ML19260C557

Westinghouse Nuclear Fuel Div,Pu Fuels Development Lab,Pu Oxide Shipment Rept Per 10CFR71.61

ML19260C557
Person / Time
Site:	07109009
Issue date:	12/05/1979
From:	WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML19260C556	List: ML19260C555 ML19260C557
References
14978, NUDOCS 8001070500
Download: ML19260C557 (45)

Text

{{#Wiki_filter:. 'dESTINGH00SE NUCLEAR FUEL DIVISION PLUTONIUM FUELS DEVELOPMENT LABORATORY PLUTONIUM 0XIDE SHIPMENT REPORT PER 10 CFR 71.61 CECEMBER 5,1979 1694 048 4=====* fCO 8001070 1

TABLE OF CONTENTS Section Title Pace I DETAILS OF SHIPMENT 1 II METHOD OF PACKAGING 2 III RECEIPT OF MATERIAL AT SRP 4 A. Unloading Procedure 4 8. Unloacing Schedule 4 C. Conditions Observed by SRP Personnel 6 IV INVESTIGATION AT SRP 7 A. Chemical Analyses 7 B. Atmosphere Analysis 8 C. Evaluation of the Inner Canister Weld 10 V INVESTIG4 TION AT WESTINGHOUSE 12 A. Verification of Decay Heat Loads 13 B. Thermal Analysis of Shipping Containers 14 C. Mechanical Testing of Food Cans 18 D. Thermal Exposure of Mock-Up Food Cans 19 E. Internal Pressurization Tests 21 F. External Pressurization /Depressurization in a 24 Pressure Chamber G. Tests on Food Can Coatings and Gasket Materials 26 H. Interviews with Personnel 28 VI CONCLUSION 30 VII RECOMMENDATIONS 31 )b9k 11

LIST OF SKETCHES, FIGURES, AND TABLES Number Ti tle Pace Sketch 1 Cross Section of the FL-10-1 Shipping Container 32 Figure 1 Inner Canister Loading Pattern for FL-10-1 S/N RLO88 33 Figure 2 Inner Canister Loading Pattern for FL-10-1 S/N RLO67 34 Figure 3 Inner Canister Loading Pattern for FL-10-1 S/N RLO61 35 Figure 4 Inner Canister Loading Pattern for FL-10-1 S/N RLO70 36 Figure 5 Permanent Deformation Versus Internal Pressure for 37 Food Cans Table 1 Contents of FL-10-1 Shipping Containers 38 Table 2 Cbservations by SRP Personnel of the Condition of the 39 Food Cans and Their Contents Table 3 Results of Chemical Analyses of the Pu02 Powder Shipped 40 in FL-10-1 Number RLO70 Table 4 Original Versus Verification Decay Heat Loads for Each 41 Shipping Container Table 5 Calculated Temperature Distribution Within FL-10-1 42 Shipping Containers Table 6 Sunnary of Internal Pressurization Tests 43 Table 7 Description of Specimen Cans for Pressurization / 44 Depressurization Tests Table 8 Summary of Visual Observations of Specimens After 45 Removal From the Pressure Chamber iii 1694 050

I. OETAILS OF SHIPMENT A. Type of Material: 1 lot of conmercial LWR Pu02 powder contained in 3 shipping containers, and 1 lot of ARHCO-produced virgin Pu0 2 powder contained in 1 shipping container. B. Total Quantity Transferre.d: 11,966 grams of plutonium contained in 14.027 grams of Pu0

• 2 C.

Transferred To: DuPont Savannah River Plant (SRP). D. Transfer Dates: Shipped from PFDL October 10, 1979. Received at SRP October 11, 1979. E. Type of Shipping Container: FL-10-1 (see Sketch No. 1, attached). F. Total Number of Shipping Containers: 4. G. Regulatory Requirements: NRC Certificate of Compliance No. 9009, Revision 3; 10 CFR Part 71. H. Opening Instructions: Prior to packaging of the Pu0, a SRP 2 representative visited PFDL to review packaging techniques in order that personnel at SRP could develop opening procedures. 1694 051 II. METHOD OF PACKAGING The pug 2 p wder was packaged per PFDL Operating Procedure No. PFDL-0P-0212. The sequence of operations was: A. Transfer the Pu02 p wder from vault storage to a low humidity glove box. B. Transfer the Pu02 p wder fr m the storage container to paint cans. Detemine the net weight for each. C. Apply a tamper safe seal to the paint can. D. Bag the paint can out of the glove box and enclose in a second layer of plastic. E. Place the double-bagged paint can in a crimp-sealed food can. F. Load the powder-filled food cans into an inner canister with a bottem end plate welded in place using empty food cans and aluminum spacers to complete the stack. G. Weld the top end plate onto the inner canister. H. Flush the inner canister with helium through the purge holes in each end plate. I. Plug the purge holes with filler wire and seal weld. 1694 052 -?-

J. Transfer the inner canister into the pressure chamber of a FL-10-1 shipping container. X. Leak test the inner canister using a helium mass spectrometer. L. Pressurize the FL-10-1 pressure chamber with helium and verify the integrity of the 0-ring seal with a sniffer probe connected to a helium mass spectrometer. M. Vent the FL-10-1 pressure chamber to atmospheric pressure and complete the assembly of the shipping container. N. Tamper safe the FL-10-1 shipping container and hold in storage for the shipment. O. Transfer the FL-10-1 shipping container to the transport vehicle and secure for the shipment. Table 1 and Figures 1 through 4 describe the contents and loading patterns of each FL-10-1 shipping container. )b II:. RECEIPT OF MATERIAL AT SRP The shipping containers were received at SRP on October 11, 1979. Shipping Containers No. RLO88, RLO67, and RLO61 were unloaded on October 15 and 16,1979. Shipping Container No. RLO70 was unloaded on October 25, during which time SRP personnel experienced problems. The following information relative to the unloading operations was discussed at a November 16, 1979, meeting between SRP and Westinghouse personnel. A. Unloading Procedure 1. Unload the inner canister from the FL-10-1 shipping container and transfer to a " hut" (plastic tent). 2. Place the inner canister in a horizontal position in a cradle-type fixture. 3. Using a pipe cutter, cut through the well of the inner canister. 4. Remove the food cans from the inner canister and transfer them to storage. B. Unloading Schedule 1. The inner canisters from Shipping Containers RLO88, RLO67, and RLC61 were unloaded per the above procedure without any problems. ~4-1694 054

2. While proceeding through Step A.3, above, with the inner canister from Shipping Container RLO70, the following series of events occurred: a. After ccmpleting the cut through the inner canister, the pipe section fell to the floor. The cut was made approximately 3" from the bottom. b. Three empty food cans in the bottcm of the inner canister " ejected" and fell to the floor. c. Personnel observed a " puff of smoke." d. The continuous air monitor outside the hut alarmed. e. Personnel outside the hut evacuated the room. f. Contamination was detected in the hut and the room. g. Special procedures were developed to decontaminate the area and complete the unloading operations, h. Cleanup and unloading operations were completed by October 31, 1979. 1694 055 C. Conditions Observed by SRP Personnel The observations made by SRP personnel after the food cans were removed from the canister are summarized in Table 2. Also included in this table are their observations of the condition of the plastic bags and paint cans after opening the food cans. 1694 056

IV. INVESTIGATION AT SRP SRP personnel have conducted an investigation of the problems that were encountered during the unloading operation of FL-10-1 Number RLO70. A "Special Hazards Report" has been issued by SRP in draft fann and will be transmitted to the NRC when it becomes available. This report will be fonnalized after review by v*rious SRP committees and is not expected to be issued until early 1980. Data related to the shipping package part of the problem have been extracted from the report and discussed below. The discussion includes: Chemical analyses of the Pu02 p wder Detennination of atmospheres found in the plastic bags Evaluation of the canister wel :s. A. Chemical Analyses 1. Purcose The purposes of the chemical analyses were: a. To determine the composition of the received Pu02 p wder for a comparison with the as-shipped Westinghouse data. b. To detennine any differences in the compositions of powders from breached and intact containers. )(30 k YJ7

2. Method Standard analytical techniques were used to develop these data. 3. Results/ Discussion The data are sumarized in Table 3 and show that there is no significant difference between the powder that was shipped (Westinghouse data) and the powder that was received (SRP data). In addition, there is no difference between the powder from Product Can #3 (breached) and Product Can #2 (intact). 4. Conclusions These data are discussed in Section V.B. B. Atmosobere Analysis 1. Purpose The purpose of the atmosphere analyses was to determine the composition of the gases. found inside the plastic bags in Product Cans #1 and #2 and one of the swollen empty spacer cans. 2. Method Gas chromatography was used for a qualitative identification of the gases while mass spectrcmetry was used for the final analyses. All samples were diluted with an unknown quantity of air because of the sampling technique. 1694 058

3. Results/ Discussion a. The gas chromatographic analysis of the sample for Product Can #2 showed approximately 4 volume percent of a light gas, probably hydrogen or helium, and the balance as air. The sample was consumed prior to identification of this light gas, b. The mass spectrometric analysis of the sample from Product Can #1 showed approximately 4 volume percent as heliuni with the balance as air. c. The sample from the empty food can was found to be air only by gas chromatography. 4. Conclusion Helium leaked into the food cans and permeated through the PVC bags during the shipment. It leaked from the empty spacer food can between the time of opening the canister and sampling the atmosphere. The plastic bags prevented the rapid diffusion out of the product food cans. The bloating indicated a higher than atmospheric pressure inside the plastic bags. 1694 059 _g.

C. Evaluation of the Inner Canister Weld 1. Purpose The purpose of evaluating the integrity of the welds of the inner canister was to determine the possibility of a leak path through a weld. 2. Method Standard dye penetrant tests were used to identify suspect areas which were subsequently examined at 20X magnification followed by metallographic sectioning. For one sample the metallographic technique was to cut the sample % 20 mils from the dye penetrant indication. The sample was then mounted and polished incrementally until the defect was observed. Incremental polishing continued until solid metal was again observed. Measurements taken during polishing were used to quantify the size of the defect. The second sample was ground until the defect was observed. One photograph was taken of each defect. 3. Results/ Discussion a. Positive dye penetrant indications were found at two locations s 120 apart in the weld at the top end of the inner canister, i.e., the weld made after loading the canister. 1694 060

b. The first defect was measured to be N 6 mils in diameter. The second defect is estimated to be about the same size as the first. 4. Conclusion Positive leak paths from the outside to the inside of the welded inner canister were confirmed. -n_ 1694 061

V. INVESTIGATION AT WESTINGHOUSE Upon notification of the problems encountered during the unloading of Shipping Container RLO70 at SRP, various actions were initiated by Westinghouse. These actions included: Verification of the decay heat load calculations Thermal analysis of the FL-10-1 shipping container Interviews with personnel involved in the packaging of the shipping containers Compilation of pertinent historical information for the Pu0 2 powder Mechanical tests on food cans Internal pressurization tests on cans Thernal exposure tests on cans Pressure chamber tests on cans' materials Identification and thermal stability tests of food can coating and gasket materials. 1694 062 _12

The objective of these actions was to gain insight into the cause of the problem and provide explanation for the various ostensibly con-flicting observations made at SRP, e.g., crushed spacer can, bulged product and spacer cans, and one normal can. A. Verification of Decay Heat Loads 1. Purcose The purpose of verifying the decay heat load calculations for each shipping container was to determine if a gross error was made in these calculations at the time of shipment. 2. Me thod Services of an independent organization were enlisted to per-fom the verification calculations. Using the isotopic analysis for each lot of Pu0 shipped, the energy released 2 during the radioactive decay of each isotope was calculated. From these values the heat released per unit weight of Pu0 was 2 obtained and the total decay heat load for each shipping con-tainer was calculated. These values were then compared to similar calculations performed at the time of shipment. 3. Results/ Discussion Results of the original and the verification calculations of decay heat loads for each shipping container are compared in 1694 063

Table 4. The original calculations were confirmed within 3.8 to 5.3%. The differences are attributed to differences in the values of energy release associated with the radioactive decay reactions for each isotope. 4. Conclusions It is concluded that no gross errors were made in the original decay heat load calculations. The differences noted above are considered insignificant for purposes of heat transfer calculations. B. Thermal Analysis of Shiocing Containers 1. purpose The purpose cf this analysis was to determine the equilibrium temperature distribution within the shipping containers. This information was then used to evaluate the relative importance of various postulated sources of pressure buildup within Shipping Container RLO70. 2. Me thod Since the paint can is placed within two plastic bags and, subsequently, within a food can, it was assumed that the primary made of heat transfer between the two cans was conduc-tien through the gaseous medium. Convective heat transfer was 1694 064 9 assumed from the food can to and from the inner canister, the FL-10-1 pressure chamber, and the carbon steel liner for the insulating foam. Conduction through the foam was the assumed heat transfer mode. Two cases were examined: a. Case 1, which is representative of Shipping Containers RLO88, RLO67, and RLO61, assumes four cans per shipping container, each containing 800 grams of Pu0 with a heat 2 generation rate of 6.75 watts per can. b. Case 2, which is representative of Shipping Centainer RLO70, assures 5 cans per shipping container with a heat generation rate of 2.55 watts per can. 3. Resul ts / Discussion Key temperature values of interest for each case are summarized in Table 5. The condition of the PVC bags as observed at the time of unloading RLO70 at SRP tends to corroborate the reported temperature distribution for that container (see Case 2 in Table 5). PVC is reported to degrade at temperatures greater than 100*C as evidenced by the release of hcl and an 1694 065

EI) accompanying darkening in color No such darkening in color was observed for the PVC bags of RLO70 at SRP, indicating the temperature of the bags was rrnst likely 100 C or less. Re fer-ring to Table 5, the temperature of the PVC bags was analyzed to be between 80 and 108*C. The maximum temperature of lil*C for the centerline value for the Pu0 -filled product cans is significant. If it is assumed 2 that the temperature and pressure of the atmosphere within the product cans at the time of loading was 25*C and 0 psig (14.7 psia), respectively, then it can be shown that the pressure will increase to 4.4 psig (19.1 psia) with an attendant increase of temperature to 111 C. Results of internal pressuri-zation tests on paint cans and food cans (see Section V.E) indicate this pressure is well belcw that required to cause permanent can deformation (end bulging). Nessure buildup in the product cans as a result of release of [olatile components of the Pu02 p wder was also evaluated at lll*C. The analytical results presented in Table 3 indicate loss-on-ignition (LOI), moisture, and carbon values which are considered typical for powder produced at ARHC0 (Richland, Washington). The two most likely volatile species which might be released are moisture and carbon dioxide, the latter as an EI3" Rupture of Plutonium 0xide Storage Container, March 13, 1979," Page 86, Battelle Pacific Northwest Laboratories, May 29, 1979. -is-1694 066

oxalate decomposition product or as an absorbed gas. If it is assumed that a sufficient quantity of moisture in the Pu0 is 2 in a fom such that it can be liberated at 111 C, the maximum partial pressure attainable would be the equilibrium vapor pressure at lil*C, which is % 21 psia. However, the effects of the temperature gradient which exists within the inner canister outward, must be considered since condensation commensurate with equilibrium considerations will occur along the gradient, tnus reducing the pressure increase due to moisture. The likelihood of C0 being released at 111 C is nil as, 2 evidenced by the fact that none was detected in the atmosphere sample taken from within the PVC bags of the product cans (see Section IV.8). 4. Conclusions Based on the themal analysis reported in this section and results of tests reported in other sections of this report, it is concluded that: a. Observations concerning the condition of the PVC are con-sistent with the temperature distribution reported for the shipping containers in Table 5. b. Pressure buildup due to themal effects alone is not sufficient to cause pemanent deformation (end bulging) of cans. -'7-1694 067

c. Pressure buildup due to liberation of volatiles from within the Pu0 at 111 C is not sufficient to cause permanent 2 deformation (end bulging) of cans. C. Mechanical Testing of Food Cans 1. Purpo e The purpose of this test was to determine the deformation behavior of food cans subjected to compressive end loading and to provide specimens for comparison with the crushed empty spacer can observed in RLO70 at SRP. 2. Method Four empty food cans with mechanically crimp-sealed lids, identified as Numbers 1 through 4, were end loaded in compres-sion using a "MTS" testing machine. Loading rate was controlled by controlling cross-head speed at 5 and 10 mils /second. Graphs of load versus deformation were recorded and photos were taken at selected intervals to document the mode of deformation. 3. Results/ Discussion Can #1 was loaded at a cross-head speed of 5 mils /second. The load built up to a maximum of 1,040 lbs. without any visibly detectable deformation of the can. The load then dropped dramatically, and deformation was observed at one set of the 1694 068

stiffening ribs on the can. Deformation at the ribs continued with further loading until both sets were folded like a bellows, at which time the smooth surfaces of the can began to crush and buckle. Can #2 was loaded at a cross-head speed of 10 mils /second and behaved essentially the same as Can #1 except the maximum load was 1,130 lbs. Cans #3 anc #4 were stacked and loaded at a cross-head speed of 10 mils /second. The maximum load was 1,065 lbs. Deformation occurred at the ribs of the bottom can and continued until this can sti.ffened sufficiently to initiate deformation at the ribs of the top can. Conclusions Comparing the cans crushed during these tests with photos of the crushed empty spacer can of RLO70, it was concluded that the empty spacer can was not crushed by compressive end loading. D. Thernal Excosure of Mock-Uo Food Cans 1. puroose The purpose of these tests was to determine the long-term (s 4 weeks) effects of thermal exposure at 100 C on mock-up food cans designed to simulate the product cans in RLO70. 1694 069

4 2. Method Two crimp-sealed food cans, both with coated inside surfaces, were placed in an oven at 100 C. Can #1 contained an empty paint can which was double-bagged in dielectrically sealed polyvinylchloride (PVC) plastic with the outer bag ends taped down with duct tape. Can #2 was identical except that the paint can was filled with sand. Oven temperature was monitored with a digital pyrometer and controlled at 100 +5'C. The time of exposure was 28 days for Can #1 and 30 days for Can #2. 3. Results/Di scussion No end bulging was observed for either can. Upon opening the cans, the coated inside surfaces were essentially unaffected except for the lids, which were in contact with the duct tape. The black backing of the tape had decomposed to a molten, tacky residue. The PVC bags were discolored, ranging from a trans-parent, reddish-brown tinge in some areas to a dark, opaque reddish-brown color in others. The outer zinc-coated surfaces of the paint cans were observed to be blackened in areas which were in contact with the dark opaque PVC. This is attributed to attack by hcl, which is reported to be liberated frem PVC when exposed to temperatures greater than % 100*C (see Section V.8). 1694 070 4. Conclusions a. It was concluded that a thermal exposure to 100 C for 4 weeks will not cause permanent deformation of the food cans, including contributions to internal pressure buildup from thermal expansion of gases and from volatilization of the PVC bags, the duct *. ape, and the coating on the interior surfaces of food cans. b. It was also concluded that the PVC bags in Shipping Container RLO70 did not experience temperatures of 100 C. This conclusion was based on observations made at SRP and the results of this test. E. Internal Pressurization Tests 1. Purpose The purpose of these tests was to determine the mechanical response of food cans and paint cans to interncl gaseous pressurization. 1694 071 2. Method Food cans and paint cans were modified by drilling a hole in the side of each can and installing a bulkhead connector pres-sure fitting. The lids were then installed on the cans and the cans were connected to a regulated high pressure bottled helium supply so that the internal volume could be pressurized. The mechanical response of the cans was then observed as a function of internal pressure. A cursory leak test of the lid seals using " snoop" solution was performed to determine the degree of leak tightness of the sealed cans. 3. Results/ Discussion a. Food Cans No permanent deformation was observed until a threshold pressure of 30 psig was reached. Subsequent cans were pressurized up to 70 psig in 10 psig increments. All deformation took place in the ends of the cans in the form of bulging or doming. The amount of bulging / doming, as expected, was detennined to be proportional to the internal pressure as determined by measuring the overall length of each can after pressurization (see Figure 5). It was also observed that as the amount of bulging increased, the number of folds on the ends increased to accommodate the bulging. Table 6 is a tabulation of the number of folds observed on the ends of each can. None of the food can lids were blown off during these tests. 1694 072 b. paint Cans Two paint cans were also internally pressurized. The bottom end of one can bulged at 29 psig and the lid blew off at 30 psig; while the bottom end of the second can bulged at 26 psig and the lid blew off at 34 psig. c. Leak Test of Food Cans and Paint Cans Results of the bubble leak test performed on both paint cans and food cans indicate that the lids leak profusely at pressures as low as 6 psig. Tests at lower pressures were not performed because of the insensitivity of the pressure regulator being used. 4. Conclusions a. A threshold differential pressure of 30 psig is required to cause permanent deformation of food cans from the inside out. b. Based on the number of folds observed on the ends of the product and spacer cans of Shipping Container RLO70, the pressure within the FL-10-1 pressure chamber, inner canister, and product and spacer cans was on the order of 50-70 psig at the time of opening at SRP. Paint can and food can lid seals leak profusely at p[qj{)tps Q[} c. as low as 6 psig and, most likely, lower. F. External Pressurization /Deoressurization Tests in a Pressure Chamber 1. Purpose The purpose of this test was to determine the response of simulated product and spacer cans to a cycle of gradually increasing the external pressure, holding at pressure for a period of time, followed by a sudden venting of the external oressure to ambient. This test was aimed at simulating conditions which the inner canister with the defective weld (Shipping Container RLO70) was subjected to during pressurization at Westinghouse and unloading at SRP. 2. Method A total of twelve specimen cans were made up to simukate the product cans and spacer cans contained within Shipping Container RLO70. A description of each can is provided in Table 7. The cans were then loaded into a pressure chamber which was to simulate the inner canister of RLO70. The pres-sure chamber was sealed and connected to a regulated helium supply bottle. The pressure within the chamber was then gradually increased to a pressure of 60 psig at an average rate of 1 psig/ minute, held at 60 psig for 100 minutes, and rapidly vented to O psig within s 6 seconds. The cans were then removed from the chamber and visually examined. This pressuri-zation/depressurization cycle was intended to simulate a small 1694 074

leak in an inner canister during the 150 psig pressurization step at Westinghouse, and the sudden venting which would have occurred at SRP when the end of the inner canister was cut off. 3. Results/ Discussion Table S sunnarizes the visual observations made on the twelve specimen cans following the pressurization /depressurization cycle. It should be noted that all conditions for the food cans of Shipping Container RLO70 observed at SRP were also observed herein. These include crushed cans, unaffected or normal cans, and deformed cans with bulged or domed ends. Whether a can crushes, bulges, or remains normal is dependent upon the leakage rate across the can seal. Cans with good seals will crush, cans with medium integrity seals will bulge at the ends, and cans with poor seals will be unaffected. It was also observed that none of the PVC bags were ballooned. This is not surprising since the mode of pressure buildup within the PVC bags is diffusion-controlled and times much longer than 100 minutes would be required for appreciable permeation to occur. Apparently, the elapsed time of three weeks between loading of the FL-10-1 at Westinghouse and unloading at SRP was sufficient to allow significant permeation of the PVC bags with helium. The results of Section IV.B confirm this supposition. - 1694 073

In all cases where the food can was opened, the paint can inside the PVC bags was observed to exhibit flattened areas on the circumference and all lids were found to be intact. This is not surprising since the PVC bags act as a flexible, collapsible seal around the paint can. Upon pressurization due to leakage into the food can, the paint can will begin to collapse. Similar behavior was observed for paint cans removed from product cans at SRP for Shipping Container RLO70. 4. Conclusions It is concluded that che pressurization /depressurization mechanism is a viable explanation for the observed behavior of the contents of Shipping Container RLO70 during the unloading at SRP. G. Tests on Food Can Coatings and Gasket Materials 1. Purcose The purpose of these tests was to identify the materials used to coat une inside surfaces and to gasket the lids of food cans and to determine the thermal stability of same. 2. Method Samples of the coating ana gasket materials were analyzed by infrared spectroscopy to determine their identity. Samples were also subjected to thermogravimetric analysis (TGA) wherein 1694 076 he weight change of the sample is recorded as a function of temperature to determine the temperature of instability. 3. Resul ts/ Discussion The coating on the interior of the can was identified as a polyisobutylacrylate polymer which is an isobutyl ester type derivative of acrylic acid. The gasket was identified as a styrene-butadiene copolymer heavily filled with China clay, It also contained approximately 20% of an ester polymer. Results of the TGA indicate both materials are stable up to 250*C as evidenced by the fact that no weight change was observed for samples heated to that temperature. Above that temperature, however, rapid weight changes were observed and at approximately 600*C the bulk of the organic material had been consumed. 4. Conclusions Based on the results of this section and on the results of the thermal analyses (Section V.8), it was concluded that pressure buildup from the food can coating and gasket materials was not sufficient to cause permanent deformation (end bulging) of cans. 1694 077

H. Interviews with Personnel 1. Purpose The purpose of the operator interviews was to determine any unique occurrences associated with the packaging of the four containers. 2. Resul ts/ Discussion Two items resulting from the discussions with the operators were considered to be significant differences between the packaging of RLO70 and that of the other shipoing containers. a. Leak testing of the canister (Section II.K) required longer than normal periods of tide for the evacuation in addition to multiple openings of the leak test chamber either to check the 0-rings or to replace them. b. Pressurization of the FL-10-1 (Section II.L) to test the 0-ring seal was repeated several times due to equipment probl ens. 3. Conclusions a. The helium was removed from the inner canister during a series of evacuation-backfill with air cycles that occurred. 1694 078 b. The inner canister was exposed to high pressure helium for a time sufficient to allow internal pressurization through the weld defects. 1694 079 4 VI. C0tlCLUSION As a result of the investigations conducted by Westinghouse and SRP detailed in previous sections, themal effects, food can coatings, and fuel volatiles were eliminated as the cause of the problem. Helium used to leak test the RLO70 shipping container seals entered the inner canister through two weld defects resulting in a pressurization of the contents. When the end cap was removed, the inner canister vented rapidly, the food cans did not, thus creating a differential pressure across the food cans. This caused the food cans to swell. 1694 080 VII. RECOMMENDATIONS A. Amendment to Certificate of Comoliance 9009 Add a dye penetrant test of all inner container welds after completion of the leak test specified for the inner container under Item 9 of the Certificate of Compliance. B. Opening Instructions Westinghouse will recommend the following special instructions to the consignee: 1. Pricr to opening the FL-10-1 pressure chamber, it should be vented through the valve provided on the top flange. 2. Prior to opening the inner container, it should be vented. 3. When performing the steps described in 1 and 2, above, normal health physics procedures must be applied for control of potential radioactive material release. 1694 081 \\ 's N \\ \\f'-- r--, \\ l Two 55-gallon drums 3"' ~ h\\ Stacked and Welded a e ~ Double 0-Ring Aluminum Flange Seal i Spacers s' j l Inner _ Canister N ' s Pressure Chamber Food Cans Either Empty as Spacers or Filled with Pu02 Powder Fire Resistant Foam ~ s \\ \\ \\- \\ \\ SKETCH 1 CROSS SECTION OF THE FL-10-1 SHIPPING CONTAINER 1694 082

FIGURE 1 INilER CANISTER LOADING PATTERil FL-10-1 Serial Number: RLO88 Inner Canister Nunber: 3 Total Weight Pu0 : 3,200 Grams 2 Total Weight Pu: 2,736 Grams (s-Total Decay lleat Load: 28.5 Watts g=-- =g -2 3 Item Description Contents Identi fication 4 1 Inner Canister Assembly 10 Food Cans + Al Spacers

1. 3 2

Aluminum Spacer As Required N/A ,g -S 3 Food Can Enpty N/A --6 4 Food Can 800.0 Grams Pu02 Product Can #4 5 Food Can Empty fl/A 6 Food Can 800.0 Grams pug 2 Product Can #2 -8 7 food Can Empty N/A 8 Food Can 800.0 Grams Pu02 Product Can #3 9 Food Can Enoty N/A -10 10 Food Can 800.0 Gra u02 Product Can #1 11 Food Can Empty fl/A 12 Food Can Empty N/A 12 Dottom

FIGURE 2 IllilER CAtllSTER LOADitlG PATTERil FL-10-1 Serial flumL r: RLO67 Inner Car.irter Nuibi:r: 4 Total Weiglit Pu0 : 3,300 Grams 2 Total Weight Pu: 2,1331 Grams g \\.. Top 9* % D Total Decay lleat Loau: 29.2 Watts ss ss g i__ . -J -2 W, z ..- - 3 Item Description Conten ts Iden ti fica tion 4 1 Inner Canister Asseir.bly 10 Food Cans + Al Spacers

1. 4 2

Aluminum Spc.cer ' As Required fl/A <g 1 5 3 Food Can Empty ti/A E ' ~ 4 Food Can 800.0 Grams Pu02 Product Can #8 5 Food Can Empty fl/A ~ _,.7 6 Food Can 800.0 Grams Pu02 Product Can #6 8 7 Food Can Empty il/A 8 Food Can 900.0 Grams Pu02 Product Can #7 ~~ ~ 9 9 Food Can Empty fl/A 10 10 Food Can 800.0 Grams Pu02 Product Can #5 11 Food Can Onpty fl/A CN 11 12 Food Can Empty N/A 4 . - 12 00 s s Bottom

FIGURE 3 IllflER CAfilSTER LOADIllG PATTERfl FL-10-1 Serial fluniber: RLO61 Inner Canister flunber: 5 Total Welulit pug :. 2,823 Grams 2 Total Weight Pu: 2,414 Grams fSs Total Decay lleat Load: 25.1 Watts s 2 {:. __ q 3 I tem Ibscription Contents Identification 4 1 Inner Canister Assenhly 10 Food Cans + Al Spacers

1. 5 2

Aluminum Spacer As Required fl/A g1 5 3 Food Can Enoty il/A 6 4 Food Can 423.2 Grams Pu02 Product Can #12 5 Food Can Empty N/A 7 6 Food Can 800.0 Grams Pu02 Product Can #11 - -8 7 Food Can Empty il/A 8 Food Can 800.0 Grams Pu02 Product Can #10 e 9 -p= 9 Food Can Empty fl/A CD , 10 10 Food Can 800.0 Grams Pu02 Product Can #9 CD W 11 Food Can Empty fl/A 11 12 Food Can Enoty fl/A _ - 12 "g C9 0

• ]Dg ] Y hl 3

Bot tom

FIGURE 4 IllflER CAtilSTER LOADIllG PATTERil FL-10-1 Serial fiumber: RLO70 Inner Canister flunber: 6 Total Weight Pu0 : 4,704 Grams 2 Total Weight Pu: 3,995 Grams b Total Decay lleat Load: 10.5 llatts cc m -2 3 Item Description Contents Identi fica tion 4 1 Inner Canister Asserbly 10 Food Cans + Al Spacers

1. 6 2

Aluminum Spacer As Required fl/A j 's 5 3 Food Can Empty fl/A _ -6 4 Food Can Empty fl/A 5 Food Can 324.7 Grams Pu02 Product Can #5 . -7 6 Food Can 1,066.6 Grans Pu02 Product Can #4 _ -8 7 Food Can 1,104.2 Grams Pu02 Product Can #3 8 Food Can 1,102.5 Grams Pu02 Product Can #2 . -9 9 Food Can 1,105.9 Grams Pu02 Product Can #1 _ - 10 10 food Can Empty fl/A m 11 Food Can Enp ty fl/A e p ,j j 12 Food Can Enpty ti/A cD Q - 12 Os Bottom

%Rngge M I **. f.".7 ~~ Z2 U "" " ~ 7M"j f**

• ~

~ * * * * * ' _ ",,2* _ i ,4- ~ - j;**:."_O.~.**t--- - _ --1.d ;-- t-- ! 9 -- .1 .T ___ + --t---. m-; .. - -. - ~ _p_

• • ~ !*. *l""._;_.

T 4.__ --' n-2 --, _ { -- ~_._-{_--_+- "~ ] ;.= ,,, g.,; t - -- t----- ---l-a -.c b h / _M --r. W r 1 ^ - - - ' 4 99N$ 6 ,' h .1 i s . f_, D e-4 es.e N N r i _ g' r- + s,l Me 6 ~f-+- ~-- ,p - A

~

'I p-- w ~7 o. h e n, ~~~~ 1 Ma "f @1 6& 4i Wi 3 _Z'.'- J = 8 v i 4 a 1 ~

• ,,e Qi$

e 3 2 +

• g e

M* 8 W ene t I ge = __ 6 3m-i._ I ~2 MC .b so S % EDb ". -5M$5%.-,% G 3 6 l

TABLE 1 CONTENTS OF FL-10-1 SHIPPING CONTAINERS FL-10-1 Serial Number RLO88 RLO67 RLO61 RLO70 Inner Canister Serial Nutrber 3 4 5 6 Total Weight Pu02 (Grams) 3,200 3,300 2,823 4,704 Total Weight Pu (Grams) 2,736 2,831 2,414 3,995 Total Decay Heat (Watts) 28.5 29.2 25.1 10.5 . 1694 088

TABLE 2 OBSERVAT10ftS BY SRP PERSOfillEl. OF lilE C0flDITI0fl 0F Tile FOOD CNIS AtlD TiiEIR C0flTEllTS Food Can Position Food Can from Bottom of I den ti fica tior. (c.i Gn Ccndition As Removed Canister or Contents from Canister Condition of_ Food Can Contents 1 _Enp ty jwollen at Both Ends II/A t 2 Empty Swollen at Both Ends _fi/A 3 Empty Swollen at Both Ends il/A 4 Product Can #1 Swollen at Both Ends Plastic bags intact but ballooned after removal of lid. Paint can opened and contents spilled. S Product Can #2 Swollen at Both Ends Plastic bags intact but ballooned af ter removal of lid. Paint can opened and contents 1 sp1]1ed. 6 Product Can #3 Breached at One End Plastic bags stretched and ruptured. Paint can opened and c.ontents spilled. 7 Product Can #4 Swollen at Both Ends flot yet opened. 8 Product Can #5 flonnal - flo Apprent Deformation __ flot yet opened. 9 Engty Swollen at Both Ends _fl/A 10 Empty Crushed il/A l T[w.A N 1UR{L

• 10
• ]O %

g e

TABLE 3 RESULTS OF CHEMICAL ANALYSES OF THE Pu02 POWDER SHIPPED IN FL-10-1 NUMBER RLO70 SRP Data Westinghouse Data Analysis Product Can #3 Product Can #2 Lot Average Pu Assay (w/o) 86.77 84.94 85.40 Pu-238* (w/o) 0.023 Pu-239* (w/o) 90.98 90.98 90.959 Pu-240* (w/o) 8.33 8.34 8.337 Pu-241* (w/o) 0.61 0.60 0.605 Pu-242*(w/o) 0.08 0.08 0.076 I Am-241* (w/o) 0.12 0.10 Am-241 (w/o) 0.15 l Metallic Impurities Clean Clean Clean l l t Carbon (ppm) 975 1,280 954 loss-On-Igni tion w/o @ 200*C 1.10 0.80 w/o @ 700'C 2.71 2.48 Mois ture w/o @ 450*C 1.43

• 0n a Pu-metal basis. All other data are on an oxide basis.

~# ~ 1694 090

e TABLE 4 ORIGINAL VERSUS VERIFICATION DECAY HEAT LOADS FOR EACH SHIPPING CONTAIflER Decay Heat Load (Watts) Shipping Percent Container No. Oriainal l Verification Di fference RLO88 28.5 27.0 5.3 RLO67 29.2 27.8 4.8 RLO61 25.1 23.8 5.2 RLO70 10.5 10.9 3.8 e 1694 091 TABLE 5 CALCULATED TEMPERATURE DISTRIBUTION WITHIN FL-10-1 SHIPPING CONTAINERS Temoerature (*C) Location Case 1* Case 2** Pu0 Centerline 231 111 2 Paint Can Surface (Containing Pu0 ) 223 108 2 Food Can Surface (Containing Pu0 ) 148 80 2 Spacer Food Can Surface (Empty) 117 68 Inner Surface of Foam 68 45 Outer Surface of Drum 27 26 Ambient 25 25

• Representative of RLO61, RLO67, and RLO88.
  • Representative of RLO70. 1694 092

TABLE 6

SUMMARY

OF INTERNAL PRESSURIZATION TESTS Number of Folds on Bulced Ends Internal Deformed can Too Bottom Pressure Length No. Type (psic) (in) Major Minor Major Minor 1 Food Can 30 5.025 3 0 1 0 2 Food Can 35 5.140 3 2 3 1 3 Food Can 40 5.218 4 0 4 2 4 Food Can 50 5.360 7 2 6 0 5 Food Can 60 5.432 7 0 6 0 6 Food Can 70 5.550 7 1 6 2 7 Paint can 29* N/A 0 0 1 0 8 Paint Can 34* ff/A 0 0 2 2

• Lid pe; ped off.

TABLE 7 DESCRIPTION OF SPECIMEN CANS FOR PRESSURIZATION /DEPRESSURIZATION TESTS Specimen Identification Description Food Can Seal A, B, C A sand-filled paint can Acceptable enclosed within two layers of PVC and placed within a sealed food can. D Same as A, B, C. Intentionally Defective E,F,G Empty paint can enclosed Acceptable within two layers of PVC and placed within a sealed food can. H Same as E, F, G. Intentionally Cefective I, J, K, L Empty sealed food cans. Acceptable 1694 094

3.- w TABLE 8

SUMMARY

OF VISUAL OBSERVATIONS OF SPECIMENS AFTER REMOVAL FROM THE PRESSURE CHAMBER l Specimen Condition of Paint Can 1 Identi fication Condition of Food Can and PVC Bag A Bulged at both ends; 3 major Unknown. Food can not and 1 minor fold in bottom; opened. 4 major folds in top. B Slightly bulged at both Paint can crushad on sides, ends; 1 minor fold in top. lid intact; PVC bag normal, i.e., not inflated. C Bulged at both ends; 4 major Paint can crushed on sides, folds in bottom; 2 major lid intact; PVC bag normal, folds in top. i.e., not inflated. D Top end popped off; no Paint can crushed on sides, bulging of ends. lid intact; PVC bag normal, i.e., not inflated. E Bulged at both ends; 5 major Unknown. Food can not and 1 minor fold in bettom; opened. 5 major and 1 minor fold in top. F Crushed slightly on sides Paint can crushed severely, and ends bulged slightly; lid intact; PVC bag normal, can appears to have taken i.e., not inflated. form of the severely crushed paint can inside. G Crushed on sides. Unknown. Food can not opened. H Top end popped off; no Paint can crushed on sides, bulging on ends. lid intact; PVC bag normal, i.e., not inflated. I Bulged at both ends; 5 major Not applicable, and 3 minor folds in battcm; 4 major and 3 minor folds in top. J,K,L Crushed on sides. Not applicable. 1694 095 iC%}}