Note: Descriptions are shown in the official language in which they were submitted.
106Z5~0
This invention relates to a method of decontaminating
a ,heavy water moderated and cooled reactor or an ordinary water-
cooled reactor, including the fuel assemblies and associated
heat transport systems. The invention also relates to novel
reagent compositions which may be used in the aforesaid method.
More particularly the invention relates to the injection of a
relatively small quantity of a reagent composition directly in-
to the circulating reactor coolant - either water (H2O) or
heavy water (D2O) - so that the coolant acts as a carrier for
the reagent in dilute solution therein. The reagent dissolves
the active corrosion products containing radioactive contamina-
tion, and these are removed by passing the coolant over a
cationic exchenge resin which becomes "loaded" and is then dis-
carded. Reagent which is regenerated by the cationic exchange
resin is recirculated through the reactor system. Finally,
the reagent may be removed from the system by passing the coolant
through an anionic exchange resin and depositing the reagent
thereon.
In the past, decontamination of nuclear reactors
has been effected by circulating various reagents, in a concentra-
tion of about 3-Z0%, through the equipment and then discharging
the spent reagents and contaminants. Reagents such as oxalic
acid (2-25 g/l), ammonium citrate (5-100 g/l) and EDTA (0.4-4 g/1)
have been employed. Phosphoric acid and dilute dichromic acid
have also been employed. Such methods all have the inherent
disadvantage that heavy water coolant becomes contaminated, and
use of a conventional decontamination reagent results in serious
downgrading of the expensive primary heavy water coolant, and
probable corrosion of parts of the e~uipment. In order to
- 1 -
~F
___ rm/~`
106Z590
minimize downgrading of the D2O, the D2O would have to be replaced
with H2O before decontamination with the relatively concentrated
and corroSive conventional reagents took place. This is not only
time consuming but could be hazardous since the workers are
likely to be exposed to radiation. As the radioactive wastes
from the conventional treatments are in liquid form and large
quantities are involved, disposal is very costly. Many of the
disadvantages of the prior art were overcome by Motojima et al
in United States Patent 3,737,373, issued June 5, 1973, which
describes a process in which deuterated 0.1% oxalic acid was
added to the D2O reactor coolant to dissolve the contaminants.
The dissolved radioactive contaminants were removed from the
coolant flow by irradiation of the oxalic acid causing decomposition
thereof and precipitation of the dissolved metals and contaminants.
~he contaminants can then be recovered by simple filtration
and/or ion exchange techniques. The Motojima method, as per the
aforesaid United States Patent, is limited to the use of deuterated
oxalic acid which is destroyed during the process, so that additional
acid must be added to continue decontamination. The use of oxalic
acid may be disadvantageous owing to the formation of insoluble
oxalates. Furthermore, the method is discontinuous in that the
reactor must be cycled between cool, subcritical conditions and
hot, critical conditions.
An object of the present invention is to provide a
simple process for continuously decontaminating shutdown heavy water
moderated and cooled reactors which minimizes corrosion, down-
grading of the expensive heavy water, the volume of radioactive
waste for disposal, and decontamination time.
Another object of this invention is to provide a
novel composition of matter for use as a decontaminating reagent
106Z590
in the process of the present invention.
Thus, by one aspect of the present invention there is
provided a method of decontaminating a nuclear reactor coolant
system which comprises in;ecting an acidic chemical reagent into
circulating coolant in said system to thereby form a dilute
reagent solution, circulating said solution to dissolve contam-
inated deposits therein, passing said dilute solution through a
cationic exchange resin to thereby collect dissolved cations
and radionuclides thereon and regenerate said chemical reagent
solution, recycling said regenerated solution through said
system, and subsequently passing said coolant solution through
an anionic exchange resin to remove said reagent from said
coolant system.
In another aspect of this invention there is provided a
decontaminating reagent for use in the decontamination of nuclear
reactor systems comprising a binary or ternary mixture of (a) at
least one of sulfuric acid, sulfamic acid, dithionous acid,
phosphoric acid and hydroxylamine, (b) at least one of ethylene-
diaminetetraacetic acid (EDTA) and nitrilotriacetic acid (NTA),
and (c) 0 - 50% citric acid. -
A particularly preferred reagent for use in the decontamin-
ation method comprises a mixture of ethylenediaminetetraacetic
acid (EDTA), oxalic acid and citric acid.
In the process of the present invention it has been found
that the reagent may be used in concentrations as low as 0.01~
and consequently the risk of corrosion of reactor components is
considerably reduced. As the reagent is regenerated on the
cationic resin exchange bed the overall cost of some reagents
j?l _3_
., ~
106Z5~0
which are relatively costly is reduced considerably. On the other
hand, the cationic exchange resin is sufficiently inexpensive
to permit discardal following loadiny with radioactive contaminants.
As the radioactive contaminant waste is in a low bulk solid form,
disposal of the radionuclides is simplified.
Although decontamination factors are smaller than
can be achieved with more aggressive conventional methods, the
present process is inexpensive, simple to operate, relatively
non-corrosive and generates much less waste than the conventional
. 10 chemical cleaning methods. These features permit it to be used
more frequently than conventional methods, so that radiation
fields need never be allowed to become very high, with a conse-
quent saving in man-rem exposure.
As the reagents are added directly to the coolant to
provide a low concentration of reagent, it is merely necessary
to shut down the reactor before effecting the decontamination
process. It is not necessary to defuel the reactor nor is it
necessary to drain the system. Because the decontaminating reagent
is regenerated the process can be continued as long as activity
is still being removed and because the reagents are dilute, the
formation of gases from corrosion and resin degradation is slight.
Many reagent compositions have been found to be
effective to dissolve contaminants from nuclear reactor systems
and amenable to removal by ion exchange resin. Most such comp-
ositions are based on mixtures of organic acids, such as oxalic
and citric acids, with or without complexing or chelating agents
such as EDTA, HEDTA and NTA. The precise mechanism of dissolution
is not entirely understood but it is believed to include diss-
olution as well as complexing of the contaminants. Use of oxalic
- 4 -
jC~
~06Z590
acid alone is believed not to be entirely beneficial due to the
possibility of the formation of insoluble oxalates. It has
also been found that inorganic acids such as sulphuric acid are
effective in combination with either complexing agents or oxi-
dizing agents, depending upon the type of contamination under
consideration. Thus, the reagent composition is generally a
mixture containing at least two compounds selected from the
group consisting of sulphuric acid, oxalic acid, acetic acid,
thioglycolic acid, citric acid, ethylenediaminetetraacetic acid
(EDTA), nitrilotriacetic acid (NTA), hydroxylamine, hydrazine,
dithionous acid, sulphurous acid, sulfamic acid, phosphoric acid
and ethylene diamine, with at least one acidic compound being
present. Pref~rably a reducing agent component is included.
One suitable reagent composition which has been proven effective
in large scale testing is that sold under the trade mark NUTEK
L-106 by Nuclear Technology Corporation, of Connecticut, U.S.A.,
which is a modified polyfunctional organic reagent comprising
about 40% EDTA, 22% oxalic acid and 20~ citric acid, the balance
being water of hydration and impurities (all percentages in this
specification are by weight unless otherwise indicated).
Other particularly suitable compositions include:
(a) 20-50% citric acid
10-50% ethylenediaminetetraacetic acid (EDTA)
20-50% hydroxylamine
(b) 0-50% citricacid
10-50% NTA or EDTA
20-50% oxalic acid or hydroxylamine
(c) 25-75% ethylenediaminetetraacetic acid (EDTA)
25-75% hydroxylamine
- 5 -
rm/'~
~06Z590
(d) 20-40~ sulphuric acid
60-80% hydroxylamine
(e) 20-70% sulfamic acid
30-80% ethylenediaminetetraacetic acid (EDTA)
(f) chromic acid
All of the above reagents or mixtures can be regener-
ated by the use of suitable cationic exchange resins to remove
the contaminants, and the reagents or mixtures themselves can
be removed from the reactor systems by anionic exchange resins
; 10 when the level of activity of the coolant system has been reduced
sufficiently to warrent such removal. Suitable anionic and
-cationic resins include POWDE ~ anionic resins and AMBERLIT
XE-78 anionic resin and POWDE ~ cationic resins and AMBERLIT
XE-77 cationic resin. Other cationic and anionic resins are
also operative.
It s desirable that the exchange resins be in forms
which tend to restore the original cooIant composition. In
the case of heavy water moderated and cooled reactors a preferred
cationic resin is in D form, and a preferred anionic resin in
OD form.
The method of the present invention will be described
in more detail hereinafter by particular reference to the Examples
and to the accompanying drawings in which:
Fig. 1 is a simplified flow diagram of the path of
the decontaminating reagent through a CANDU-BLW (Canadian Deuterium
Uranium-Boiling Light Water) reactor described in Example l;
Fig. 2 is a graph showing 60Co in South HTS (Heat Trans-
port System) against time;
-- 6 --
;
rm/~
106Z590
Fiq. 3 is a graph showing 54Mn in South HTS against
time;
F~g. 4 is a graph illustrating removal of crud from
South HTS; and
Fig. 5 is a graph illustrating corrosion of steel
during decontamination of North HTS.
Example 1
A shutdown of a CANDU-BLW reactor gave the opportunity
to install additional in-cbre flux detectors and control ab-
sorbers and anchor some steam-separator cyclones which had
become loose in the steam drums. Designers estimated that
installation of the in-core equipment would require 125-400
man-hours of work inside the reactor outlet feeder cabinet where
the radiation field was about 1 Ry/h in July. Modifications
in the steam drums, where the radiation field was about 3 Ry/h,
would require 20-120 man-hours. Decontamination of the steam
drums and outlet feeders was recommended to reduce the ex-
posure of personnel to radiation.
The heat transport system (HTS) was predominantly
carbon steel, and radiation fields resulted from 54Mn and
60Co contamination in deposits consisting of approximately
equal proportions of Fe2O3 and Fe3O4. The distribution of
corrosion products and contaminants in the HTS at the time of
the reactor shutdown was estimated to be as follows:
Crud 60co 54Mn 59Fe
kg Ci Ci Ci
.
Reactor Core 24-44 30-50 60-120 80-350
Steam drums 13 2 0.4 0.3
Outlet feeders 1.5 0.2 - -
- 7 -
106;~590
Chemical decontamination of the reactor by the method
of the present invention using 0.1~ Nutek~ L-106 was the method
chosen for reducing radiation fields. Experiments using contam-
inated materials from the reactor indicated that L-106 could lower
fields by factors of up to 13 without causing excessive corrosion
or damage to components of the system.
The two loops of the HTS, north and south, each received
two separate decontaminations. Each decontamination was applied
in two stages, designed to minimize the spread of radioactive
material from the reactor core to the steam drums. Figure l is
a Simplified diagram of the path of L-106 solutions through
the HTS during the decontaminations. Each of the four decontam-
inations of the HTS was done with the following sequential steps:
a) removal of impurities from the HTS using the ~urification
system,
b) heating the water in the loop to about 85C with the main
pumps 3 followed by drain-down of the water in the elevated
steam drum,
c) injecting L-106 into the reactor inlet header and circulating
it through the reactor core, feeders, and purification system
using the shutdown cooling pumps 4 and the purification booster
pump 5. The ion exchange units 6 were coated with 9/l cation/
anion H-form Powde ~ resin to remove dissolved activity and
regenerated the L-106,
d) refilling the steam drum with water (end of phase 1) and
circulating L-106 through the whole HTS loop and purification
system (9/1 c/a resin) using the main pumps 3.
e) removal of L-106 from the loop using 3/7 cation/anion
H-form Powde ~ resin (end of phase 2).
jc~
1062590
Decontamination of the South HTS
The south HTS was decontaminated on October 2 and 4
by injecting 5~ kg of L-106 into valve 9 each time and following
the procedure outlined above, except that during the second
decontamination an additional 34 kg of ~-106 was injected during
step (d) (phase 2). For L-106 regeneration, the Powde ~ units
were coated with 3 kg/m2 of 9/1 cation/anion resin and were operated
at 20 to 30 ~s for 40 min or longer. For the most part, one
Powdex~ unit at a time was used while the other was being
recoated. Decontaminating solution from the reaction inlet
header (sample post 301~ and from the Powdex~ effluent (post 302
- or 304) was sampled about every 15 minutes and analysed to follow
the progress of the decontamination.
Phase 2 of the decontamination (flow to the steam drums)
was started after several cation-rich Powdex~ units were used to
lower the concen'ration of activity in the HTS to about 20% of its
peak value observed just after L-106 injection. After the regen-
erations were stopped, 2 Powdex~3 units,coated with 3 kg/m2 of
3/7 cation/anion resin operated at about 20Q/s were used to remove
the L-106 and residual radio-elements from the HTS. Phase 1 lasted
for about 10 hours, and Phase 2 for about 7 hours for each decontam-
ination.
Decontamination of the North HTS
Decontaminations of the north HTS were done on October 23
and 26 by injecting 57 kg of L-106 into the reactor inlet header
on each date. The general procedure outlined above was followed for
both decontaminations. The first Powdex~ unit used for regeneration
on October 23 was coated with 3 kg/m2 9/1 (c/a) resin and operated
~c:
106Z590
at 30Q/q for about 20 minutes. Thereafter, all L-106 regeneration
in the north loop was done with the Powde ~ units coated with 1.5
kg/m2 of 9/1 (c/a) resin operated at a nominal 30Q/s for 20 minutes.
During the regenerations, samples of L-106 solution from the reactor
inlet header (post 305) and Powdex~ outlet (302 and 304) were sampled
at about 5 minute intervals.
Results and DisCUSSion
_
General
Partial results of chemical and radiochemical analyses
for 60 Co and 54 Mn, radiation surveys, and corrosion coupon
examinations are given in Figures 2 and 3 and illustrate the effects
of the Powdex~ purification system on the chemistry of the
decontaminating solution.
Adding L-106 to the HTS immediately released some of the
deposited iron and radio-elements to the water. The amount of
material liberated immediately by a given quantity of L-106 appeared
to decrease with successive injections of the chemicals. Injecting
about 50 kg of L-106 into either HTS loop at 80C caused the
release of 7 to 8 Ci 60 Co (160 ~ci/l) after the first injec:ion
and 4 to 5 Ci 60 Co after the second injection. 37 kg (0.05~)of L-106
used for a third injection into the south HTS liberated only about
1 Ci 60 Co on October 5. The contaminated surfaces also released
1 to 5 Ci 54 Mn (20 to 100 ~Ci/l), 4 to 6 mg crud/kg and 200 mg Fe
(total) /kg shortly after each of the first two L-106 injections
into either loop. Starting the main HTS pumps and circulating
L-106 solution through the steam drums caused the release of an
additional 1.4 to 3.5 Ci 60 Co (20 to 50 ~Ci/1),0.6 to 3.2
Ci 54 Mn, and up to 100 mg Fe/kg.
-- 10 --
~c:?,~
- 1062590
Removal of Impurities from the HTS
The quantities of radio-elements and Fe removed from
the HTS were calculated and the results appear in Table 1. The
values given in Table 1 probably represent the minimum quantities
- removed from the HTS, since sampling was not frequent enough to
permit accurate determination of the quantities of iron and
radioactivity removed during short-lived crud bursts.
The decontaminant removed most of the radioactive
deposits from the fuel. At least 70% of the 60 Co (36.8 Ci) and
35~ of the 54 Mn (22.6 Ci) were removed from the estimated
30-50 Ci 60Co and 30-60 Ci 54 Mn resident on the fuel.
Figures 2 and 3 illustrate the effect of the purification
system on radio-elements in the HTS, and Figure 4 shows that the
Powdex~ units retained at least 1/3 of the crud (>0.45~ ) which
entered them. The resin removed up to 3/4 of the particles which
were in the purification feed immediately after the pumps were
started on October 23, indicating that hydraulic shock dislodged
larger particles than did the L-106 alone.
TABLE 1
QUANTITIES OF RADIO-ELEMENTS AND IRON REMOVED FROM THE HTS
SOUTH HTS NORTH HTS
Oct. 2 Oct. 4 Sum Oct. 23 Oct. 26 Sum TOTAL
-
60 Co Ci 10.1 0.1 10.2 14.1 12.5 26.6 36.8
54Mn Ci 2.4 1.8 4.2 7.6 10.8 18.4 22.6
65 Zn Ci 2.5 - 2.5* 3.4 3.1 6.5 9.0
Fe kg 10.5* 15 25.5 27.4 8.4 35.8 61.3
* minimum quantity
-- 11 --
jc:-\2.,
106Z590
Table 2 shows operating conditions for Powdex~) units.
TABLE 2
CONDITIONS FOR 9/1 CATION/ANION POWDEX~
RESIN MIX OPERATED IN 0.03-0.1% NUTEK~ L-106 AT 80C
- Powdex~ Resin Coating Q/sw Q/(s.m ) Loading2
kg/m *
Normal for south HTS 20 to 30 .47 to .85 3
Experiment .05 .75 3
1st coat for north HTS 30 .75 3
Normal for north ~ITS 30 .75 1.5
*each Powdex~) unit has 40.3 m2 of area for resin support
The benefits of L-106 regeneration are estimated from
Table 3, which compares the totals of nuclides removed from the
HTS with the quantities which might have been removed had no
regeneration been made. The latter quantities were calculated from
the concentration increases which resulted from the injection of
L-106 and the pump starts.
TABLE 3
CURIES OF R~DIOELEMENTS REMOVED BY REGENERATION
Oct. 2 1 Oct. 23 Oct. ;~6
60Co 54r~ 60Co 54Mn 60Co ~
Total removed from HTS, A 10.1 2.4 14.1 7.6 12.6 10.8
Amt. released by injection of chemicals 7.4 1.0 7.8 4.8 4.3 2.6
Quantity loosened by pump start 1.7 0.6 1.4 0.8 3.5 3.2
Amt. removable without
regeneration, B 9.1 1.6 9.2 5.6 7.8 5.8
Amt. removed by regeneration, C 1.0 0.8 4.9 2.0 4.9 5.0
(B+c)/B 1.11 1.5 1.53 1.36 1.62 1.86
Regeneration rate, kg NTL-106/min* 0.16 0.32 0.43
* regeneratlon rate - ((kg L-106 in HTS). (purification rate, kg/m).
(no. regeneratlon Powdex coats used). (20 min.))/
((60,000 kg HTS vol.). (duration of decontam.,min.))
j c ~
1062590
There appears to be a positive correlation between the
rate of L-106 regeneration and the extra quantity of radio-
activity removed by regeneration. No correlation exists for
the Mn data alone, possibly because the dissolution rate of
Mn increases with decontamination time.
The decontaminations were terminated by using resin
coats of either 1/2 or 3/7 cation/anion Powde ~ resins, which
removed at least 75% of all impurities entering the purification
system.
Corrosion
The corrosion of HTS materials during the decontam-
ination was monitored with coupons and with a model L-3 Corrosomete ~ -
manufactured by the Magno Corp., Santa Fe Springs, California. Both
pre-oxidized and newly pickled coupons were exposed to the
decontaminating solution in autoclaves in:
1) a south HTS outlet feeder,
2) the inlet to the purification system, and
3) the outlet from the Powdex~.
The dry coupons were weighed before and after the decontamination,
after loose deposit was brushed away, and again after adherent
deposit was removed with Clarke's solution. The average loss of
thickness (uniform penetration) of each coupon was calculated
from the weight of metal lost. The Corrosomete ~, indicated
differential corrosion rates by changes in electrical resistance
through corroding metal "probes". The Corrosomete ~ was located
in the HTS shutdown cooling circuit and was fed from the outlet
of the shutdown cooling pump 7. Table 4 summari~es the information
obtained from the coupons, and Figure 5 presents the Corrosometer~
results for October 23-24.
j c ~
1062590
.
.
TABLE 4
CORROSION OF COUPO~S DETER~IINED AFTER t~ASHING
SOUM HTS NO~ TS
Auto- Oct. 2 C3ct. 4 Oct.2 & 4 Oct. 23 6 26
~ clave l~m ~m/dl~m llm/d llmllm/d llm l~m/d
Powdex ~ Inlet
M~ld steel " 2.7 4.14.9 13.47~6 7.79.2 35.3
" " 2.7 4.25.1 14.1 to to*8.6 32.9
- 5.514.98.2 8.28.0 30.6
403--SS " 0.120.18 ------------------ ------ -- --
410--SS 0.010.01 ------------------ ------ ------ ------
Powde~ Outlet 3
Mild steel " 8.212.7 2.36.410.1 10.317.3 66.4
" " 7.712.0 2.46.7 to to*16.1 61.8
" " - --- 2.57.010.6 10.815.1 58.0
Outlet Feeder
Mild steel " 7.8 5.6
" " 7.7 5.6
" . " 7.8 5.6
~ " 7.0 5.1
" " 5.6 4.0
" " _7.6 5.4
410-SS " 9.8 7.0
~ " g.O 6.5
" " 9.0 6.5
2 1/4 Cr, 1 ~lo " 6.7 4.8
" " 6.4 4.6
.
~Unlform pcnctrntion assumcd
*Rfltc bflscd on rnn~c of pcnctrntlon in (0.646 ~ 0.365)d of flow
throu~h autoclnvcs 2 and 3 on Oct. 2 and 4.
-- 14 --
rm/_lc
106Z590
TEST RESULTS
The decontamination lowered radiation fields on most
HTS equipment and saved radiation exposure during subsequent
maintenance.
- Regeneration of the L-106 increased the removal of
60 Co by about 60%. The L-106 reacted rapidly with 1 Fe3O4/
1 Fe2O3 corrosion products and continuous regeneration was
essential for efficient decontamination. Corrosion of carbon
steel was greatest in the purification outlet (17.6ym, 67.6 ym/d),
but the average uniform penetration for all carbon steel surfaces
was 3.6 ym. Materials other than carbon steel were unaffected.
Removal of most of the fuel deposits should significantly lower
the rate at which radiation fields increase in the future.
The following examples illustrate bench scale tests
. . .
showing the efficacy of alternative reagent compositions in the
practice of the present invention.
EXAMPLE 2
.
In a series of tests in autoclaves, the decontamination
potential of ~ilute solutions of each of ethylenediaminetetraacetic
acid, sulfurous acid, sulfuric acid, and phosphoric acids was
measured for the removal of 60 Co from oxide on Zircaloy- ~,
carbon steel, Inconel-600~, and Monel 400~. Decontamination factors
for these alloys when treated by the different solutions at tempera-
tures ranging between 50 to 150C were:
Zircaloy-4~ 1.2 - 25
Carbon Steel 2 - 100
Inconel-600~ 1.3 - 4
Monel-400~ 1.2 - 6
- 15 -
jc:~ ~
106Z590
EXAMPLE 3
.
An in-pile recirculating loop has been decontaminated
at different times by dilute phosphoric acid; by ethylenediamine-
tetraacetic acid plus oxalic acid and citric acid; by ethylene-
diaminetetraacetic acid plus hydroxylamine; by sulfuric acid
plus hydroxylamine; and by sulfuric acid plus oxalic acid. In
all the decontaminations, the active ingredients were regenera-
ted on cation resins. The loop is constructed of carbon steel
and stainless steel.
Decontamination factors for irradiated corrosion
products ranged between 1.5 - 3.5 for the two steels.
~0
- 16 -
rm/J~'
