herein is much more
selective as to what will pass through the nickel layer or
window and react inside the device with zirconium alloy.
Addltlonally, the prior art alloy is used to remove moisture
and other impurity gases from a fuel rod prior to reactor
operation, whereas the device disclosed herein performs its
functlon subsequent to reactor startup and durlng the life
of the fuel rod. The devlce disclosed herein further has
significant benefits in terms of tritium recovery and separa-
tion subsequent to reactor operation.
SUMMARY OF THE INVENTION
.
This invention provides a device for removing and
retaining triti.um from a gaseous medium, and also a method
of manufacturing the device. The device, in the preferred
embodiment, consists of an inner core of zirconium alloy,
B deslrably an alloy known commonly as Zircaloy~ , and an
outer adherent layer of nickel which acts as a selective and
protective window for passage of tritium. The tritium then
reacts with or is absorbed by the zirconium alloy, and is
retained until such time as it is desirable to remove it
during reprocesslng. In the main embodiment, a small elon-
gated annular shaped device is incorporated within a re-
tention spring in the upper plenum of a nuclear fuel rod,
such that it will remove tritium formed within the rod
during the fissioning process which migrates outside the
-5-
. . .
45 ,970
~OB1673
fuel pellet matrix.
BRIEF DESCRIPTION OF THE DRAWINGS
The functions and advantages of this invention
will become more apparent from the following description and
accompanying drawings, in which:
Flgure 1 is a simplified schematic, in cross
section, of a fuel rod;
Flgure 2 is an elevation view, in cross section,
of a tritium removal and storage device;
Figure 3 is a view, in cross section, taken at
III-III of Figure 2;
Figure 4 is an elevation view of a spring used in
the upper plenum of a fuel rod;
Figure 5 is an elevation view of the device of
Figure 3 contained within the spring of Figure 4;
Figure 6 is a schematic of a test furnace; and
Flgure 7 is an elevation view, in partial cross
section, of a test capsule.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
During the operation of a nuclear reactor tritium
i8 formed. Once tritium forms HTO in the reactor coolant,
it is technically difficult and economically costly to
- separate. Also, there are stringent regulatory restrictions
today which limit the release of tritium to the environment.
For these reasons, it is highly des~rable to minimize the
amount of tritium combining with the reactor coolant. In
all nuclear reactors, tritium is produced as a byproduct of
ternary fissions. This is the largest source of tritium
production in many reactors. It is also formed as a product
3 of other reactions, such as reactions with boron-10, lith-
-6-
45, 970
10816~73
ium-6 and llthium-7, and deuterium.
As the largest source of tritium production is
within the fuel itself, typically comprising uranium, or
plutonlum, or thorium, among others, it is desirable to
remove the tritium by means in close proximity to this
source. A typical form of nuclear reactor fuel is a stack
Or solld sintered pellets 10 of uranium dioxide encased in a
sealed metal cladding 12, as shown in Figure 1. End plugs
14 hermetlcally seal the cladding 12 at the top and bottom.
10 The most widely used cladding materials are stainless steel
B and alloys of zirconium, such as Zircaloy-4. Tritium is
formed ln the fuel pellet 10 matrix, and migrates in a gas-
eous phase to the void volume between the cladding 12 and
the pellets 10. Because of its small atomic size, a signi-
- ficant portion of the tritium in the void volume may diffuse
through the fuel rod cladding 12, and into the reactor cool-
ant. Also, tr:ltium may react to replace hydrogen atoms in
the fuel cladding 12 or react with the cladding 12. It has
been found that tritium diffuses through stainless steel in
20 a reactor environment at a high rate, the rate being signi-
ficantly hlgher than its diffusion through zirconium alloys.
Tritium also reacts with the zirconium alloy cladding to
form an hydride, lessening the release of tritium to the
reactor coolant. An ideal device which will remove and
store thls ternary produced tritium should have the fol-
lowlng characterlstics: (1) it should remove and store
tritium in a gas phase within a fuel rod during the operat-
ing life of the rod, (2) the removal function should not be
limited by residual air, water vapor, or other gases nor-
30 mally present in fuel rods, such as C0, C02 and CH4, among
--7--
, .
45,970
10~1673
others, (3) it should reduce the reaction of tritium with
the rod cladding, (4) it should be inexpensive to manufac-
ture as compared to the cost associated with dealing with
excess tritium in the reactor coolant, (5) the device
should be easily adaptable to current and future ~uel rod
designs, and (6) it should provide a relatively inexpensive
source of trltium during fuel reprocessing, as compared to
removal of tritium from an aqueous solution, for medical,
tracer and other uses.
The apparatus disclosed herein meets all of these
characteristics. The apparatus, for use in a fuel rod,
consists of a two-layered composite of materials, and can be
produced in almost any geometric shape desired. The inner -
core 16 (Flgures 2 and 3) can be a number of materials, as
long as the material meets the criteria of removing tritium
in a gaseous phase within a reactor environment and retaining
it through absorption or chemical reactlon until such time
as it is specifically desired to remove the tritium. Tests
performed and discussed below were based upon an inner layer
¦ ~ 20 of zirconium alloy, such as Zircaloy-4 which is the pre-
ferred material of the inner core 16. Pure zirconium, as
well as other zirconium alloys such as Zircaloy-2, among
others, may also be used. The outer layer 18 of the device
is an adherent layer of nickel bonded to the inner core 16.
The nickel layer 18 acts as a selective and protective
barrier and allows passage of tritium as well as hydrogen
and deuterium, at reactor operating temperatures. In a
higher temperature environment, other materials, such as
dissociated hydrocarbons, could pass through the nickel
window if a sufficient quantity of these materials were
.:
45, 970
~08~G~73
available. Tests have shown, for the size device necessary
for incorporation in fuel rods, on the order of 1.5 grams,
that the nickel window should constitute roughly five to
twenty percent by weight of the device, with a more ideal
range between eight and twelve percent by weight. The
nickel should be evenly distributed over all surfaces of the
lnner core 16, such that about four to six percent by weight
is on each slde of the inner core 16. Below this level,
experimental results have shown that the removal rate is
lessened. It further may allow buildup of an oxide layer on
the device which also partially poisons its tritium removal
function. This poisoning would be the effect if only a
surface of zirconium alloy were placed inside the fuel rod,
wlthout the overlaying protective nickel window. Although
the apparatus will function above the preferred weight
percent level, to lncrease reactor efficiency, lt is desirable
to minimize the amount of neutron poisoning material in the
reactor core. As there are typically in excess of 20,000
fuel rods in a typical reactor, even a small device in each
rod will have an effect on neutron absorption. It is there-
fore preferable not to exceed the eight to twelve percent by
weight level. For the device for use in a fuel rod, an
; inner core 16 of thickness between 0.01 and 0.03 inch will
be consistent with an outer layer of eight to twelve percent
by weight. It should be noted that if the inner core is not
completely covered with a nickel layer, the device will
still operate to perform its removal function, but with
decreased efficiency.
Since the device consists of two adherent layers,
the bonding of these layers is critical, and must be care-
_g_
. , . .. , . ~ .
45,970
108~6~3
fully controlled in manufacturing. The heat treatment iscrucial. The method disclosed herein includes cleaning the
surface of the inner layer of zirconlum alloy to nuclear
specifications. Allowable impurity levels of the zirconium
alloy are as typically standard in the industry for fuel
rods, and are defined in ASTM V-353. Subsequent to clean-
ing, hlgh purity nickel is deposited upon the surface of the
lnner layer by commercially well known manufacturing techni-
ques. These techniques may include electroplating, vacuum
deposition, or a liquid dip technique, among others, as long
as the amount of the deposit is controlled. Controlled
sputtering techniques may also be used. Subsequently, the
zirConium alloy core 16 with nickel deposit 18 is thermally
treated in a vacuum maintained at about 10 6 millimeters of
mercury. It is heated to a temperature between 775C and
825C, and maintained for a minimum of three hours. It
should not be heated more than several hours beyond this
amount of time. This treatment activates the surfaces of
the zirconium alloy by diffusion of the nickel into the
zirconium alloy surface. This thermal vacuum implantatlon
provldes the protective and selective layer of nickel 18,
which is, as shown by testing discussed in the examples
below, unreactive in the presence of water vapor and fission
product gases, but permeable to tritium, hydrogen, and deu-
terium in a reactor environment. The time and temperature
relation of the heat treatment is critical as an excess of
either would allow the material to form an homogeneous
alloy, and an insufficiency would not provide sufficient
bonding. As discussed above, an alloy would be poisoned by
the other available gases within a fuel rod, thereby limit-
t -10-
i
45,970
1081f~73
ing its tritium removal and storage function.
Among the prime characteristics desired of a tri-
tium removal and storage device for use in a nuclear fuel
rod is that it does not add significant costs to the manu-
facturing process, and that it does not in any way adversely
affect reactor operation. An apparatus as hereafter des-
cribed provides such desirable results.
Most fuel rods of the type discussed include a
void plenum 20 (Figure 1) in the fuel rod, typically in the
upper regions to allow for the buildup of fission product
gases. The plenum 20 area may also be used for inclusion of
mechanical components, most notably a retention spring 22
(Figure 4) or other retention device to maintain proper
axial positlon of the fuel pellet 10 stack and allow for
fuel axial expansion. An elongated annular shaped tritium
removal and storage devlce 24 may easily be placed wlthin
the spring 22, as shown ln Figure 5, and function to remove
and store ternary fission produced tritium during the operating
life of the fuel. The device 24 shown is approximately two
20 inches in length with a 0.2 inch outer diameter and a 0.03
inch wall thickness. This device 24 may be placed within
the spring 22 in a fuel rod without complication, during
fuel manufacture. A typical spring 22 as used in pressurized
water reactor fuel rods is approximately seven inches long
with a 0.35 inch outside diameter and a 0.22 inch inside
diameter. An end cap 26 affixed to one or both ends of the
spring 22 to retain the device 24 in the plenum 20 area may
also be used. It may be a stainless steel disc with~ or
without, a central aperture 28 to provide a free path for
!
transport of tritium to the device 24. For example, the
--11-- ,
'``~
45, 970
1081t~73
device 24 may be placed inside the spring 22 and then two
end caps 26 spot welded to each end of the sprlng 22. The
spring 22 would then be placed in the fuel rod as is presently
done, with perhaps the added step of mere visual inspection
to ascertain that each spring 22 does contain a tritium
removal device 24. Alternatively, the device 24 could be
placed above the spring 22, or in rods not using a spring or
other retention devlce, it could be placed in the plenum
with means, such as a small plate, separating the device 24
from immediate contact with the fuel pellets 10.
In accordance with the invention, a series of
tests were performed to ascertain the ability of the inven-
tlon to remove and store tritium. The tests were arranged
to simulate a reactor environment, including placing a
tritium removal and retention test device in competition
with the zirconium alloy cladding for the tritlum. Early
tests also simulated the abillty of the inventlon for
tritium removal and retention in competition with several
other mediums.
It should be noted that in all of tests, deuterium,
whlch can be more easily obtained, was used as opposed to
tritium, which is a typical laboratory technique. Deuterium
is easier to work with in a laboratory environment and posed
less of a health concern than would tritium. Tritium and
deuterium are similarly sensitive to surface barriers and
; isotopic exchange reactions. Also, well recognized in the
art, is that similar recovery and detection techniques may
be used for tritium and deuterium. As among any isotopes of
a given element, the kinetic relationships of tritium and
deuterium are similar. Further, briefly statedg the dif-
,,~
f
45,970
~081~73
fusion coefficient o~ deuterium and tritium through materials
as taught herein is similar, with tritium having a somewhat
lower coefficient than deuterium.
EXAMPLE I
The first laboratory test involved a competition
among eight differing devices. All of the samples were of
slmilar mass. They were cleaned with acetone and then dried
and weighed prlor to being inserted into a quartz furnace
tube 60 (Figure 6). The samples were either in a thin foil
sheet ~orm, approximately 10 mils thick, or powders, as
lB noted below, and the Zircaloy-4 cladding samples were cuts
from actual fuel cladding. The powders were contained in
high purlty platinum crucibles which, when subsequently
analyzed, had essentially no deuterium content. The furnace
tube was then evacuated, and placed in a furnace 62. The
eight samples were suspended within the tube 60 by a quartz
sample holder 64. The samples were then heated to 650C
while the gas pressure was observed and the furnace 62
composition analyzed by mass spectrometry. When the gas
atmosphere in the furnace 62 showed little or no change~ the
furnace 62 temperature was reduced. When the temperature of
the furnace tube 60 reached 310C, deuterium gas at a pressure
of 1.4 millimeters of mercury was added to the furnace tube
60, corresponding 'co about 1.2 cubic centimeters. The
pressure was monitored continuously by a metal capacitance
manometer and gradually decreased to 0.44 millimeters of
~ mercury after forty-two hours. The furnace 62 was then
; cooled to room temperature and mass spectrometric analysis
; performed on the gas atmosphere. It showed that 0.16 cub~c
centimeters of deuterium remained in the system. The samples
. ~ :
,. . .
~ ............................... . . .
45, 970
108~673
were then weighed, the deuterium extracted from each sample
by a hot vacuum extraction technique and mass spectrometrD
analysls. This involved heating each sample to about 1050~C,
which is above the temperature range (800~C-850C) at which
hydrogen and its isotopes dissociate from zirconium and
Zlrcaloy-~. The amount of deuterium was quantitatIvely
determined by mass spectrometry. Before each experiment
discussed herein, the experlmental system was calibrated
wlth National Bureau Or Standards (NBS) Hydrogen Standards.
1~ The results are shown in Table I. The letters "A"
through "H" representing each sample correspond t~ the
letters on Figure 6 showing the relative locatlon of the
samples ln the furnace tube 60. Sample "A" represents a
zirconium-titanium powder, with 6.2% by weight nickel;
sample "B" a zirconium-titanium powder with a 3.9% by weight
nickel; samples "C" and "D" were Zircaloy-~ cladding; sample
"E", as discussed herein, a Zircaloy-~ core with a 5.7% by
weight outer nickel layer; sample "F" a zirconium metal core
with a paladium outer layer; sample "G'l a zlrconium-titanium
alloy with a paladium coating, and sample "H" a zirconium
core with a ten weight percent vanadium coating.
As shown from the three data columns of Table I,
the Zircaloy-4 core with a nickel outer layer proved far
superior to the other samples in removing and retaining the
deuterium, even in competition with a wide variety of other
samples:
,
~: :
~ -14-
~. .
'
45, 970
~081~73
TABLE_I
D2 D2 D2
Sample (ppm) (cc)(cc/gm sample)
A 747 .24874.180
B 669 . 21203.744
C 4.8 .oo56 .026
D 2.9 .0032 .016
E ?243 528612.556
11.5 .oo58 .o64
G 34.2 .0102 .191
H 2.8 .oo36 .016
EXAMPLE II
A second competltion test was run, using the same
experimental procedure as discussed in reference to Example
. I. Among the samples here, however, were included three
comprising oores of Zircaloy-~, wlth varying weight percentage
nlckel outer layers. Samples "B-2", "C-2", and "E-2" com-
prised a ten (10%) percent, a 5.7%, and a 3.3% nickel layer,
respectlvely. Sample "A-2" was Zircaloy-4 cladding material;
20 sample "D-2" a zlrconium-titanium powder with 6.2 weight
percent nickel; sample "F-2" a zirconium-titanium powder
with 3.9 weight precent nickel; sample "G-2" Zircaloy~in a
thin foil (.005" thick) form; and sample "H-2" Zircaloy-~7
claddlng materlal.
~'' ~'.
,
.
--15--
~ '
45,970
~081~73
TABLE II
D2 D2 D2
Sample (ppm) (cc) (cc/gm sample)
A-2 1.2 .0013 .0067
B-2 122 02~3 6832
C-2 45.1 0123 2526
D-2 4.6 .0018 .0258
E-2 2.4 ooo6 0134
F-2 2.3 .0007 .0129
G-2 11.2 .0026 .0627
H-2 0.6 .0007 .0034
As shown from Table II, the samples including an
inner core Or Zircaloy-~7and outer layers of nickel dld
quite well in deuterium adsorption. Further, it is most
evident that the deuterium removal ability significantly
increased with increasing weight percentage of nickel.
EXAMPLE III
A third competition test was run, utilizing the
. ~/77
same procedure, and again the Zircaloy-4 core with a ten
20 weight percent nickel outer layer showed far superior. In
Table III, sample "A-3" was Zircaloy-4 cladding; "B-3"
Zircaloy-/~7 foil with a ten percent nickel outer layer;
sample "C-3" zirconium-titanlum powder with 7.75 weight
percent copper; sample "D-3" zircor~ium-titanium powder with
12.1 weight percent nickel; sample "E-3" zirconium-titanium
powder with twelve weight percent copper; sample "F-3"
zirconium-titanium powder with 6.5 weight percent nickel;
sample "H-3" Zircaloy-~ foil; sample "I-3" Zircaloy ~ clad-
ding.
--16--
45,970
~81673
TABLE III
D2 D2 D2
Sample tppm) (cc)(cc/gm sample)
A-3 0.9 .001 0.005
B-3 407 100 2.279
C-3 29 .003 0.160
D-3 57 . oo8 o . 319
E-3 11 .002 0.062
F-3 19 .003 0.106
H-3 17 .003 0.095
I-3 1 .001 o.oo6
::
EXAMPLE IV
A fourth competition test was similar to those
discussed above. Here, however, all the samples were an-
nealed at 660C for fifteen hours in a vacuum before the
deuterium pressure addition. Again, the Zlrcaloy-4 with a
ten percent by weight nickel outer layer proved by far to be
~ superior, and the adsorption significantly increased.
; Sample "A-4" was a Zircaloy-~ cladding sample; sample "B-4"
the Zircaloy-~ with nickel outer layer; sample "C-4" a
zirconium-titanium powder with 7.75 weight percent copper;
sample "D-4" a zirconium-titanium powder with 12.1 weight
, . . .
; percent nickel; sample "E-4" a zirconium-titanium powder ~ ~ .
with twelve weight percent copper; sample "F-4" a zirconium-
titanium powder with 6.5 weight percent nickel; sample "G-4"
Zircaloy-4 foil; and sample "H-4" Zircaloy-~ cladding
material.
.
~ -17-
.~
45, 970
1(~81~73
TABLE IV
D2 D2 D2
Sample(ppm) (cc) (cc/gm sample)
A-4 4.3 .0049 .0241
B-4 1413 .325 7.92
C-4 40 .0051 .224
: D-4 14 .0018 .079
E-4 16 .0021 .090
F-4 8 .0012 .045
G-4 6 .0015 .034
H-4 2 .0026 .011
EXAMPLE V
Later tests were performed and arranged to simu-
late a reactor environment, including placing a tritium
removal and retention test device 30 in competition with
B Zircaloy-4 cladding. A test apparatus was arranged, and is
shown ln Flgure 7. The test apparatus, referred to here-
inafter as the "test capsule" 40, included fuel rod test
Tr,q
cladding 32 of Zircaloy-4. The capsule 40 was approximately
20 ll-l/2 inches ln length. Also included in the test capsule
40 were end plugs 34 o~ Zircaloy-4, a test tritium removal
and storage devlce 30, a deuterium gas generator 36, and a
glass tube spacer 38. The test device 30 was a rod prepared
as discussed above, with a nickel layer of twelve percent by
weight, and was approximately one and one-half inches in
length and 0.2 inches in outer diameter. As shown, it was
placed in the upper end of the test capsule 40. At the
lower end of the test capsule 40 was placed the deuterium
generator 36. The generator 36 was approximately one inch
45,g70
~0816'73
in length with a 10 mil wall thickness of nickel and a 3/16
inch outside diameter. The generator 36 was made by taking
a 3/16 inch diameter high purity nickel rod, and drilling
the inside of the rod to give the desired wall thickness.
The bottom surface of the nickel rod was not drilled through.
Then, a controlled amount of deuteriated water (D20) was
placed in the nickel shell. Also placed within the shell
was high purity lron (Fe) wire, in coil form. While main-
taining the lower portion of the deuterium generator 36 in a
liquid nitrogen solution to solidify the deuteriated water,
the upper portion of the nickel shell was welded shut.
After cooling, the generator 36 was then placed in the lower
portion of the test capsule 40, which previously had been
~ealed by welding on one of the end plugs 30. The tube
spacer 38 was a sealed glass tube approximately 7-1/8 inch
long that made a loose sliding fit with the test rod clad-
ding 32. The test device 30 was then inserted in the cap-
sule 40, and the upper end plug 34 was welded in place,
sealing the capsule at approximately one atmosphere of
helium. A second test capsule was constructed, the only
difference being the inclusion of a capillary tube placed
ad~acent the test device 30. The capillary tube contained
260 micrograms (~4gm) of water. The tube ruptured at test
temperature, releasing high temperature water vapor.
To run the test, the capsule 40 was placed in a
gradient furnace that heated the cladding wall 32 opposite
I the glass spaoer 38 and the deuterium generator 36 to slightly
higher temperature than the test device 30. The device 30
t` ran at approximately 320C while the cladding 32 wall tem-
30 perature varied from 380C to 320C. The higher temperature
1, --19--
. ~
45, 970
1081~73
area was between the deuterium generator 36 and the device
30. The iron wire reacted with the deuteriated water to
form a combination of Fe3O4 and Fe2O3 at approximately
300C, and liberated the deuterium, which freely passed
through the nickel wall of the generator 36. The glass
spacer 38 formed a small annulus for transport of the deu-
terium to the test device 30, simulating the annulus between
the fuel pellet 10 stack and the cladding 12 inside diameter
in an actual fuel rod. The test was run in a controlled
argon atmosphere that was monitored for escaping deuterium;
none was observed. The test capsule was held at temperature
for seven days, and then cooled to room temperature.
Multiple analyzes were then performed upon the
test capsules. Puncture and recovery of the internal gas
atmosphere showed the only gases present to be helium and
traces of hydrocarbons. Hydrogen and deuterium analyzes
were then per~ormed on the test device 30 and at selected
locations of the test cladding 32 represented by the arrows
on Figure 7. The results are summarized in Table V. The
letter "H" denotes the capsule with the 260 ~4gm of water
addition.
,~ :
,~
.
-20-
:.
45,970
~)81~73
TABLE V
Deuterium Hydrogen
Capsule 1 lH 1 lH
Device
ppm, wt. 7.0 5.3 22.9 35
~4gm 30.4 23.0 96.8 149
percent 51.6 55.1 22.9 36.2
Claddin~
ppm, wt. o.8 0.53 9.2 7.5
~gm 28.1 18.6323.0 262.0
percent 47.7 44.6 76.6 63.6
Generator
ppm,~ wt. 0.3 0.1 1.4 0.55
~gm o.4 0.2 2.0 o.8
percent 0. 8 0.35 0.5 0.2
As shown from Table V, the test device 30 con-
tained about 52 percent of the initial deuterium. Less than ~ -
1 percent of the deuterium remained in the generator 36.
The tests further showed that the added moisture had very .
20 little effect on the ability of the devlce 30 to remove the
deuterium. In fact, it increased the removal and retention
of deuterium, by the device 30, by several percent. This is
believed due to buildup of an oxide film on the inner sur-
face of the test cladding wall 32. The film could be seen
by visual inspection, and was especially evident in the
upper area of the cladding 32, where the water was released.
j There was no such film on the device 30 itself, as there was
no reaction with the protective adherent nickel layer. As
there is typically excess moisture on the surface and within
; 30 the fuel pellets 10 during manufacture, this same effect can
;~ be expected to be experienced during operation of the fuel
within a reactor. An oxide film will be built up on the
inner surface of the fuel cladding 12 early in the operating
;~ life of the fuel, thereby forming somewhat of a barrier to
~ - 21 -
~, . .
,,
45,970
108~673
the interaction of tritium with the cladding 12. This will
increase the efficiency of the tritium removal and retention
device.
As a further result, the device 24 may perform a
safety related function during plant operation. In the un-
likely event that the cladding 12 of a fuel rod fails,
reactor coolant water reacts with the inner surface of the
fuel rod. The tritium removal and retention device 24 not
only is lnert to the coolant but also retains its inventory
in the presence of steam formed by the reactor coolant. In
the unlikely event of fuel rod failure, the device will act
to absorb free hydrogen, and will not act to catalyze the
incoming coolant water, as might an alloy type device.
Another benefit of the device 24 disclosed herein
is its ability to provide a source of tritium relatively
less expensive than obtaining tritium from an aqueous solu-
tion. Tritium has been used as a tracer element in many
functions. It is also used in medical treatment. After a
fuel rod containing the disclosed device is removed from a
reactor, the device 24 can be easily removed and separately
processed. Heating the device to a temperature in the range
of 1100C in a vacuum maintained at 10 6 mm Hg releases the
,
entrained tritium, and also any entrained hydrogen, in a
~; gaseous phase. Separation of the tritium from this medium
is significantly easier than separation from water.
It is therefore seen that the device disclosed
herein provides a means to remove and store gaseous tritium.
It is particularly applicable to use in nuclear fuel rods,
where its function is not reduced by residual water vapor or
other fission product gases within a rod. It further limits
-22-
~'`'' ..
45, 970
10l3~673
the reaction of tritium with the fuel rod cladding, and canbe easily manufactured and incorporated in existing fuel rod
types. It poses no additional problems in the unlikely
event of fuel rod failure, and may provide tritium for
medical, tracer, and other uses. It is apparent that many
modifications and variations are possible in view of the
above teachings. It therefore is to be understood that
within the scope of the appended claims, the invention may
be practiced other than as specifically described.
-23-
"~
