Cleanup of Demineralizer Resins - ACS Publications - American

14 Downloaded via UNIV OF CALIFORNIA SANTA BARBARA on July 13, 2018 at 03:04:29 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

Cleanup of Demineralizer Resins 1

1

1

2

3

W. D. Bond , L. J. King , J. B. Knauer , K. J. Hofstetter , and J. D. Thompson 1

2

3

Oak Ridge National Laboratory, Oak Ridge, TN 37831 G P U Nuclear Corporation, Middletown, PA 17057 E G & G Idaho, Inc./TMI, Middletown, PA 17057

Radiocesium is being removed from demineralizers A and Β (DA and DB) by a process that was developed from labora tory tests on small samples of resin from the demineralizers. The process was designed to elute the radiocesium from the demineralizer resins and then to resorb i t onto the zeolite ion exchangers contained in the Submerged Demineralizer System (SDS). It was also required to limit the maximum cesium activities in the resin eluates (SDS feeds) so that the radiation field surrounding the pipelines would not be excessive. The process consisted of 17 stages of batch elution. In the i n i t i a l stage, the resin was contacted with 0.18 M boric acid. Subsequent stages subjected the resin to increasing concentrations of sodium in NaH BO -H BO solution (total boron = 0.35 M) and then 1 M sodium hydroxide in the final stages. Results on the performance of the process in the cleanup of the demineralizers at TMI-2 are compared with those obtained from laboratory tests with small samples of the DA and DB resins. To date, 15 stages of batch elution have been completed on the demineralizers at TMI-2, which resulted in the removal of about 750 Ci of radiocesium from DA and about 3300 Ci from DB. 2

3

3

3

As a consequence of the accident at the Three Mile Island Nuclear Power Station, Unit 2 (TMI-2) on March 28, 1979, the two demineral izers (DA and DB) in the water makeup and purification system were severely contaminated with fission product radionuclides. The resin beds in the demineralizers were significantly degraded both radiolytically and thermally by the decay of most of the radionuclides during the period since the accident. The principal gamma-emitting radionuclides remaining on the resin beds when demineralizer cleanup activities began were due to the relatively long-lived C s (ti/2 • 30.1 y) and *Cs ( t ^ " · Inception of the present investigation, nondestructive assay (NDA) methods had been employed to estimate the cesium activity and material content of 1 3 7

131

2 e 0 6

P

r

l

o

r

t

o

t

h

e

0097-6156/ 86/ 0293-O250$06.00/ 0 © 1986 American Chemical Society

Toth et al.; The Three Mile Island Accident ACS Symposium Series; American Chemical Society: Washington, DC, 1986.

B O N D ET

14.

Cleanup of Demineralizer Resins

AL.

251

each demineralizer (Table I) (1). Comparison of the post-accident with pre-accident resin bed voTumes Indicated that the beds had undergone severe shrinkage (~55%) and s i g n i f i c a n t degradation had occurred. It was known that the beds had not only been subjected to high radiation dosages (~10 rad) but had also been exposed to high temperatures because of the radioactive decay heat. The necessity to i s o l a t e the demineralizers from l i q u i d flow about 19 h after the accident prevented e f f e c t i v e removal of the decay heat, and e s t i mates indicate that center-line bed temperatures may have been as high as about 540°C (1000°F). The demineralizers were sampled by GPU Nuclear personnel i n early 1983, and i t was observed that the DA vessel contained only dry, caked, r e s i n , whereas l i q u i d was s t i l l present i n DB. However, the caked bed (DA) was apparently deagglomerated after water addition and sparging so that resin samples were obtained i n a l a t e r sampling e f f o r t . The absence of l i q u i d i n the DA vessel i s contrary to the NDA estimate of 3 f t ^ (85 L ) . 9

Table I.

Estimated Demineralizer Vessel Loadings Based on NDA C h a r a c t e r i z a t i o n s 3

Post-accident Loadings Resin Volume, f t Weight, lb l ? C s , Ci l * C s , Ci

3

3

3i

Liquid Volume, f t Weight, l b

3

Pre-accident^

DA

DB

50 2,139 0 0

22 1,025 3,500 270

22 1,025 7,000 540

44 2,746

3 193

3 193

Debris U, l b Core debris, lb l ? C s , Ci * *Cs, Ci

5 95 177 16 21 106RU, Ci 28 i ^ C e , Ci 116 125sb, Ci 0.5 TRU, Ci D a t a obtained from M. K. Mahaffey et a l . (1). DA and DB vessel loadings are i d e n t i c a l . A l p h a a c t i v i t y only. 3

3i

a

b

C

1 19 35 3 4 5 23 0.1

c

c

Conceptual studies of the various alternative methods for cleanup of the demineralizers (2) Indicated that the most desirable method was to elute the cesium and subsequently sorb i t on the z e o l i t e s i n the Submerged Demineralizer System (SDS) C3-j>). This concept a l l e v i a t e d the high-level radiation problems associated with eventual removal of the degraded resin beds and with management of the r e s i n wastes. Since the effects of the r e s i n bed degradation were unknown with regard to the quantitative elution behavior of cesium and the quality of the eluates, this concept required


THE THREE MILE ISLAND ACCIDENT

252

experimental determination of i t s f e a s i b i l i t y and the development of a s a t i s f a c t o r y chemical flowsheet* A program was established i n early 1983 involving the collaborative e f f o r t s of Oak Ridge National Laboratory (ORNL), General Public U t i l i t i e s Nuclear Corporation (GPU Nuclear), and EG&G/Idaho, Inc./TMI (EG&G) to: (1) establish the technical f e a s i b i l i t y of cesium e l u t i o n , (2) develop a chemical flowsheet that would meet processing requirements for the SDS at TMI-2, and (3) provide for cleanup of the demineralizers. ORNL was responsible for establishing the f e a s i b i l i t y of cesium elution and for developing the chemical flowsheet using demineralizer samples provided by GPU Nuclear. GPU Nuclear, with some assistance from EG&G, was responsible for i n s t a l l a t i o n of the process at TMI-2 and subsequent cleanup of the demineralizers. Several requirements had to be met i n the development of a s a t i s f a c t o r y process flowsheet for cesium elution and f i x a t i o n . In e l u t i o n , i t was necessary that chemical reagents be employed that were compatible with the ionic solution chemistry of SDS feed solut i o n s . This requirement dictated that elution of the cesium be accomplished by displacement with sodium ions using sodium borate or sodium hydroxide solutions. The elution process was also required to l i m i t the cesium a c t i v i t y i n eluates to levels that were no greater than about 1 mCi/mL to avoid excessive radiation f i e l d s i n subsequent SDS operations. Even with compatible ionic solution chemistry, i t was by no means certain that eluates could be s a t i s f a c t o r i l y processed i n the SDS. Cesium loading of the SDS z e o l i t e beds might be seriously Impaired by the presence of degradation products i n eluates, depending on their quantity and nature. Finely dispersed solids (or c o l l o i d s ) from resin p a r t i c l e breakage and degradation might not be readily separable and therefore could cause plugging of the zeolite bed. F i n a l l y , soluble organic compounds and/or emulsified o i l s might sorb on zeolites and foul or block the exchange sites for cesium. Laboratory Development Studies Laboratory-scale experiments on the development of the process flowsheet Included: (1) batch elution tests to determine the f e a s i b i l i t y of cesium elution by sodium ion displacement; (2) tests with small zeolite beds to evaluate the f e a s i b i l i t y of s a t i s f a c t o r i l y processing eluates i n the SDS; and (3) studies of eluate c l a r i f i c a t i o n by s e t t l i n g and by f i l t r a t i o n through sintered-metal f i l t e r s . The results of these experiments provided the technical basis for the process flowsheet for the cleanup of the demineralizers. The DB sample was available about 1 year before a s a t i s f a c t o r y DA sample was obtained; therefore, most of the elution flowsheet conditions were developed using the DB r e s i n . Description and Analysis of Demineralizer Samples. Samples from the DA and DB vessels consisted of resin mixed with l i q u i d . The DA sample contained about ~20 mL of resin and ~50 mL of l i q u i d , and the DB sample contained about 40 mL of resin and 80 mL of l i q u i d . The l i q u i d s were separated from the resins by s e t t l i n g and décantation; most of the resin settled very rapidly. The separated resin was then dried i n a i r at ambient conditions (25 to 30°C). In some


14.

BOND ET AL.

253


cases, the separated l i q u i d was further c l a r i f i e d by centrifugation before analysis. The resin and l i q u i d were analyzed by chemical and radiochemical methods. The resin was also examined by v i s u a l microscopy at ~20X magnification to assess i t s physical charac teristics. Radionuclide and chemical analyses of the resin and l i q u i d phases of the DA and DB samples are given i n Table II (only the p r i n c i p a l constituents are l i s t e d ) . Small quantities (10 to 1000 ppm) of many other metal cations were also present i n the resin, but they are not relevant to the work reported here. Table I I .

Radionuclide and Chemical Analyses of the Demineralizer Samples DA

Analyses Radionuclides, pCi/g 137 133.48 E5

1.50

E2

90sr 5.11

C s

E5

137

Cumulative DF

2-mL mixed z e o l i t e bed (60 vol % Ionsiv 9640 vol % Linde A-51); residence time » 8.4 min

Cumulative Decontamination Factors for i n Zeolite Bed Tests

Conditions:

Table V.

η δ m H

α

r >

m

m m

Χ 70

m H

s

14.

B O N D ET

Table VI.


AL.

261

Results of C l a r i f i c a t i o n Tests with Sintered-Metal Disk F i l t e r s using DA Liquid

Conditions:

9.5-mm-diam f i l t e r disk, f i l t r a t i o n rate « 2.2 mL/cm -min; DA l i q u i d diluted 20:1 with 0.18 M H3BO3 2

Filter Final Run rating Turbidity Time Volume pressure drop No. (min) (mL) (NTUs) (psi) (ym) 1 0.5 2.1 19 30 20 2 2 0.5 2.3 16 25 3 10 170 14 2.2 270 «Turbidity of feed was 12.6 NTUs. ^Feed supply was exhausted. Extrapolation of the pressure-drop-vstime curve indicated that 20 p s i would be reached after 190 min and a f i l t r a t e volume of ~300 mL. a

K

b

S e t t l i n g tests (Table VII) showed that c l a r i f i c a t i o n could be s i g n i f i c a n t l y improved by using longer s e t t l i n g times. Therefore, r e l a t i v e l y long s e t t l i n g times after batch contacts of resin and eluents could be employed to reduce f i l t e r loadings i n eluate c l a r i f i c a t i o n s and to increase l i q u i d throughputs before excessive f i l t e r plugging occurred. Relatively long s e t t l i n g times (24 to 72 h) were used i n the processing of the eluates at TMI-2 (discussed later). Table VII. Effect of Settling Time on DA Liquid Turbidity S e t t l i n g Time (d) 0 1 2 3 4

Turbidity (NTUs) 12.6 6.1 5.5 4.6 4.0

An attempt was made to increase the s e t t l i n g rate by using fresh anion exchange resin (Amberlite IRA-400; 20-40 mesh) as a f l o c c u l a t i o n aid i n one test; however, i t did not s i g n i f i c a n t l y improve the rate unless p r o h i b i t i v e l y large quantities were added. In that experiment, 1-g additions of resin were made each day to the same 35-mL aliquot of l i q u i d u n t i l a t o t a l of 4 g had been accumul a t e d . The resulting t u r b i d i t i e s of the l i q u i d after s e t t l i n g for 1 d i n the presence of 1, 2, 3, and 4 g of the resin were 5.2, 4.4, 3.5, and 2.8 NTUs, respectively. Process Flowsheet Description and Process Operations at TMI-2 The schematic process flowsheet i s shown i n Figure 4. After the e x i s t i n g l i q u i d i n the demineralizers had been removed, cesium was eluted i n multistage batch contacts of the resin with the eluent solutions. As elution proceeded, the sodium concentration of the



V E N T TO WASTE GAS SYSTEM

TIE-IN POINT

, EXh

J —

Q-

55 GAL DRUM

-krj

INEUTRALIZER I TANK

Γ

J

i

MU-F-5A I

-CX3-

RADIATION DETECTOR-

PROCESSED WATER

BACKFLUSH A B L E FILTER

NITROGEN •

PROCESSED WATER A N D CHEM. ADD.

PROCESSED WATER

Figure 4. Flowsheet f o r E l u t i o n of the TMI-2 Makeup and P u r i f i c a t i o n Demineralizers.

H— -•--MXI-

B A C K F L U S H TO OPPOSITE DEMINERALIZER

14.

B O N D ET

AL.


263

eluent solution was increased from 0 to 1 M (0 to 23,000 ppm sodium) to f a c i l i t a t e elution of the residual cesium from the r e s i n , A l i q u i d / s o l i d phase volumetric r a t i o of ~1.5 was used i n each stage and i s governed by the available free volume i n the demineralizer vessels. Batch contacts of the l i q u i d eluent solutions with the r e s i n was accomplished by upflowing the l i q u i d into the demineral i z e r and then a i r sparging the l i q u i d above the resin bed to promote mixing for 8 h. The a i r sparging was then stopped, and 24 to 72 h was allowed for the suspended solids to s e t t l e . Typically, three to four drums (55 gal each) of eluent solution were added per batch contact. Eluates were withdrawn through a suction hose from about 1 f t below the l i q u i d surface to prevent the withdrawal of any s o l i d s which might be f l o a t i n g on the l i q u i d . The removal was accomplished by using either an eductor or a pump to l i f t the l i q u i d . The eluates were then f i l t e r e d through a sintered, stainless steel f i l t e r f r i t (pore s i z e , 20-ym rating) before being d i l u t e d with processed water (800 ppm boron), transferred to i n plant neutrallzer tank storage, and f i n a l l y processed through the e x i s t i n g SDS (3-^)· The f i n a l f i l t r a t i o n polish of the eluates avoids the unnecessary introduction of fine particulate matter to the SDS and subsequent water management systems. Eluate d i l u t i o n was necessary to minimize personnel exposure during transfer through piping to in-plant tanks. Most of the eluates were transferred from the demineralizer using the eductor that dilutes the eluates (up to 20:1) before f i l t r a t i o n . Flow t o t a l i z e r s were used to measure volume of l i q u i d transfer, and the flow of eluate from the demineralizer continued u n t i l the pump or eductor loses suction, leaving a considerable residual volume (V ) of eluate heel i n the demineralizer. The values of V remaining i n the DA and DB vessels after transfers were 145 and 175 g a l , respectively. Samples from the neutrallzer tank were analyzed after each batch e l u t i o n . Routine analyses Include determination of C s by gamma spectrometry, S r by beta-counting, and sodium by atomic absorption. The progress of the batch elution was monitored from the determination of the concentration and volumes i n the neutral l z e r tank before and after transfer of batch eluates. r

r

1 3 7

9 0

Processing Results. After 15 stages of batch elution, approximately 750 and 3300 Ci of C s have been eluted from DA and DB, respect i v e l y (Table VIII). Process operations have been generally s a t i s factory, but some d i f f i c u l t y was encountered due to frequent f i l t e r plugging when eluting DB with sodium concentrations greater than about 0.25 M (8000 ppm). F i l t e r plugging was minimized and s a t i s factory f i l t r a t i o n rates were obtained by increasing the s e t t l i n g time of the eluates from the nominal 24 h to 72 h and by the addition of boric acid to reduce solution a l k a l i n i t y . It i s believed that f i l t e r plugging was due to degraded resin product part i c l e s that were more e f f e c t i v e l y dispersed at the highly alkaline conditions; this phenomenon was not observed i n the DA e l u t i o n . Results from the processing of the f i l t e r e d eluates through the SDS system have been completely s a t i s f a c t o r y . Based on the NDA estimates of the * C s demineralizer loadings i n Table I, about 22 and 46% of the C s have been eluted from DA l 3 7

3 7

1 3 7



264

Table VIII. Cesium Removals Accomplished after 15 Stages of Batch Elution of the TMI-2 Makeup and P u r i f i c a t i o n Demineralizers Demineralizer A 1 3 7

Cumulative Cs removal Stage No. (Ci)

Demineralizer Β 1 3 7

Sodium cone. of eluent (M)

Cumulative Ce removal (Ci)

a

0 119 1 0 201 2 0.07 206 3 0.34 395 4 464 0 5 0.23 499 6 532 0 7 0.23 562 8 0.74 608 9 0.38 630 10 0 646 11 0.38 706 12 0.90 711 13 1.0 14 745 1.0 745 15 Proceesed water containing 0.18 Μ Η Β 0 · Removal of f i l t e r backflush l i q u i d . a

b

b

b

a

3

b

Sodium cone. of eluent (M) a

0 0 0.07 0.07 0.34 0.34 0 0.30 0.30 0.23 0.30 0.51 0.51 0.59 0.59

811 932 1190 1510 1940 2000 2180 2420 2640 2840 2920 3070 3170 3240 3260

a

b

3

1 3 7

and DB, respectively (Figure 5). The removal of C s was about one-half of that expected on the basis of the h o t - c e l l tests at ORNL, i f i t i s assumed that the samples tested were representative of the bulk resin from each demineralizer resin bed. The l i q u i d r e s i n contact was d e f i n i t e l y better i n the laboratory h o t - c e l l tests because the resin was s l u r r i e d with the l i q u i d v i a shaking. Nevertheless, the trends i n C s elution behavior with increasing sodium concentration were expected to be the same. The in-plant data from the DB resin elution show the expected trend; however, the DA data do not, p a r t i c u l a r l y at the higher sodium concentrations approaching 1 M. The results to date indicate that l i t t l e addi tional C s can be removed by continuing the DA elutions with 1 M sodium concentrations. Thus, i t appears that either the o r i g i n a l cesium loading estimates for DA (Table I) are high or the resin sample used i n the h o t - c e l l elution tests i s not representative of DA resin bed. Resin sampling procedures with the demineralizers using quartz f i b e r o p t i c a l examination of the DA r e s i n bed showed that i t was dry, which i s contrary to the wet-resin-bed model used to interpret the NDA data. The p o s s i b i l i t y that the o r i g i n a l estimate of the C s loading on DA i s high i s currently being explored i n c a l c u l a tions using a revised source term model to interpret the NDA data. No attempts at further elution of DA are planned, pending the out come of the reassessment of the C s loading estimate and radiation l e v e l surveys of the DA v e s s e l . However, elutions w i l l continue for DB at higher sodium concentrations to further reduce the cesium a c t i v i t y . Following completion of the cesium elution operations, 1 3 7

1 3 7

1 3 7

1 3 7



BOND ET AL.

25

• ι ' ι ' I I DEMINERALIZER A 1

20 k 15h

O CUMULATIVE • INSTANTANEOUS

•Α. -ι

I

ι

I

'

I

'

I

I

Λ

I r-

DEMINERALIZER Β

η - 0-Ô-9

2en 40 UJ

ο

30 h

p-o'

20 10 1

5 7 9 11 BATCH NUMBER

13

15

1 3 7

Figure 5. Percentage C s Eluted i n the Multistage Batch Elution of the TMI-2 Demineralizers.


266


dose-rate surveys of the demineralizer cubicles w i l l be performed to confirm the f i n a l resin e l u t i o n e f f i c i e n c i e s and the dose rate reductions that were achieved. Acknowledgments The authors wish to express their sincere appreciation to several individuals at ORNL and GPU Nuclear for valuable assistance i n this work. Gratitude i s expressed for the excellent services provided by the ORNL A n a l y t i c a l Chemistry Division personnel under the d i r e c tions of J . A. Carter, J . H. Cooper, D. A. Costanzo, and J . M. Peele. Special thanks are due the ORNL Metals and Ceramics D i v i s i o n personnel under the supervision of R. Lines, who provided the photomicrographs of the demineralizer r e s i n s . Sincere appreciation i s also acknowledged to GPU personnel under the d i r e c t i o n of B. G. Smith for capably performing the demineralizer cleanup operations at TMI-2 and the associated a n a l y t i c a l work.

Literature Cited 1. M. K. Mahaffey et al. "Resin and Debris Removal System Conceptual Design--Three Mile Island Nuclear Station Unit 2 Makeup and Purification Demineralizer," Hanford Engineering Development Laboratory Report HEDL-7335, Richland, Washington, May 1983. 2. E. J. Renkey and W. W. Jenkins, "Planning Study of Resin and Debris Removal Study, Three Mile Island Nuclear Power Station Unit 2 Make-up and Purification Demineralizers," Hanford Engineering Development Laboratory Report HEDL-7377, Richland, Washington, June 1983. 3. D. O. Campbell, E. D. Collins, L. J. King, and J. B. Knauer, Evaluation of the Submerged Demineralizer System (SDS) Flowsheet for Decontamination of High-Activity-Level Water at the Three Mile Island Unit-2 Nuclear Power Station, Oak Ridge National Laboratory Report ORNL/TM-7448, Oak Ridge, Tennessee, July 1980. 4. D. O. Campbell, E. D. Collins, L. J. King, and J. B. Knauer, "Process Improvement Studies for the Submerged Demineralizer System (SDS) at the Three Mile Island Nuclear Power Station, Unit 2," Oak Ridge National Laboratory Report ORNL/TM-7756, Oak Ridge, Tennessee, May 1982. 5. L. J. King, D. O. Campbell, E. D. Collins, J. B. Knauer, and R. M. Wallace, "Evaluation of Zeolite Mixtures for Decontamination of High-Activity-Level Water in the Submerged Demineralizer System (SDS) Flowsheet at the Three Mile Island Nuclear Power Station, Unit 2," in Proceedings of the Sixth Internatinal Zeolite Conference, Reno, Nevada July 10-15, 1983, D. Olson and A. Bisio (Eds.), Butterworths, 1984, pp. 660-68. 6. K. J. Hofstetter, C. G. Hitz, T. D. Lookabill, and S. J. Eichfield, "Submerged Demineralizer System Design, Operation, and Results," Proc. Topl. Mtg. Decontamination of Nuclear Facilities, Niagara Falls, Canada, September 19-22, 1982, Vol. 2, pp. 5-81, Canadian Nuclear Association (1982). RECEIVED July 16, 1985


Cleanup of Demineralizer Resins - ACS Publications - American

Recommend Documents