ML20086Q100

Proposed Tech Specs,Deleting Requirement for Chlorine Detection Sys from Tech Spec 3.13

ML20086Q100
Person / Time
Site:	Prairie Island
Issue date:	12/13/1991
From:	NORTHERN STATES POWER CO.
To:
Shared Package
ML20086Q086	List: ML20086Q083 ML20086Q091 ML20086Q100
References
NUDOCS 9112270382
Download: ML20086Q100 (57)

Text

_ . . . _ _ _ .. _ _ _ _ .

4 -

4 Exhibit B PRAIRIE ISl AND NUCLEAR GENERATING PLANT License Amendment Reauest dated December 13. 1991 MARKED UP TECllNICAL SPECIFICATION PAGES Exhibit B consists of marked up pages for' the Prairie Island Nuclear Generating Plant Technical Specifications with the proposed changes incorporated as listed.below:

EDE&

TS-iv TS.3.13-2 B.3.13 1 f

9112270332 911213 PDR ftDOCi: 03000232 P. pga

, . . , ~. -

TS-iv REV 94 3/20/91 TABLE OF CONTENTS (Continued)

TS SECTI2H III,13 PAGE 3.10 Control Rod and Power Distribution Limits TS.3.10 1 A. Shutdown Margin TS.3.10 1 B. Power Distribution Limits TS.3-.10 1 C. Quadrant Power Tilt Ratio TS.3.10-4 D. Rod Insertion Limits TS.3.10-5 E. Rod Misalignment Limitations TS,3.10 6 F. Inoperable Rod Position Indicator Channels TS.3.10 6 G. Control Rod Operability Limitations TS.3,10-7 l

H. Rod Drop Time TS.3.10-7 I. Monitor Inoperability Requirements TS.3.10-8 J. DNB Parameters TS.3.10 8 3.11 Core Surveillance Instrumentation TS.3.11-1 3.12 Snubbers TS.3.12-1 3.13 Control Room Air Treatment System TS.3.13 1 A. Control Room Special Ventilation System TS,3.13-1 D. Chleriuo Deteet4+n4yattms TS.3.13 3.14 Fire Detection and Protection Systems TS.3.14-1 A. Fire Detection Instrumentation TS.3.14-1 B. Fire Suppression Vater System TS.3.14-1 C. Spray and Sprinkler Systems TS.3.14-2 D, Carbon Dioxide System TS.3.14-3 E. Fire Hose Stations TS.3.14 3 F. Yard Hydrant Hose Houses TS.3.14 4 G. Penetration Fire Barriers TS.3.14-4 3.15 Event Monitoring Instrumentation TS.3.15-1 A. Process Monitors TS.3.15-1 B. Radiation Monitors TS.3.15-1 C. Reactor Vessel Level Instrumentation TS.3.15 - - -

1 l TS.3.13-2 REV 91 10/27/89 3.B. Chloribe Detection Systems l

Two independent chlorine detection systems, each consisting of two t annels of instrumentation shall be OPERABLI at all times except l as ecified below. The alarm / trip setpoint shall be adjusted to actuat at a chlorine concentration of less than or equal to 5 ppm.

1. If one h Qorine detection channel for one train of ventilation is inoperabi , then within 7 days: l

a. Rescore the operable channel to OPERABLE status, or j

b. Operate the redun aqt ventilation system in the normal (non recirculation) m Qe, and close the cutside air supply dampers for the affected crain of ventilation.

2. If both chlorine detection channe - for one train of ventilation are inoperable then within 6 hour6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />s: I

a. Restore at least one channel to OP LE status, or l

b. Operate the redundant ventilation system in he normal (non-recirculation) mode and close the outside it supply dampers for the affected train of ventilation.

~

3. If all chlorine monitors for both trains of ventilation a

(. inoperable then within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> close all Control Room venti -

[

tion outside air. supply dampers.

l 4

1

. . . _ _ _. .__ _ _ _ __ _ .. . _ . . ~ . _ . _

B.3.13 1 REV 91 10/27/89 3,13 CONTROL ROOM AIR TREATMENT SYSTEM papes The Control Room Special Ventilation System is designed to filter the Control Room atmosphere during accident conditions. The system is designed to automatically start on a high radiation signal in the ventilation air or when a Safety Injection signal is received from either unit. _Two completely redundant trains are provided.

Each train has a filter unit consisting of a prefilter, HEPA filters, and charcoal adsorbers. The HEPA filters remove particulates from the Control Room. atmosphere and prevent clogging of the iodine adsorbers.

The charcoal adsorbers are installed to remove any radioiodines from the Control Room atmosphere.

'l?t PERABILITY of the chlorin tection system ensures that suffi- I cier.t . bility is available to pro detect and initiate protec-tive action the event of an accidenta orine release. This capability is req ed to protect the control r ersonnel and is consistent with the re mendations of Regulatory Gu'iBt 95 "Protec-tion of Nuclear Power Plant ontrol Room Operators Against a ciden-tal Chlorine Release" February .

The ol Room Special Ventilation System ns OPERABLE if the l s

ventilation em can be operated in the recircu Lqn mode.

m,. --...m.. - . - , . , _ , . , , . ,- . _ . . , , , ,.,4-

b' .) f Exhibit C PRAIRIE ISLAND NUCLEAR GENERATING PLANT License Amendment Reauest dated December 13. 1991 REVISED TEClINICAL SPECIFICATION PAGES

.i Exhibit C consists of revised pages for the Prairie Island Nuclear Generating Plant Technical Specifications with the-proposed changes incorporated as listed.below:

EBZ9.

TS iv B,3.13-1 l

l

_ - - . . . _ - . _ ._. .._ ._ _. _ , . . . . _ . _ . _ . . _ _ . . ~ _ _ . . _ . _ . .

o.

TS iv 1

TABLE OF CONTENTS (Continued) )

g TS SECTION TITLE PACE I

l 3.10 Control Rod and Power Distribution Limits TS.3.10-1  :

A. Shutdown Margin TS.3.10-1 l B. Power Distribution Limits TS.3.10 1 l l C. Quadrant Power Tilt Ratio TS.3.10-4 1 D. Rod Insertion Limits TS.3.10-5 E. Rod Misalignment Limitations TS.3.10-6 i

.F. Inoperable Rod-Position Indicator Channels TS.3.10-6 l C. Control Rod Operability Limitations TS.3.10-7  ;

H. Rod Drop Time TS.3.10-7

1. Monitor Inoperability Requirements TS.3.10 8 J. DNB Parameters TS.3.10-8 l 3.11 Core Surveillance Instrumentation- TS.3.11-1 i 3.12 Snubbers TS.3.12-1 l

3.13 . Control Room Air Treatment System TS.3.13-1  !

l .A. Control Room Special Ventilation System .TS.3.13-1 g  !

1; :3.14 -Fire-Detection and Protection Systems TS.3.14-1 1 l A. Fire Detection Instrumentation TS.3.14-1 '

l B. Fire Suppression Water System TS.3.14-1 L .C. Spray and Sprinkler Syste. , TS.3.14-2 D. Carbon Dioxide System TS.3,14-3 l E. Fire Hose Stations TS.3.14-3 F. Yard Hydrant Hose Houses TS.3.14-4 C. Penetration-Fire Barriers TS.3.14-4 L 3.15- Event Monitoring Instrumentation TS.3.15-1 A Process Monitors TS.3.15-1 l 8. Radiation Monitors TS.3.15-1 E C.~ Reactor Vessel' Level Instrumentation TS.3=15 L.

I' L

l l

. - , ~ -. _ .... - ..- _ -. . - ... - _ . - , ,, , _ . - . _ , , . . , , ,. . ,, .-. -- , ,-

)

B.3.13-1 I

3.13 CONTROL ROOM AIR TREATMENT SYSTEM h.0lta The Control Room Special Ventilation System is designed to filter the Control Room atmosphere during accident conditions. The system is designed to automatically start on a high radiation signal in the ventilation air or when a Safety Injection signal is received from either unit. _Two completely redundant trains are provided.

Each train has-a filter unit consisting of a prefilter, HEPA filters, and charcoal adsorbers. The HEPA filters remove particulates from the Control Room atmosphere and prevent clogging of the iodine adsorbers. The charcoal adsorbers are installed to rem *ve any radiciodines from the Control Room-C atmosphere.

l l

4

w. m- -e- -e-> e -,-

• _ . .m.....___... _ _ . - . _ . _ _ _ _ . . _ . . _ _ . . _ . . . _ _ . . . . _ _ _ . . _ . _ _ . _ . . _ . _ . _ _ . . _ _ . . .

'l

~

Exhibit D PRAIRIE ISLAND NUCLEAR GENERATING PLANT License Amendment Reauest dated December 13. 1991

-Control Room Habitability Toxic Chemical Study 9

d P

l^

I:

i

..a.- ,_ r _ ,. -- .-m, ,..4, ym,.mm,... . - , _

i NORTHERN STATES POWER COMPANY PRAIRIE ISLAND NUCLEAR GENERATING PIANT Control Room IIabitability Toxic Chemical Study Prepared by TENERA,LP.

1221 Nicollet Mall, Suite 320 Minneapolis, MN 55403 612 340-9228

. _ -. _.__m._ . . . _ _ _ _ . _ . - _ . . _ _ . _ _ . _ _ . _ _ _ _ _ _ . _ . _ _ _ _

._ Nonbern States Power Company Control Room II:bitability Prairie Ist:nd Nucle:r Generating Pt:nt Tosie Chemic:t Study

?

TABLE OF CONTENTS SECTION- PAGE EXECUTIVE SUht hiARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii 1.0 INTR O D UCTIO N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . 11 2.0 TOXIC CHEhilCAL STUDY hiETHODOLOGY AND SURVEY RESULTS 21 '

2.1 NRC Regulations for Control Room Habitability . . . . . . . . . . . . . . . . . . 21 '

2.2 Toxic Chemical Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 9 2.3 . Toxic Chemical Spill hiodels . . . , . . . . . . , , . . . . . . . . . . . , , . . . . . . . . 2 15 2.4 Incapacitation Assessment Criteria , . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 3.0 FINAL R ES U LTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1

. 3.1 - Toxic Chemicals Stored Onsite . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . 3-1 3.2 Toxic Chemicals Stored Within Five hiiles of the Plant . . . . . . . . . . . . . 31 3.li Toxic Chemicals Transported by Truck . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2 3.4 Toxic Chemicals Transported by Rail . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2 3.5 Toxic Chemicals Transported by Barge , . . . . . . . . . . . . . . . . . . . . . . . . . 3-11 '

3.6 Ammonia and Chlorine Spill Releases - Probabilistic Results . , . . . . . . . 3-11 4.0 RE FE R EN C E S , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 - 1 4,1 t1RC Regulations and Other Regulatory Requirements . . . . . . . . . . . . . . , 41 4.2 . Toxic Chemical Survey Information . .-. . . . . . . . . . . . . . . . . . . . . . . . 4 2 -

- 4.3 Technical and Descriptive References . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2 19EMR120/20 i TENERA

, Nonheca States l'uwer Compuy Control Room Iltbit:bility Prairie ist:nd Nucleer Generating I'!:nt Tetic Chemical Study LIST OF TAllLES Table 2-1 Regulatory Requirements for Control Room Habitability . . . . . . . 2 3 Table 2 2 1-lazardous Chemicals Stored Onsite at Prairie Island that Exceed SARA Reportable Limits . . . . . . . . . . . . . . . . . . . . . . . . 2 10 Table 2-3 City of Red Wing Major Industry . . . . . . . . . . . . . . . . . . . . . . . . . 2-11 Table 2-4 Chemicals Transported By Rail On The SOO Railroad Line 30 Shipments or More Per Year . . . . . . . . . . . . . . . . . . . . . 213 Table 2-5 Chemicals Transported by Rail on the Durlington Northern Line 30 Shipments or More Per Year . . . . . . . . . . . . . . . . . . . . . 2-14 Table 3-1 Final Results for Surveyed Chemicals . . . . . . . . . . . . . . . . . . . . . 3-14 19EMR120/20 ii TENERA

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

Northern St:tes Power Company Control Room liabitability Prairie Island Nuclesr Generating Plant Toxie Chemical Study EXECUTIVE

SUMMARY

Due to recent design changes at the Prairie Island Nuclear Generating Plant (PINGP),

chlorine is no longer stored onsite and regulations requiring early warning of onsite chlorine releases no longer apply. NSP recognized that removal of onsite chlorine may climinate the need for the control room llVAC chlorine detectors. NSP also realized that the detectors were installed in response to a Control Room liabitability Study based on survey results now ten years old. NSP decided to revise the sun'ey and reassess the need for toxic chemical detectors. The results of this reassessment are contained in this report.

A new toxic chemical survey was performed which identified toxic chemicals either stored onsite in sufficient quantities or shipped near the plant at sufficient frequencies to warrant further evaluation. These toxic chemicals were evaluated in accordance with applicable regulatory requirements. It was deterministically concluded that all chemicals stored onsite and transported near the plant, with the exception of SOO Railroad Line chlorine and anhydrous ammonia (ammonia) shipments, do not pose a significant threat to control room operators. No early detection equipment is required for postulated chemical releases as sufficient time (at least two minutes) is available for the control room (CR) operators to don protective breathing equipment.

For the case of SOO 1.ine railcar releases of chlorine or an"nonia, sufficient time could not be demonstrated using the conservative regulatory guidance for all possible combinations of weather conditions and distances from the spill to CR intake. For such releases, a probabilistic model was developed which accounts for the frequency of various weather conditions and the likelihood of a chlorine or ammonia railcar accident which results in a toxic chemical release. Calculated probabilities were compared to the criteria of Standard Review Plan (SRP) Section 2.23 [Ref. 4.1.5] and USNRC Regulatory Guide (Reg. Guide) 1.70 [Ref. 4.1.12]. Reg. Guide 1.70, Section 2.23.1 states: " Design basis events external to-the nuclear plant are defined as those accidents that have a probability of occurrence on the order of about 104per year & 3reater and have potential consequences serious enough to affect the safety of the plant to tue extent that Part 100 guidelines could be exceeded." The SRP indicates that offsite hazarcous releases need not be considered if "a conservative calculation showing that the probability of occurren e of potutial exposures in excess of the 4

10 Part 100 guidelines is approximately 10 per year..." This "is acceptable if when combined with reasonable qualitative arguments, the realistic probability can be shown to be lower." The probability of either a chlorine or ammonia spill resulting in control room operator incapacitation to the extent that Part 100 guidelines could be exceeded is approximately 104 per year. This probability was determined with multiple conservatisms in the analysis approach. Therefore, the acceptance criteria of both the Reg. Guide and i SRP were demonstrated.

l l

L Based on this study, potential hazardous releases from all identified sources in the new l survey need not be considered in the design of the PINGP and no special CR HVAC i detectors are required, l

l l

l 19EM R120/20 ni TENERA

North:rn States Power Company Control Room liibitability Prairie Islad Nucle:r Gener: ting PI:nt Toxic Chemical Study

1.0 INTRODUCTION

Draft Atomic Energy Commission (AEC) 1967 General Design Criterion (GDC) [ Ret.

4.1.11) Number 11 requires facilities to provide a " control room from which actions to maintain safe operational status of the plant can be controlled." The criterion also requires that the control room provide adequate radiation protection of personnel during both normal and accident conditions. This criterion is part of the Prairie Island Nuclear Generating Plant (PINGP) operating license.

Following the Three Mile Island (TMI) accident, the Nuclear Regulatory Commission (NRC) reinforced the requirement of adequate CR operator protection by requiring all plants to perform a Control Room Habitability Study in response to Item llI.D.3.4 of NUREG-0737 [Ref. 4.1.2]. This study was to evaluate both the effects of postulated toxic chemical spills and post. accident radiation on CR operators. For toxic chemicals where it could not be demonstrated that adequate time exists for CR operator response, special design features such as early chemical detection and CR HVAC automatic isolation were to be provided.

Bechtel Corporation was contracted by NSP to perform the Control Room Habitability Study (Original Study) in response to NUREG-0737 requirements. The Original Study

[Ref. 4.3.1] was completed in 1981 and concluded that four chemicals either shipped on nearby railcars or stored onsite posed a significant enough risk to require detection and automatic CR HVAC isolation in 1984. Bechtel was again contracted by NSP to re-evaluate the incapacitation effects of the four chemicals using more current modeling techniques [Ref. 4.3.1]. It was concluded that only one of the original four chemicals, chlorine, still required early detect..,n. The principal source of chlorine was onsite as it was used in the water treatment process.

Due to recent design changes [Ref. 4.3.3) at the PINGP, chlorine is no longer used in water treatment and is no longer stored onsite. Therefore, regulations requiring early wtrning of potential onsite chlorine releases are no longer applicable. NSP recognized that such a design change may eliminate the need for the chlorine detectors. NSP also realized that the original study was based on a survey close to ten years old. Therefore, it was decided to revise the survey and reassess the need for toxic chemical detection.

This report presents the results of thh reassessment. The implemented approach is in accordance with applicable regulatc:y requirements as outlined in Section 2.1 below.

The remaining parts of Section 2.0 describe the analysis approach and present the survey results. Final incapacitation assessments are presented in Section 3.0.

References are itemized in Section 4.0.

l

{ 19fMR120/19 11 TENERA

.. ._ , .. __ - __ _ _ . . _ _ _ . ~ _ _ ._ _ - _ _ _ . _ _ . _ ,_

,. Northern States Power Company Co: trol Room lisbitability Prairie Islaid Nucle:r Ge:erating I't:nt : Tosic Chemic:t Study

. 2.0 TOXIC CIIEMICAL STUDY METilODOLOGY AND SURVEY RESULTS The toxic chemical study approach consisted of the following steps:

. Define regulatory requirements,

. Perform a survey to identify hazardous materials,

- Develop a toxic chemical spill analysis model,

. Define incapacitation assessment criteria, and

. Assess the effects of toxic chemical spills on control room operators.

' The first four steps, including the results of the survey, are presented below. The effect of various toxic chemical spills on CR operators are presented in Section 3.0, 2.1 NRC Regulations for Control Room liabitability The first step of the study was to review current and past regulations on control room habitability and their applicability to the PINGP license. The objective of this review was twofold: 1) to ensure the approach implemented complies with all applicable regulations, and 2) to determine if any recent regulatory changes have occurred which might provide improved criteria upon which to base this study.

Results 'of both objectives are -presented below. In summary, the modeling

. techniques employed have not changed since the Original Study,-however, additional tools are now available to model the probabilistic aspect of chemical-spills. As shown in Section 3.0, such tools were employed in the evaluation of postulated chlorine and ammonia releases.

NRC regulations on control room habitability are contained in a number of documents including NUREG-0800, Standard Review Plan (SRP) [Ref. 4.1.5), and '

various regulatory guides (see Table 21). These regulations all ultimately refer to 10CFR50 Appendix A General Design Criteria [Ref. 4.1.1] as the basis for implementation. Though PINGP is not a 10CFR50 Appendix A plant (the PINGP is a draft AEC GDC plant), control room habitability analysis is a requirement as a result of the Three Mile Island (TMI) accident. All plants were required to examine control room habitability for both toxic chemical spills and post design -

l-basis accident radiation effects. These requirements are contained in NUREG-L 0737,Section III.D.3.4.

The effects of toxic chemical vapors on control room operators depend on a number of factors including chemical type, spill distance from the control room, volume of spill, wind speed, etc. The NRC position on many of these factors is contained in two Regulatory Guides (RGs). These are RG 1.78," Assumptions for -

Evaluating the liabitability of a Nuclear Power Plant Control Room During a 19I11R120/19 2-1 TENERA

. Northern States Power Compary Control Room II:bitability Pralrie Isted N:cle:r Generating Plant Toxic Chemic:1 St:dy 4

Postulated Hazardous Chemical Release," Jime 1974 [Ref. 4.1.3), and RG 1.95,

" Pre;ection of Nuclear Power Plant Control Room Operators Against an Ac;ldental Chlorine Release," Revision 1, January 1977 [Ref. 4.1.4]. The accepted NRC model for calculating toxic chemical spill concentrations in the control room is contained in NUREG-0570, " Toxic Vapor Concentrations in the Control Room Following a Postulated Accidental Release," June 1979 [Ref. 4.1.6]. Additionally, modeling techniques for control room operator incapacitation are contained in NUREG/CR-1741, *Models for the Estimation ofIncapacitation Times Following Exposures to Toxic Gases or Vapors," December 1980 [Ref. 4.1.7]. The calculations generated for this study use the criteria of RG 1.78 and the models of NUREG-0570 and NUREG/CR-1741 to select which hazardous materials to model and to assess the potential for CR operator incapacitation.

The probabilistic aspect of the chlorine and ammonia calculations used two NUREOs, issued since the Original Study, for guidance. These are NUREG/CR-5042, " Evaluation of Egrnal Hazards to Nuclear Power Plants in the United States," December 1987 [Ref. 4.1.10), and NU REG /CR 2650," Allowable Shipment Frequencies for the Transport of Toxic Gases Near Nuclear Power Plants,"

October 1982 [Ref. 4.1.8].

Table 2-1 provides an overview of the existing regulatory documents for control room habitability and discusses the specific requirements of these documents. The table ident!'ies which documents are part of the PINGP license as requirements or as guidance used to meet the requirements.

i l

l i

19EMR120/19 22 TENERA

h e t _J m Control Room Itaisitability Northern States Power Company Toxie Chemical Study .

Prairie Island Nuclear Generating Plant Table 2-1 NRC Regulatory Requirements for Control Room Itafiitability If not PINGP Commitment used as Guidance? Oserview of Requirements Regulatory Document Commitment Yes -- Per 10CFR5034 a3.i, all nuclear power plants Atomic Energy Commission (AEC) shall be designed and evaluated during the license Draft General Design Criteria j application against a set of Principal Design Criterion 11 - Control Room: The facility shall be Criteria. As PINGP design was compl-te prior to i l provided with a control room from which actions to the issuance of 10CFR50, Appendix A, veneral l uaintain safe operational status of the plant can be Design Criteria, these criteria are not part of the controlled. (Abridged) plant license. A draft version of these criteria was issued in 1%7 by the Atomic Energy Commic%

for redew and comment. This draft criteria set i form the Principal Design Criteria for the PINGP. l Criterion 11 defines requirements for a control room with a safe operating environment. Later regulations, such as RG 1.78 invoke 10CFR50 Appendix A as the basis for their application.

Though this implies such regulatory guides are not part of the plant license, the election to invoke such requirements ensures the intent of the Draft AEC GDC and later NUREG-0737 requirements, which apply to all plants, are met.

n TENERA

-,u mn, 1

Northern States Power Company Controi Ree IIabitability Prairie Island NucIcar Generating Plant Toxic t,nemical Study o

2-1 (Continued)

NRC Regulate., <ments for Control Room Itabitability if not PINGP Commitment used Regulatory Document Commitment as Guidance? Overview of Requirements NUREG-0737 Yes - This Toxic Chemical Study is an update of the Section ill D.3A - Control Room IIabitability Original Study completed in May of 1981 to Requirements address NUREG-0737 requirements. As a result of TMI, the NRC required a*1 licensees to re-evaluate control room habitability. This review for previously liccused plants, such as PINGP, was to focus on the effects of the accidental release of toxic gases and radioactive materials. The objective was to determine that control room operators are adequately protected in the event of such accidents, and that the plant could be safely operated or shutdown. The NUREG identified RGs 1.78 and 1.95 as acceptable guidance for meeting CR habitability analysis requirements.

RG 1.78, " Assumptions for Evaluating she IIabitability No Yes This RG provides the guidelines for performing of a Nuclear Power Plant Control Room During a the toxic chemical analysis. It specifically outlines ,

Postulated llazardous Chemical Release," June 1974 a position acceptable to the NRC for sekcting and analyzing the effects of hazardous chemical spills on CR operators.

The RG provides acceptance criteria when,if met, require no special automatic detection or IWAC cutrol equipment installation. Specifically, if there is at least two minutes between the time t!.at the toxic chemical is detected in the control room and the control room operatoss become incapacitated, no special monitoring er control equipment is required. Two minutes i:, considered sufficient time for a trained operator to don any necessary protedive breathing apparatus.

mmi, u TENERA

Nawthern States Pe=re Company C==tr=1 Renee Itabitability Prairie Istnd Nuclear Generating Plant Tesie Ctemiest Study oi Tabte 4-1 (Continued)

NRC Regulatory Requir_wents for Centrol Rtwwn Itabitability If nat PINGP Cawnmitment i. sed ,

Regulatory Document Commitment as Guidance? (heniew of Rw _.s ts RG 155, Protectkm of Nudcar Power Plant Control No -

This RG was issued to prmide guidance on,bew Room Operators Against an Accident Chknine to design acceptaNe CR operator protectkm f- wa Release," Revism 1. January 1977. pmtulated onsite chkwine releases. If chkwine is stored ensite, the RG indicates (a) the cimtrol room is to be capable of being iscla -d on a high chkwine level, (b) chivine detedors shaII be installed, (c) the chlorine detectors should meet i single failure crit-ria, be qualified Seismie Category I, and be environmentaTIy qualified "fo.

aII expected e ...u. .. cats that could clearly kao to or be a result of a chkxine release.' In i additkm, technical specificatica limits were to be specified.

'Through part of the original study, thi RG no longer applies to the PINGP s;nce chlorire is no kmger stored onsite. 'ihis is a result of a design 1 change that replaced chkwine as the biocide agent ,

for the cooling water system with a sodium bromide, sodium hypochkvite minure (see Design i Malification 89YO15, [Ref. 43.13D.

RG 1.70,' Standard Porinat and Content of Safety No Yes Though the PINGP has no expbcit commitment to Analysis Reports for Nuclear Power Plants" meet this RG,it des pres

• acceptance criteria Redsion 3, Novernber 1978. for use when evaluating new plant Final Safety -

Analysh Reports _ The RG indicates a prtability l of 10'/ year as the cutoff for constaing man- l made external event haards in power plant safety.

)

1 .

i

,,nm,f,. 1, TENERA l t

- Northers St tes Power Coenpany Ceotrol R:<,as Itabita* iiity Prairie Island Nucle:r Generating Plant Toxic Chernical Study .

Table 2-1 (Cm.tinued)

NRC Regulatory Requirements for Control Room Itabitability If nd PINGP Commitment used Regulatory Document Commitment as Guidance? Overview of Re:;uiren. cats

NUREG4MR " Standard Review Plan" (SRP), No Yes Though the PINGP has no expliot commitment to Redsion 2. July 1981 meet these sections of Ihe SRP, they do outline Secthm 2.2.1 ? ? ? Identification of Potential liarards guidance to be used by the NRC in the review of in Site Vicini

y" license arplications. IIence, they were redew -d to ensure the intent was being met. This section of the SRP requires that for ewry site, a redew l be performed to identify potential h: =ds. These harards are to include external har2rds or j harardous materials that could present a prob!cm during the lifetime of the g bnt. This review is ta encompass a five-raile ra6cs of the plant.

&ction 2.23,' Evaluation of Potential Accidents- This sect:on <( the SRP discusses the offsite toxic accidents as an initiating event that could impact control room habitability ar d potentially result in offsite dose releases tc, the general public. The intent of this secthm is to prodde guidance to determine what type of potential accidents need to be considered. This determination is based on the expected cum.ma rate for the accident. The I

obgective is to show that 7he pctential for causing onsite accidents leadine to the release of significant quantities of radactive fission

! products, and thus pose an undue risk of public  !

expowre is significantly faw. SWeantly km is defined as arz occurrence rate of approximately ,

i 10-* per year when combined mth reasonable l quaintatin arguments.

i twwR12 Dip 24 TENERA

Northern States Power Ccmpany Centrol Renen ItaMtaWiity Pra*rie Island Noelcar Generating Plant Tosie Chemical Study .

2 Table 2-1 (Continued)

NRC Regulatory Requirements Inr Centrol Rmwn IIabitaMiity j tr not PINGP Cwmitment used j Regulatory INcument Ccmmitonent as Guidance? Overview of Requirements

NUREG41570 No Yes This NUREG describes an b7C acceged

• Toxic Vapor Concentratbns in the Contn4 Room appro;xh for determining toxic chemical vapor Following a Postulated Accklental Release ~ June 19 a=tcentrations both outside and i-tside the controi

room.

NUREG/CR-1741 No Yes This NUREG remides the malcling techniques Models for the Estimation n' Incapacitation Tim s for assessing the potential for CR operator Folksaing Expmures toToxic Gases or Vapors incapacitation resuking from toxic chemical December 1980 exposure. The model . elected depends on how a the chemical effects a human and how it is ultimately detected. These modeling techniques

=cre used by Bechtel in 1%4 to justify the climination of all CR HVAC toxic chemical monitors except chkwine.

NUREG/CR-26'O No Yes This NUREG provides more realistie (less i *Alkwable Shipment Frequencies for the Transport of conservative) criteria for when evaluation cf Toxic Gases Near Nuclear Power Plants

• October Chemicals based on shipment frew.% is 1982. required. The NUREG concludes that 3

i equencies of 2fN per year for railroad traffe and 1820 per year for truck traffe are more realistic criteria than 30 and 10, respectively, as K@J by Reg. Guide L7&. The value of 0.1 kw the 2 probability of a chemical release causing eperator incapacitation resulting in exceeding 10CFRitU Iimits was taken from this ht* REG.

1 i

-,. 1. rENERA

Northem States Power Company Centrol Reeer ItaMtability Prairie Island Nuclear Generating Plant Tosie Chemical Study ,<

l Talde 21 (Continued) ,

NRC Regulatory Rew;u...n.ts for Centrol Raam liabitability i t

t

?

i I Ifont PINGP Commitment used i Regulatory Document Commitmer.t as Guidance? Oversiew of Requirements i

1  !

I NUP.EG/CR-5042 No Yes His NUREG prmides a probaNiistic evaluathm i

• Evaluathm of External llazards to Nuclear Iber method to deal with the potential for nuclear [

Plants in the United States December 1987 pmrer plant accidents initiated by external ewnts, including td chemical spills f External ewnt acceptance criteria are prmided based on the contributim of total etwe melt I accident frequency for the event. liv the ,

purposes of this study, only the railroad accident frequency data and approach for calculatkm equivalent track miles were implemented.

ProbaNristic accertance criteria were obta'med from the SRP.

i b

P t

t

-,w ,. u TENERA '

i

. Northera Stain l'outr Connpa y Controt Room llibit:btlity Pralrie Ist :d Nale:r Generating PI::t Toxic Chemle:I Study 2.2 Toxic Chemical Suncy The objective of this sun'ey was to identify all hazardous materials stored or shipped near the PINGP and to determine which materials from this survey required further evaluation. The steps employed to determine which materials required modeling for comparison evaluation and the resultant list of hazardous materials are presented below.

The first step in performing the survey was to develop a comprehensive list of hazardous materials. The Original Study used 1101.78, NUl(EG 0570 and data provided by the Association of American Itailroads and Committee on Safety of Nuclear Installations Organization to develop a list of hazardous materials. His study used both l(G 1.78 and NUllEG 0570 but used the list of hazardous ma'erials from Title 29, Code of Federal Regulations, Part 1910 [Itef. 4.1.9) in place of railroad data. This list was used as a basis for confirming that the list of hazardous materials provided by different survey contacts included chemicals defined as hazardous.

The next step in the survey was to perform a review of all fixed sites and shipping routes near the PINGP which may store or transport hazardous materials. The guidance of RG 1.78 was applied to define the scope of this review. In accordance with the Reg. Guide, any hazardous chemicals stored or shipped beyond a five-rnile radius of the plant requires no evaluation. Therefore, this review was limited to all transportation routes and fixed sites within the five mile radius. Per the PINGP Emergency Planning Zone drawing [Ref. 4.3.6) these me:

The PINCP (Section 2.2.1)

• Red Wing area industry urers of hazardous materials (Section 2.2.2)

. Truck transport along liighway 61 (Section 2.2.3)

. The SOO and Burlington Northern rail lines (Section 2.2.4)

. Mississippi River harge traffic (Section 2.2.5)

Each rail company noted above was contacted to obtain information on all hazardous materials shipped on routes within five miles of the plant, frequency of shipment, and shipment size. The local authorities, chamber of commerce and U.S. Army Corp of Engineers provided hazardous material data for all other transport routes and industry users.

Once the list of chemicals shipped and stored was obtained, the remaining crhria of RG 1.78 were applied to determine which chemicals required further evalunt.on.

These criteria include frequency of shipment, stored or shipped quantity and ter m u20/19 29 TENERA

. Northern States Power Compary Corttel Moon Ilibitability Pruirle litard Nucle:r Generatirg Pl::1 Tosic Chemical St:dy chemical toxicity. The results of this survey are presented below and form the basis for the modeling and comparison evaluations presented in Section 3.0, 2.2.1 Toxic Chemicals Stored Onsite (PINGP)

As required by Federal law, every major industrial manufacturer must keep a list of hazardous substances maintained at the facility. The

• llazardous Substance Inventory" at the PINGP, effective as of January 15, 1991 [Ref. 4.2.5j, was used in the survey. This list identifies substances, storage location, maximum storage capacity, average daily amount and if the substance is SARA (Superfund Amendments and Reauthorization Act) reportable. Chemicals which are not stored onsite in capacities exceeding the SARA limits need not be reported and were eliminated from further consideration. Chemicals whose storage capacity are below SARA levels are not considered to pose a significant toxic spill hazard to the control room operators. Table 2 2 lists all chemicals maintained on the

'llazardous Substance Inventory" stored at the PINOP hi excess of SARA limits.

Table 2 2 - llazardous Chemicals Stored Onsite at the PIN (;P that 1:scred SARA Reportable Limits Chemical Storage Location Maximum Capacity

1. Sulfuric Acid E side of plant tank 121 chemical 5,(XX) gal.

storeroom, cold lab 695 level. bot lab Auxiliary lluilding

2. Diesel Fuel #2 E of screenhouse, NW corner of plant, 142,(KA) gal.

E side of plant 3.13oric Acid Laundry area 735 level 36,(K10 lbs

4. Liquid Nitrogen By demineratiict 695 level 3,000 gal.

5. Sodium hydroxide 715 level above demineralizer 5,(XX) gal.

6.11ydrazine 35% Warehouse #2 outside cold lab 250 gal.

7. Sodium Bromide Chlorine llouse 400 gal.

8. Sodium flypochlorite Chlorine !!ouse 1,142 gal.

19muumi9 2.w TENERA

. Northern States Power Corupary Co: trol Room llibitability Prairie Island Nale:r Generating Pl::t Totle Chemic:1 St:dy 2.2.2 Toxic Chemicals Stored Within Five Miles of the PINGP To obtain information regarding industrial producers or users near the plant, the City of Red Wing was contacted. In response, a Community Profile [Ref. 4.2.3] was provided by the city for the area around lled Wing.

The profile lists all major industries in the Red Wing area, their products / services, and the number of employees. This information is summarized in Table 2 3.

Table 2 3 - City of ited Wing hlajor Industry Employer Products /Serikes Employees ited Wing Shoe Company Work . hoes and boots 1NO Northern States Power Ca Utilities 676 Josten's Diptorna Division Diplomas 11aques 325 St. John's llospital hiedical 322 P.L hicyer Industries Transmission Poles 2M S.11. Foot Tanning Co. 1 cather Processing 140 Durkee.Atwood llubber Products 262 Hed Wing llealth Center hiedical 231 Interstate hiedical Center hiedical 19t Riviera Cabinets Kitchen Cabinets 170 St. James llotel Food & Lodging t$3 Iliedet! Shoes. Inc. Sports Footwear ,

130 Central Itesearch hianufacturing htechanical Arms 58 rah! Control Inc. Robotics $$

i -

Due to the relative size of the industry, as indicated by number of employees, and the types of products and services it was concluded in this study that hazardous materials used by industry within the five mile radius poses no threat to the PINOP. Validation of this assumption was provided by the Red Wing Fire Department [Ref. 4.2.2]. Per request, the fire department reviewed the associated chemical risks from manufacturing facilities in the Red Wing vicinity.13y law, facilities are required to provide Material Safety Data Sheets (MSDS) for all chemicals that they maintain at their facility. The MSDS summarizes the hazards associated with the particular chemical. This information is kept on file by the fire department for emergency response purposes. The Red Wing Fire Department concluded that none of the hazardous ma'mials stored in the Red Wing vicinity pose a threat to the PINGP.

Red Wing is located between three and four miles from the plant. At this distance, a chemical with a toxicity of 50 mg/m' would need to be stored in quantitles exceeding 33,000 lbs to require additional consideration per RG 1.78 (see Table C 2 of RG 1.78). Less toxic chemicals could be stored l

wumu20/n 2.n TENERA

~_ . -- . - __. __ _ . _ _ . _ _ _ - _ . . - .

. Northern States Power Compa:y Co: trol Room llibitability Prairir Istard Nattar Generating P!::t Totic Chemle:1 St:dy in even larger quantitles. For example, gasoline, with a toxicity of 890 mg/m', would need to be stored in quantitles larger than 5.9 x 10' lbs to require further consideration.

Finally, as shown later in this report, none of the hazardous materials shipped on the Burlington Northern (UN) Line pose a threat to the CR operators. As the Hurlington Northern Line is within a two mile radius of the plant and as it ships large toxic chemical quantitics, chemicals stored by industry at three to four miles should pose no threat as the UN shipments pose no threat.

2.2.3 Toxic Chemicals Transported by Truck Within a Five hille Itadius of the PINGP Transport of hazardous materials by truck within the five mile radius is limited to the only major highway, liighway 61 [Ref. 4.3.6 and 4.3.7].

There are no interstate highways within the five mile radius At its closest approach, liighway 61 comes within approximately 2.5 miles of tha PINGP.

The highway connects the city of Red Wing to the greater hiinneapolis/St.

Paul metropolitan area located to the northwest of Red Wing. The highway continues to follow the hilssissippi River to the southeast of Red Wing, and eventually connects to Interstate 90.

As discussed in Section 2.2.2 of this study, there are no major producers or users of hazardous chemicals within a five mile radius of the plant.

Therefore, frequent hazardous materials shipments along 1lighway 61 are not anticipated. hiajor shipments would occur along major interstate routes.

The Red Wing Fire De"artment [Ref. 4.2.2) indicated no accidents involving a truck carryin ..azardous materials on liighway 61 has occurred in over 10 years. The fire department concluded that hazardous materid shipments on liighway 61 pose no threat.

Finally, as shown later in this report, DN rail car shipments come closer to the plant and ship larger quantities than truck traffic m accommodate.

None of the BN shipments pose a threat. Therefort any shipments on liighway 61 should also pose no threat.

2.2.4 Railroad Traffic Within a Five.htile Radius of the PINGP The SOO and Burlington Northern rail lines are within five miles of the PINGP [Ref. 4.3.6]. The SOO line comes to within approximately 0.5 miles of the plant. This line was originally owned and operated by the Chicago Milwaukee St. Paul Railroad Company. The Burlington Northern line travels past the PINGP on the opposite side of the hiississippi River in the m m uom 2. u TENERA

• l

, Northiru States Power Company Cor. trol Room llibitability Pralrle ist=d Nrele:r Ge:rrating Itnl Tosic Checle:I St2dy state of Wisconsin. Its point of closest approach is approximately two miles.

I Table 2 Chemleals Trunsported Ily Rail On 'ihr SOU Railroad 1Jue 30 Shipments or More Per Year Oscinical Nurnber <{Shipmenu

1. Chlorine 96

2. Ammonia, Anhydrous 732 1 Isobutane 144

4. Liquiried Petroleum Gas M

5. Styrene, Monomer 1296

6. Vinyl Acetate 72

7. 11eniene 72

8. Denatured Alcohol 72

9. Ethyl Alcohol 240

10. Ethyl Acetate 48

11. Methanol 948

12. Toluene 36 11 11ammable Liquid, N.D.S. (Pulp Mill Liquid) 36

14. Petroleum Naphtha 144

15. Ammonium Nittate Fortihrer 36

16. Ilydrogen Peroxide 168 17, Phenol M 18, Phosphoric Acid 96

19. Ilenzene Phosphorous Dichloride 156

20. Molten Sulfur 6588

21. Nickel Sulfate 48 Note: Chemicals listed by the railroad line are typically given by a stock code number and shipping name. A few of the chemicals (or stock code numbers) provided in the list are generie in nature. These chemicals are transported as either liquids or solids. By their title they do not indicate a specific huard and are excluded from the analysis.

m air e /i9 2.n TENERA L _ _ . _

- Northern States Power Compa:y Co: trol Room II:131tahllity Pr:Irle Isl::d Nucle:r Gr:rruting Plznt Tosle Chemical Study Tohte 2 5 - Chemicals Transported by Rail on the llurlington Northern IJnr 30 Shipruents or blorr Per Year orndcal Number of Shijvnens

1. Chlorine 140

2. Sulfur l oxide 31

3. Carbon Dioxide 327

4. Ilydrogen SulGde 43

5. Butane 51

6. Liquined Pctroleum Gas $(o

7. Vinyl Chloride 790

8. Ethylene Oxide 275

9. Styrene hionomer 54

10. Denzene 577

11. Denatured Alcohol 64

12. Ethyl Alcohol 165

13. hiethyl Alcohol 905

14. Paint 49

15. Resin Solution 44

16. Aromatic Concentrates 578

17. Fuel Oil Diesel 1393

18. Petroleum Naphtha 32  !

19. Calcium Carbide (llammable Solid) 82

20. Sodium hietal (llammable Solid) 119

21. Sodium Chlorate 71

22. Chloropictin hiiuure 47

23. Sodium Cyanide (Solid) 51

24. Sulfuric Acid 95

25. Phosphoric Acid 81

26. Acetic Anhydride 162

27. Pctric Chloride Solution 62

28. Silicon Chloride 37

29. Titanium Tetrachloride 50

30. Potassium flydroxide 36

31. Sodium flydroxide 30

32. Sulfur, hlotten 9454 Note: Chemicals listed by the railroad line are typically given by a stock code number and shipping name. A few of the chemicals (or stock code numbers) provided in the list are generic in nature. These chemicals are transported as either liquids or solids. By their title they do not indicate a specific hazard and are excluded from the analysis.

Frem information provided by the two railroad lines [Ref. 4.2.1,4.2.4,4.2.6, l

and 4.2.7], Tables 2-4 and 2-5 were campiled. These tables indicate the chemicals that are shipped in excess of the RG 1.78 criterion times per l

year. The railroad companies consider each individual tanker as a separate l shipment. Therefore, the number of shipments indicated in Tables 2-4 and l

l l m uu2cm 29 TENERA

. Northern States l'ower Corupaiy Control Rooin Ilibitability Pralric Ist::d Nuticar Gen:r: ting PI:nt Totic Chernir:I St dy 2 5 are conservative as they do not account for the potential of multiple car shipments of a chemical on a single train.

No specific information on the average weight of each shipment could be obtained from the railroad companies. Ilowever, a tank car manual [Ref.

4.3.4] was obtained that provides information on chemical shipping weights.

This information is used in various calculations to provide the basis for chemical shipment weights evaluated.

2.2.5 Chemicals Transported by llarge Within a Five Mile !!adius of the Plant Prairie Island is located next to the Mississippi River. The Mississippi River is a navigable body of water and, is within approximately 0.25 miles of the plant site. Per the Original Study, the closest navigable point of approach to the PINGP CR intake for barge traffic is approximately 0.5 miles.

Ixcks and dams have been constructed and are maintained at various points along the river. Lock and Dam 3 is located close to the plant and is controlled by the U.S. Army Corps of Engineers. The U.S. Arr / Corps of Engineers was contacted [Ref. 4.2.8] to obtain barge traffic information for the 1990 navigation season. Only chemical fertilizers were shipped in excess of RG 1.78 criteria (50 times per year) for the navigation season.

Each snipment averaged 2000 tons.

2.3 Toxic Chemical Spill Models During the initial stages of the study,it was recognized that many chemicals woald require further evaluation beyond the application of RG 1.78 shipment frequency criteria. Two models were developed to determine the buildup rate of a postulated chemical spill in the control room: 1) Model 1 for compressed gases and chemicals whose boiling point is less than ambient temperature and 2) Model 2 for chemicals with boiling points higher than ainbient temperature. These models were used, in combination with appropriate incapacitation models, to determine if the two-minute criteria of RG 1.78 could be met. Not all chemicals identified in Section 2.2 were modeled - only those chemicals which could not be climinated from further consideration t. sing the criteria described in Section 2.4 were modeled.

Both toxic chemical models simulate spill release, atmospheric dispersion and l eventual buildup in the control room. These models were developed in l

accordance with the requirements of NUREG 0570 and are described in detail in l TENERA Calculation 1934 2.2 001 [Ref. 4.3.8]. Many conservatisms exist in the rrodels including:

mNR120/19 2 15 TENERA

Northern States l'ower Compa:y Co: trol Room liabitab!Ilty I'.elrie Ists:d Nuclear Generati g l*la:t Totle Chemitt Study

. One complete container of the chemicalis assumed immediately released in the accident,

. Y %d direction carries the spill directly to the control room intake,

. Worse case 5% atmospheric dispersion conditions are assumed (Pasquill Category F),

. No credit for intervening structures or topology is assumed which would dilute concentration at the contro! room intake,

. No credit is assm.ied for an elevated control room intake which redaces the concentration at the intake, and

. No credit is assumed for spill absorption or dilution in surrounding ground or water.

Summary descriptions of both models are provided below. Specific equations and design inputs are contained in the detailed TENERA calculations [Refs. 4.3.8 to 4.3.12].

2.3.1 Model 1: Chemicals Whose lloiling Point is Less Than Ambient According to NUREG 0570, two release phenomena are assumed for chemical spills when the chemical boiling point temperature is below ambient temperature. These phenomena are instantaneous puff and vaporization. Each spill phenomena acts independently. Therefore, each phenomena is modeled separately and the contributions to control room vapor concentration are summed at corresponding time steps.

An ambient temperature of 70*F was used in the calculations [Refs. 4.3.8 to 4.3.12]. Ilowever, significant change in ambient temperature would not change the model applied as chemical boiling points were typically well in excess or well below ambient temperature.

Instantaneous Puff:

The percentage of total spill which is instantaneously releasec depends on the chemical's heat capacity, boiling point, and heat of vaporization. The total amount released is dependent upon the temperature difference between the boiling point and cmbient temperature, the above chemical properties, and the total spill mass.

The instantaneous puff assumes the shape of a three dimensional exponential bell curve with the maximum vapor concentration (top 01 the bell) at the point of spill release. The maximum vapor concentration and l

19e,uu:o/i9 2.is TENERA

. Northeru States l'onr Compary Co: trol Room II:bitabitity l*ralrle istrd Nucle:r Ge:erating I' lint Totic Chemic 1 Study puff shape are dependent upon the mass of the instantaneous release and the assumed atmospheric distribution coefficients (Pasquill Category). He l puff travels in a straight line directly to the control rcom intake. As no  !

deviation in wind direction is assumed, the problem becomes two- l dimensional. The shape of the puff cur e is conservatively assumed to remain unchanged during its transport to the control room intake, Therefore, the maximum vapor concentration at the point of release will ultimately reach the control room intake depend!ng on wind speed and distance to Cit intake.

Chemical buildup and dissipation in the control room will assume a similar exponential bell shape response over time for an unisolated control room, llow quickly control room vapor concentration increases and dissipates depends upon the air exchange rate of the control room, the shape of the outside Cit puff at the intake, and the speed at which the puff passes the Cll intake.

Vaporization:

The remaining spill is released through vaporization. De rate of vaporization depends upon the spill area rate of change, atmospheric and solar heat fluxes, ambient and chemical boiling point temperature difference, and ground and chemical boiling point temperature difference.

The spill area is assumed to spread uniformly otitward in a circular fashion to a maximum spill depth of I cm. The spill then remains at this maximum area until completely vaporized. Typically, the spill expands to its maximum area quickly with respect to total time to complete vaporization.

This spill area model is conservative as it assumes no credit for surrounding topology which might impede spill growth. As mass release is largely dependent on spill area, any impediments to spill area growth wotild slow chemical release rate.

Atmospheric dispersion is modeled similar to the instantaneous puff release. The applicable Pasquill Gifford Category is combined with distance to spill to determine the appropriate dispersion coefficients.

These coefficients are then combined with the vaporization model, wind speed, spill area model, and remaining spill mass to determine the rate of chemical mass release and its progression to the Cil intake. The vaporization cloud assumes the shape of an exponential bell curve in the dimensional plane perpendicular to the direction of propagation. This shape is similar to the instantaneous puff liowever, in the direction of propagation, the cloud shape is different due to the interactive effects of .

the spill area and vaporization models. The cloud assumes a step function shape at the time of spill release. This step function then grows and finally dissipates over time in an exponential manner until the entire spill has 19f31R120/19 2-17 TENERA

6

. Northern States l'ower Comge:y Control Room ll:bitability I'ruirle 151=d Nucle:r Generating I' lint Totic Chemical St:dy vaporized. An end step function then occurs and the mass release rate drops back to zero.

Control room buildup is modeled using the identical equations implemented for the puff release. Buildup and dissipation in the CR is dependent upon the sarne factors.

As the vaporization cloud is not instantaneously distributed,like the puff release, its contribution to CR chemical buildup will not take effect until the cloud has traveled to the control room at the wind speed rate. By the time it reaches the control room, detection and corrective actions in response to the instantaneous puff may have already occurred. If left unisolated, the CR will experience two separate maximums resulting from each release phenomena's contribution. Typically, the hi;her l maximum is contributed by the puff.

2.3.2 Model 2t Chemicals Whose Holling Point is Greater than Ambient A simpler model was developed for chemicals whose boiling point temperature exceeds ambient temperature. For these chemicals, only evaporation exists as the mass release phenomena. The evaporation rate is dependent on the chentical's molecular weight, vapor pressure, Reynolds N'imber, Schmidt Number, diffusion coefficient, spill area model, and ambient temperature. The spill area model, atmospheric dispersion model, and control room buildup model are all identical to those developed for the vaporization release as described above. The remaining terms are combined into one of two equations for calculating mass release for either turbukat or laminar flow. The selection of laminar or turbulent flow is done conservatively by selecting that equation which results in the highest mass release rate. These equations are described in detail in Reference 4.3.8. For evaporation releases the spill will first reach the control room intake only after it is carried by the wind. Due to rnuch slower release rates than chemicals with below ambient temperature boiling points, these chemicals do not typically pose a threat to CR operators and release over a much longer period.

2.4 Incapacitation Assessment Criteria The final step in the Toxic Chemical Study was to evaluate the chemicals identified in the survey for potential CR operator incapacitation. A set of incapacitation assessment criteria were developed and applied to each chemical from the final survey described in Section 2.2. These criteria are:

- Is the chemical lis'ed in 29CFR Part 1910 as an air containment? Also, does the chemical have a TLV or other limit (s) specified by NIOSIl in Reference 4.3.5 or by ACGill in Reference 4.3.157 19iwuzon9 m TENERA

- ~

Northern States l'ower Cocpany Control bm it: bit %Illty Pralrie bl::d N- cr Gr: truth, Itat Tosic Cheml Stud)

Hused or. avaDaW data in the Original Study, is there seffic.ient evidence to demo *tstrate through cotr.parison analysia that CR operator incapacitation wil! not occur'!

Based on known properties at the chemical as compared to spill model results for computable chemical types, can it be demonstraitd that incapacite son w,ll not occur?

Based on spill inodel results, will the CR chernical concentration exceed the chemical's lhreshold Limit Value (TLV) or the immediately Dangerous to IJfe or lieslth (IDLil) W.lue?

Hased on spill mooel results, will the Cil chemical concentratien result in tncapacitation t. ing the modeling techniques of NUlV.EG/CRJ7417 Based on prehilistic model, is the likelihood of a spill accident resulting in CR operator incapitat oni and exceedin;; 10CFR130 limits greater (Lan W /7 year?

Each successive criteria was applied until a rese of "rio incapacit ation" tould be demonstrated. N. those chemicals v here :,uch a con &ien could not be reached, a fina; probah!!ist:'; analysis was performed to determine the likelihood of accit.:rital release.

The f.cceptance criteria used rete in accotaance with the regulatmy requirements outlined in Section 2.1. Is.ial results Sr all cl emicals are presen6cd ir Section 3.0 and a more detailed Wscussion of each incapacitation assessment criteria is provided below.

2.4.1 29CFR Part MO, NIOSil or ACGill Listed Tht; Grst step in det.n nining CR operator inacapacitation was to review the tot of chemicals from the survey against 29CFR Part 1910. An extensive list of hazardous materials is provided in Title 29 and is in accordance with federed standards. Chemicals identified in the survey which are not listed as air contaminants in Title 29 were removed from further consideration provided they also were not contained in the 1990 to 1991 American Conference of Governmental Industrial flygienist (ACGill) Threshold Limit Value (TLV) Guidelines (Ref. 4.3.15] or in the Pocket Guide ta Chemical liazards by the National Institute for Occupational Safe n and llealth (NIOSil) [Ref. 4.3.5). The ACGill and NIOSil lists were also t: sed to further substantiate the removal of any chemical from further consideration on the basis of 29CI R Part 1910. Ilowever, no instance was found where a chemical from the survey was not listed in Title 29 but was listed in ACGill or NIOSil.

191N R120/19 2 19 TENERA

s

• Northern States h.wer Co:npa:y Co: trol Itoom llibliability I'r:Irir th d Nude:r Geirr: ting I'l:nt Totle Chemic-1 Strdy 2A.2 Comparison to Original Study The second step in determining potatial Cit operator incapacitation was to compare each of the surveyed chemicals against results presented in the Original Study. As most che;1icals evaluated in the Original Study were not presented with sufficient infermation (i.e., chemical quantity, storage or shipment container Jocation, etc.) an accurate comparison could not be developed and this step was skipped. Where an accurate comparison could be made the chemical was eliminated from further consideration if the Original Study reached such a conclusion. Typically, this step only proved beneficial for chemleals stored onsite where the original stored quantity and contaher type remain urwhanged.

2A.3 Chemical Properties "oniparison i

The third ttep was to determine if the chemic:d could be eliminated from further consideration based on its physical properties. This included a review of the chendcals' toxicit! end properties which affect mass release.

For exainple, molten sultur was not evaluated using the toxic spill models as it has an extremely low vapor pressure and an extremely high boiling point. Though sh%ed in a heated tank car, the chemical's tendency is towards extrer 3 low dispersion and becomes a solid at arabient temperature.

2AA Spill Model itesults Comparison to TIN and 1D1.11 Chemicals which did not meet the first three criteria were modeled and CII concentrations calculated. All calcult.tions were carried out over the first eight hours following the spill. The CIIIIVAC remained unisolated over the entire period with normal outside air intake rates assumed.

The calculated Cit vapor concentration was first compared to its TLV. If it could be shown that for the entire eight hcur period the TLV was not exceeded, incapacitation would not occur and protective breathing equipment would not be required. This acceptance criteria is consistent with itG 1.78. The TLV i:, defined by the American Conference of Governmental industrial llygienist ( ACGIli) as that time weightea average chemical concentration that a worker could be exposed to with no long-term health effects. TLVs published by the ACGill and NIOSil were used.

For those cases where the TLV was exceeded, a comparison to IDLil values was made. The IDLil is defined by the NIOSil as that concentration to which a worker may be exposed for 30 minutes without any escape impairing symptoms or any irreversible health effects. If it could be demonstrated that two minutes subsequent to CR operator reome 22o TENERA

' Northern States Power Compa y Control Roma llibitability I'rdrie Istud Nalcar Ge:erating I'lant Totle Chemie-I Study detection the IDLil has not been exceeded, it was concluded that incapacitation would not occur. The total tim inten d of concern was from the point the TLV is first exceeded to two rali u subsequent to detection. This total time span could be more or les. I tan a total time of two minutes, as the TLV could be greater or lower than the e oncentration at v^ilch detection would occur. This comparison was extremely conservative as constant exposure at the IDLH vah;c i allowed. In most cases the IDLil was never approached and ihe totsl tirre was we.Il within 30 minutes.

.t4.5 Incapacitation Assessment Using NUREG/CR.1741 Models If TLV or IDLil values were exceeded o. , wid not be faurid, .

comparisonusingappropriateNUREG/CR417a1incap cit losimodels w ,n performed.

NUREG/CR 1741 presents five separate moueh for hte nlang cheaiire.1 incapacitation. The selection of which model to impleiae..t depends an tk properties of the chemical and the biologimi mechanisms ofincapacitation, i.e., concentration or accumulated .h se dependen if it could be demonstrated through the use of the appropriate med , that incapacitatio-in two minutes would not occur, the chemical was clin.inated fren further consideration. Specific incapac'ta% 1 enodel' t 'd design inputs used are included in the various calculations. These mouels ar identical to those used in the Bechtel reanalysis of incapacitathn effats perfortned in 19. L 2,4,6 Probahilistic Assessment of Remain.'ng Che.n'nis As discussed in Section 3.0, other than ammonia 14. chloi4ne, a wo.s.: case analysis of all chemicals from the surve) did not result in incapacitation.

For chlorine and ammonia a probvailist.c study was a formed to assess the likelihood of a spill event a.id the li';e' hood tl at the 9ill would result in control room operator incapacitalinn. As the wra case meteorological conditions rarely exist at the p'unt, a grametric st.dy using the toxic chemical spill model and appropiinie NUREG/rR.1741 incapacitation model was petlot med Meteorological conditic e,were varied to determine what comb nation of Pasquill Categcry, w.nd speed, wind direction and distance to (.e CR intake would result m less than two minutes for oprator response. T ice this set of conditions was known, the conditions were weighted by their frequency of ocet.rrence using historical raeteorological data far the site [Refs. 4.3.16 t o 4.3.:8.. This was then combined with n'ttional data on frequency of railroad car accidents and probability ol chemical release in an accident [Ref 4.3.14] to determine overall probabrity of an external event and likelihood of incapacitation.

It was funher, conservatively assun i d, that if CR operator incapacitation occuried, the probability of exceeding 10CFR100 limits is 0.1 [Refs. 4.1.8 191:MR120/19 2 21 TENERA

Northern St:tes Power Compa:y Co trol Room II:bitability Prairic Ist::d Nucle:r Generatlog Pir:t Totle Chemic-1 Study and 4.3.18). Such an assumption gives little credit for operator response in less than two minutes, response by others, or automatic safe shutdown ft atures of the plant. It was also conservatively assumed that the actual number of chlorine or ammonia shipments is twice the actual number shipped. 'lhis assumption was included to allow for some future variances in the quantities shipped without the need to recalculate the probability.

The following formula was developed for determining probability:

Probability of a - Total Number of Chemical Shipments /yr X 2 (Chlorine / Ammonia) Release Execeding 10CIH100 Limits X Probability of a X Total Train Miles X Probability of Execeding Ilarardous Release / of Incapacitation / 10ClH100/ Operator Train Mile Shipment Incapacitrdion This formula was only applied to railcar accidents. As shown in Section 3.0, only ammonia and chlorine spills on the SOO Rail IJne pose a potential incapacitation threat to the Clt operators.

The acceptance criteria of Standard Review Plan Section 2.2.3 [Ref. 4.1.5]

and Reg. Guide 1.70 [Ref. 4.1.12] was used to determine if the likelihood of an ammonia or chlorine : pill warranted any special control room design features for operator protection. If it could be shown that the probability 4

of an accident resulting from an external event is less than 10 per year or 104per year with conservative assumptions, special design features would not be required, i

l I

I 1913f1U20,/;p 2-22 TENgns I

• Norticrn Stat:s l'ower Company Control Itoorn ll:bitability I ruirle Isl::d Nucle:r Generating I'l:nt Toste Chemle:l Stud) 3.0 FINAL RESULTS The Toxic Chemical Survey identified a number of chemicals that exceeded the screening requirements of RG 1.78. These chemicals were evaluated using the assessment criteria of Section 2.4. The results are summarized in Table 31 and presented below in chemical groupings by locale or transport mechanism similar to Section 2.2. Table 31 includes each chemical, identifies the applicable elimination criteria, and summarizes the final analytical results. Each chemical is discussed in more detail below based on the most limiting, conservative, calculation. Many of the chemicals listed below were evaluated in the Original Study and no special detectors were required. Tne additional bases used in this study further substantiates the conclusion that special detectors for the same chemicals are still not required.

Ilowever, credit for results of the Original Study were only used on a limited basis as reanalysis provides added assurance early detection is not required.

3.1 Toxic Chemicals Stored Onsite The survey identified a total of eight chemicals stored on site. Four of these chemicals (boric acid, liquid nitrogen, sodium bromide and sodium hypochlorite) were. climinated from further consideration as they are not identified in 29CFit 1910 or by the ACGill [Ref. 4.3.15) or NIOSil [Ref 4.3.5) as toxic. Of the four remaining chemicals, three (sulfuric acid, hydrazine and sodium hydroxide) were evaluated in the Original Study and were eliminated on the basis that they posed no threat to CR operators. With the exception of hydrazine, these chemicals are stored in the same locations and quantities as the Original Study. Though hydrazine is now stored in a larger container (250 gallons), it is stored much farther from the CR intake. Therefore, it was eliminated from further consideration.

The final chemical is diesel fuel oil. Diesel fuel oil stored onsite is not specifically identified in the Original Study. It does not pose a threat to the CR operators as a majority of the fuel oil is stored underground. A rupture of the main underground tank would result in ground seepage and not vapor release. That portion stored above ground is approximately 500 gallons in each of two separate day tanks. These tanks are located indoors which would further delay and reduce the buildup of fuel oil in the CR. Therefore, fuel oil was eliminated from further consideration.

3.2 Toxie Chemicals Stored Within Five Miles of the Plant As noted in Section 2.2.2, no toxic chemicals are stored within a five mile radius of the plant which,if spilled, would results in CR operator incapacitation.

i9tsmuon, u TENERA

NorGern Staten Power Compa:y Control Itoom II:t,itatellity Pr.arle 161ud Nalear Generating Pl:nt Tusic Chemic:1 Study 3.3 Toxic Chemicals Transported by " ruck As noted in Section 2.2.3, no toxic chemicals transported on liighway 61 pose a threat to Cil operators.

3.4 Toxic Chemicals Transported by 1(all The major source of toxic chemicals transported by the plant are carried by two rail lines. The survey results indicated that 21 chemleals carried by the 500 IJne Railroad and 32 chemicals carried by the Burlington Northern (llN) Itallroad which exceed the shipping frequencies of 30 per year defined in 11G 1.78 required additional evaluation. These chemicals are listed in Section 2.2.4 and on Table

31. Each chemical, by railroad line, is discussed separately below.

SOO Line Chlorine:

llecause of the frequency, size of shipment, and proximity to cit intake chlorine shipments on the SOO Line could only be eliminated through the application of the probabilistic criteria. As shown on Table 31, the probability of a incapacitating chlorine release is 1.16 x 104 per year. The conservatisms associated with this analysis are described in more detail in Section 3.6 below.

SOO Line Ammonia Anhydrous:

Similar to SOO Line Chlorine shipments, Ammonia shiprnents were eliminated using the probabilistic approach. Tne frequency of incapacitation per year derived 4

for Ammonia is 1.47 x 10 . See Section 3.6 for further discussion.

SOO Line Isobutane:

Iso' atane was modeled by analyzing butane releases from the SOO Line. As nott.d in Iteference 4.3.11, tha TLV for butane is exceeded and no IDlJI is available. llowever, butane acts as an asphyxiant and reaches a concentratiots of 2,214 ppm in the Cil two minutes subsequent to detection. As Carbon Dionide, which also acts as an asphyxiant, has an IDLil of 50,000 ppm, it is concluded that early detection equipment for butane releases is not required as at least two minutes is available without operator incapacitation. Also, it states in Iteference 4.3.17 that: "lligh exposure produces drowsiness but no other evidence of systemie effects."

SOO Line LPG:

LPG is primarily a combination of butane and propane. Demonstrating no incapacitation in two minutes subsequent to detection for butane and propane will, therefore, satisfy the requirements for LPG.

i muuma u TENERA

0

. Northern States Power Colupary Control Roorn it:bitability Prairir l$ta:d Nuclear Gerraiting I'li:t Tosic Chrinic 1 Study According to Reference 4.3.11, the concentration of propane will reach 32,440 ppm at two minutes subsequent to operator detection. This is based on closest distance and other modeling conservatisms. This value exceeds the 30 minute IDLil of 20,000 ppm but was still eliminated from further consideration based on IDLll for the following reasons:

. LPG was not assumed detected untilit reached the ppm value for propane.

As lpg is a combination of both propane and butene,it is reasonable to expect detection at the lower butane limit which was not assumed. Also, a foul smel!!ng odorant is typically added to LPG [itef. 4.3.5) which would enhance the possibility of early detection.

. The IDLil is not the level of incapacitation. Incapacitation is at much bigher levels.

- The IDLil is not execeded for the entire two minutes subsequent to detection but only the last 45 seconds, if a time weighted average ofIDLil is plotted the total area under the thirty minute cun'e is substantially greater than the area under the two minutes post detection curve.

SOO Line Styrene hionamer:

The maximum concentration of styrene in the Clt over the entire 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> period subsequent to release is 43.4 ppm [Itef. 4.3.9]. This value is well within the 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> TLV of 100 ppm. Therefore, early detection of styrene is not required.

SOO Line Vinyl Acetate:

According to Reference 4.3.9, two minutes subsequent to detection the concentration of vinyl acetate in the CR has climbed to 69.2 ppm. The TLV is 10 ppm. As no IDLil was available, Model E of NUREG/CR 1741 was used to determine if at least two minutes were available subsequent to detection.

Substantially more than two minutes was found to be available assuming a constant concentration of 69.2 ppm. According to Reference 4.3.17, the " vapor causes a slight smarting of the eyes or respira:ory system if present in high concentrations.

The effect is temporary" Therefore, early detection of vinyl acetate is not required.

SOO Line llenzene:

Ilenzene reaches a CR concentration of 166.7 ppm two minutes subsequent to detection [Ref. 4.3.9). This value is reached 261 seconds after the TLV is exceeded, lloth the 261 seconds and 166.7 ppm are well within the 30 minu o and 2,000 ppm allowable by the IDLil, Therefore, early detection is not required.

muuwm n TENERA

y

• Erthern States l'ower Compa;y Control Hoom II:bitability l'rulrie 1612:d Mclear Gentruting I'la t Tale Checleil Sig SOO Line Denatured Alcohol:

Denatured alcohol is not listed as a toxic vapor in 29 CFR 1910 nor does it .

an established TLY in the ACGill or NIOSil references. Though potentially harmful, no incapacitation threat is posed. Therefore, early detection is not required.

SOO Line Ethyl Alcohol:

The maximum concentration of Ethyl Alcohol reached in the CR is 617.7 ppm over the 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> period subsequent to the release [Ref. 4.3.10). The TLV is 1,000 ppm which is never exceeded. Therefore, early detection is not required.

SOO IJne Ethyl Acetate:

Two minutes subsequent to detection the concentration of ethyl acetate in the CR has climbed to 110.2 ppm [Ref. 4.3.10). This value is well below the 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> TLV of 400 ppm. Therefore, early detection is not required.

SOO IJne Methanoh Methanol reaches a CR concentration of 798.1 ppm two minutes subsequent to detection [Ref 4.3.9]. This value is reached 553 seconds after the TLV is exceeded, lloth the 553 seconds and 798.1 ppm are well within the 30 minutes and 25,000 ppra allowable by the IDLil. Therefore, early detection is not required.

SOO Line Toluene:

Two minutes subsequent to detection the concentration of toluene in the CR has climbed to 31.5 ppm [Ref. 4.3.10]. This value is well below the 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> TLV of 100 ppm. Therefore, early detection is not required.

SOO Line Flammable Liquid N.D.S. (Pulp Mill Liquid):

Pulp mill liquid is not listed as a toxic vapor in 29 CFR 1910 nor does it have an established TLV in the ACGlil or NIOSil references. Though potentially harmful, no incapacitation threat is posed. Therefore, early detection is not required.

SOO Line Petroleum Naphtha:

According to Reference 4.3.10. the maximum control room concentration reached in a 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> period is 821.7 ppm. This value exceeds the TLV of 500 ppm but does not exceed the IDLil limit of 10,000 ppm. The TLV is also not reached until well over 1/2 hour into the accident. Though detection values could not be found, the odor is similar to that for gasoline or kerosene [Ref. 4.3.17]. Gasoline has a m.mu20m u TENERA

9 Nodhern States Power Compey Co: trol Ibann it: bit:btlity Prairie Isind N: clear Generating I' lint Totle Chemical Study recognition threshold of 240 ppm (for octane), if recognition occurred at 240 ppm, more than two minutes is available before the TLV of 500 ppm is exceeded.

Therefore, early detection is not required.

SOO ljne Ammonium Nitrate Fertilizer:

c ertilizers were climinated based on larger shipment quantitles at closer distance to the plant in the Original Study. Therefore, early detection is not required.

SOO IJne Ilydrogen Peroxide:

According to Reference 4.3.9, the highest concentration reached by hydrogen peroxide in the CR is 23.9 ppm. The TLV is first exceeded at 27,9 minutes and the IDLil of 75 ppm is never reached, if it can be assumed that the IDLil can be time weight averaged, a concentration of 23.9 ppm would be allowed for at least 1.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> in comparison. No detection data for Ilydrogen Peroxide could be found, llowever, according to Reference 4.3.5, hydrogen peroxide has a "slightly sharp odor.' Chemicals with sharp odors have detection levels in the 0.1 to 9 ppm range [ Pef. 4.3.9]. Detection of hydrogen peroxide either directly by the operators or by notification within the first two hours of the accident should occur and appropriate corrective actions, if needed, wauld be taken. Therefore, early detection is not required.

SOO Line Phenol:

According to Reference 4.3.10, two minutes subsequent to detection the concentration of Phenol in the CR has raised to 0.4 ppm. This value is well below the TLV of 5 ppm. Also,if the CR remains unisolated for the entire 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, the highest concentroion reached is only 3.8 ppm, as noted on Table 31. This maximum concentration is also below the TLV. Therefore, early detection is not required.

SOO Line Phosphoric Acid:

According to Reference 4.3.9, the highest concentration reached over an 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> period is 0.45 ppm. This value is low due to the relatively low vapor pressure of phosphoric acid. This value exceeds the TLY of 0.245 ppm and no data on IDLil could be found. Use of Model E from NUREG/CR-1742 resulted in substantially more than 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> available at 0.45 ppm without resulting in incapacitation [Ref.

4.3.9). Therefore, early detection is not required.

SOO Line Benzene Phosphorous Dichloride:

Benzene phosphorous dichloride is not listed in 29CFR Part 1910 as an air contaminant nor could NIOSil (Ref. 4.3.5) or ACGill [Ref. 4.3.15) TLVs be found. The- Jore, :arly detection is not required.

191NR120/19 3-5 TENERA

• Northern States Power Cornpany Control Roorn Ilibliability Padrir Istrid Nurirar Gener: ting Pl:nt Toile Chemic:1 Study SOO Line hiolten Sulfur:

hiolten sulfur is not listed in NIOSil [Ref. 4.3.5] or ACGill [Ref. 4.3.15]. Due to the physical nature of sulfur it is a solid at room temperature and is shipped in a liquid state in heated tank cars. According to Reference 4.3.9, sulfur has a vapor pressure of only 1 mmlig at 184*C. As there is no physical means to keep sulfur heated once it has spilled, it does not pose a threat to the CR operators.

Therefore, early detection is not required.

SOO Line Nickel Sulfate:

Nickel sulfate is a crude ',roduct of copper refineries. It is not listed in 29CFR Part 1910 as an air contaa . ant nor could NIOSil [Ref. 4.3.5) or ACGill [Ref.

4.3.15] TLVs be found. Therefore, early detection is not required.

Burlington Northern Chlorine:

A probability assessment was not required to demonstrate that there is no need to monitor for IlN chlorine releases. The DN line runs on the opposite site of the Mississippi River from the PINGP and is approximately four times farther from the CR intake than the 500 Line. As the distn' ice is much larger, atmospheric dispersion factors are greater ar the ultimate chlorine concentration and CR buildup are less. According to Reference 1.3.11, exactly two minutes are available subsequent to detection using NUR"./CR 1741 hiodel A. Therefore, early detection is not required, llurlington Northern Sulfur Dioxide:

Sulfur dioxide was only modeled on the UN line as it is not shipped in excess of Reg. Guide 1.78 criteria on the SOO Line. According to Reference 4.3.11, approximately 4.3 minutes subsequent to detection is evailable prior to incapeitation. This is conservatively based on hiodel A of NUREG/CR 1741 using the highest CR concentration reached two minutes subsequeat to detection (335.6 ppm) as constant over the entire period. Therefore, early detection is not required.

Durlington Northern Carbon Dioxide:

No detection data for carbon dioxide could be found. However, according to Reference 4.3.11 over an eight hour period subsequent to the spill the CR concentration only exceeds the 5,000 ppm TLV for approximately 300 seconds to a maximum of 5,901 ppm. As the 300 seconds and 5,901 ppm are well below the IDLII of 50,000 ppm, incapacitation due to carbon dioxide releases will not occur.

Also, the time weighted average concentration of carbon dioxide over the entire 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> period is well below the 5,000 ppm TLV. Therefore, early detection is not required.

nouumn u TENERA

U Northern States l'ower Company Control Room tirbitability l'r ltle Isted Nuclear Gr:crati:g Itut Tosic Chemic:1 Study Burlington Northern flydrogen Sulfide:

According to Reference 4.3.11, hydrogen sulfide climbs to a CR concentration of 448.9 ppm two minutes subsequent to detection. This value exceeds both the TLY (10 ppm) and the IDLil (300 ppm). Using the Model D approach of NUREG/CR-1741 and conservatively assuming a constant concentration of 448.9 ppm,it was shown that more than two minutes is available prior to incapacitation.

Therefore, early detection is not required.

Burlington Northern 13utane:

Enveloped by the SOO Line butane analysis provided above. Therefore, early detection is not required.

Ilurlington Northern LPG:

Enveloped by the SOO Line butane analysis provided above. Therefore, early detection is not required.

Burlington Northern Vinyl Chloride:

According to Reference 4.3.11, vinyl chloride reaches a CR concentration of 288.1 ppm at 200 seconds subsequent to detecuan. This value, and the point of detection, both cxceed the TLV. No IDLil could be found. Using Model D of NUREG/CR 1741, subatantially more than two minutes is available at a constant concentration of 288.1 ppm prior to incapacitation. Therefore, early detection is not required.

Burlington Northern Ethylene Oxide:

According to Reference 4.3.11, the CR concentration of ethylene oxide has climbed to 1,189 ppm two minutes subsequent to detection. Approximately 16.7 minutes have elapsed up to this time from the point the TLV is first exceeded.

Though the concentration of 1,189 ppm exceeds the IDLil of 800 ppm, incapacitation should not occur as the total time of 16.7 minutes is less than the 30 minutes at IDLil and the total time at a concentration in excess of 800 ppm is approximately 3 minutes. Also, the IDLil is not the incapacitation level.

Therefore, early detection is not required.

Burlington Northern Styrene Monomer:

Results for the llN Line are enveloped by the SOO Line. Therefore, CR concentration will not exceed the TLV limit for a BN spill and early detection is not required.

19nMn120/19 57 TENERA

s a

Northen: Statrs l'ower Compa:y Co: trol Room II:hltability Prairle 16trd Nrclear Gr:erating Pl::t Totle Chemic 1 St dy Ilurlington Northern Ilenzene:

Results for the IIN Line are enveloped by the SOO Line. Therefore, CR concentration will not exceed the IDLil limit for a llN spill and early detection is not requhed.

Ilurlington Northern Denatured Alcohol:

Denatured alcohol is not listed as a toxic vapor in 29 CFR 1910 nor does it have an established TLV in the ACGill or NIOSil references. Though potentially harrnful, no incapacitation threat is posed. Therefore, early detection is not required.

Ilurlington Northern Ethyl Alcohol:

Results for the llN Line are enveloped by the SOO Line. Therefore, CR concentration will not exceed the TLV limit for a 13N spill and early detection is not required.

Ilurlington Northern Methyl Alcohol:

Results for the llN Line are enveloped by the SOO Line results for methyl alcohol. Therefore, CR concentration will not exceed the IDLil limit for a 13N spill and early detection is not required.

Ilurlington Northern Paint:

Paint is not listed in 29CFR Part 1910 as an air contarninant nor could NIOSil

[Ref. 4.3.5) or ACGlH [Ref. 4.3.15] TLVs be found. Therefore, early detection is not required.

Ilurlington Northern Resin Solution:

Resin Solution is not listed in 29CFR Part 1910 as an air contaminant nor could NIOSil [Ref. 4.3.5] or ACGill [Ref. 4.3.15) TLVs be found. Therefore, early detection is not required.

Ilutlington Northern Aromatic Concentrates:

Aromatic Colicentrates are not listed in 29CFR Part 1910 as an air contaminant nor cou,d NIOSH [Ref. 4.3.5] or ACGilI [Ref. 4.3.15] TLVs be found. Also, due to the chemical nature of aromaties they would most likely be bounded by results for benzene. Ilenzene, as noted above, does not require early detection.

Therefore, early detection is riot required.

191NR120/19 u TENERA

v Northern States Power Company Co: trol Room llibitability Prairir 161rd Nuclear Generating Plait Tale Chemicil Study llurlington Northern Diesel Fuel Oil:

Diesel fuel oil was eliminated in the Original Study. It was shipped on barges which contain much larger quantities and potentially ship closer to the Cil intake than rail traffic. Therefore, early detection is not required, llutlington Northern Petroleum Naphtha:

llesults for the UN Line are enveloped by the SOO Line results for petroleum naphtha. %erefore, Cil concentration will not exceed the IDLH limit for a llN spill and early detection is not required.

Ilurlington Northern Ctdeium Carbide:

This chemical is a solid at ambient term erature and has an extremely low mass release rate due to a low vapor pressure. Therefore, early detection is not required.

Durlington Northern Sodium Metal:

His chemicalis a solid at ambient temperature and has an extremely low mass release rate due to a low vapor pressure. Therefore, early detection is not required.

Burlington Northern Sodium Chlorate:

This chemical is a solid at ambient temperature and has an extremely low mass release rate due to a low vapor pressure. Therefore, early detection is not required.

Burlington Northern Chluropierin:

According to iteference 4.3.10, three minutes are available prior to incapacitation using the Model A technique from NUllEG/ Cit 1741. The calculated three minutes is conservative as it assumed the CR concentration at two minutes subsequent to detection for the entire exposure period and assumed the spill was from the closer SOO Line. Should this chemical be shipped on the SOO Line,it is already analyzed. Therefore, early detection is not required.

Burlington Northern Sodium Cyanide:

This chemical is a solid at ambient temperature and has an extremely low mass release rate due to a low vapor pressure. Therefore, early detection is not required.

m.uuma u TENERA

u Northern States Power Company Control lloom II:t ltability Prairie istrd Nuclear Grueratire Pl=t Totle Chemic:1 Study Burlington Northern Sulfuric Acid:

According to iteference 4.3.10, the 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> TLV for sulfuric acid is never exceeded for a postulated spill. The TLV is 0.25 ppm and the highest CR concentration reached is 0.021 ppm. This analysis conservatively assumed a spill distance for the 5001Jne. Therefore, early detection for both UN and SOO Line releases is not required, Burlington Northern Phosphoric Acid:

Results for the BN Line are enveloped by the 500 Line results for phosphoric acid. More than two minutes was demonstrated for SOO Line releases using Model E of NUREG/CR 1741 as noted above. Therefore, early detection is not required.

Burlington Northern Acetic Anhydilde:

According to Reference 4.3.10, the 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> TLV for acetic anhydride is never exceeded for a postulated spill. The TLV is 5 ppm and the highest CR concentration reached is 4.1 ppm. This analysis conservatively assumed a spill distance for the 500 Line. Therefore, early detection for both UN and SOO Line releases is not required.

Burl'ugton Northern Ferric Chloride Solution:

Ferric chloride solution is not listed in 29CFR Part 1910 as an air contamim'nt nor could NIOSil !Ref. 4.3.5) or ACGill [Ref. 4.3.15) TLVs be found. Therefore, early detection is not required.

Burlington Northern Silicon Chloride Solution:

This' chemical is a solid at ambient temperature and has an extremely low mass release rate due to a low vapor pressure. Therefore, early detection is not required.

Burlington Northern Titanium Tetrachloride:

Titanium tetrachloride is not listed in 29CFR Part 1910 as an air contaminant nor could NIOSil [Ref. 4.3.5] or ACGill [Re.* 4.3.15] TLVs be found. Therefore, early detection is not required.

Burlington Northern Potassium llydroxide:

Potat.am Hydroxide is not listed in 29CFR Part 1910 as an air contaminant nor could NIOSH [Ref. 4.3.5] or ACGil1 [Ref. 4.3.15) TLVs be found. Therefore, early detection is not required.

IEMRn0/19 3 10 TENERA

_y __ __ _ > . - . ~ . _.- _ .-. . _ . _ _ _ . _ . . _ - . _ _ . _ _

-Northern States Power Company Co

• trol Room Hrbitability Pr-trie Isis:d Nrcle:r Ge:erating Pirnt Toxic Chemical Study Burlington Northern Sodium Hydroxide:

Sodium hydroxide is stored at the PINGP site Sob im hydroxide was eliminated from further consideration onsite based on the Original-Study as noted above.

Also, sodium hydroxide has practically no vapor pressure at ambient temperature and has a 2,534*F boiling point according r.o Reference 4.3.5. Any rail release would have practically no dispersion into the atmosphere. Therefore, early detection is not required.-

Burlington Northern Molten Sulfur:

As noted above, molten sulfur is a solid material will almost no vapor pressure at ambient temperature. As there is no mochanism for heating the sulfur to vaporize it during its transport to the CR intake, it poses no threat to the CR operators.

Therefore, early detection is not required.

3.5. Toxic Chemicals Transported by Barge As noted in 'Section 2.2.5, the only hazardous material shipped on this part of the Mississippi River is chemical fertilizers. Barge transportation of chemical-fertilizers was assessed in the Original Study. As the Original Study concluded that -

such shipments post no threat to CR operators, chemical fertilizers transported by barge are eliminated from further consideration. Additionally, barge accidents would likely spill into the Mississippi River which would both dilute and minimize

the atmospheric release of any chemical spill.

3.6 Ammonia and C.,.1 # 3 pill Releases -- Probabilistic Results The probability of a chemical accident involving either chlorine or ammonia was -

evaluated in TENERA Calculadons 1934-2.2 005,"PiNGPToxic Chemical Analysis -

- Chlorine and Ammonia Probability Analysis" [Ref. 4.3.12] end 1934 2.2-006,-

! -*PINGP Toxic Chemical Analysis - Revised Chlorine and Ammonia Spill-Estimates" [Ref. 4.3.18]. The results from ths calculation show that the probability- of a SOO Line' railcar accident releasing chlorine that would incapacitate control room operators was 1.16 x 107 per year. - For SOO Line 4

L _ ammonia accidents, the probability is 1.47 x 10 per year for ammonia. These calculations included many _conservatisms, such as:

l L

E .- The total contents of the rail car are assumed immediately K.ailable _for-release. In reality, some rupture with slower dispersion of release is:

expected. Such slower release would reduce the CR buildup rate and provide operators additional time to take any required corrective measures.

. No credit for the difference in CR intake elevation versus the ground level release is ocluded. Such consideration would further increase the "Z" 1

29e.ini20/i9 3-t i TENERA l

Northern St:tes Power Compaly Control Room ll:bitability

!?alrie Ist=d N clear Generating P1:nt Toxic Chemical Study direction dispersion coefficient, thus reducing the control room concentrations.

. No credit is taken for the topology between the point of release and the CR intake. Additionally, no credit is taken for any intervening buildings.

Structures do exist onsite between the CR intake and point of release which could furth:r dilute the release.

. The number of chlorine a "i ammonia shipments wa!, conservatively assumed to be twice th". ac * .m ual shipment frequencies of 96 and 732, respectivay.

. According to Reference 4.3.12, the wini direction used in the analyses is divided into the same number of sectors as wind direction is divided into for measurement. This reference .dso 'emcastrates that if the wind direction was outside half the angle of a measured sector, CR incapacitation would not occur. Therefore, if it is assumed that a wind's direction within a sector is evenly distributed, an additional factor of 2 reduction in the probability of an incapacitating chlorine or ammonia spill results. This factor was not included in the probabilities calculated as furthe: refm' ement in wind direction measurements were not available.

- The model assumes that the wind direction remains unchanged over the course of the accident. Any shift in direction could eliminate incapacitation altogether.

- No credit is taken for the 20% fresh air CR intake that diverts to the computer room. This flow diversion will slow the rate of chemical buildup in the control room. The diverted flow will not have an immediate incapacitating effect on the CR operators as it enters a different room. I

- The release rate / train mile is conservatively calculated based on available information. Some of the inherent conservatisms include:

- The train accidents per train-mile traveled rate includes switching yard -

accidents. Such accidents are expected to occur more frequently and for shorter total miles traveled, thus increasing the accident rate. The SOO Line track near the PINGP includes no switchyards and is a fairly straight run of track.

- The releases / accident with hazardous material rate assumes that, of the 19 accidents involving hazardous material near PINGP, one resulted in a release. In actual fact, no releases occurred.

Consistent with the data proGded a he NRC in the development of Reference 4.1.8, it was assumed il at if the CR operator becomes i'

manu2on9 u2 TENERA

v Northern States Power Company Control Room II:bitability Pr:lrie islind N clerr Generating I'lant Toxic Chemic:1 Study incapacitated, the probability of exceeding 10CFR100 guideline exposure limits is 0.1. As noted in Reference 4.3.18, such a probability is conservative. This figure assumes little credit for actual safety system design and automatic operation.

In conclusion, the SRP and Reg. Guide acceptarr , criterion of 10-7 has been met.

Further reduction in calculated probabilities is possible should the above conservatisms be quantified further. The above probabilities are valid for up to twice the number of current actual shipments of chlorine and ammonia on the S00 Line.

19anu2on9 >u TENERA

c yy t d A

il u R bi t aS E t

i Jr F mu t N b I a ai c

'i c t

E i

l nr e

• a Eac p T mh oC n I

o Ri c s l

a o rl t 2 m

e n

a E "*deol^

C h T Mlep

{

r  ? y te a teT l

e =

r c G M 4?

d Ne I de Uce Dx I E ta s "

c s

m e

h C

d 1

3 -n e Su 4

1 3

Ma r )

1 Tfo '? s 3 s

e n

• o t

l u i si e

s

• rep t

R e YcSe la

" o r e

'e n " "" P S e

l

(

?

d y ed s

)

1 y )

t a u t

3 3 3 BS n o

n o

dl a " n

• s e g te m a g e Yc o n n g S e e imO i

g e

S e

Dno ( (

t n

l a

l

$ ) ) ) )

I' i 1 1 1 1 g "S 0 3 3 3 3 yn IO91 1 m n i m m ni t *N ,R sk e 1c s

e x sk e o o a

pr a I 1 I Y e Yc o Y eo N N men 1 I C S e

S S S oe G29 e e e e e e

e e

'Cr CG Ao S

( S( S

(

S

(

r r ea

%wlce li O s s 1

u e d r m

n aN e ls i

t

< ic e e l

e sh k k

s cu i o

t d l

a a : n e

e k

i A n d f u l a

I ne) l a r tc an ic c e d g mdn mh c n m no ic int n iTc e) t S a m imits iu o r u n i<m ur l

u t ir u

a r iu r l

e s mh t la S2 e3 my S3 e3 9

1 l e e n qt i p f i. d dy e eiP e eb e /

nI r

s h C

hO C

i. i 1 N SB d

%l iy S

lu ly oi Sl i2 D#

hWf C o S dn

(

h Cde S f

dn 0 2

c d s a e a f

c de 1

c p e eie r uro ie x r e n2 i p x i n3 R ltro iar ot TS 2

3 4 5 6

7 8

. otoli TSM o2 N2 TS oh N2o1 M I

1 9

N I' 1

c yy it d l

i u 4 4 b t t

aS i i F m kt # s 3

1. s 3

b1i tc Tfot a 0 1

n) o6 0 nk )6 1

i I 'ra aa md s 3 a t 3 I m

• e p 6 ed 7 cd e e ya 1 S na 4 S na mh c i e  ! e oC o

I n

S e

S e

( (

Ri sc o o l

t rT 2 1

4 l

)

4 n n 7 e 1 d 3}

o a R- Ml o  !

C h l C[ e se k* eI l r #* / ? w N e

t a "# EeT Gl d

YMdMo e

r Ro [

G UM N'

(

N

) ) ) )

4 4 C 4 4 t

3 3 3 3 o n . W ,

m N o o s w s s ;o ss i

N N e

Yc e t e

• e, YcSe et Ye c j

l s Y%e ^' S D,

I g e e

e e

e e

S 9 N' S S

_ ( ( ( (

_  ? ) ) )

d 4 4 4 e 3 3 3 e

c , n x ,

o , o E 5 s ; si Yeo o C ; o o e, t

o N 9 M N' Y ,c, N N Y, g

c M 3 S N , , c V S g

L g g (

1 5

1 1

' la? c s e imi e r t

o g o o o o o o

' h ep N M N N N N N N N

" Cor

• on P d

s e

a Il g d o o o o .o o o o o o te a N N N N N N N N N N n

img i

l

?

I t

n a

l )

P I

" IS 0

' 4 g 3 yn IO91 1 n i

ni sk o a

t #NR o o o o o e o o pr a I,T I

N .g N W N N N Y tce N N N S

men I K G 'z e o e Cr e CG Ao S

(

r r e

wea Podu e f o e h ta sN e s l a u s

c r ta te d o c

t e

l a la a i e c e c o

a a dn e e t

in or r n

ce c R: n oa sa e n A n u o l

lA A t

S a n im y )e ir myh u n o e tah t ht a 9 n o ly 1

l r

e e t> n o b G e r n 3 n' n i y 1 nI s h h i l

h mn m P ty o e el c h

t ht e /

0 r

e e C Cdle c p0 C AA I I SM Vi D DA E E M 2 1

t hi r i p i ohi05 x

toiar M

r . 0 1 7.

TS( 1 2 3 4. 5 6 8 9. 1 1 E

_ NP _ 9 1

Northern States Power Company Centrol Room IIaltitchility Prairie Island Nuclear Generating Plant Tmic Chemical Study c Tabte 31 I~ma? Reschs for Sunewd Chemicats Greater Ihan 2 Not listed in I!bminated 11ased M:nutes Using Probatniity per Chemical ACGill. NIOSil on Chemical RN Not lSceed? NUREG/CR-1741 yest of on Original Study,. g >ece y 7gg 39p p -

gg,g7 gy,$,g goe,g,%

Typel Yes

12. Toluene No No No (see secik, 3 4)

13. Ilammable I mtuh!, N.D S . Yes (l'ulp Mill (See Secten 3.4) 1.iquid)

14. resruleum ..

• Yes 9 ';n We Mm 3.4)

Naphtha

15. Ammonium y,3 Nitrate No (g , 3,ygg, y 4) g I'ertituer p,

N') N' "

(See Se km 34)

Yes

17. Phenot No No  %+

gg,, gecg 3.4)

Yes

18. Phnphoric No No No Not Available (see Section 3 4)

Acid

[Model I!]

19. iteniene 9, Phu phorous ,9 g 3y Dkhloride t

Yes

20. ) olten Sulfur -

No p,3,g 3,4) i Yes

21. Nickel Sulfaie g 3,) g 19131R120/19 3-16 TENERA

e yy it d A l

i 2 R b t r E aS ep m t

ib dc tyf"t ta u

N ai l s

'a a E T

I i I m b p e a mh + Iacn aC o

• n f I Ri c s l

o o rT l 3,

) )

t n

2 g7 4 l )

4 4 4 -

in.

s 'i a o n 3 3) 3 eDj l

a 3 , p C

i

, A hU I

M )g s , s mt sg g snl e r de

'r I e es! c e s I Yt c

,t a Ye e o te O

? y Ye te a u T Y %, SM Sy c

S i e M

em *l ( ,g el e

i c e ,

r G MUM N p S e 3 4

S

(

(

) )

4 4

~

'I 3

'd

• m 5 " sk d o C o e t

1. M N T 'c C g M Ye c S S

'I e e e e S S

( (

?

d e

h c a c u

ic E o g g o o

m. t M N N N C

E o N c

y 1 1- ev 1 3 r 7 1

le Su 3 b d a r e Tfo s l ts tah?se I

l u

s de emir t g ,

o o e ta h ep M  % g 9 N N l

f nCo 1

r r

im onPr

'i il l E d I' e

b I

d o o o

go o

te a

" M N y' N N n

.u

.ne t

n a

l l

i ? s s, s, I' 0 t ins lu4 l

ys tu) l )

g 1 tu s )4 y dI O9 y s 4 y n 1 b e3 b e3 b e3 ni r r r o o d d n o o d n a a pr t

i te s N.R l N g y* we e ynk e e o yitni N N e e yit ni o l l i IC a t men mG2 9 kl r e t

h I t

c e k l e

c e

oe NCr m0S h OS vOS n

CG Ao 0e e I( 5 s Oee t( S s EOe

( S s e

r r ea we ol c N e e

!c d P u I I

(

k d n

i x

sN e l kl i L n

y n l o O '

d a a a e D e h e #

ta n i c c iR n n de ye e C n '

e

- t my n "

n i

r 9 S a r r

> u mi nd a t

s ly ' 1

. l e e > h f d w Igf i

G n' ht 'Y /

nI s t t r C h h Cde h

C lu S

an CD 1 S l

u iu l

I.

1  % E b' "

0 2

_ e e ic p)

1 R

_ hi r n ipen tro iar oh i . M E

7.

- TSl 1 2_ 3 4 5 6 8 9

NP 1

t yy d A i u l

i bt R

aS r e

p n

o E

i t l hai ac ityret ita N .

E t

s r ci T

t I m ba aa e p .

e b ya mh oC o r

c n

o P I Ri c s o o l

rT 1 t

n 2 g7 4l e o n a n1 4 i -

C hUsR<l I

C[M e "re s/  ? w ta teGteT nI I d er oRo G MUM N d

'I

?

d ls e c

c a

k m

u e t a

h C N d V.

e i l 1 l 3 - wr '

s i

le Su s-b d ) )

a r o e 4 4 l s 3 3

' f s ta la? c se n*

i t

Ide imt r sk g u n-R e

taco m

he esp e t Ye c S

yg l

a n P r

e e n mo S e

p i

l I l (

)

• 4

• " 3

' em s

Y g' N o

k

."" e t

n a

l I'

f i ? s

, )

4 s. s, )

3 y )

4 s ins 01 t t t t )

y n g

dI O9 l

ysu4

) 3 y lu4) s y lus )4 3 3 3 lu4 s

1 b e3r m b e3 r b e3 r m m e3 r

ni t d n sk d d sk n o a a ite N. R s

e e t e e m e e m e 1 N e ni o g' N pr l ! I !

C yien or t Y ec S

Tnk it pn u loit g c Ye c S i tc men t oG2 ! 9 h l e

c hl c e e

l c e e e f.

e o e nOS w e wOS *0S

mOS NCr e e e n

CG Ao EOee S s S

( l l Oe

( S s e 1 Oee S s g

(

S

( 0e 5 s e

rr e a

( (

_ we ol t d

c o m l' u lo h k s O la t

o e e e sN e d h

o lc t

u t a

l e

m le b Mbn l

d l

a e e

r lc A d ic trn 'u u a med m i

_ ta n c n uo l A ly S t a

l e mm n ec i

t e

l vi ) 9 n

l m tah e o i + md) i u md S a l e i n n o i

y h t tn ise wn s r k tr ai d ai /

1 nI s h k el c h D

e i a ne AC D ie t e, aaBo l

oll l

o 0 r

e e C I DA M P R P CC( S S(S 2 1

hi r R tr ii 0 1

2 3

4. 5 6

7. . .8 9. 0 M

onl' 1 1 1 1 1 1 1 1 1 1 2 1 9

N 1 a

)

yy !d it c .I l

i ;

u

b. ;"

z it

• b .r mk 4

t m

a '

I a en t' c

8 MC I

e P mh 8 oC o N 3" Ri cs o o l

t rT 2

)

4 n n M 3}

o

• C h Ta%M!

'r e

r s

e te e mlA e m l? y c4 a u t leT Ye% S e W Grh M k(

e[

gd ge

,d y

g g .:

g g

? ) )

d 4 4 e 3 3 k c a c , m ci i s , ' "

m l

e ,c e 'c e t & Ye Y e h A s S C

d V g e

S r

y L ( (

1 'I

- t 9 3 n 1 eu 1 H S a r d ) ) ) )

4 e 4 4 4 Tf o s -

3 3 ts ta la? c 3 3 n

o a m a

Ideimtt N l

u sk g

5 , >

e Y5 tc C i Ye tc s e e R

e taCp h Y ed  % M"c' S N e S

e S

n o S l

a n

imi e mPr c e

e e S e

e e

e

~r 2 S S ( S(

! 1 ( (

. )

4 3

• ." g ,

g' .*' n sg e

' " . g M . y Y g t

n l

a ) ) )

P l ? 4 4 4 g inSi 0 y( 4) 3 3 3 yn d 1091 1 b 3 n n n ni t ten R " ,

d m s

e i o se s o e

r g_

a pr a s

y' e g' Y en e tc Y tce -

logt i . T t  %) p w Ye l i S S S men r lG2 W e 9 c

e e e CG oe WCr Ae bgsc e S

(

c S

(

e S

(

e I(

r e r a me l )

e Pic u d a d d

i e r

u n r d f sN n ic o n l

e w e mde e u l

l c S( A r d h o d S t

a dn a e pre e o Cn o e mih iu i t ic mt a t mc er h

p ce ir dy e i d

ni r u c s w m uo n e 9 S a m u r r u iui n u o ni a s n u r iclho s t 1 l e io lota e teh ntu a tr ta ev i d l o /

= ya dv f i rt nI s h 4lh h tu c n e - 1 ti e oi 0 r

e c C SC CM NC S h^

l' AA l' S %C TT l' l Sl i M 2 1

ii R l r M

troiar 1

2 3

4. 5 6

7 8 9. 1 L

1 2 i 2 2 2 2 2 2 2 2 2 3 3 3 t 9

NP 1

o Northern States l'ower Cornpany Control Room Italiitalfility ,

Prairie Island Nuclear Generating I'lant "lotic Chemical Study C Tabi 't 1 Final Results for Sarwyed Chemicals Greater han 2 -

Not IJsied in ,g Eliminated Itased '""*" "E

• IF IDLII Not Chemical ACGIII. N10Sil " ' #** I F' on Original Study? Exceeded?

or 29Cllt 1910' Pmperties? Model? [Model Incapacitatirwi Type)

Toxic Cheenicals Shipped by ILstge:

1. Chemical Yes I ertilizers (See Section 35) 19EM R120/19 3 20 TENERA

, u v N

~O ~ Northern States Pour Company. Control Room liabitabtlity Pralrle Islind Nuclear Generating Plant Toxic Chemical Study 4.0 - REFERENCES 4.1 NRC Regulations and Other Regulatory Requirements 4.1.1 10 CFR 50 Appendix A - General Design Criteria for Nudear Power

. Plants.

4.1.2 NUREG-0737, " Clarification of TMI Action Plan Requirements," Item III.D.3.4, "Contial Room Habitability," November 1980.

4.1.3 - Regulatory Guide 1.78," Assumptions for Evaluating the Habitability of a Nuclear Power Plant Control Room- During a Postulated Hazardous Chemical Release," June 1974.

4.1.4 Regulatory Guiue 1.95," Protection of Nuclear Power Plant Control Room Operators Against an Accidental Chlorine Release," Revision 1, January 1977. ,

4.1.5 NUREG 0800, Standard Review Plan," Revision 2,- July 1981.

a. Section 2.2.1-2.2.2," Identification of Potential Hazards in Site Vicinity,"

b. Section 2.2.3, " Evaluation of Potential' Accidents." '

c. Section 6.4," Control Room Habitability System."

4.1.6 -NUREG-0570, " Toxic Vapor Concentrations in the - Control Room Following a Postulated Accidental Release," June 1979.

4.1.7 NUREG/CR-1741, "Models for the Estimation of Incapacitation Times-Following Exposures to Toxic Gases or Vapors " December 1980.

" 4,1.8 NUREG/CR-2650, " Allowable Shipment Frequencies for the Transport of -

Toxic Gases Near Nuclear Power Plants," October 1982.

4.1.9 29 CFR, Part 1910, Subpart Z, Toxic and Hazardous Substances, Section 1910.1000, Air Contaminants," Revised July 1,1985. <

l l '4110 NUREG/CR-5042 " Evaluation of External Hazards to Nuclear-Power

! Plants in the United States," December 1987.

i 4.1.11 Draft Atomic Energy Commission General Design Criteria,1967.

.4.1.12 Regulatory Guide 1.70," Standard Forms and Content of Safety Analysis

- Reports for Nuclear Power Plants," Reymon 3, November 1978.

19EMai20/19 4-i TENERA r-, w, m , , < - ,

4 - ,- -

v

}W

• Northern States Power Company Control Room lixbitability f Pr:Irle Islind Nucle:r Generating Plant Toxic Chemical Study I 4.2 Toxic Chemical Survey Information 4.2.1 Letter from E.P. Murphy, TENERA, to P. Marabut, Director Hazardous ,

hiaterials SOO .Line Railroad, " Requesting Information on 11azardous 1 Chemicals Type, Weight, and Frequency of Shipment Past Prairie Island,"

dated January 29,1991.

4.2.2 Letter from E.P. Murphy, TENERA, to D. Kosec, Red Wing Fire Chief, "Pequesting Information on Toxic Chemicals in the Red Wing Area and Truck Accidents on Highway 61," dated January 29,1991.

4.2.3 Community Profile, City of Red Wing, Minnesota Department of Trade and Economic Development, dated May 16,1990.

4.2.4 Letter from E.P. Murphy, TENERA, to M.B. Henry, Manager llazardous Materials Burlington Northern Railroad, " Request for Information on Chemicals Transported on the Burlington Northern Line which runs past Prairie Island."

4.2.5 Prairie Island Plant, " Hazardous Substance Inventory," dated January 15, 1991.

4.2.6 Letter M.B. Henry, Manager Hazardous Materials Burlington Northern, to E.P. Murphy, TENERA. " Hazardous Material Traffic Statistics," dated February 21,1991.

4.2.7 -Letter from E.P. Murphy, TENERA. to P. Marabut, Director Hazardous Materials SOO Line Railroad, " Request validated information sent to T.

Verbaut, NSP, or :hemicals shippet past Prairie Island and average train size," dated March 5,1991.

4.2.8 Letter from D. Erickson, U.S. Army Corps of Engineers, to E.P. Murphy, TENERA," Response to request on chemicals transported through lock and dam No. 3," dated February 28,199' 4.3 Technical and Descriptive References l

4.3.1 " Prairie Island Nuclear Generating Plant - Unit 1 and 2 - Main Control Room Habitability Study - NUREG-0737," prepared by Bechtel Power Corporation, dated May 1981.

4.3.2 " Prairie Island Generating Station Toxic Chemical Study - Incapacitation Levels," prepared by Bechtel Power Corporation, dated June 20,1984.

l. 4.3.3 PINGP Design Modification 89YO15, " Cooling Water Treatment System Upgrade," July 19,1989.

l ..

tenazon9 +2 TENERA i

[ s,.

Q' Nath:rn States Power Ccmpany

_ Pr:Irle Ist:nd Nuclear Ge:erating Plant Control Room H:bitability Toxic Chemical Study 4.3.4 "G ATX Tank Car hianual," published by General American Transportation -

Corporation, Fifth edition, dated February 1984.

4.3.5 Pocket Guide' to Chemical Hazards, U.S. Department of Health and Human Senices, National Institute for Occupational Safety and Health, September 1985.

4.3.6 Drawing NF 108578, " Prairie Island Area Emergency Planning Zone (EPZ)," Revision 2, dated April 16,1990.

4.3.7 hiap of hiinnesota and Wisconsin, published by the American Automobile Association,1990 edition.

4.3.8 TENERA 'alculation, 1934 2.2-001, " Accidental Chemical Spill Analysis Technical Approach, Example - and Lotus . (Ver 2.01) Spreadsheet Verification."

4.3.9 TENERA Calculation, 1934 2.2-002, "PINGP Toxic Chen ical Analysis -

Seven Chemicals Evaluated."

4.3.10lTENERA Calculation, 1934 2.2-003, "PINGP Toxic Chemical Analysis -

Eight SOO Line Chemicals."

4.3.11. TENERA Calculation, 1934-2l1-004, "PINGP Toxic Chemical Analysis -

Nine Boiling Point Below Ambient Chemicals." ,

4.3.12 TENERA Calculation, 1934 2.2-005,"PINGP Toxic Chemical Analysis Chlorine and Ammonia Probability Analysis." -

4.3.14 TENERA hiemorandum from Andy hicClymont to Ed Murphy,

Subject:

, - Prairie Island Railroad Accident Hazardous hiaterial Release Frequency Derivation, dated May 10,1991.

4 4.3.15 1990-1991 Threshold Limit Values for Chemical Subst.mces and Physical -

Agents and Biological Exposure Indices, copyright American Conference . <

on Goverm tental Indatrial Hygienists. .

I - 4.3.16 Prairie Island Nuclear Generating Plant, Updated Safety Analysis Report.-

4.3.17 Department of Transportation Coast Guard (CHRIS) Hazardous Chemical

' Data, October- 1978 4.3.18 TENERA Calculation 1934-2.2-006, "PINGP Toxic Chemical Analysis -

Revised Chlorine and Ammonia Spill Estimates."

L 19EMR120/19 4-3 TENERA

..