ML20153G221

Toxic Chemical Study

ML20153G221
Person / Time
Site:	Trojan File:Portland General Electric icon.png
Issue date:	01/31/1988
From:	PORTLAND GENERAL ELECTRIC CO.
To:
Shared Package
ML20153G219	List: ML20153G214 ML20153G221
References
RTR-NUREG-CR-1741 NUDOCS 8809080082
Download: ML20153G221 (10)

Text

.. - . _ _ _ _ . . ._ . _ _ .- - .-._ _. - - _ -. -. . - - .

Tro}anNuclearPlant Document Control Desk

- Dochet 50-344 Attachment 1.icenso NPF-1 Page 1 of 10 l

\

l l

L i I 1

TROJAN NUCLEAR PLANT  !

l l 1

)

i TOXIC CHEMICAL STUDY l i -

4 L

I.j ,

!. - Chlorine-  !

1 l

l i i 1 i i

) r j

{

! t l

I 1

i 1

[

i

(

I

! l l

J January,1988 I l t i l 1

1 l

I I l

I i

l ( r 1

(

1 <

I 8809080082 880902 PDR  !

l P

ADOCK 05000344

) PDC i

i i I

l I

I r

_ _ _ . _ _ _ _ . _ , _ _ _ _ _ _ , _ _ . _ _ _ _ . _ . _ _ _ . _ _ _ , _ . . . _ . _ _ . . _ _ _.___..___..___.._______2

Tro'jan Nuclear Plant Document Control Desk

. Dockst 50-344 Attachment License NPF-1 lage 2 of 10 Table of Contents

1. INTRODUCTION 1_

2. CONCENTRATION MODELS 2 3.NUREG/CR 1741 MODELS 3

,4. CHEMICAL DESCRIPTION FOR CHLORINE, Cl2 3

5. ANALYSIS PERFORMED 4 5.1. Control Room isolation Not Consbered . . . . . . . . . .. 4 5 2. Control Room isolatkm Considered .............. 4 6.RESULTS 6

• 6.1.Onsas cNorine ......................... 6 6 2 Offs!!e Worine (B.N.R.R.) ................... 7

7. CONCLUSION 7 8.REFEMENCES 8

8 Trofcn Nuc1 car Picnt Docunort Control D2ek

• pockst 50-344 .

Attcchmtnt

,g I.icense NPF-1 Page 3 of 10

/

1. INTRODUCTION

(

In the updated FSAR Section 6.4 (Amendment 6 dated July 1987), Habitability Sys-tems, chlorine is identified as a potential hazard to the control room operators if an ac-cidental spill occurred. As stated in the FSAR, chlorine detectors have been installed in both the Chlorine Building and the control room for purposes of detection and sub-sequently isolation of the control room. De chlorine detector at the control room air intake will automatically isolate the control room ventilation system upon detection of chlorine concentrations at 5 ppm. De possible sources of these chemicals are: on site storage and Burlington Northern railroad on the west side of the plant.

De criteria of Regulatory Guide 1.78 is used specifying that operators must have at least two minutes to put on self contained breathing apparatus.

In early 1981, NUREG/CR 1741 entitled "Models for the Estimation ofIncapacitation Times Following Exposures to Toxic Gases or Vapors" was published. The report presents a methodology to predict operator incapac.itation for one time exposure to toxic chemicals. It consisted of using 5 models (A E) covering significant physiologi-cal and toxicological effects to humans.

he Chlorine Analysis was then undertaken to insure that the present control room ventilation system and chlorine monitor adequately protect the control room operatord from an accidental chlorine spill. The concentration models were based on NUREG-0570, and the incapacitation models on NUREG/CR 1741.

(

1

Trojan Nucl0:r Plcnt Docun nt Control Desk

, pock 3t 50-344 Attecha nt 1.icsnsa NPF-t Page 4 of 10

2. CONCENTRATION MODELS

(-

The analytical models used to calculate the concentration of chlorine in the control room atmosphere in the event of a spill are consistent with those described in NUREG.

0570 (Ref.1),

nese models include the following bases and assumptions:

a. Consistent with the criteria of Kegulatory Guide 1.78, one container of chlorine (tank car or cylinder) was assumed to fall, releasing all ofits contants,

b. That fraction of chlorine which would flash to a gas at atmospheric pressure is assumed to be released as a puff, ne remaining chlorine is assumed to spread uniformly on the ground and evaporate over time. it is assumed conservative-ly that no losses of chlorine occur as a result of absorption into the ground, flow into the river, cleanup operations, or chemical reactions.

c. The initial puff due to flashing, as well as the continuous plume due to evapora-tion,is transported (and diluted) by the wind to the control room air intake.

d. Atmospheric dispersion factors are calculated using the methodology of Regulatory Guide 1.78 and NUREG 0570. Dilution due to building wake ef-fects from the plant structures using Meteorology and Atomic Energy (Reference 9) methodology (Section 3.142 page 112) is conservatively con-sidered. Re building cross sectional area orthogonal to the wind speed is ap-

' plied to account for the building wake correction factor in the atmospheric dispersion factors calculation.

l 1

e. Conecntrations in the control room as a function of time were calculated for both normal control room ventilation without automatic isolation and control room normal ventilation with automatic isolation. Subsequently, the latter case assumed inleakage from the adjacent areas following isolation.

I i

]

I

i 2

l

Trojcn Nuclear Plant Docuntnt Control Dask

• pockat 50-344 Attachment

, License NPF-1 Page 5 of W

3. NUREG/CR-1741 MODELS Human exposure to airborne toxic materials produces a wide range of physiological and toxicological effects. For incapacitation effects there is a threshold concentration below which the body can eliminate, transform or otherw:se act on the chemical to negate its effects. Above this threshold, there are two principal physiological modes which dominate: concentration dependence and dose dependence. For concentration ,

dependent chemicals, the total dose received is not as important as the concentration of the chemical during exposure. Dose dependent chemicals produce an effect that is directly related to the total exposure regardless of the concentration at any given time.

NUREG/CR-1741 presents 5 models to describe incapacitation. However Chlorine at the Trojan Nuclear Plant belongs in the first or "A" model which is described below:

Model A. Concentration Dependent -Immediste Sensory Irritants:This model describes a procedure for predicting the time to incapacitation for immediate sensoryirritants (e.g. Ammonia, Chlorine, and Sulfur Dioxide).The effects are concentration and not dose dependent.

Chemicals classified as immediate sensory irritants are corrosive or desiccant l

in their action.They inflame skin or mucous membrane especially when moist.

Rey stimulate nerve ending in the cyes, nose, and oral cavity and inhibit respira-tion. ney have essentially the same effect on animals as on humara and the ex-posure concentration is of greater significance than the duration of the exposure.

4. CHEMICAL DESCRIPTION FOR CHLORINE, Cl2 Presented below is a descriptive summary for chlorine .The information is taken from literature on the subject, ne summary focuses on the effects chlorine has on human beings.

Chlorine in its gaseous form is greenish yellow in color. It has a disagreeable, suffocat-ing and irritating odor which is readily detectable at 3 5 ppm. Irritant effects to eyes, nose, throat and/or face were noted at low concentrations. Effects on the upper and lower respiratory tracts and pulmonary edema were reported at high concentrations. '.

l Chlorine becomes highly dangerous for exposures of 30 minutes at 40-60 ppm;it be-ccmes fatal for exposures of 30-60 minutes at 800 ppm and fatalin a few breaths at 1000 ppm (Reference 2).

I 3

i

Trojcn Nuclocr Plant Docu;Gnt Control D3sk Dockot 50-344 Attach: nt 1.icense NPF-1 #

5. ANALYSIS PERFORMED he Bechtel standard computer code TOXG AS (NE314) was used to calculate the con-centrations of chlorine and the resulting incapacitation times in the control room. Two types of analyses w:re performed:

1. No control room isolation.nis analysis was performed to determine the need for chlorine detection and automatic isolation.

2. Controt room isdation.This analysis was performed to insure that the chlorine detectors and control room ventilation systems adequately protect the operators.

5.1. Control Room isolation Not Considered An analysis of a control room without isolation was performed to determine which of the following sources of chlorine pose a hazard to the control room operators if ac-cidentally spilled:

1. One ton chlorine tank located inside the Chlorine Building.

2. 90 ton railroad tank car transported on the Burlington Northern railroad on the west side of the plant.

1 Results of this analysis indicate that both onsite and offsite chlorine require further in-1 vestigation.

5.2. Control Room Isolation Considered Further analysis was performed for onsite and offsite chlorine assuming automatie isolation of de control room. A monitor setpoint of 5 ppm was assumed. A total of 6 seconds delay time prior to isolation was used; 3 seconds for the monitor to actuate after the concentration at the intake reached the setpoint level and 3 seconds for the damper to close. Conservatively, no travel time from the control room intake to the control room was assumed. Utilizing Standard Review Plan 6.4, this analysis conser-vatively assumed that the inleakage rate into the control room following isolation con-sisted of one half of the pressurization make up flow plus 10 cfm or leakage from control room ingress and egress. i

{

Evaluation showed that chlorine transported through the Burlington Northern Rail- l

' road on the west side of the plant still pas a threat to the control roem operators if

( all of the controlinleakag: rau is assumed to come directly from the outside air,i.e.

4 l

Trojen Nuclect Picnt Docun nt Control Dosk

, pockit 50-344 Att*ch $"t

, , 1,1 cons,e NPF-1 Page 7 of 10 the control building envelope is not considered. In reality, following control room isolation, the leakage into the control room will be primarily from adjacent building areas, and not directly from the outside air. Therefore, the analysis was modified con-servatively to take into account leakage from adjacent areas to the control room, t

If chlorine concentrations at the air intake exceed the monitor setpoint, the control room isolation dampers DM 10501 A & B and DM 10504 A & B are closed and the normalventilation system (CB 2) supply fans (V-141 A & B) and recirculation fan (VC-148) are shut off automatically on the High Toxic Gas isolation signal, ne emergen-4 cy ventilation system (CB 1) is then initiated in a full recirculation mode through the l charcoal filters. De ventilation systems for the other areas in the control building are also automatically secured, thus no direct outside air enters the control building and, consequently, the control room. '

l Since the control room is isolated, inleakage into the control room can only be from the adjacen: areas. A detailed description of the adjacent areas is shown below:

{ s North:

The north wall of the controlioomis the outsfde wall of the control build-

{ ing.nis is a concrete block wall 2' 10" thick with no penetrations to the

outside. '

! East: ne east wall is adjacent to the auxiliary building. Dis is also a concrete i

( block wall 2' 10" thick with no significant air leakage between the build.

ings. '

West

The west wall is adjacent to the turbine bu0 ding (2' 10' thick concrete i

j block). One leaktight door leads from the turbine building to the view- i ing gallery.

South:  !

De south wallis adjacent to the remainder of the control building.ncre are two leaktight doors and one security door.ne upper portion of the control room is adjacent to the mechanical equipment room. Scaled i

HVAC penetrations and one leaktight door connect the mechanical i

equipment room to the control room.

Ceiling:

l De ceiling of the control room is also the roof of the control building.

This roof is a concrete structure with sealed pentrations for the roof drains.

i Floor:

ne cable spreading room is directly below the control room. A large number of sealed penetrations exist between these rooms.

1 Since the outside wall, the wall to the auxiliary building, and the ceilins of the control j (

room are thick concrete structures with no penetrations, r'o leakage from the outside i

I

Trojcn Nuclect Plcnt Docur nt Control D ck pockst 50-344 Attechtsnt Page 8 of to LicGns? NPF-1 air is expected to occur. Leakage is expected to occur frcm the outside to the mechani-(, cal equipment room through the air intake louvers. Since the normal ventilation sys-tem for the control building is not drawing air, no driving force exists, therefore this leakage is expected to be small. However, to bound uncertainties in the analysis, one half of the total control room inleakage was conservatively assumed to come through these sources and assumed to be at outside air concentrations.

Most of the leakage into the control room will occur through the doors and penetra-tions to the control building. Since the control building ventilation system is secured, the only mechanism available for the chlorine to enter the control room is first to leak into the control building, mix slowly in the free air space available, then leah into the control room at a lowered concentration. Some leakage can occur through the door leading from the turbine building to the viewing gallery. The ventilation system of the turbine building is secured manually by operations as a subsequent action in ONI 54.

Prior to securing of this system, there will be no more than 10 air changes per hour.

This air change rate was conservatively used to represent the leakage into the areas ad-jacent to the contrcl room. One half of the total control room inleakage was then as-sumed to come fu,m these areas.

Time between detection and incapacitation for both onsite and offsite chlorine were calculated to determine if the R.G.1.78 limit of 2 minutes was met.

(

l 6. RESULTS 1

De study showed that in order to adequately protect the control room operators and to meet the two minute requirement, chlorine detection and automatic isolation of the nu mal ventilation sytem is required. '

An evaluation of the potentialleakage pathways into the control room following isola.

tion showed that direct inleakage from the outside air is not likely, thus the leakage is ,

expected to come from areas adjacent to the control room. However, to bound uncer-tainties in the analysis, one half of the total control room inleakage was conserva assumed to come from adjacent areas and the other I f from outside air.

6.1. Onsite chlorine Results showed that the incapacitation times for all the cases considered are over 10 minutes, thus above the required 2 minute limit. The cases analyzed used different control room inleakages (150,190,235, and 450 efm) and monitor setpoints of 1 ppm and 5 ppm. Dese cases bounded the actualleakage rate into the control room. ,

, i, l

i

Trojan Nucioer Plant Documsnt Centrol Dask

, , Docket 50-344 Attachment License NPF-1 Page 9 e' 10 6.2. Offsite chlorine (B.N.R.R.)

Figure 1 shows the concentration in ppm as a function of time for each of the control room inleakage sources and the resultant total control room concentration. Also shown on Figure 1 is the incapacitation function for model A (a dimensionless function of con-centration). Incapacitation, as defined by this model, is reached when the incapacita-tion function reaches unity. As shown on Figure 1, the incapacitation function reaches unity 127 seconds after the chlorine monitor alarms. Since more than two minutes are available before incapacitation can occur, the control room operators are adequately protected against a chlorine tank sceldent.

iD000. i.ne.t 0  % -1 I I

\:

~. . m -; ,

. .. io ,

l / s /

/ \ t -

7 I

p

{g

- N

[

/ + -I p

l /u.

\

W ' >

ai

/

' / s i

ai 11 Y g a.$ .ci i $. io. .

Figure 1 - Chlonne Concentration for 5 ppm Monitor Shtpoint -

l t

i

7. CONCLUSION As discussed above, the control room operators are adequately protected with the present chlorine detectors and controt room ventilation system design. Chlorine detec-tors can be set at a monitor setpoint of 5 ppm.

j 7

. Trojan Nuclocr Plant Docursnt Control D3sk

,. ,D6ckot 50-344 Attech;Gnt

,' Liconso NPF-1 Pegs to of to

8. REFERENCES

1. NUREG-0570 "roxic Vapor Concentrations in the Control Room Following a Postulated Accidental Release", James Wing,USNRC ONRR, dated June 1979

2. Effcets of Exposure to Toxie Gases - First Aid and hiedical Treatment by Braker, hiossman and Siegel, Second Edition

3. Patty's Industrial Hygiene and Toxicology by George D. and Florence E. Clayton, Volumes 2A & 2B, Third Edition.

4. Dept. of Transportation, ' Coast Guard CHRIS Ilazardous Chemical Data", Oct.

1978.

5. "Documentation of the Threshold I.imit Value", Fourth Edition,1980, American Conference of Govenunent Industrial liygienists Inc.

6. "User hianual for Bechtel S.C.P.TOXGAS NE314 Release 2", Bechtel Power Cor-poration, hiay 1982. (Bechtel Proprietary)

( 7. NUREG/CR-1741 *htodels for the Estimation ofIncapacitationTimes Following Exposure to Toxie Gases and Vapors", Gordon J. Smith, David E. Bennet, Sandia National 1.aboratories, Dec 1980.

8. FSAR Section 6.4 (Amendment 6 July 1937).

9. hieteorology and Atomic Energy 1968 by David Slade, U.S. Atomic Energy Com-mission, published July 1968 i

( l i

1 8

\