ML19322C291
| ML19322C291 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 06/30/1974 |
| From: | Boswell J, Evans A, Larry Jones US ATOMIC ENERGY COMMISSION (AEC) |
| To: | |
| References | |
| TASK-TF, TASK-TMR DP-1355, UC-80, NUDOCS 8001160810 | |
| Download: ML19322C291 (8) | |
Text
r p
DP-1365
- J uc-80
('
s
- t 9
' d' t I C00]FDDJEMG0JT OF AGRBORPdE RAD 00 ACTIVITY PROSOEGO REPORT: JULY - DECEMBER 1973 by f A. G. Evans L. R. Jones ,g b 'l-Approved by i J. fl. !!osuoli, Research !?.anogor Rocctor Engincoring Division L. n Publication Dato: June 1974 Esn nossce This report was pr reted as sa a:cosat of watt specserad by the Unit J 51 stas Govaramar.t. Neither the Uc!!ad States por the Unitsd St&Its Atomic Enerdy 1 Commluloa, set any of their emp!cytss, not any of their contrr,tters, subcontractors, er insir employees, saskas say warranty, ez;rtss er implied, or er.uses say J s' traal 1:stCity er respons:bl: sy for tse accura.y. seat. $1staatse or usefutacts of any leformistloa, apparatus, predact et pruersa disclosed, et teptatants that its use wovid act infrtNt ritweety owned rights. 800II6pEto
y /
SUMMARY
'I Confinement system studies to determine basic gas-phase reaction mechanisms of iodine with activated carbon indicate that potassium plays a significant role in high-temperature iodine retention. Low-potassium charcoals (such as coal anil petroleum-base carbons) retain iodine less effectiyely than coconut-shell charcoal, which has a high natural K content. impregnated) coconut-shell carbons Iodized (KI, I2, or KI-12 tend to perform more poorly when the atom ratio of iodine to potassium (I/K) exceeds a value of approximately 0.4. High-temperature (180 C) iodine penetration of some commercial iodized charcoals exceeds 1% when the I/K ratio exceeds 1.0. Carbons of high pH also tend to retain iodine better than low pH carbons; thus, the basic reaction mechanism in iodine retention is probably to an I~ complex. the conversion of I2 2 loading followed The high-temperature desorption test (I by 4 hours desorption at 180 C) can be used as a quality. control performance test for commercial iodized carbons. Small HEPA filters,7 designed for 10 to 15 cfm air flow, were irradiated by an ~3 x 10 -rad /hr "Co source before their per-formance characteristics were measured in a new test facility-containing a " Teflon"*-stainless steel moisture separator upstream of the test filter. Accident conditions were simulated by brief exposure of the filter to mixtures of steam and air at several times filter design flow capacity followed by ~5-hour exposure to steam and air at decaying temperatures; flow was limited by pres-sure drop - across the. test filter. Similar test filters,- fabri-cated:from filter media with 7 to 48 months service in the con-finement system, were exposed to the same test conditions. Although measured pluggage with moisture of. test filters made from- . service-aged media was:more severe than pluggage of irradiated filters potential reduction of confinement system air flow from filter pluggagi during an accident _would not prevent adequate removal.of'radioiodine decay heat from carbon adsorbers in the system if at_least 3 to 5 filter compartments were'on line and radioiodine adsorption was: dis-- tributed evenly among the-on-line compartments. More severe test-conditions were -required to rupture a. test ' filter than could be postulated during any Savannah River reactor accident.,' a _J
DISCUSSION CARB0ft TESTIflG High-Tercperature Desorption Tests The high-temperature desorption test was initially designed as a screening test to aid in selecting potential carbon types suitabic for use in the confinement system.' The test conditions (10-minute loading of elemental iodine at ambient temperature and humidity followed by 4 hours desorption at 180 C) were selected to prevent damage to the " Teflon"-coated test apparatus and " Neoprene"* "0"- rings used as seals in the apparatus. The 180 C temperature is also the upper limit of usefulness for TEDA-im-pregnated carbons because TEDA boils at 174 C and flashes at ~190 C.12 In the original test series,10 13 of the 21 candidate ad-sorbers were subjected to the high-temperature test and had penetration values ranging from 0.003% to 18.08% (Table I). The values fall into three broad penetration catagories: (1) very low (<0.010%), which were coconut charcoals either unimpregnated or impregnated with TEDA or TEDA + KI; (2) intermediate (<0.10%), which were coconut carbons impregnated with I2, KIs or PbI ; and 2 (3) very high (>1.0%), uhich were some coconut carbons impregnated with KIa and petroleum-base carbons with KI3 or TEDA impregnation. 2 The anomaly in the data is that four of the 1500-m /g coconut-shell carbons impregnated with KI3 use the same source of base carbon, yet the penetration values range from 0.023% to 18.08% (the ex-tremes are two different lots from the same vendor). Preliminary investigation of the cause of differences in carbons indicated that different the performance of the KI 3 methods of impregnation were used by the three vendors;10 however, confirmation of the techniques used could not be obtained because of the proprietary nature of the processes. Variation in the iodine content of the different carbons is also a possible cause (particularly because the greatest performance variation was observed between different lots from a single vendor). The sodium, potassium, and iodine content of each sample was determined by neutron cctivation analysis. The results of these analyses indicated a Registered tradename of E. I. du Pont de Nemours and Company. t
7~ ^ ~ ~~ ~ ~ " ~ ~ ~~ - - - - ~. w ETABLE 1 High-Tegerature Test Data" Surface Arek, Iodino Penatration,b Hfg Base Carbon as /g Inpregnant 5 2 -k 1\\- ,f 6 A Coconus 1100 None 0.004 p A' Coconut 1500 KI 0.052 B Coconist-1100 None 0.004 8 Cocenut 1500 K!s 0.028 8 Coconut 1100 KI 0.056 B Coconut 1100 la 0.070 8 Coconut 1500 Pbla 0.004 D Cost 1300 Pbt 0.420 D Cocenut 1100 KI + TEDA 0.006 B Coconut 1100 KI + TEDA 0.005 B Coconut 1100 TEDA 0.003 C. Cocenut 1500 K!: 18.00 C Cocenut 1500 KI: 0.023 D Coconut 1500 KIs 2.412 E Petroleum 1500 XIs 6.484 E Petroleum 1100 TEDA 12.03 E Petroleum 1500 TEDA 4.560. a. Source of data is Reference 10, except where noted. b. See text for test conditions. c. New data. L:. 10 -
strong correlation beti:cen the atea ratio of'icdine to potassium cnd iodino penetration during the high-temperaturo tcst. Tha sodium c ntcnt of tho itpragnated carbons was c:nsistantly <0.2 wt t End scs not related to penetration. Chcalcal Analysis of Carbons Rosults of the neutron activation analyses of soveral of tho KI and KI3 carbons are shoun in Tabic II. Iodino and potas-siu:a contents cro compared with the iodine penetration values frca Tablo I and uith pII values for water extracts of cach carbon (5 g carbon entracted with 20 al distilled unter; pH censured after 10 minuto contact tico). The comparison botueen I/K ratics cnd iodino penetration is shoun graphically in Figuro 1. TE!.E !! Ccep3rl:en of Ch:-! cal Anslysis with p!! and Iodine Penetration loEno," Potani:c,# A::n Intio Iciinc Fenete:ticn,' y Carbcn uts ut3 I/K ph 5 A 3.70 1.75 0.665 0.36 0.052 0 D-1 3.02 1.55 0.601 9.62 0.023 I 0-2 1.31 1.04 0.336 9.E6 0.006 0 3;r <0.01 2.09 <1.5 x 10 9.93 0.003 b C-l 4.01 0.05 1.26 7.05 IS.0S C-2' 4.23 2.23 0.536 0.02 0.023 I II 4.42 1.14 1.19 9.56 2.412 E 3.12 0.96 1.00 0.78 6.48 a. Activation analysis of undried sacple. b. See text for cethod. c. Ilich-te perature test. J. KI: on 1500 u'/g coconut carbon, canufacturer A. a c. K!: on 1500 m /g coconut carbon, manufacturer D. f. K! + TEDA on 1100-m'/g coconut carbon, canufseturer D. go Unicpregnated 1500-u'/g base carbon of the same type used to prepare sazples D-1, C-1. C-2, and II. h; K13 on 1500 n'/g coconut carben, canufseturer C - first lot tested. i, K13 on 1500-n'/g coconut carben, canufacturer C - second tot tested. J. KI: on 1500 n'/g coconut corben, canufacturer D.
~ r ggg ;, i j l O Coconut Corben o Petreacum Ccikon i; ~ 10 + o ~ g
- -[
' 0 f R q .hI! o.i :- J 0.01 e / u i I I I I I I 0.0 01O O.2 0.4 0.6 0.0 1.0 1.2 1.4 Atom Ratio, I/K FIGURE 1. Effect of I/K Ratio on liigh-Temperature Iodine Penetration Figure 1 indicates a logarithmic increase in high-temperature iodino penetration with a linear increase 'in the I/K ratio (above a threshold of ~0.35) and suggests that potassium plays a signifi-cent role in iodine retention mechanisms. This probably explains why coconut-shell charcoals (which are rich in K+) are more offective for iodino removal than coal or petroleum-base charcoals. D:ccuse variability in performance between different manufacturing -19ts of KI -impregnated carbons could be caused by variations -in 3 K ccntent of bas 6 carbons as well as differences in impregnation tcchnique, sampics of base carbons from 10 different manufacturing 1 cts from the same vendor (covering a 5 year period) were analyzed far potassium content (again by neutron activation analysis). Re-Es_ulto are summarized in Table III.
,0,,ti i g, Ch:=1 cal Cent:nt cf Cccenut-shall Daso Carbons 47rc:Maca Dwfac2 Arca# e//g pl/2 g, oc 3 f.'[, c3' S 13C0 10.0 0.09 0.13 1100 0.0 1.10 0.51 1100 0.0 0.C5 0.05 1100 10.1 0.70 0.10 1100 10.1 0.00 0.07 1500 0.0 1.13 0.11 1500 10.0 1.17 0.07 1500 0.0 1.10 0.03 1000 0.0 1.30 0.00 1500 10.3 1.27 0.05 a. Sco text for de::criptica of othod. Tho data in Tchio III in:iicato that thoro is littlo vari-cbility in the baso carbon, particularly the 1500-o /g carbon 2 fron which the cc:rerical KIs carbons cro canufactured.
- Thus, the variability in performnco (and the I/K ratio) is probably a function of itpregnation techniquo.
The best carbons cro those with a pII greater than 9 and which havo I/K ratios loss than 0.7. To validato theno conclusions, a series of four iodino solutions ucre prepned with reagent grado Ia, K0ll, cnd KI. Three solutions ucre prepared uith a constant In content (6.2 grams in 100 al unter) and a varying KGil concentration. Tho f urth solution was made by dissolving 10.5 g KI in 100 al unter. ThD first two solutions (4.6 g K0li with I/K = 0.47 and 5.8 g K011 with I/K = 0.60) woro predictably clear and coloricas (oven cfter standing in fluorescent light for 3 uceks in a scaled polyethyleno bottlo) because of the absence of tho I -cocplex or 3 frco la in colution. The pli of both colutions uns 12.3. The third solution (2.9 g KOII with I/K = 0.95) was a very dark brotm
color (charactoristic of 13 solutions) uhich stained the plastic .bottlo on standing and had c p!! of 8.9. The fourth solution (KI uith I/K = 1.0, p11 a 7.3) uns initicily clear end colorless, but gradually turned a faint broun color on standing in tho light. The colutica behavior cf the iodino stroagly suggests a sinilar reaction on the carbon surfaco uhcro soco la cay bo liberated on heating cnd accotetts for pooror perfortanco of tho hi@ I/K ratio corbens. Tho high p!! solutions favor for- ~ pII carbons (such as coconut carbon) should favor In to I,High-ration of I~coplexos uhich aro relativoly nonvolatilo., convor- . alon, uhorcas lou p!! ccrbons (ncutralized coconut or coal and patrl51cun carbons) should fasfor the formation of loss stablo 13 co plexes, particularly uhen tho I/K ratio oncceds 1. Several car 7 cs of nou and servico-aged carbons uoro testod 1 for pil and high-tcrporaturo iodino penotration to datoraino if any corrolation existed botwcon residual alkalinity (availablo excess K') and iodino rotontion. The results are shown in Tchio IV. A good corrointion between p!! and iodino penetration is not choun by tho data oncept for the general trend toward highor panotration with lower pII (Figuro 2). 17hilo the I/K ratio in service-aged carbons does not chango significantly uith tico, the pli does change uith tho accuculaticn of acidic attospheric pollulants (such as N02 and S0 ). In ad-2 dition, other otrospheric pollutants (such as hydrocarbons) accumulating on the service-aged carbon are corpoting uith iodino for reactivo sorption sites (or cocpounds in the carbon uhich uill react with I2). Thus, the rapid decrease in perforcanco (Figure 3) appears independent of carbon type, and an icprognated carbon (G-615, KI + TEDA) deteriorates as rapidly in service as an unimpregnated carbon (416) whop avaluated by the high-tem-perature desorption test. Use of the High-Temperature Desorption Test i The high-temperature desorption ~ test has been a valuabic cid in selecting an impregnated carbon for use in the Savannah River confinement system. The test should prove equally valuable for other investigators interested in screening potential con-tainment system carbons. In addition, the test can be used as a quality control test by carbon manufacturers. -Because the I/K rctio correlations are still in the development stage, direct measurement of the carbon performance is the preferred method of ovaluation of different lots of carbon.. The high-temperature tast has been proposed for inclusion in the ASTM and ANSI standard}}