ML20126B345

Insp Rept 70-1257/92-08 on 921015-17 & 29.Violations Noted. Major Areas Inspected:Overflow of Low Enriched Uranyl Fluoride to U Hexafluoride Vaporization Chest

ML20126B345
Person / Time
Site:	Framatome ANP Richland
Issue date:	12/04/1992
From:	Reese J, Wenslawski F NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
To:
Shared Package
ML20126B320	List: ML20126B317 ML20126B331 ML20126B345
References
70-1257-92-08, 70-1257-92-8, NUDOCS 9212220062
Download: ML20126B345 (17)

Text

.

U.S. NUCLEAR REGULATORY COMMISSION REGION V i

Report No. 70-12S7/92-08 Docket No. 70-1257 1

l License No. SNM-1227 Licensee: Siemens Nuclear Power Corporation 2101 Horn Rapids Road Richland, Washington 99352-0130 Facility Name: Siemens Nuclear Power Corporation Inspection at: Richland, Washington ~

InspectionConducte! Onsite Oct r 15-17 and 29, 1992 Team Leader: /OAtu; V I 97-Qi.Reese,Chieft Date Signed FacVlities Radiological Protection Branch l

Approved by: -

~f F. A. W6nslawski, Deputy Director MkfA Date ' Signed i

Division of Radiation Safety and Safeguards i

Team Members: C. A. Hooker, Region V D. L. Proulx, Region V l

C. H. Robinson, NMSS/IMSB H. W. Lee, NMSS/SGTB Summary:

Areas inspected: This was a special announced inspection to review the circumstances surrounding a licensee events involving: (1) The overflow of low enriched uranyl fluoride to a uranium hexafluotide vapo"ization chest, and (2) bulges in favorable geometry low enriched uranium powdei storage slab hoppers.

The inspection also included walk-downs of the systems involved. Inspection -l proceduras 30703, 90712, 93702, 88015, 88020, and 88025 were addressed. '

Results: In the areas inspected, two ' apparent violations were identified that involved:- (1) failure to consider credible accident scenarios in Nuclear Criticality Safety Analyses (Section 3.2), and (2) failure to follow procedures relative to evaluating and reporting requirements (Section 3.3).

The second violation is considered a repeat of a violation identified in NRC Inspection Report No. 70-1257/92-06. Although two apparent violations were identified, the licensee's corrective actions appeared extensive and indicated an improvement in managements commitment to improve their criticality safety program.

9212220062 921204 PDR ADOCK 07001257 C PDR=

1

4 DETAILS

1. Persons Contactgd 1.1 Siemens Nuclear Power Corporation (SNP)

• B. N. Femreite, Plant Manager,

• R. E. Vaughan, Manager, Safety, Security and Licensing

• R. L. Feuerbacher, Manager, Plant Operations

• J. W. Helton, Manager, Plant Engineering

• H. K. Valentine, Manager, Manufacturing Engineering

• W. E. Stavig, Staff Engineer

• L. J. Maas, Manager, Regulatory Compliance

• W. G. Keith, Manager, Mechanical / Chemical Plant Engineering C. D. Manning, Criticality Safety Specialist

• J. B. Edgar Staff Engineer, Licensing J. H. Phillips, General Supervisor, Chemical Operations T. C. Probasco, Safety Supervisor

• J. J. Payne, Chemical Shift Supervisor

• Denotes those attending the exit meeting on October 29, 1992.

In addition to the individuals noted above, the NRC representatives met and held discussions with other members of the licensee's staif.

2. Backaround Information The primary activities at the facility involve the processing of low enriched uranium (equal to or less than 5.0 wt % U-235) for the Following chemical conversion p(roduction of PWR and BWR reactor fuel.two process lines, Line 1 and Line 2),

heated in a calciner (rotary kiln) and reduced to uranium dioxide (VO,)

powder. This process is located in the VO, Building of the licensee's facility. The ADU is derived from either uranium hexafluoride (UFs ) or uranyl nitrate (UNH) feed. Normally UFs is received as a solid in 81 inch long and 30 inch diameter Model 30 stainless steel cylinders shipped in steel over-packs. Each cylinder contains a maximum of about 1500 kilograms (kg)_of UF filled with UFs is about136 I. The kg. approximate weight of each cylinderUFs has

• F (132 *C) and sublimation point of about 134 *F. The UFs cylinders are electrically heated to vaporize the UFs in vaporization chests. Line 1 has two vaporization chests (Chest I and Chest 2), and Line 2 has 3 chests, two for chemical conversion and one for the dry conversion process. Line I chests are about 100 inches long, 36 inches wide, and 40 inches in height. Line 2 chests are slightly larger. Line 1 chests are equipped with two crane operated doors on top of each chest for loading and unloading cylinders, and an a small door at the front for connecting the cylinder to the process system. The depth of the inside cavity (basin) below the front door of the chests ranges from about 15 (Line 1) to 17 inches (Line 2). The cavity of each chest contains a cradle that supports the cylinder and heating elements used to vaporize the UFs .

2 The UF a mixi,gn nozzle which converts the UFgas from a heated cylinder is contacted with solution and hydrogen fluoride (HF), 6a gas to a uranyl fluoride highly toxic corrosive ga(sU0,F,)The liquid 00 formation is of co U0,F,llected in aeometry favorable gis called The the hydrolysis ph 9.56 inch inside diameter an

,F,d 120 inch tall tank (Tank 10 line 1 - Tank 102 L the HF system. gases are vented to a process off-gas (P0G) exhaust ventilation (first floor level) and extends several feet into the second floorT levcl. The UD ammoniumxi hydro,F,de in a precipitation tankAlso, to produce UNH ADU. produce derived feed material can be precipitated in this tank.

The POG system collects gases in a common exhaust system from each respective chemical conversion line's (Line 1 and 2) UFs vap M zation exhaust from the ADV dryer and calciner. chests, vaporization room, proc in a water scrubber The vented gases are treated discharged via a sam, pled exhaust stack.a dryer and double HEPA filters before being In addition to the normal P0G system, each vaporization chest and vaporization room is equipped with an independent emergency scrubbing system to control any UF 6 releases.

3.

NRC Inspection Team Review and Evaluation of Events. Procramn Procedures, October and Personnel Performance As Related to the Event of 13 1992.

At about 12:00 pm, on October 15, 1992, an inspection team of members from the NRC Region V and Headquarters offices arrived onsite to review and evaluate all aspects of the October 13, 1992, if licensee programs, procedures, and performance were conducted withinevent and to de regulatory requirements, and if good safety practices were implemented.

3.1 lime Sequence of Activities Related to the Event of October 13, 1992 October 13. 1992 At about 2:30 am, on October 13, 1992, the Line 1 conversion area, an operator observed U0,Fwhile making a routine chec r solution (determined by color of solution) on the Line 1 vaporization room floor.

This the UFindividual immediately went to the Line I control room and shut off time. 6 gas flow from the cylinder in chest I which was on line at this 12:45 am. Chest 2 contained an empty cylinder that had been valved off at cooling of the empty cylinder.The top doors After shutting offand front door of Chest 2 were open to all the UF operator immediately notified his Shift Supervisor and th, e flow, shift the Chemical Technician Specialist (CTS) of the incident. The CTS inspected the vaporization room and saw UO F leaking from cracks (caused by deterioration) in the flexible h,os,es that connected each chest to the P0G exhaust duct work in the vaporization room.

By 2:50 was am secured. the deionized water (OlW) flow to Tank 10 (hydrolysis tank)

All during this time Tank 10 indicated a normal operating

{

L . ,

3 3

level of about 45 percent full.

from the floor and At the3:00 CTSam the operators visually checked began cleaning the inside up the U0,F, Chest 2 and no of vaporization t liquid was observed. A'oout 40 gallons of UO was removed from the floor. The airborne concentration of HF did,F, no t exceed 5.0 parts per million (ppm) and the airborne radioactivity did not exceed 3.3 E-Il microcuries per cubic centimeter. Contamination levels after cleanup of the U0 F solution did not exceed 1000 disintegrations per 100 square centim,et,ers. -

P At about 3:30 am the shift instrument technician was called by the CTS-to check the level transmitter on Tank 10. No-problems were identified during a calibration check of the level transmitter and its associated-high level alarm. Although_the valves on the instrument air. supply to the pneumatic level sensing probe (dip tube) inside of Tank 10 were found not to be fully open, the indicator was still showing a level for Tank 10. Since no problems were identified with Tank 10-instrumentation, the instrument technician suspected that there may be a physical problem with the pneumatic level sensing tube (dip tube) inside of the tank.

3 M about 3:55 in concentration am thethe U0,F,ization room was at 2.0 ppm.

vapor spill 5:00 At about cleanup am- was comple the dip tube (1/4 inch PVC tubing) was removed and inspected. The inspection of the dip tube, which included a quick air pressure check '

for leaks and plugs, revealed no problems. Another calibration check-4 was performed on the level transmitter and no problems were identified.

Tank 10 was filled about one-half full with DIW to check the level . -

control transmitter,-which indicated the correct tank level. The tank.

was filled to about 70 percent full with DlW. Since there was no tank overflow during these tests, the not fully o)ened i_nstrument air _ valves '

were suspected as being the cause of the Tan ( 10 overflow. After running DlW flows for about 20 minutes, the problem was thought to be -

corrected.

- At 6:00 am the UF, flow from Chest I was restarted. At 6:10 am the Shift Supervisor notified the General Supervisor, Chemical Operations of the event._ At about 6:50 the CTS visually checked the Line_1 leakage. No leakaga was_

-observed.

vaporization room'and At 7:00 am theChest 2 for U0,F,isor notified _the Manager, Plant Shift Superv

- Operations _ of- the event and the oncoming day shift crew assumed duty. e At 8:00 am the day shift-CTS went to the Line I vaporization room to remove the empty UF s cylinder from Chest 2.

The CTS closed the top doors of the chest and-prepared to disconnect the UF, transfer line from the cylinder. After-he disconnected the UF6 transfer line and reopened the top doors, he heard the movement of liquid in the chest and subsequently noticed a liquid solution at the bottom of the. chest.

{ .

By 8:30 am the Line-1 chemical conversion process was shut down by the -

. =, =-- -_;,. - - . = - _ . - - . - - ,a.. -, ,

i i

I Shift Supervisor. At about 9:00 am an operator checked the specific  !

gravity of the solution which correlated to about 75 gm U/1 and the  !

removal of the liquid (VO F r ) was initiated. No liquid was found in l Chest I and the liquid fr,om Chest 2 was pumped to a uranium liquid waste  !

recovery system. '

At 10:00 am the operators began inspecting the P0G system for UD,F, by opening drain lines on various portions of the duct work. Approximately ,

12 gallons of UO F liquid was removed from a section of ducting (in the  !

overhead of the ,se,cond floor level) that once exhausted heat from the Line 1 calciner shroud. It was later determined that this section of duct was flanged off within about four feet of main P0G duct, this prevented the overflow of UO,F, to the calciner. The Tank 10 vent was connected to the main P0G system ahead of the P0G down-riser to the vaporization room. When the short section of duct (to calciner) . filled, the solution flowed down to the vaporization room.

At 11:00 am the heaters were removed from the bottom of Chest 2 for additional cleaning. -At this time, the operator noted that there had been several inches (about 4 inches) of UO F, in bottom of the chest, as determined by the chemical stain marks on khe sides of the chest.

At about 11:30 am, the Shift Supervisor notified the General Supervisor, Chemical Conversion and the Manager Plant Operations who had been in a meeting with the Plant Manager from 9:00 am to 11:30 am.

At 12:15'pm operations notified Safety, Security and Licensing management of the b0,Fr over flow problem. At 1:54 pm the licensee convened an Incident Investigation Board (IIB) to review the matter.

Operations concluded that the liquid was U0,F, based on the color of the solution and a specific gravity check grab sample, which was later confirmed by the analytical laboratory as being U0,F rat 80-gms U/1 and the same enrichment (4.0 weight percent U-235) of 1.he material being processed.

The IIB postulatei that the U0,F rwas due to an overflow from the Tank  !

10 vent to the P0G system on the second floor level and back down the P0G duct work to tha vaporization chests on the first floor level.

Because no problem had been-detected with the Tank 10 level control -

instrumentation, the IIB decided to recheck the dip tube _in Tank 10. As a precautionary measure, the IIB decided to place the Line 2 chemical conversion on hot standby (UF6 vaporization and _ hydrolysis tank secured) . .

until the IIB could resolve the question of how.the level indicator.in Tank 10 appeared to working properly (indicating a normal o)erating level) with the tank overflowing. The line 2 UF6vaporizati on-and-hydrolysis phase of the operation was not on'line at this time. The operations and engineering department were tasked to investigate the

- cause of the overflow from Tank 10. .

At-2:55 pm, on August 10, 1992, the Manager Regulatory Compliance-and the Criticality Safety Specialist notified the Region V office of the event and the NRC Operations Officer at 3:05 pm. .

y .-a--w- -w.e-__-w,.e,-- -,%,ev-,..I,- - -c,- -

w .e-

. ,n;-, ,te, ,-r.y--.v, -

---

.,e,,,r-n,- ,- >-h- y g-

5 October 14. 199&

At 8:30 am the !!B convened to review the findings of the investigating staff. The licensee's investigation, after a more thnrough inspection of the dip tube in Tank 10, discovered a small vertical crack in the top threaded portion of the dip tube. Such a crack would allow the leakage of the air that provided the head pressure for the dip tube. The air leak would provide a false input to level control transmitter which would indicate that the VO,F, level in Tank 10 was significantly lower than it actually was. The investigation revealed that the U0 F had overflowed from Tank 10 to the P0G system, with some of the s,ol,ution flowing to the scrubbers on the second floor level and some flowing down to the vaporization room on the first floor level. The solution only flowed to Chest 2 and not into Chest 1 because its P0G valve (butterfly valve with a plastic seat) was closed. This valve (one for each chest) was interlocked with the vaporization chest's front door. When the door is open, the valve is also open. The valve closes when the door is closed to maintain heat in the chest when a cylinder is being processed.

When the Tank 10 overflow occurred, the Chest 2 door and the P0G valve were open to allow cooling.

During several telephone conversations, the NRC Region V Office discussed the status of the licensee's investigation, and the circumstances surrounding the event with the licensee. Due to the potential safety significance of this event, the Region V Office issued a Confirmatory Action Letter, dated October 14, 1992, to the licensee-confirming the Region's understanding that SPC had implemented or would implement certain corrective actions, determine the cause of the event, and receive NRC concurrence of those corrective actions prior to restart of conversion Line 1 or processing UF, in Line 2.

3.2 Criticality Safety Analyses (CSAs)

The Inspection Team reviewed CSA No. U-1.2, "UFs Vaporization Chests,"

Revision 73-1, dated August 1973 (for Line 1); CSA No. U-1.7,

" Hydrolysis Tank," Revision 3, dated May 15,1980(forline1);CSANo.

U-1.26,"P0G Ductwork," Revision 1, dated October 4, 1982 (for Line 1) to determine whether adequate limits, barriers, controls, and procedures had been established and were being implemented to safely control UF, and U0,F, hydrolysis process activities.

The Inspection Team, in its review of the above referenced Criticality Safety Analyses (CSAs), determined that CSAs failed to evaluate.

interfaces between process operations and equipment; discuss the safety aspects of the interfaces; and identify all nuclear criticality safety (NCS) controls requiring continued attention with respect to maintenance and/or surveillance testing. Also, the te m determined that the CSAs lacked accident analyses to demonstrate that contingency conditions were not credible. As such, credible accidents were not identified and analyzed and NCS controls related to interfaces were not identified.

Subsequent to the above evaluation The Inspection Team reviewed the

.g ,.p.n,.

6 licensee's commitments contained in the license application and associated documents and made the following observations:

1) License Condition No. 9 of SNM License No.1227 authorizes the use of licensed materials in accordance with the statements, representations, and conditions contained in Part I of the licensee's application dated July 1987, and supplements dated November 12, 1987, through March 25, 1992.

2) Section 2.5, " Operating Procedures, Standards and Guides", Part I of the license application states, in part: (1) the licensee conducts its business in accordance with a system of Standard

• Operating Procedures, Company Standards, and Policy Guides; and (2) the licensee is committed to controlling activities involving special nuclear materials in accordance with these approved written procedures, standards and guides.

Section 4.2.5.1, Part 1, of the license application. Section 4.2.1, Fart 1, of the license application requires two physical barriers to a criticality accident.

3) The introduction to " Chapter 4 Nuclear Criticality Safety," Part I of the license application states:

" Nuclear criticality safety shall be assured through both administrative and technical practices. Administrative practices clearly include establishing the responsibilities for nuclear criticality safety, providing adequate and skilled personnel, preparing written standards and procedures, process analysis, materials and operational controls, operational and incident reviews, and emergency procedures. Technical practices include exercising control over the mass and distribution of significant quantities of special nuclear materials, and the mass, distribution, and nuclear properties of all other materials with which special nuclear materials are associated."

4) Section 4.1.1, " Process Analysis (Criticality Safe Determinations)," Part I of the license application states in part:

"Before any operation with special nuclear material is begun or changed, it is determined that the entire process will be subcritical under both normal and credible abnormal conditions, and within the technical requirements specified in Section 4.2. Criticality safety analysis are performed on all applicable processes...."

5) Section 4.2.1, " Double Contingency Policy," of the license application states:

i

" Process and equipment designs and operating procedures

d I

7 ,

incorporate sufficient factors of safety to require at least

two unlikely, independent, and concurrent errors, accidents,

equipments malfunctions, or changes in process conditions before a criticality accident is possible."  !

6) Section 4.1, " Purpose and Scope," Chapter 3, " Nuclear Criticality  !

Safety Standards," of the licensees Safety Manual (EMF-30) states .

in part: '

"The CSA is a study of equipment / operations' involving fissile material at normal conditions and at credible accident conditions to determine if the criticality safety criteria are satisfied."

7) Section 4.3, " Criticality Safety Criteria," paragraph 4.3.2 of EMF-30 states in part:

"No single credible accident condition shall be capable of causing an accidental criticality."

Based on the above observations, the Inspection Team concluded that CSA ,

U-2.1 did not adequately consider all credible accident conditions.

Specifically, flooding the vaporization chests with uranium bearing solutions from other process systems that vented to_the P0G system was'a credible accident condition that was not analyzed.

CSA U-1.7 did not address any potential consequences and/or any potential criticality safety concerns that could result from overflows from various process vessels that vent to the P0G system. In '

particular, CSA U-1.7 failed to discuss back-flow prevention and failed -

to identify the controls associated with back-flow prevention.

Since accident conditions were not analyzed-for the hydrolysis tank, the controls required to maintain the system subcritical and ensure NCS ,

under contingency conditions were not stated. In addition, the evaluations did not identify controls requiring maintenance and/or surveillance testing, such as the level indicator. Since the back-flow contingency condition was not considered-in an accident analysis, the use of the level indicator for back-flow prevention was not recognized .

as a control method or designated as safety-related equipment.

CSA U-1.26, showed that- the P0G duct work was safe for SNM accumulations within the duct work but did not show that equipment interconnected with the duct work'was safe for contingency conditions related to flow through cross connections such as vents, overflows, or chemical addition lines; accidental transfers resulting from valve leakage; unauthorized piping changes; or changes in geometry resulting from leakage or mechanical ' failure.

Failure of CSAs to adequately address credible accident scenarios that-

• could impact criticality safety was identified as an apparent violation of License Condition No. 9(70-1257/92-08-01). .

. .nu.._ ._ .n. 2 ; . c _ _ --a . ,. w .;_ ;_ s, . . .r. ~ u -, - ,. ., , # - .

~

8 One apparent violation was identified in this area.

3.3 Qperations Performance as Related to the Octohlr 13. 1992 Event As described in Section 3.1 above, the overflow of a liquid solution into the Line 1 vaporization room was observed by an operator at about 2:30 am on October 13, 1992. The operating shift determined that the liquidsolutionwasUO/,0amonOctoberthat (Tank 10). At about 8:3 13, had overflowed 1992, from the the operating hydrolysis tank shift observed UOf in the basin (unfavorable geometry vessel) of vaporization khest 2, and the overflow of this material was not reported to appropriate Safety, Security and Licensing management personnel until 12:15 pm on October 13, 1992.

Section 11.1.1, Responsibilities," under Section 11.1, " Reporting of Incidents," Chapter 3.0, " Nuclear Criticality Safety Standards," of the licensee's Safety Manual (EMF-30) states in part:

"All Employees: It is the responsibility of each employee to immediately report to his or her immediate supervisor any incident or off standard condition related to criticality Safety...."

" Supervision: It is the responsibility of the first line supervisor to document all reported criticality safety related incidents or conditians. The supervisor is also responsible to ensure the reported incident or condition is evaluated and reported to [ Safety, Security and Licensing) as required by this procedure."

" Safety. Security and Licensina: It is the Responsibility of Safety, Security and Licensing to make the final determination of whether or not an incident or condition is reportable to the NRC."

Section 11.1.2, " Evaluation," states in part:

"This evaluation must be made promptly. If the incident or condition is determined to be reportable to the NRC this report must be made within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> of discovery."

"The supervisor shall immediately conduct an evaluation to determine if the reported incident or condition may be reportable to the NRC using the criteria given in Section 11.2."

"If, after completing the evaluation, the supervisor determines that the incident may be reportable to the NRC or is not sure, he or she shall inmediately report the incident to Safety. Security and Licensing...."

Section 11.2 of this procedure describes events potentially requiring

-NRC notification within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.Section II, Item 5 states " Criterion An unusual event or condition for which the severity and remedy are not readily determined," and provided the following example:

9 "An unexpected accumulation of large amounts of uranium in an unfavorable geometry process system such as ventilation duct work."

D4F-30, items 6 and 7 of Appendix 5, specify conditions which require reporting to SS&L:

" Conditions occur in the process, which are not controlled by the procedure."

"An abnormal situation is discovered where uranium has accumulated in or has been released to unsafe geometry ducts, lines, or tanks."

The Team made the following observations related to this event:

1) At about 2:30 - 2:45 am on October 13, 1992, a shift supervisor was aware of a significant process upset that resulted in shutting down the UF3 gas flow to the hydrolysis tank. The shift supervisor also believed the process upset condition was associated with an overflow from Tank 10 containing UO F, with an enrichment of 4.0 percent U-235 to the P0G system. Allhough immediate actions were taken to shut the process down, the severity and remedy of the overflow were not fully known at this time. These conditions represented an unusual condition for which the severity and remedy were not readily determined.

F overflow to the vaporization Specifically, (1) system) room via the P0G this event (U0,d ,not to have previously occurred appeare during plant operations (2) this condition (overflow of Tank 10) was not covered in the operating procedures, and (3) the P0G system had not been previously analyzed for U0,F2 accumulation.

2) At about 3:00 am on October 13, 1992, the CTS checked inside of Chest 2 for the presents of U0,F, and none was observed. Based on discussions with licensee personnel and by direct observations, the inspection Team, with the agreement of cognizant licensee representatives, observed that it would be very difficult to see 4 inches of liquid solution in the bottom of the chests when they contained a UF 3 cylinder, unless a detailed inspection of the chest was performed. Since the P0G valve was assumed to have been open (interlocked with front door of chest which was open) at the time when Tank 10 overflowed and U0,F, was observed leaking from the horizontal P0G flex hose at the back the chest, it is reasonable to assume that some U0,Fr had entered Chest 2. In addition, since no more leakage was observed from the P0G flex hose following the initial overflow of Tank 10, it is reasonable to assume that the 4.0 inches (determined by stain marks on the walls of the chest at 11:00 am) of U0,F,, had been in the chest from the time the overflow from Tank 10 was terminated at about 2:30 am on October 13, 1992.

Although U0,Fr solution was observed leaking from the Chest 1 flex

4 10 hose, no liquid was found in this chest. Since the P0G valve (not a water tight valve) was closed on this chest and its flex hose had more cracks than the flex hose on Chest 2, the licensee assumed that the leak rate of the solution from the flex hose cracks was sufficient to preclude solution leaking through the closed valve from entering Chest 1. The Team had no concerns relative to this assumption.

3) The quick air pressure check of the Tank 10 di) tube for leaks and plugging was not adequate. Specifically the c1eck for air leaks was performed by the technician running his hand along the tube to check for leaks, as opposed to pressure check at a specified aressure with the dip tube submerced in water and checking for subbles (licensee identified).

4) The cause (partially opened instrument air valves) had not been identified for the Tank 10 overflow and an inspection was not performed to determine if any U0,F, solution remained in various portions of the P0G duct work, before the UF6 flow from vaporization Chest I was restarted at 6:00 am.

5) At about 8:00 am on October 13,1992, U0,Ft solution was discovered in vaporization Chest 1. Line 1 UF process flow was shut off at 8:30 am and inspection and drainin,g of the P0G duct work was initiated. The level of U0 2 accumulation in the Chest 1 basin (unfavorable geometry vessel)F was not known until 11:00 am after the material was pumped out, and the cylinder cradle and heating elements had been removed from the chest.

6) The General Supervisor, Chemical Conversion and the Manager, Plant 0?erations were not notified that U0,F, soluticn had been found in tie vaporization chest until 11:30 am on October 13, 1992.

Safety, Security and Licensing was not notified until 12:15 pm on October 13, 1992. The event was subsequently reporteJ to the NRC at 3:05 pm on October 13, 1992.

Based on the above observations, the Team concluded that, the failure to immedf ately evaluate the event for reportability and to report the event in accordance with Section 11, Chapter 3 of the Nuclear criticality Safety Standards is considered an apparent violation of License Condition No. 9(70-1257/92-08-02). However, based on a review of the licensee's corrective actions submitted in response to the CAL, described in Section 3.5 below, the Team. oncluded that the licensee had corrected this matter.

Other observations:

1) The licensee identified that Tank 10 had no overflow and was vented to the-P0G system. The hydrolysis tank (Tank 102) on Line 2 was equipped with an overflow system that drained to the floor and a vent to its respective P0G system. It appeared that any overflow of U0,F, from Tank 102 would be directed to the floor and

11 not affect the P0G system. The licensee also identified several other process tanks on the Line I system that were not equipped with an overflow system and vented to the P0G system. It appeared that all of Line 2 process vessels that vented to the P0G system were additionally equipped with overflows to preclude flooding of the P0G system.

2) On October 16, 1992, after a thorough inspection of Line 1 and Line 2 vaporization chests, the licensee discovered that Chest I and 2 in the Line 1 vaporization room had 7, one quarter-inch, holes about 5 inches from the bottom of the chests. These holes '

apparently were the result of a shroud which was eliminated in the past. Also, small holes for electrical conduit penetrations near the bottom of the va)orization chests in the Line 2 system were observed. None of t1ese holes were made for the purpose of draining liquids from the basin of the vaporization chests.

One apparent violation was identified in this area.

3.4 Criticality Safety Evaluation of Event The licensee performed a criticality safety calculation to evaluate the significance of the unintended transfer of U0,F r solution to the vaporization chest. The calculation revealed that the incident resulted in a subtritical condition. Following the incident, the vaporization with an enrichment 4.0 weight percent (wt.%) U-235 chest contained U0,F, and a concentration o f 80 g U/1.The estimated quantity of U0 F transferred to the vaporization chest resulted in a slab heighk #of about 4 inches with a total mass of about 19.6 kg of uranium compounds. Given these conditions, the incident resulted in a sufficiently subcritical condition in which no greater than a safe batch (about 25 kg of uranium at 4.0 wt.% U-235) was accumulated.

The margin of safety could have been less under normal operating conditions. UO r could have been processed at an enrichment of 5 weight percent ,FU-235 and the licensee's operating procedure (S0P P66,811) allows a concentration of 250 g U/1 to be achieved in the hydrolysis tank. The overflow from the doors of the vaporization chest is approximately 50 centimeters. Under these conditions, the licensee calculated a k-effective of less than 0.85 using KENO. ARH-600 gives a minimum critical concentration of 350 g U/1 for uranium dioxide solutions at a slab height of 50 centimeters and at 5 weight percent U-235. The allowed 250 g U/1 constitutes a safe critical concentration.

The team agraed with the licensee calculations. However, the slab height of the vaporization chest was not established as a safety control. ANSI /ANS-8.1-1983 gives a subcritical limit of 261 g U/1 for uranyl fluoride at-5 weight percw+ U-235. If not for the random factor of safety (the doors of the vaporization chest limiting the slab height), the margir. of subtriticality would have been inadequate.

l l

12 3.5 Response to Confirmatory Action letter (CAL)

The Region V staff reviewed the licensee's response, dated October 21, 1992, to the NRC's Confirmatory Action Letter issued to the licensee on October 14, 1992, relative to the October 13, 1992, event. The licensee's response included the cause and detailed the corrective actions taken and those planned to prevent recurrence of similar a event. The following is a summary of the cause and corrective actions:

1) from Tank 10 to the Line 1 vaporization room The and Chest overflow 2 wasof UO,F, d by a crack in the dip tube caused from cause either being (1) flawed at the time of installation, or (2) b6ing improperly threaded (region of crack). The crack resulted in a -

malfunction of the tank level control system which provided a false reading of a less than half full tank, when the tank was actually overflowing.

2) Corrective actions included but were not limited to the (1) replacement of the dip tube with a stronger material (2) examination of dip tubes in other tanks, (3) a review of maintenance calibration of such instruments, (4) installation of overflow lines on tanks that were not equipped with overflows, (5) modification of the P0G system so that any accidental overflows of solutions from tank vents would not have a pathway to unfavorable geometry vessels, (6) modifications as practical so that unfavorable geometry vessels such as vaporization chests cannot hold uranium bearing sclutions (holes placed in basin of chests),

and (7) addition of liquid monitoring alarms to the low side of POG systems. In addition, an independent task team was commissioned to review the conversion process and potential pathways for uranium-bearing solutions to enter non-geometrically safe vessels. The task team's findings and recommendations were also incorporated into these corrective actions.

3) In addition to the engineering actions provided in the licensee's response, the licensee provided a memorandum regarding SPC's commitment to improve management oversight relative abnormal event reporting, based on lessons learned from the October 13, 1992 Tank 10 overflow event.

o This memorandum, from the Manager, Plant Operations, to all plant operations supervisors / potential delegates, dated October 21, 1992, described the need for Plant Operations to clarify procedures for (1) abnormal event reporting by shift supervisors, and (2) defining abnormal ennditions which require higher level authorization for resuming operation of the plant. The Memorandum to be used as an interim procedure until a new Standard Operating Procedure is prepared included (1) a description of unusual events and a form for documenting all levels of such events, (2) a description of events that require shutting down the affected operations, and (3) the authority needed to resume operations after abnormal events.

1 1

?

.. r 13 5 l

Other than as previously noted with the CSAs, the inspection Team ,

concluded that the licensee had adequately determined the cause and its  :

corrective actions appeared extensive and a)propriate. The inspection Team had no further questions relative to t1e corrective actions taken or planned.

7 4.0 Slab Hopper Buloino Event I This event involved the licensee's identification of bulging in I favorable geometry slab hoppers used to store low enriched U0 powder.

The licensee's investigattor, and corrective actions related t,o this '

event were reviewed. >

4.1 Backaround Calcined U0 powder is transferred to favorable geometry slab hoppers where the p,owder is confined until it's moisture content has been verified to be within specified limits prior to being transferred to  !

unfavorable geometry blending vessels. The inside thickness is a primary criticality control parameter for the slab hoppers. The licensee has a total of seven slab hop)ers on three powder preparation systems-(PPSs). Line 1 PPS has three loppers ( Nos. -11,12, and 13 ).

The Line 2 PPSs has two slab hoppers (Nos. I and 2) and the Line 3 PPS hastwoslabhoppers(Nos.3and4) The Line I slab hoppers are 48 inches wide,108 inches long, and a nominal 4 inches thick. The Line 2 _,

and 3 slab hoppers are 48 wide,138 inches-long, and a nominal 4 inches '

thick.

On theirOctober 5,1992,ofthe identification licensee bulging of anotified small the area NRC Regional (about I ft on p)ffice Slab of 1 llopper No.12 and subsequently reported the matter to the NRC Operations Officer in accordance with NRC Bulletin 91-01.

By letter dated October 8, 1992, the licensee informed the NRC Region V Office incidentof that additional included: information related (1) details to the of the slab hopper at found conditions. bulging)-

(2 >

current surveillance program for determining bulging, (3) a description of the mode of operations and safetyarecautions which allow continued plant operations without the use of tie slab hoppers, and (4) ,

preliminary corrective actions for resolving' the identified problems and : )

returning the slab hoppers to service. '

4.2 CSAs-The CSA for the slab hoppers in _the Line 1-limited the thickness of each - H hopper to 4 inches, and the CSA for the slab hoppers in the Line'2 and-3 ,

PPSs limited the thickness of these hoppers to 4.28 inches. ,

4.3 Licensee'sinvestication By_0ctober 6,1992,- the licensee had taken all of the slab hoppers out -

of service and completed its inspection of each hopper. All but one i

)

d i

I

,=c-iMEIrt,- >c w.m. . . - - , . ,r 5 E , w, - , r,..~e..%..--,,---,w,--,-m.._.m --. - ,

14 4.25 inches in Line 1 and a maximum thickness of 4.5 inches in Line 2) that exceeded the limits provided in the respective CSA. The buiging was in the same area as Slab Hopper No. 12.

As an alternate mode of operation, the licensee elected to continue operations of powder handling and storage using 5-gallon buckets for transferring powder and 45-gallon poisoned drums for storage. The alternate mode of operation included sampling and analysis of the moisture content of the powder from the First and tenth 5-gallon bucket of powder out of the calciner and each 45-gallon container before being placed in storage.

Although the licensee had not completed its investigation, their preliminary conclusion was that bulging occurred due to metal fatigue caused by extended service, and heat and powder expansion that occurs The when U0,' powder oxidizes (burn-back) 3 to U 0 while in the hoppers. licensee s final inve inspection and is considered as an inspector followup item (70-1257/92-08-03).

4.4 Corrective Actions The licensee's long term corrective actions involved replacement of the slab hoppers with new units that incorporate design criteria which inhibit oxidation of the U02 powder. Interim corrective actions to allow use of the existing slab hoppers required a temporary reduction in the enrichment of uranium powder entering the slab hoppers, and a number of maintenance inspection and monitoring actions to reduce the potential for oxidation or to detect oxidation in early stages. All interim corrective actions were to be evaluated and approved by the licensee's Startup Council prior to reuse of the existing slab hoppers.

4.5 NRC Onsite Review During facility tours, the Inspection team noted that the licensee had installed bracing to the area of bulging on Hopper No.12. As a result,  ;

only a very slight bulging (less than 1/4 inch) was noted. The NRC observed no apparent structural failure in any of the seven slab hoppers. The Inspection Team's findings are summarized as follows:

1) The licensee had no records of material specifications for the slab hoppers and could not determine the grade of steel from which the hoppers were made.

2) Because of poor weld detail around the spacers (tie rods) inside powder escapes through the gap between the of the hoppers, stiffener angle anU0,d the hopper shell.

3) The licensee did not have documentation of the design loading and design criteria for the existing slab hoppers. However, they had contacted a consultant to develop design specifications for all of the slab hoppers.

4 I

15

4) The licensee had no acceptance criteria with regard to materials and welds.

5) There was no quality assurance (QA) program for the slab hoppers. k

6) As part of the licensee's ongoing investigation, Siemens performed ultrasonic (UT) examination on No.12 slab ho)per and found that the hopper wall thickness was only 0.122 incies. This was thinner than the design specified wall thickness of 0.164 inches. It was not clear whether the hop)er was fabricated with thinner material or material thinning had )een taken place due to 15 or more years of continuous operation.

Based on discussions with cognizant licensee representative regarding the above observations, although no specific commitments were made, the licensee acknowledged the benefit of performing the following tasks for related to the use of slab hoppers:

1) Conduct UT examinations of all of the other hoppers to determine if material thinning has been taken place.

2) Take thickness measurements on the main body to confirm that the bulging was indeed confined to a small area near the bottom of the hopper.

3) Develop a design specification that specifies design loadings, design criteria, methods of design, and acceptance criteria for the slab hoppers.

4) Develop a QA program for slab hoppers to address all- activities including design, purchase, fabrication, inspection, operation and maintenance.

5) Develop a Preventive Maintenance Program to perform routine surveillance and periodic dimensional check of the slar hoppers.

The P.M. program should also cover the liquid slab tanks.

The final corrective actions taken by the licensee will be examined in a subsequent NRC inspection and is considered as an inspector followup item (70-1257/920-8-04).

On several occasions, the Inspection Team observed the licensee's alternate mode of operations for handing, transferring, storage, and moisture sampling of U0, powder. No concerns were identified during these observations No violations or deviations were identified.

16 5.0 Insoection Exit Meetina The inspection scope and findings were summarized with the individuals denoted in Section 1, on October 29, 1992. At this time the Leputy Director, Division of Radiation Safety and Safeguards, emphasized concerns to the findings relative to apparent generic deficiencies in older CSAs and the apparent repeated finding regarding evaluating and lack of timely reporting potential criticality safety problems to Safety, Security, and Licensing.

The observations described in the report t.are acknowledged by the licensee. The licensee was informed of the apparent violation -

identified in the report.

I

,,,,,, , . ,-.m