ML20155A434

Insp Rept 70-7001/98-205 on 980915-18.No Violations Noted. Major Areas Inspected:Paducah Gaseous Diffusion Plant High Risk Chemical Release Possibilities That May Occur as Result of Seismic Initiating Event

ML20155A434
Person / Time
Site:	07007001
Issue date:	10/22/1998
From:	NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
To:
Shared Package
ML20155A419	List: ML20155A415 ML20155A434
References
70-7001-98-205, NUDOCS 9810290095
Download: ML20155A434 (10)

Text

. . .. . - ___ . . . _ . _ _ . _ . _ . _ . _ . , _ _ . _ . _ _ . . _ . _ _ _ _ . _ . _ - . . _

t l

l l l U.S. NUCLEAR REGULATORY COMMISSION ,

1

' OFFICE OF NUCLEAR MATERIAL SAFETY AND SAFEGUARDS  !

l 1

I Docket No: 70-7001 Certificate No: GDP-1 I

Report No: 70-7001/98-205 Certificate Holder: United States Enrichment Corponsion )

1 Location: Paducah Gaseous Diffusion Plant Paducah, Kentucky Dates: September 15-18,1998 Inspectors: Garrett Smith, Inspector, NRC Headquarters David Hurley, Contractor, Battelle Approved By: Philip Ting, Chief, Operations Branch, Division of Fuel Cycle Safety and Safeguards, NMSS Enclosure l

9810290095 PDR 981022 P ADOCK 07007oog-PDR ,

.~.

l. '

? .

i 2

UNITED STATES ENRICilMENT CORPORATION PADUCAH GASEOUS DIFFUSION PLANT NRC INSPECTION REPORT 70-7001/98-205 EXECUTIVE

SUMMARY

i Areas Inspected A Nuclear Regulatory Commission (NRC) lleadquarters team conducted an announced Chemical Process Safety inspection of the Paducah Gaseous Diffusion Plant (PGDP) in Paducah, ,

Kentucky, on September 15 - 18,1998. The inspection was conducted using NRC Headquarters ,

staff and a contractor. The inspection focused on PGDP's high risk chemical release possibilities that may occur as a result of a seismic initiating event.  ;

l Results i

! e Based on conversations with the regulatory alTairs manager the inspectors concluded that  !

because Part 76 does not specifically require the high risk chemical facilities at PGDP to l be evaluated with respect to a seismic event, no specific seismic analysis has been 1 completed for those high risk chemical areas. The process hazards analysis (PHA) for l these facilities indicate a vulnerability in that a seismic event could result in levels of certain high risk chemicals above those immediately dangerous to life and health (IDLH) both onsite and offsite.

e USEC has assessed the seismic vulnerabilities with respect to the uranium hexafluoride (UF.) facilities and is upgrading the facilities and equipment to eliminate these vulnerabilities. However, the potentially updated seismic response spectrum with a 0.165g zero' period ground (ZPG) could uncover some new vulnerabilities. Additional review and inspections will be required if the new spectrum is accepted. l l

L.

l l

4

I

! 3 l

REPORT DETAILS I

1.0 Etocess Safety Information and Process Hazards Analysis - Seismic Initiating Evqnt

a. Scone of the Inspection 1

1 The inspectors reviewed the use and storage of high risk chemicals onsite (UF , chlorine, fluorine, and CIF 3). Specifically, the systems were evaluated to determine the consequences of a 250 year retum earthquake event. The inspectors reviewed the Process Safety Information (PSI) and PHA for these facilities and performed walk downs of the areas to evaluate the likely effects of a seismic event.

b. Observations and Findings General l

10 CFR 76.85, Assessment of Accidents, requires the certificate holder to perform an analysis of potential accidents and consequences to establish the basis for limiting l

conditions for operation of the plant with respect to the potential for plant releases of radioactive material. Additionally, the Statements of Consideration for Part 76 (59 FR 48944), published on September 23,1994, states that in assessing the adequacy of protection of the public health and safety from potential accidents, the NRC will consider whether the potential consequences of a accident exceed 25 rem, or a uranium intake of 30 milligrams, taking into account the uncertainties associated with the modeling. The Statements of Consideration for Part 76 also states that the NRC will evaluate consequences of potential accidents resulting from natural phenomena (i.e., seismic event).

The evaluation basis earthquake (EBE) for the Paducah Site was determined by the Beavers et al. study (1981) to have an estimated peak ground acceleration of 0.18g. At the time, an engineering evaluation determined that there was a 10 percent probability that the 0.18g value would be exceeded during the remaining life of the plant (25 years).

As part of the Department of Energy (DOE) Safety Analysis Report (SAR) Upgrade program, a revised EBE has been developed. The revised EBE has an estimated peak ground acceleration of 0.15g, and a 250-year return period. Currently, to support the SAR Upgrade program needed to fulfill the commitments in the compliance plan,

additional seismic analyses have been completed and submitted to the NRC. These analyses are currently being reviewed by NRC. To support this inspection activity, the l

j inspectors reviewed all of the available seismic information, but evaluated the seismic risk at the site against the requirements published in the existing NRC approved SAR.

l

1*

l 4

Seismic Event Chemical Conseguences The inspectors reviewed the facilities, PSI, and PHA for the high risk chemicals at PGDP.

The specific areas were chosen because of their increased potential to release large l quantities of highly hazardous chemicals to the environment in the event of a catastrophic event. Specifically, chlorine and chlorine tri-fluoride facilities were reviewed. The PHA, "Paducah Gaseous Diffusion Plant Process Hazards Analysis for Water Treatment Plant  !

j' and Pump House Chlorination Systems," KY/S-263, dated May 1996, determined for i l each of the six areas where at least 4000 lbs of chlorine are used (at least 2 cylinders), a I seismic event would result in a potential gaseous release of chlorine to the environment.

These releases have no preventative safeguard measures and the mitigative measures are  ;

the see and flee policy. Also the PHA takes credit for the ventilation and detection ,

systems, if operable. ]

The PHA," Process Hazards Analysis for Paducah Gaseous Diffusion Plant - USEC Balance of Plant Facilities," K/GDP/SAR/SUB-23, dated March 1996, determined that a seismic event could release large amounts of CIF3 and F2 from the storage tank and/or the feed bottle. The document lists no preventative safeguards for this initiating event.

Next the inspectors reviewed two atmospheric dispersion analysis documents generated I

by PGDP personnel. The document," Bounding Chemical Release Atmospheric Dispersion Analysis for the PGDP Safety Analysis Upgrade Program," document number DAC-M0848401-SAR-68, date started March 26,1996, discussed the l consequences of a chlorine release. The document," Chlorine Tri-Fluoride Atmospheric l Dispersion' Analysis for the PGDP Safety Analysis Upgrade Program," document number l DAC-M0848401-SAR-69, date started June 1,1996, discussed the consequences of a l chlorine tri-fluoride release. These reports use computer models to predict the levels of l toxic chemicals resulting from the releases with respect to the distance from the release.

Several assumptions must be made to baseline the meteorological conditions, release rates and durations, and physical location and structures surrounding the release model.

! In these cases major assumptions were made as follows:

l

- The chemical release rates and meteorological conditions are constant over the release L duration so that steady-state models are applicable.

- Even though the chemicals are released into the building, no credit is taken for the effects of the buildings and the release is assumed to be a jet going directly into the atmosphere.

I' - The 10 meter elevation meteorological data is Stability Class F, wind speed 1 m/s, temperature 4.44*C for chlorine and 21.1 C for chlorine tri-fluoride, humidity 70%,

pressure 1.0 atm, and surface roughness 0.03 meter.

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

.._..7___._

• l l

l I' 5 l l 1 l

Based on the reviews of these documents the inspectors determined that significant levels j L of these high risk chemicals could be released to the environment as a result of a seismic i event. The results from these two atmospheric dispersion analysis documents are summarized in Table 1.

l-

?

Table 1. Dispersion Model Results Summary Distance from release i

IDLH Value concentrations >lDLH l Chemical (ppm) (meters)

Cl 2 10 1600 ClO2 ' 10 50 liF' 30 50 During walk throughs of the chlorine buildings, the inspectors made several observations regarding these systems. The 2000 lb chlorine cylinders are positioned approximately l 1 foot off the floor in either concrete or metal saddles. The cylinders were resting in the saddles and were not secured in place by any other means except for the weight of the l

cylinder. In some cases, the' concrete saddles had been cracked or the edges had broken away. The systems located in the recirculating cooling water buildings were configured so the chlorine gas entered the system at a positive pressure until it was mixed v ith water.

The piping in this area was securely fastened to the wall. In this building, the irispectors

' were concerned that a seismic event could cause the cylinder to move and cause a breach of the piping system. This event could cause the release of up to 4000 lbs of chlorine to the environment.

Seismic Event Uranium Hexafluoride Conseauences  !

As stated earlier, the current SAR indicates that the radioactive facilities could withstand an earthquake with a ZPG acceleration of 0.18g horizontal. This early analysis was based on two dimensional seismic analysis. In 1994, DOE produced DOE-STD-1020-94 as an updated acceptance criteria. This changed the acceptance criteria and generated a new and rigorous analysis of the radioactive facilities. These new analyses were performed with a reduced ZPG of 0.15g horizontal and with three dimensional modeling. These new analyses found several weak spots. Buildings C-331 and C-335 have the potential for releases if the columns don't vibrate in unison. These facilities are being modified to I

p 'CIO 2and HF are the reaction products when CIF3 is released in air.

L

.\~

6 l

corr'ect this problem and in the interim, the process is being operated below atmospheric pressure. Operating below atmospheric pressure will greatly reduced the offsite dose as air would leak into the system, instead of UF6 leaking out into the environment. In Buildings C-310 and C-315, the UF 6condensers, accumulators, and the Normetex pump discharge piping do not have adequate earthquake capacity. These facilities are being upgraded and in the interim, the liquid levels in the accumulators are administratively controlled to a low level to reduce to an acceptable level the postulated dose exposure at the fence line (below the 30 milligrams Uranium intake levels). These analyses also showed that Buildings C-333/337, C-333-A/337-A, C-400, and C-720 are right at the limit for the 0.15g earthquake.

PGDP was designed and built in the early 1950's. No seismic criteria was required at that time. To show the plant does not pose a high risk to the public, numerous efforts have been perfhrmed to review the plant's capacity to withstand a postulated earthquake.

Many DOE facilities have the same problems and to help with the analysis that was needed, DOE instituted a database of equipment that can withstand earthquakes, and so identifying equipment that falls within that database simplifies the seismic evaluation process. Part of the seismic evaluation process required walking down various systems or subsystems of the facility and evaluating these systems. If equipment could be identified that was covered by the database of experience, then the evaluation was relatively simple.

At other times additional analysis had to be performed. As part of this review, the inspectors chose two random Design Analysis Calculations to audit this walkdown process. The inspectors technically reviewed the analysis and then physically inspected the systems to verify the original walkdown. The analyses were accurate, and no discrepancies were found.

Recent studies have been performed to predict a realistic ground motion based on history and geology. A new seismic analysis that documents a ground motion with a ZPG of 0.165g horizontal nas been submitted to the NRC for review,

c. Conclusions Based on conversations with the regulatory affairs manager th:: inspectors concluded that because Part 76 does not specifically require the high risk chemical facilities at PGDP to be evaluated with respect to a seismic event, no specific seismic analysis has been completed for those high risk chemical areas. The PHA for these facilities indicate a vulnerability in that a seismic event could result in levels of certain high risk chemicals above those IDLH both onsite and offsite.

USEC has assessed the seismic vulnerabilities with respect to the UF 6 facilities and is upgrading the facilities and equipment to eliminate these vulnerabilities. However, the

l l

)

l

! - I l

7 i l

l pot 6ntially updated seismic response spectrum with a 0.165g ZPG could uncover some j new vulnerabilities. Additional review and inspections will be required if the new spectrum is accepted.

2.0 Management Meetings The inspectors met with USEC management periodically during the inspection. The inspectors presented the inspection scope and findings to members of USEC's staff at the conclusion of the inspection on September 18,1998. USEC acknowledged the fimdings presented at the meeting.

l l

l I

l 1

i t

l l

l I-

. = _ . - _ _ _. . ... . . . _ .

l 1

8 4

ITEMS OPENED. CLOSED. OR DISCUSSED i

l None l:

i PARTIAL LIST OF PERSONS CONTACTED Lockheed-Martin Utility Services David Stadler .

Allen Burnett i Howard Pulley Tony Canterbury.

Danny Schneider W. Don Dontson J.D. Sohl, Jr.

Jack Howard Tim Reynolds Donald Elrod Keith Potter Larry Jackson U.S. Enrichment Corocratior.

Steve Cowne (via telcon)

Russ Wells (via telcon) '

Rusty Rawson (via telcon)

NRC John Jacobson, Resident Inspector Garrett Smith, Chemical Safety Inspector David Hurley, PNNL l

r 1~

me .a 1_.em. x .c-

f,.

F

(..; . .

r.c ,

9 .

i .

I l

ACRONYMS USED l

l. DOE Department of Energy

.EBE Evaluation Basis Earthquake H IDLH !Immediately Dangerous to Life and Health j

-NRC Nuclear Regulatory Commission PGDP Paducah Gaseous Diffusion Plant PHA: Process Hazards Analysis L PSI Process Safety Analysis .j i

SAR. Safety Analysis Report j I, ,

UF6 Uranium Hexafluoride

ZPG ' Zero Period Ground

- CIF3 Chlorine Trifluoride Cl 2 Chlorine DOCUMENTS REVIEWED l

• - KY/S-263,'?Paducah Gaseous Diffusion Plant Process Hazards Analysis for Water l Treatment Plant and Pump House Chlorination Systems," May 1996 l

• K/GDP/SAR/SUB-23," Process Hazards Analysis for Paducah Gaseous Diffusion Plant -

L USEC Balance of Plant Facilities," March 1996

- DA.iM0848401-SAR-68," Bounding Chemical Release Atmospheric Dispersion Analysis for the PGDP Safety Analysis Upgrade Program," date started March 26,1996

• - DAC-M0848401-SAR-69," Chlorine Tri-Fluoride Atmospheric Dispersion Analysis for the PGDP Safety Analysis Upgrade Program," date started June 1,1996

• United States Enrichment Corporation," Safety Analysis Update Certificate Amendment -

Request," proposed October 31,1997

= ' DOE /ORO-2026," Plan for Achieving Compliance with NRC Regulations at the Paducah Gaseous Diffusion Plant, Issue 36," Rev. 7, March 20,1998 -

• ' . DOE-STD-1020-94," Natural Phenomena Hazards Design and Evaluation Criteria for

' Department of Energy Facilities," Change Notice #1, January 1996

s;

+

10-1

• . K/GDP/SAR-122," Gaseous Diffusion Plant Safety Analysis Report Upgrade Program-Natural Phenomena llazards evaluation of Selected Equipment, Piping, and Components

'in Paducah Gaseous Diffusion Plant Buildings C-310, C-310A, C-315, C-331, C-333, C-333A, C-335, C-337, C-337A, and C-360," June 1996 j

. K/GDP/SAR-57, Rev. 3, " Walk-through Screening Guidelines for Gaseous Diffusion  !'

Plant Piping Systems," April 1996'

. K/GDP/SAR-58, Rev. 2," Structural Assessment Methods for Gaseous Diffusion Plant Piping Systems," April 1996

. DAC-19045-CCA-13,"Walkdown Package for Building C-337A at Paducah"

. DAC-19045-CCA-76," Building C-337A Process Piping and Equipment NPli Evaluation," April 2,1996

. DAC-19045-CCA-83," Building C-337A Process Piping and Equipment NPH Evaluation," May 23,1996

. ESO-CE848401,"C333 Process Bldg NPH Eval," September 30,1994

. Purchase Requisition #ES-CMA-15781-Al, January 6,1983 )

l

.- Interoffice' Memo, Allen Burnett to David Stadler,"NRC Inspector Questions,"

September 17,1998,14:34 CST

-- e , * = = v-