ML20203A429
| ML20203A429 | |
| Person / Time | |
|---|---|
| Site: | Fort Calhoun  |
| Issue date: | 01/26/1998 |
| From: | NRC (Affiliation Not Assigned) |
| To: | |
| Shared Package | |
| ML20203A377 | List: |
| References | |
| NUDOCS 9802240032 | |
| Download: ML20203A429 (7) | |
Text
p uou O
UNITED STATES j
NUCLEAR REGULATORY COMMISSION
(
WASHINGTON, D.C. acesMoot
\\,...../
l SAFETY EVALUATION BY THE nFFICE OF NUCLEAR REACTOR REGULATION RELATED TO AMENDMENT NO.183 TO FACILITY OPERATING LICENSE NO. DPR-40 OMAHA PUBLIC POWER DISTRICT EORT CALHOUN STATION. UNIT NO.1 DOCKET NO. 50-285
1.0 INTRODUCTION
By application deted January 9,1995, as supplemented by letters dated October 17,1996, and January 26,1998, Omaha Publi : Power District (OPPD) requestec changes to the Technical Specifications (Appendix A to Facility Operating License No. DPR-40) for the Fort Calhoun Station, Unit No.1. The requested chances would modify Sections 2.22 and 3.1, Table 3-3, item 29 of the plant's technical specifications (TS) by deleting some of the requirements for the toxic gas monitoring system (TGMS). This system was instslied as a part of the post-TMI requirements c' NUREG-0737, Item 111.D.3.4, " Control Room Habitability." However, it proved to be very cumbersome to maintain in operatYg conditions. Also, most of the toxic chemicals stored on-site were removed and the main function of the TGMS was limited to monitoring accidental releases of hazardous chemicals that occurred in the vicinity of the plant. The licensee proposed, therefora, that as long as it can be demonstrated that the accidental re' eases of hazardous chemicals occurring inside the plant and or in its vicinity will not endanger the plant's sefety, the requirement for having the TGMS could be deleted. The licensee justified its request by providing analyses to demonstrate that the potential releases of the hazardous chemicals inside or outside of the plant either will not result in the conditions endangering Ks safety, or the probability of their occurrence will be below the level considered by the NRC to endanger safety of the plant.
The October 17,1996, and January 28,1998, supplemental letters provided additional clarifying information and did not change tha initial no significant hazards consideratin determination published in the Federal Registar on March 1,1995 (60 FR 11137 ).
I 9802240032 980124
{DR ADOCK 05000285 PDR
S t'
. t 2.0 EVALUATION 2.1 B :ckground The TGMS in the Fort Calhoun plant is required to monitor, in all operating modes, relesses of five hazardous chemicals: chlorine, hydrochloric acid, sulfuric acid, hydrazine and ammonia.
Currently, chlorine and ammonia have been removed from the plant site and the only chemicals left are sulfuric acid, hydrazine and ethanolamine, which is a new chemical used for vater treatment. The licensee is planning to remove sulfuric acid in the near future. However, within
- a five mile radius of the plart there is a number of sources of hazardous chemicals which when accidentally released could endanger the plant's safety. ~ These chemicals are either stored or transported by rail or on highways. Some of these chemicals are stored or transported to and from the Cargill Corn Milling and Pacessing Facility located in the vicinity of the plant. They include: sulfur dioxide, ammonia, carbon dioxide, hydrochloric acid, sulfuric acid, gasoline, and ethanol. The consequences of accidental release of tt ese chemicals on control _ room habitability have been evaluated h the licensee's Mmittal of August 19,1994, which was reviewed and approved by the NRC. None of these chemicals could cause hazardous
- conditions in the planti in the present submittal, in addition to the hazardous chemicals stored
_ on the plant's site, the licensee evaluated al! hazardous chemicals which were transported or stored within a five-mile radius of the plant and were not included in the August 19,1994 submittal. The following chemicals are evaluated:
. ammonia
. chlorine hydrazine e
ethanolamine sulfuric acid
=
2.2 Deterministic Analygg Using the computer code TOXIC 5.2, the licensee has calculated the rates of toxic gas buildup in the control room foWowing an accidental release of the above hazardous chemicals stored or transported in the vicinity of the plant. The staff has performed independent verification of the licensee's calculations using the HABIT code and found that the licensee's models included safety margins that provided results with a significant degree of conservatism.
2.2.1 Ammonia Ammonis is stored in several facilities within a five mile radius from the plant. The largest tank, holding 25,000 tons, is located 1.8 miles from the plant. The deterministic analysis performed by the licensee has indicated that release of ammonia, caused by tank failure, will result in a buildup of ammeda concentration in the control room which would exceed toxic limit in a time shorter than 2 minutes. For this case, therefore, the licer.cze performed a probabilistic analysis to determine the risk caused by the accidental release of emmonia.
Ammonia is also transported on highways and by railroad within a five-mile radius from the plant. The largest single loads for highway and railroad transportation are 40,000 lbs and 78 tons, respectively. The deterministic analyses performed by the licensee have ind%ted that in both these cases, accidental release of ammonia during an accident will cause the concentration in the control room to exceed toxic limits in less than two minutes. Therefore, the licensee performed a probabilistic analysis for these cases.
In addition, ammonia was transported by two pipelines passing 1.8 miles from the plant. The licensee performed a deterministic analysis of conseriuences of ammonia release, ass'aming break of the larger diameter pipeline. The analysis los indicated that the resulting aminonia concentration in the control room would never reach toxic concentratior', levels.
2.2.2 Chlorine The only source of chlorine within a five-mile radius from the plant is chlori te transported by railroad in 90 ton containers. The deterministic analysis, performed by the licensee, has indicated that an accidental release of chlorine from the container would result in unacceptable concentrations in the control room. Therefore. the licensee performed a probabilistic analysis for this case.
2.2.3 Hydrazine Thirty-five percent hydrazine solution is stored inside the plant in a 365 gallon container located 218 feet from the ventilation system.nir intake to the centrol room. Another 300 gallon 4
container of hydrazine solution is stored in the warehouse. The licensee has demonstrated, using deterministic analyses, that catastrophic failure of the container with instantaneous release of twica its contents would not result in formation of tc4c concentrations at the ventilation air inlets.
2.2.4 Ethanolamine Forty percent ethanolamine solution is stored inside the plant in a 365 gallon container, located 150 feet from the ventilation systein air intake to the control room. Another 300 gallon container of ethanoLmine solution is stored 'n the warehouse. The licensee performed a deterministic analysis using s!milar assumptions as for hydrazine and has demonstrated that concentration of the chemical at the ventil&Nn inlet to the control room was considerably below toxic limit for ethanolamine.
2.2.5 Sulfuric Acid Sulfuric acid is stored in a 225,000 gallon tank inside the plant. How2ver, several factors make it impossible to affect safety of control room operators during its accide ital release. A cement brem under the tank limits the aree of spilled acid. In addition, at the temperatures existing at the plant site, 'tapor pressures of sulfe.ic acid are extremely low, hence the rate of evaporation is negligible. The licensee concluded, tha' the presence of sulfuric acid, until its projected removal, does not pose any danger to the operation of the plant. The staff cgrees with the licensee's conclusion.
1
4 2.3 Probabilistic Analyses Deterministic analyses have indicated that the accidental releases of ammonia and chlorine l
outside of the plant site can potentially cause unacceptably high concentration of toxic gases in the control room. The licensee evaluated the potential hazard caused by these releases using a probabilistic approach. The study conc uded that the calculated risk is small and supports removal of the monitors for ammonia and chlorine. In order to verify the licensee's conclusions, the staff requested additional information which was provided in a letter dated October 17, 1996. The staff's evaluation of the licensee's probabilistic analyses is based on the information provided in this letter as well as in the original submittal.
The staff has a concern with the large quantity of ammonia stored at the Terra Nitrogen Company (a staalier amount is also stored at Cargill Com Processing Plant) within 5 miles of the Fort Calhoun Station. Ammonia is also transported via railroad and highway. Chlorine is transported by railroad within 5 miles of the plant but is not stored in large quantities near the plant. The potential toxic gas release scenarios from storage tanks and transportation accidents are evaluated belcw based on the guidance and probabilistic acceptance criteria in the NRC Standard Review Plan (SRP) 2.2.3 and DG-1061 guidance.
2.3.1 Storage Tanks At Terra At the Terra facility, ammonia is stored in two 25,000 ton atmospheric tanks and two 30,000 gallon pressurized tanks. The Cargill facility stores ammonia in a 15,000 pound tank. The licensee's analysis used the generic failure rate data for single-walled tanks from the Environmental Protection Agency's (EPA)" Handbook of Chemical Hazard Analysis Procedures," dated 1989. The Terra tanks, although considered as single-walled, have an exterior wall to hold the tank insulation. In response to the staff's request for additional information, the licensee performed an analysis for tank unloading operations at both facilities, and a seismic analysis for the tanks at Terra. The licensee determined that tank unloading operations met the NRC deterministic c'iteria for control room habitability. However, seismic events could cause a common mode failure of the Terra ammonia storage tanks which results in toxic levels being exceeded in the Fort Calhoun control rocm in less than two minutes.
Seismic events are significant since their impact on off-site storage tanks correlate with the need for operator response at the Fort Calhoun Station. For inc tanks at Terra, the licensee performed seismic capacity evaluations using the same methodology and team that performed the Individual Plant Examination of Extemal Events (IPEEE) seismic capacity estimations. The study concluded that the High Confidence Low Probability of Failure (HCLPF) for these tanks is in the 0.1g range, which correlates to a mean annual probability of exceedance of 6.7d-5. This probability is based on a site specific hazard estimation from an EPRl/SOG study of the Eastem United States Seismicity issue. The licensee predicted that concurrent seismic tank failures result in a loss of control room habitability frequency of approximately 1E-6/yr. The staff determined that this prediction is subject to substantial uncertainties. (The loss of control room habitability frequency due to other tank failure modes was also approximately 1E-6/yr.)
s i
- In the Fort Calhoun IPEEE study, seismic-induced loss of off-site power (LOOP) is assumed to occur with certainty (i.e., probability = 1.0) for HCLPF seismic events in the 0.1g range. The IPEEE study details the equipment affected and the plant modifications to improve equipment reliability for HCLPF st smic events in the 0.1g range and lower. However, operator actions to mitigate the consequene.; of an accident will still be required during a seismic event which results in a LOOP. Thus, for the scenario under consideration in which a seismic event results in a large release of ammonia such that the operators are physically incapacitated and incapable of mitigating the consequences of an accident, the ability of the Fort Calhoun Station to achieve safe shutdown is substantially degraded at a time a LOOP transient is in progress.
In the event of a large leak at Terra or other facilities, the licensee indicated that the Blair industrial Park emergency notification system would ensura the event impact at the Fort Calhoun Station would oe investigated. Notification of the operators that a toxic ammonia gas release had occurred would provide operators with time to take proper actions to maintain control room habitability; however, the reliability of this notification fol lowing a seismic event has substantial uncertainty.
Section 2.2.3 of the Standard Rewew Plan (SRP) indicates that when there is difficulty in assigning accurate numerical values to the expected rate of unprecedented potential hazards, judgment must be used as to the acceptability of the overall risk presented. Although the frequency of a 0.'i HCLPF seismic event is low, the plant response is expected to be complex.
There would be substantial operator burden and stress associated with the initiating event (seismic induced LOOP) impacting the likelihood that operators would detect and respond to toxic gas in the control roor'. Therefore, little operator credit can be given to reduce the conditional core damage probability (CCDP), given a seismic induced LOOP concurrent with the loss of control room habitability.. Inability to monitor for ammonia with the TGMS will adversely impact control room habitability and control room recovery operations for these types of events. Although the frequency of such events are small, the staff cannot conclude that SRP 2.2.3 criterion of aporoximately 1E-7/yr or DG-1061 criteria for small change in risk are met.
This is due to seismic concem about the large amount of ammonia stored in the plant's neighborhood, the substantial uncertainties associated with the estimate of loss of control room habitability frequency under seismic conditions, and the role played by the TGMS as a defense in depth measure.
2.3.2 Transportation Accidents The staff noted that the licensee estimates of loss cf control room habitability frequencies for truck and railroad transportation of ammonia had decreased from those reported in the licensee's 1981 study. The 1981 study showed that it was 2.6E 5/yr for truck shipments of ammonia, and 1.16E-5/yr for railroad shipments of ammonia. The licenseet current estimate is 1.7E-6/yr for ammonia truck accidents and 1.45E-8/yr for ammonia railroad accidents. The observed decreases in the loss of control room habitability frequency is supported by current data and our understanding of the likelihood of such events.
i
. I Chlorine is shipped on the mainline rail at an expected rate of 190 rail cars of 90 ton size per year. The loss of control room habitability frequency for chlorine rail accidents is estimated by the licensee to be 2E-7/yr. As described below, the small likelihood is consistent with current data and our understanding of these types of events.
The methodology used for calculating the loss of control room habitabi;ity frequency accounted for accidents rates, spill size, distance from the plant, and weather conditions. The accident rate data, for both truck and train accidents, was obtained from the EPA's "Pandbook of Chemical Hazard Analysis Procedures," dated 1989. For train accident rates, the staff noted that the national train accident rate trend has been declining since 1978 based on Federal Railroad Administration data collected through 1995 in the August 1996 publication of the Accident / incident Bulleti.' No.164. The meteorological data used in the licensee's analysis are current and site-specific. The meteorological data was collected from semi-annual radioactive effluent release reports for the time periods from 1982 tnroagh 1991. The staffs review of the licensee's analysis found that the methodology and the data used in the analysis were.
reasonable.
The licensee's submittal also estimated a core damage frequency (CDF) for truck and train
- accidents which rssult in a large spill near the plant. The CDF is the prcduct of the loss of control room habitability frequency and the CCDP, given the loss of control room function. The licensee er" mated the core damage frequency to be well below 1E-7/yr for these potential accidents. Due to the expected large uncertainty in the CCDP, the staii was not able to independently confirm the licensee's CDF estimates reported in the Ja uary 9,1995 submittal and the response to the request for additional information. However, the loss of control room habitability frequency is expected to be very small for truck and train accidents, and the CCDP less than unity. Therefore, the staff concludes the acceptancs criteria of SRP 2.2.3 is met for potential transportation accidents involving ammonia and chlorine Based on its evaluation, the staff concludes that the licensee has provided acceptable deterministic justifications for deleting the requirements for monitoring for hydrazine,
. ethanolamine and sulfuric acid. The TGMS requirements for these chemicals can be removed i
from the TS for the Fort Calhoun Station. In addition, PRA insights support the removal of the -
requirements for monitoring for chlorine, consistent with draft SRP 2.2.3 and DG-1061 guidance.= Howevar, the s'aff found that PRA insights and the need for defense-in-depth protection against a large rolesse of ammonia gas near the site, do not support the removal of the existing TGMS TS for ammonia. Monitoring requirements for this chemical should remain in the plant's TS.
3.0 STATE CONSULTATION
in accordance with the Commission's regulations, the Nebraska State official was notified of the proposed issuance of the amendment. The State official had no comments.
_---_--------J
4.0 ENVIRONMENTAL CONSIDERATION
The amendment changes e coquirement with respect to installation or use of a facility component located within the 'estricted area as defined in 10 CFR Part 20. The NRC staff has determined that the amendment involves no significant increase in the amounts, and no significant change in the types, of any effluents that may be released of' site, and that there is no significant increase in individual or cumulative occupational radiation exposure. The Commission has previously issued a proposed finding that the unendment involves no
'significant hazards consideration, and there has been no pubiu comment on such finding (60 FR 11137). Accordingly, the amendment meets the eli ibility crite-ia for categorical 2
exclusion set forth in 10 CFR 51.22(c)(9). Pursuant to 10 CFR 51.22(b) no environmental impact statement or environmental assessment need be prepared in connection with the issuance of the amendment.
5.0 CONCLUSION
The Commission has concluded, based on the considerations discussed above, that: (1) there -
is reasonable assurance that the health and safety of the public will not be endangered by operation in the proposed manner, (2) such ac'Jvities will be conducted in compliance with the Commission's regulations, and (3) the issuance of th_e amendment will not be inimical to the common defense and security or to the health and safety of the public.
Principal Contributors : IK. Parczewski-D. O'Nea!
Date: January 26, 1998:
1
____