W3P84-2152, Forwards Update Re Toxic Chemical Survey of Industries Surrounding Plant.Protective Features Will Provide Adequate Protection for Control Room Operators.Probability of Immediately Dangerous Levels Being Reached Unlikely

Forwards Update Re Toxic Chemical Survey of Industries Surrounding Plant.Protective Features Will Provide Adequate Protection for Control Room Operators.Probability of Immediately Dangerous Levels Being Reached Unlikely

ML20095C965
Person / Time
Site:	Waterford
Issue date:	08/09/1984
From:	Cook K LOUISIANA POWER & LIGHT CO.
To:	Knighton G Office of Nuclear Reactor Regulation
References
W3P84-2152, NUDOCS 8408230179
Download: ML20095C965 (11)

Text

.

Lo uisI AN A / 442 oe-o~o raser POWER & LIGHT / NEW oALEANE LoUSANA

. ao oox.oo.

70174400. G (504)388-2345 USE0kS SYS W3P84-2152 3-A1.01.04 August 9, 1984 Q-3-A29.02 Director of Nuclear Reactor Regulation Attention: Mr. G. W. Knighton, Director Licensing Branch No. 3 Division of Licensing U.S. Nuclear Regulatory Commission Washington, D.C. 20555

SUBJECT:

Waterford 3 SES Docket No. 50-382 Toxic Chemicals

Reference:

LP&L Letter W3P84-1094 dated 4/30/84

Dear Sir:

The above reference submitted to you the results of the LP&L-conducted toxic chemical survey of the industries surrounding Waterford 3. The purpose of this letter is to give you an update on the same subject. The data collected in the survey were evaluated and our evaluation indicates that the protective features described in the FSAR will provide adequate protection for the control room operators. ,

The attachment to this letter is a complete report on the analysis performed and the model and assumptions used. The results of the analysis are summarized in Table 1 of the attachment.

In comparison with Regulatory Guide 1.70 and Standard Review Plan Subsection 2.2.3, the probability of Immediatgly Dangerous to Life or Health (IDLH) levels being exceeded is smaller than 10 per year if credit is taken for odor detection and will be reduced even further by the toxic gas detectors. Also, considering the conservatisms of the analysis, the results indicate that the level of protection is marginally adequate even without reliance on odor detection or the broad range toxic gas detectors.

If you have any questions, please advise.

Very truly yours, K. W. Cook Nuclear Support & Licensing Manager KWC/PC/ch cc: E.L. Blake, W.M. Stevenson, J.T. Collins, D.M. Crutchfield, J. Wilson, G.L. Constable, K. Campo 8408230179 040809 OO(

PDR ADOCK 05000382 A PDR

',I

Pign 1 of 10 ATTACHMENT RESULTS OF THE 1983-1984 SURVEY OF T0XIC CHEMICALS STORED OR TRANSPORTED IN THE VICINITY OF WATERFORD SES UNIT 3

- BACKGROUND AND

SUMMARY

In.its letter"of February - 24, 1983, the NRC requested that the FSAR be updated to include all significant quantities of hazardous chemicals near the Waterford 3 site. It further requested that any additions to the listing be evaluated to ensure that the consequences of any accident involving these hazardous chemicals are within the design basis envelope for Waterford 3.

In response to this request we have updated the survey of major industries within 5 miles of.Waterford 3 to determine the types and quantities of hacardous materials which are stored or transported in the vicinity of the site. Our evaluation of the materials indicates that the protective features described in the FSAR will provide adequate protection for the control room operators.

INDUSTRIAL SURVEY To determine the types and quantities of hazardous materials which are stored or transported in the vicinity of Waterford 3 we had sent detailed .

. written questionnaires to the major industrial facilities located within 5 miles of Waterford 3. Questionnaires were sent to, and replies received from, the following facilities:

- Beker Industries Occidental Agricultural Products EI Dupont de Nemours & Co. Shell Chemical, Taft Plant GATX Terminals Corp. Shell Oil, Norco GHR Energy. Inc. Shell Western E&P, Crawfish Plant Hooker Chemical Union Carbide Linde Division LP&L Little Gypsy SES Union Carbide, Star Plant LP&L Waterford SES 182 Union Carbide,-Ethylene Oxide /

ye ant, Taft Occidental Chemical, Taft Witco Chemical Co.

m

p Pzga 2 of 10 In addition, we obtained from the Missouri Pacific Railroad Co. data on the types,-quantities and shipment frequency of hazardous materials through Taft, LA in 1983. When necessary, written information provided by the

. industrial facilities and Missouri Pacific was supplemented by telephone inquiries.

ANALYSIS OF DATA AND RESULTS The survey revealed that many of the toxic chemical sources listed in the FSAR, such as ammonia and chlorine, are still present. In addition, a number of new chemicals were identified. Using the methodology described in Section 2.2.3.3 of the FSAR, all of the toxic chemical sources were examined to identify those chemicals which, on the basis of their proximity to the plant, volatility and toxicity, could pose a threat to the control room operators in esse of an accidental release.

The locations and quantities of chlorine and ammonia were found to be the same as in the FSAR. As demonstrated there, for these chemicals, the redundant, dedicated detectors will assure control room habitability. In fact, we reanalyzed the ammonia sources and found that the 5 ppm setpoint and 2 second response time of the ammonia detectors which were assumed in the FSAR can both be relaxed. The analysis showed that a detector set to isolate the control room at 50 ppm of ammonia and having a 20 second response time would provide adequate protection.

Chemicals other than ammonia and chlorine were analyzed to assess the severity and likelihood of the threat to control room operators. We also evaluated the feasibility of protective action based on odor detection or notification by offsite authorities. Since the broad range toxic gas detectors are not yet operational, no automatic control room isolation or alarm was assumed.

The results of this analysis are presented in Table 1. This table lists all the chemical sources which were found to be capable of producing IDLH*

• Immediately Dangerous to Life or Health. (Ref. 1)

I L

Page 3 of 10 concentrations in the control room within 30 minutes of the plume's arrival on site under a spectrum of meteorological conditions, including 5 percentile conditions. Though the 30 minute period is somewhat arbitrary, we considered it reasonable to assume that within this period the control room operators would be notified that an accident has occurred by either the St. Charles Parish Industrial Hotline, or other means. We expect it would take the parish between 2 and 5 minutes to notify LP&L of an emergency condition at a nearby industrial facility, after the parish were notified.

The name of each chemical is shown in Column 1. Columas 2 and 3 list, respectively, the IDLH value and-the odor threshold for each chemical.

Except as noted in the notes to the table, the IDLH values were taken from

-Reference 1 and the odor thresholds from Reference 2. Whenever both

. detection and recognition levels were listed in Reference 2, the latter value was used.

Columns 4, 5 and 6 show, respectively, the location of the source from which the release is postulated to occur, the annual shipment frequency for mobile sources and the amount of material which is assumed to be released.

These data were provided by the survey respondents. The amount released, listed in Column 6, is the quantity contained in the largest single storage tank. For the Missouri-Pacific data, it is the average lading for each chemical.

Column 7 shows the minimum time from when the release occurs to when the IDLH concentration is calculated to be reached in the control room for the spectrum of meteorological conditions considered. Neglecting the possibility of notification prior to release, the values indicate the time available for notification by offsite authorities. Column 8 shows the minimum time from odor detection to when the IDLH concentration is reached.

This indicates the minimum available response time by the operators based on odor detection. Column 9 shows the maximum concentration in the control room at two minutes following odor detection - time within which the control room operators would be expected to don breathing apparatus ( .

p lL Paga 4 of 10 h

.It should be noted that the values in columns 7, 8 and 9 may not correspond to the same meteorological conditions. Generally, IDLH is reached most rapidly'at relatively high wind speeds whereas the maximum concentration at 2 minutes following odor detection occurs at lower wind speeds and very

~

stable conditions. As such, the values presented in columns 7, 8 and 9 present the envelope of " worst" consequences.

The likelihood of transportation' accidents which can result in the consequences presented in columns 7, 8 and 9 are indicated in columns 10 and 11. These columns list the annual probabilities for IDLH levels being exceeded within'30 minutes of plume arrival on site or within 2 minutes of odor detection, respectively. These probabilities were calculated as follows: The length of railroad track, highway or shoreline within a

five-mile radius of the plant was divided into a number of segments. An accident was postulated at the center of each segment resulting in a total

- release of lading. -Using'the median windspeed for each meteorological

-stability class and the joint frequency data in FSAR Tables 2.3-126 through 2.3-132, the probability that the accident at a given location would result in control room concentration in excess of IDLH levels was

~

calculated. The resultant values of probability were multiplied by the lengths of the segments and summed to obtain the probability-weighted hazard length. This length was then multiplied by the annual shipment

!~

frequency in Column 5 and, finally, by the appropriate probability per vehicle mile of an accident involving a carrier of hazardous material which results in a loss of lading ( }.

Column 12 lists the ionization potentials (in electron volts) of the listed chemicals (5) . All chemicals with ionization potentials less than 11.7 eV would be detectable by the Broad Range Toxic Gas Detectors. The ionization potential of sulfur monochloride was not found in the literature; however, j as_' indicated in the notes to the table, we expect this chemical to also be l detectable, i

The results in Table I show that only three stationary sources can produce l IDLH levels in the control room within 30 minutes of the plumes' arrival on t

-site. Of these, thionyl chloride gives an adequate warning through odor L'

+' c z

+.~

Pagn 5 of 10-detection, and can also be detected by photoionization. Hydrogen chloride and sulfur dioxide do not provide a 2-minute odor detection warning, nor can .they be detected by photoionization. However, we were informed by Union Carbide that they plan to terminate'the process using hydrogen chloride in November, 1984, in which case this hazard will no longer exist.- In the case of sulfur dioxide, the IDLH concentration is not a reached until 23 m he'saftertheaccident. This should allow sufficient time for notifi:ati6n by the St. Charles Parish Industrial Hotline.

s Furthermore, the actual concentration is likely to be much lower. Sulfur

.s s dioxide has over twice the density of air and will tend to concentrate near the ground. Also' while above tihe IDLH ' level, the calculated concentration of 280 ppm 2 minutes aftier odor detection is not likely to be immediately incapacitating. Patty (5) and Fairhall (6) give 400-500 ppm as the

. con' centration that is dangerous for short exposures.

The total' probability for a transportation accident causing an IDLH concentration in the control room within~30 minutes of the plume's arrival

-6 on' site is 2.9x10 per year. The probability of exceeding IDLH before the operators can don breathing apparatus (2 minutes after odor detection) is

~

9.4x10 per year. This latter value includes those chemicals which cannot be ' detected by odor. If~only chemicals which cannot be detected by the Broad Range Toxic Cas Detectors are considered, the probabilities are

~7 ~

reduced'{d6.1x10 without odor detection and 4.4x10 with it (assuming adequate detector sensitivities and response times).

The above probability estimates may be compared to the guidance in Section 2.2.3 of Regulatory Guide 1.70 ( }. Also, Section 2.2.3 of the Standard j Review Plan } indicates: -

"...the identification of design basis events resulting from the presence of hazardous materials or activities in the vicinity of the plant is acceptable if the design basis events include each postulated type of accident for which the expected rate of occurrence of potential i

exposures in excess of the 10 CFR Part 100 guidelines is estimated to exceed the NRC staff objective of

~

approximately 10 per year."

l

$i ,

Pzgs 6 of 10 Furthermore,

"...the expected rate of' occurrence of potential exposures in excess of the 10 CFR Part 100 guidelines of approximately 10 ~0 per year is acceptable if, when combined with reasonable qualitative arguments, the realistic probability.can be shown~to be-lower."

As indicated before.c the model used to evaluate the consequences of postulated chemical releases is very conservative. Among the conservatisms'are the following:

1. Initial dilution due to entrainment of_ air in flashing liquids is neglected.

2. Material spilled on the ground-is assumed to evaporate from a pool which has a depth of I cm. While this conservative assumption eliminates the

'need for a detailed characterization of the spill terrain, it results in unrealistically large evaporation rates.

3.-Both source and receptor are assumed to be at ground level. In reality, the effects of gravity on dense vapors from pressurized liquids would tend to reduce the concentration at the 17 meter height of the control room outside air intake.

'4. Except for atmospheric turbulence, processes which would tend to reduce the concentration of airborne chemicals-are neglected. These processes would include condensation and dissolution in atmospheric water vapor by cr>1d gases such as hydrogen chloride and' sulfur dioxide as well as deposition on the ground.-

l5. IDLH values are appropriate for an exposure period of approximately 30 minutes. Significantly higher concentrations can be tolerated for the short time in which the operators would take protective action.

6. The possibility of notification prior to plume arrival on site is not.

considered.

y Pagn 7 of 10 Considering the additional fact that incapacitation of the operators need not result in exposures in excess of the 10 CFR Part 100 guidelines, the.

results of our analysis indicate that the' level of protection is marginally adequate even without. reliance on odor' detection or the Broad Range Toxic 6

Cas Detectors.' The probability is below the 10 per year criterion if credit is taken for odor detection and will be reduced even further by the toxic gas; detectors. Thus, the results indicate that the protective

. features described in the FSAR provide adequate. protection for the control room operators.

L c

. -. , - - . . , n , n. , n - -.,---,e- m-. ,c y

y Pcg3 8 of 10 References.

t

~

fl. . NIOSH/ OSHA. Pocket Guide to' Chemical Hazards. DHEW (NIOSH)

Publication No.78-210, (1978).

2.- Fazzalari,.F.A. ned.L. Compilation'of Odor and Taste Threshold Value Data. American Society for Testing and Materials (1978).

3. U.S. Nuclear Regulatory. Commission. Regulatory Guide 1.78.

Assumptione :for Evaluating the Habitability of a Nuclear power Plant

. Control Room-During'a' Postulated' Hazardous Chemical Release (June 1974).-

-4. -

Arthur D. Little, Inc. A Modal Economic and Safety Analysis of the Transportation of Hazardous Substance in Bulk.' COM-74-11271. National Technical Information Service.(1974).

5. Weast, R.C., ed. . Handbook of Chemistry and Physics, 64th Ed., CRC Press.-Inc. (1983).

-f

.6. Patty, F.A.,led. Industrial Hygiene and Toxicology, Second Ed., Vol.

II. 'Interscience Publishers-(1967).

> 7. 'Fairha11 .L'.T., Industrial Toxicology. Hefner Publishing Co. (1969).

8. U.S. Nuclear Regulatory Commission. Regulatory Guide 1.70. Standard Format and Content of Safety Analysis Reports for Nuclear power Plants, 1 - LWR Edition. Rev.'.3,.(November 1978).

L 9. _ U.S. Nuclear Regulatory Commission. Standard Review Plan for the Review of Safety. Analysis Reports for Nuclear Power Plants,. LWR Edition, NUREG-0800 (July 1981).

10. Sax,[N.I. Dangerous Properties of Industrial Materials, 3rd Edition.

- Van Nostrand Reinhold Co. (1968).

-w - . , w ..--a-- , -w ,n ---m=-w=n-y ,-yt+ - ryv,-wry-ww--v e g

• t-w- ~*5* -ev=-'-tr -r-*-e e f- +a > -- W-' 'm

d..

- ' TABLE'I Page 9 cf 10l

Chemicals in Vicinity of WSES-3 Posing' Potential Hazards to Control Room Personnel ~

• 1 -- 2 3, 4 'S 6 7. ;8" 9- 10 211 12' l Chemical IDLH- ' Odor- Source ** Shipment ' Amount Time of IDLH From: Conc.'at P(IDT.H) '(yr ) Ion..

- (ppm) l. thr. .'(kg.) Release. Odor. . th r. Odor +2 min dE30 min 40 dor +2 min Pot.

.(ppm)~ Frg). .

.(yr .(ppa) -(eV)

. 21 :7.1x10 -7~

Acrolein' 5 MPRR 164 2.4' min. 42 sec 36 .2.0x10 4.5x10 10.1 0 -9

-Acrylo- 4000 21 MISS 4 l'.9x10 25 min 18 min- 240 1.5x10 0 -10.9 nitrile 6 -8 Benzene 2000 4.7 MISS 70 1.8x10 25 min 19 min 110 2.4x10 0 9.2-I' Ethylene' 800 500 MPRR, 814 7.3x10 40 min 30 min 54 3.3x10

-9' O 10.6 oxide 4 -8 Formalde- 100 1 MPRR 33 7.6x10 12 min 8.6 min 24 2.4x10 0 .10.9-

., ehyde (37%)~

! Hydrogen 100 10 UC-T --

6.6x10 6.8 min 30 see 1100 -- --

12.7

-7 chloride MPRR 166 ~6.2x10 1.5 min 10 sec 4400 4.3x10- 3.5x10 4 -

Hydrogen 50 1 MPRR 1 4.7x10 4.8 min 2.2 min 47 3.5x10 0 13.8

. cyanide 4 -8 Hydrogen 20 --

MPRR 10 6.6x10 l'.4 min -- --

7.5x10 --

15.8

' fluoride 4 -0 Phosphorus 50 --

MPRR 67 8.1x10 12 min -- --

4.5x10 -

9.9 trichloride 4

Sulfur 100 .47 OCC --

7.8x10 23 min 1.3 min 280 -- --

-8 12.3 dioxide MPRR 61 7.6x10' 4 min 50 sec 950 1.0x10- 1.9x10 b 4 -9 Sulfur 10 S MPRR 130 8.6x10 ' 30 min .14 min 6 2.0x10 0 /_11. 7

• monochloride d

Thionyl 2 .47' OCC --

1.9x10 14 min 7 min- 2 --

-7 II*I chloride 3127 77 1.8x10 20 min 8 min 2 2.0x10 0 i

,

• See notes to this table.

P gs 10 of 10 NOTES TO TABLE I

a. Location Code:

MISS - Mississippi River MPRR - Missouri-Pacific Railroad OCC - Occidental Chemical Co.

UC-T - Union Carbide Corp., Ethylene Oxide / Glycol Plant (Taft) 3127 - Louisiana Route 3127

b. Odor detection threshold not listed in Reference 2. However, since each molecule of S C1 w uld yield two molecules of hcl upon contact 2 2 with water (such as in the moist mucous membrane), and since the odor threshold for hcl is 10 ppm, 5 ppm seemed to be a reasonable value,

c. Value not listed in references, but was inferred as follows: Sulfur monochloride (S C1 ) is structurally analogous to thionyl chloride 2 2 (SOC 21 ). Oxygen and sulfur have the same electronic configuration in the outermost shells, but the sulfur valence electrons are less tightly bound. Comparison of analogous 0 and S compounds shows that the sulfur compound always has the lower ionization potential. Therefore, we are reasonably certain that the ionization potential of S C1 is less than 2 2 that of SOCl and, thus, less than 11.7 eV.

2

d. Value not in references. Sax ( cites 17.5 ppm as fatal to cate in 20 minutes. Action is comparable to phosgene, in that both compounds react with water to liberate hydrogen chloride. 2 ppm, which is the same as the IDLH for phosgene, is a conservative value.

e. Not listed in Reference 2. Sax (10) describes the odor as similar to sulfur dioxide, so the odor threshold of the latter was used.

f. Value quoted by HNU Systems, Inc.

-