Note: Descriptions are shown in the official language in which they were submitted.
1~i9'~5~
This invention is directed to a method of removing neutron
poisons from irradiated uranium-base nuclear reactor fuel to permit
the fuel to be re-used to a higher burn-up of fissionable material
present. It has been fourld possible to remove selectively the bulk
of the neutron poisons without significantly affecting the uranium
-based mal:rix.
BACKGROUND AND PRIOR ART
Durillg irradiation in nuclear fission reactors, various
nuclides accumulate in the fuel as by-products of the fission reaction.
Some of these nuclides are strong absorbers of thermal neutrons
and it is the build-up of these nuclides that limits the useful
life of the fuel. In some reactor designs, the fuel can cease
having an energy-generating function even when less than 5% of
the fissile atoms in the fuel have undergone fission. When this
occurs, the fuel must be removed from the reactor and placed in a
waste storage facility. Alternatively, the fissile/fertile material
can be extracted from the neutron-absorbing nuclides in order that
new fuel can be fabricated and further energy produced from the
unused fissile/fertile material.
In conventional chemical reprocessing, the separation
of the fissile/fertile elements from other unwanted elements is
achieved by dissolving the irradiated fuel in a strong acid followed
by solvent extraction. Such reprocessing yields a pure (usually
nitrate) solution that must be converted into a
2S ceramic powder prior to fuel refabrication. Normally
~'~
'~
reprocessing is done not only to remove neutron~absorbing nuclides
but also radioactive nuclides. Removal of these latter nuclides
permits the conversion and fuel fabrication to be carried out without
the use of thick shielding.
Pyrometal]urgical separation (reprocessing) methods have
been successful in removing the more volatile fission products such
as the gases krypton and xenon, the halogens, and metals with low
boiling points such as cesium and rubidium. These methods use
heat, or oxidation plus heat, to volatilize certain fission product
elements or compounds thereof. To separate additional unwanted
fission products, the further steps of adding a reducing metal
(e.g. Zn or Cd) in molten form, heating to alloy certain other
fission product elements with the molten reducing metal, and separat-
ing the molten alloy from the solid oxides remaining, have been
15 carried out.
The above volatilization techniques are inefEective
in separating the rare earth fission products, (which would account
Eor most of the parasitic neutron absorptions in continued irra-
diation), none of which will volatilize at practical temperatures
20 or conditions. The alloying techniques can remove some of these
rare earths but usually are inadequately effective and relatively
expensive.
There are several possible ways being considered of
converting spent LWR (Light Water Reactor) fuel into a form that can
f r a ~ ~
~91 25 be further irradiated in a CANDU~(or heavy water moderated, natural
uranium typc) reactor to create a tandem fuel cycle. The basic
variations possible derive from the amount of decontamination carried
nlt on the spent LWR fuel after the first half of the cycle. One
--2--
important consequence of the degree of decontamination employed
is that it determines the burnup achievable in the CANDU half of
the cycle -- without any decontamination, burnup would be limited
to the 260-380 MW.h/kg il.E. range; with full decontamination, burnup
would be extended to the 480-620 MW.h/kg H.E. range. It would
be desirable to develop an efficient way of converting spent LWR
fuel into CANDU bundles that does not use chemical reprocessing,
but does succeed in removing most of the important neutron absorbers,
thereby permitting extended burnups. Thermal-based treatments
alone do not appear to be capable of decontaminating spent LWR
fuel to the extent where it can be handled without shielding or
taken to a significantly extended burnup in the CANDU half of
a tandem fuel cycle.
Recent work in our lab has shown that wet stirred-ball
milling after a single oxidation/reduction cycle can produce a
powder well suited for ceramic fabrication. (see Canadian Applica-
tion No.~ ~ G3filed 27 June 1985 by B.J.F. Palmer). In this work, sin-
tered natural U02 pellets were oxidized to a ll3~8 rubble that wns then reduced
back to U02. Wet stirred-ball milling then converted the coarse U02 rubble
into a powder consisting of soft granules of submicrometre particles. This
powder had extremely good processing characteristics and pellets refabricated
from it sintered to geometric densities oE over 96% of theoretical. Thus,
thermal treatment of spent U02 fuel in an oxidi~ing atmosphere followed
hy reductlon and wet stirred-ball mi]ling can produce a ceramic U02 powder
t:}n~t is i(ieal for fclbrication into higil-density CANDU fuel pellets.
SUi~MARY OF TUE INVENTION
We have developed a method of removing neutron-absorbing
12~i9;~5~
fission products from irradiated uranium-based fuel, comprising
a) contacting the fuel in a comminuted form with a weak
acid selected to prcferentially dissolvc saicl fission products,the
conditions (such as the acid concentration fuel particle size,
ternperature and contact time) being chosen to dissolvc neutron-
absorbing fission products present without significantly dissolving
the uranium-based matrix,
b) separating the acid solution of fission products
from the solid fuel, and
c) reconstituting the fuel for recycle thereof.
Preferably the acid treatment removes the various neutron-
absorbing fission products such that the number of parasitic neutron
absorptions occuring in the fuel during subsequent irradiation
is reduced by a factor of 2 (or more).
Preferably the method includes oxidizing spent uranium
oxide fuel (e.g. in air or oxygen at 400 to 800C) to form a U308
rubble, the rubble reduced to U02 (e.g. in 1l2 at 400 to 800 C),
the U02 particles wet ground in a weak acid solution (e.g. in
glacial acetic acid at 25 to 118C for a few hours), the acid
solution separated, and the remaining U02 ground powder fabricated
into fuel for re-use.
Preferably the irradiated fuel is a spent enriched uranium
fuel which after removal of at least 50% of the neutron-absorbing
fission products present is refabricated into fuel and used in
a natural uranium fueled reactor.
The invention includes the treated fuel product.
DETAILED DESCRIPTION
_ _ _
The irradiated fuel can be any spent uranium-based fuel
such as uranium oxide, or uranium alloys (e.g. with Al, Si etc).
--4--
~69~5~
Usually the fuel is U02 based although some plutonium or thorium
oxides may be present. Plutonium-enriched uranium oxide fuel
can be treated. In the case of oxide fuel, the uranium can be
predominantly or solely in either the U02 or U308 form during
contacting with the weak acid.
The comminution can be achieved by any form of crushing
and/or grinding. It has been found preferable to fracture to a
rubble form by thermal oxidation followed by wet grinding most
advantageously during contact with the weak acid. For efficient
wet grinding the fuel must be previously comminuted to a particle
size of less than about 10 ~m diameter; soft granules comprised
of such particles are acceptable. To ensurc that a ceramic-grade
powder is produced, wet grinding must be carried out until the
mass-median particle size is less than about 2Jum diameter.
It should be noted that wet-grinding is not essential:
any other form of fine grinding may be used. The advantage of
wet grinding is that it allows the decontamination and grinding
steps to be carried out together. For some types of fuel, refluxing
of rubble-form material in the selected acid will peptize the
fuel so as to reduce the particle size. Some of the fission products
are concentrated at the grain boundaries of the fuel matrix and
it is also possible that as dissolution of these fission products
proceeds the particle size is reduced further.
The weak acid must be selected to be substantially incapa-
ble of dissolving the fuel oxide matrix during the contacting.~cetic acid has been found preferred but any other weak acid particu-
larly weak organic acids such as carbonic, formic, propionic,
glycolic, malonic, and ciLric may be used. Stronger acids such
as hydrofluoric, hydrochloric and nitric, are not suitable because
--5--
92 ~1
they dissolve significant amounts of the uranium.
The solution concentration of the acid should be sufficient
to dissolve the neutron absorbers without significantly dissolving
the uranium-based matrix. The acids can also be used in their
molten form e.g. glacial acetic acid, at below their flash point.
The use of glacial acetic acid is a preferred aspect of the invention.
Suitably the Euel solids loading in the acid may be Erom about
10 to about 40 volume % although other proportions may be used.
One preferred mode of contacting with the acid is to
wet grind in any suitable apparatus e.g. ball mill or rod mill.
Thus size reduction and decontamination can proceed concurrently.
Stainless steel equipment would be very suitable: corrosion should
be minimal.
The temperature during the contacting with acid will
be selected within the range from ambient (or the acid melting
point if used in molten form) to reflux or boiling point temperature.
The time of contact will be determined by ,he irradiation history
of the fuel, the acid chosen, its concentration, and the fuel
particle size. In all cases the contact time is expected to be
less than 24 hours and iTl most cases will be much shorter. The
acid amount (and concentration), temperature and time should be
optimized to dissolve the maximum amount of neutron-absorbing
fission products present yet without dissolving significant amounts
of the uranium oxide matrix. Specific conditions will vary with
fuel type and irradiation conditions and will need to be determined
experimentally for each application. The commercial viability
of the process will be determined by subtracting the processing
costs from the economic worth of the extended burnup achieved
(as determined by the quantity of neutron poisons removed). A
12~
small transfer of uranium and plutonium to the acid phase has
been obsesrved in some tests but the amount has not been significant
(with proper control this can be minimized).
Sufficient neutron-absorbing poisons can be removed
by this acid treatment to render the fuel acceptable for re-use
in CANDU-type reactors. The neutron-absorbing fission products
which were found to be removed by the selected weak acid treatment
comprise Nd, Sm, Eu, Rh and Cs. Other fission products were un-
doubtedly removed also. Those that were observed to dissolve include the
most important ones (plus some that we could determine witl- little extra
work). While some of the Cs can be volatilized by heating (prior
art) other fission products particularly rare earth oxides including
Eu, Sm and Nd cannot, and thus there is a need for this acid treat-
ment. This treatment avoids the conventional chemical reprocessing
and conversion to oxide but permits much of the radioactivity
to remain with the fuel oxide so that the acid treatment and refa-
brication must be carried out remotely with appropriate shielding.
The avoidance of the conventional chemical reprocessing and conver-
sion is believed to outweigh the requirement Eor shielding during
the treatment in many applications.
A preferred procedure includes an initial thermal oxidation
of the spent fuel at temperatures high enough to remove some of
the cesium (volatilization) and to break down the solid to a rubble,
followed by a reduction back to the starting oxide, the reduced
material then being wet-ground in the acid until the bulk of the
rare earth and other basic neutron-absorbing fission products
are transferred to the acid phase, the acid phase separated from
the residual solids, and the oxide solids fabricated into fuel.
For instance irradiated uranium dioxide fuel is oxidized to rubble-
--7--
25~
form U 3 8' then the U308 reduced to U02, the U02 material wet-
ground in glacial acetic acid to less than 2,um diameter at about
25 C for about l.O hour, the resulting acetic acid phase separated,
and the leached solids fabricated into fuel for re-use. The solids
can be fabricated into fuel by usual pressing, sintering and finish-
ing operations.
It has been found difficult to predict the acid contacting
conditions required to give the desired extraction of fission products.
Various fuels have been found to behave differently apparently
depending on their irradiation and thermal history. However routine
testing will determine the degree of severity of extraction conditions
required.
The following example is illustrative.
EXAMPL~
An experiment was designed to confirm the feasibility
of the concept. Irradiated natural uranium dioxide Euel that came
from an outer element of a fuel bundle was sampled. This fuel
bundle had been irradiated for 1588 full power days at an average
power of 4.18 kW/kg U which corresponds to an average burnup of
159 MW.h/kgU. Burnup analysis of two samples yielded an average
of 157 MW.h/kgU.
In a shielded cell, two 30g samples were cut sequentiaily
from this fuel element. One of these samples was placed in a porcelain
crucible and oxidized in air for 20 h at 930 C.
The oxidized sample was carefully split into four repre-
sentative subsamples. One of these subsamples was refluxed in
acetic acid, two were dissolved in nitric acid so that the sampling
procedure could be evaluated and the composition before refluxing
12~9~5~
could be determined, and the final subsample was placed in archive.
The unoxidized sample was also dissolved in nitric acid so that
the composition before oxidation could be determined.
Refluxing was carried out as foLlows. A 8.71 g subsample
of oxidized l10 2 was placed in a dissolver flask along with 100
mL of reagent grade glacial acetic acid. This solution was refluxed
for 24 h. The oxidized rubble broke up during refluxing so that,
upon completion, the suspension did not settle rapidly. After
a period of days, the bulk of the U308 had settled and lay undissolved
in the bottom of the flask.
Aliquots of the various solutions were quantitatively
diluted and removed from the shielded cell. The amounts of europium
and cesium in Lhe various solutions were determined by gamma spectro-
scopy. Results are presented in Table 1. Agreement between the
two isotopes of each element is good. The anomalies that are present,
specifically the slight increase in Eu-155 and Cs-134 activities
during oxidation and the lack of agreement between the amounts
of Cs-134 and Cs-137 removed during oxidation, are understandable.
Counting statistics are responsible for experimental errors that
are significant for the less active isotopes (Eu-154, Eu-155 and
Cs-134). A potentially more important source of error is sampling.
Cesium is not distributed homogeneously in irradiated fuel. Although
care was taken, it is probable that some samples contained propor-
tionately higher amounts of cesium before treatment than others.
However, the two sets of results on separate samples of the oxidized
fuel demonstrate that the sample splitting procedure used was effective.
Liquid chromatography was used to measure the relative
cc-ncentrations of the other rare earths and uranium. . These results
_g_
1~69~51
are given in Table 2, which also includes the gamma spectroscopy
results expressed in terms of percent removed. The amount of uranium
in the acetic acid solution was also determined by emission spectroscopy
and this result is also included in Table 2.
This experiment conirmed that thermal treatment does
not result in significant decontamination. During treatment at
930C, oniy a smali amount of the europium and a fraction of the
cesium were removed. Refluxing in acetic acid proved to be a much
more effective method of decontamination. Despite the fact that
grinding was not employed, over half of the three most important
neutron poisons along with cesium, which dominates the gamma dose
associated with the fuel, were removed. Several other rare earths,
and probably many other fission products which were not examined
in this experiment, were also removed in similar amounts.
TABLE I: GAMMA SPECTROSCOPY RESULTS
SPECIFIC GAMMA ACTIVITY (~Ci/gU)
SAMPLE Eu-154 Eu-155 Cs-134 Cs-137
.
unoxidized fuel 209 124 132 15130
20 oxidized fuel (first 215 131 132 12550
subsample)
oxidized fuel (second 196 129 137 12520
subsample)
oxidized fuel (average) 206 130 135 12535
acetic acid solution 113 67 85 7770
--10--
9;~
BLE 2: SUMMARY OF T11E DECONTAMINATIONS ACIIIEVED DURINC
OXIDATION & REFLIJXING
A~OU~IT REMOYED (~ OF INITIAL)
m ss on
Gan~na Spectroscopy L~qu~d ChromatoqraDhy SDectroscoPy
OPERATIOII ~154 Eu-155 Cs-I3~- Cs~I~7 Sm ~Id Pr Ce La U U
oxldatlon 1 -5 -2 17
reflu~lng 55 52 63 62 55 56 63 58 60 9~1 4~2
In this experiment, the 24 h refluxing treatment in acetic
acid removed 50-60% of all of the major neutron-absorbing fission
products including Nd, Sm, Eu, Rh and Cs. A 10% carryover of U
and Pu was detected in the acetic acid solution. Some or all of this-
is believed to have been colloidal material which could have been
separated (ultrafiltration or centrifugation) and returned to the
solids. If the wet-grinding step were included it is expected
5 that the fission product removal would be higher than the 50-60%
in this test.
The irradiated uranium-based nuclear reactor fuel from
which substantial amounts of neutron-absorbing fission products
have been removed by selective extraction with a weak acid, is
believed to be a novel product.
