ML20235G952

Responds to NRC Re Violations Noted in Insp Rept 50-029/87-10 on 870615-19.Corrective Actions:Protective Clothing Policy Revised & Portal Monitor Set Up at Exit from Control Point to Radiological Control Area

ML20235G952
Person / Time
Site:	Yankee Rowe
Issue date:	09/23/1987
From:	Heider L YANKEE ATOMIC ELECTRIC CO.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
FYR-87-96, NUDOCS 8709300307
Download: ML20235G952 (17)

Text

'

YANKEE ATOMIC ELECTRIC COMPANY r.i.vu on.<s,s><2s-s2e, Star Route, Rowe, Massachusetts 01367 I

c

,Yauxes September 23, 1987 FYR 87-96 United States Nuclear Regulatory Commission l Washington, DC 20555 l Attention: Document Control Desk

References:

(a) License No. DPR-3 (Docket No. 50-29)

(b) USNRC Region I Letter to YAEC, dated August 24, 1987

Subject:

Reply to a Notice of Violation (Inspection No. FiO-29 / 07- 10 )

Dear Sir:

This letter is in response to the Notice of Violation resulting from Inspection No. 50-29/87-10 conducted during the period June 15-19, 1987. The Notice of Violation identified two i tems which apparently were rio t conducted in full compliance with NRC requirements. In accordance with Section 2.201 of the NRC's " Rules and Practices", Part 2, Title 10, Code of Federal Regulations, we hereby submit the following information:

Opparent Violation G A. 10 CFR 20.101(a) Iimits the exposure to the skin of the whole body of any individual in a restricted area to 7.5 rems per calendar quarter.

Contrary to the above, on May 30, 1987, a worker at the facility was contaminated with a radioactive " hot particle" and received a dose to the skin of the whole body (1 cm" of the skin of the scalp) of 10.5 rem, raising his cumulative exposure during the second calendar quarter of 1987 to 11 rem.

This is a Severity Level IV violation.

8709300307 G70923 PDR ADOCK 05000029 OI G $

PDR h/ \ \

y l

l

[ '

(

{

FYR 87-96 l

Resoonse l We acknowledge the Notice of Violation as stated and further

( explained in Inspection Report No. 50-29/87-10. Using j calculational methods such as those contained in the computer l code "VARSKIN" and the isotopic data available for the hot l particle, a dose of approximately 10.5 rem to I square l centimeter of the skin of the whole body can be calculated. -

l  :

However, it is our opinion that dosimetric evaluations, '

including those for skin dose due to hot particles, should be based on valid measurement techniques whenever possible. {

Calculational techniques should be employed when measurements )

can not be performed. Based on the measurement techniques used, we feel that the measured dose of 6.5 rem is a more accurate estimate of the dose actually received than the calculated estimate of 10.5 rem. In addition to the measurements of dose performed for the subject hot particle, (as discussed in Inspection Report No. 50-29/87-10), further measurement techniques have been developed to validate this measured dose and to assess skin dose due to hot particles.

Attachment A is a summary of the techniques developed by ;the yankee Nuclear Services Division (YNSD) Environmental Laboratory and the results obtained. This attachment is included for your review with the intent of providing conuttuctive data to aid in addressing the significant, industry wide, " hot particle" problem.

I Evaluation The root cause of this event has been attributed to personnel error. Radiation Protection management failed to have adequate evaluation techniques in place to conservatively estimate the dose rate to the skin due to the presence of the hot particle. Extensive decontamination efforts to remove the hot particle, resulting in significant physical effects such as reddening and bleeding of the scalp of the individual, were not effective. It was determined that more radical .

surgical techniques should be performed only by a physician. The dose evaluation techniques in place at the time led to a significant underestimate of the dose rate to the scalp of the individual. Based on this underestimate, the hot particle was allowed to remain on i the scalp of the individual for a period of 75 hours8.680556e-4 days <br />0.0208 hours <br />1.240079e-4 weeks <br />2.85375e-5 months <br />.

The subsequently calculcted exposure of 10.5 rem is a direct result of the residence time of the hot particle on the individuals scalp. Although this period of time 1 allowed the hot particle to be easily removed on the morning o f 6/2/87, surgical removal at a much earlier time would have avoided the over exposure.

b:

FYR 87-96 II Corrective Actions Which Have Deen Taken And Results Achieve _d This is the first occurrence of this nature. Based on this occurrence, a number of program and procedural changes were developed and implemented. These changes were effective in controlling and minimizing subsequent exposures of personnel to hot particles and were reviewed in Inspection No. 50-29/97-10. Specifically, the following changes were. implemented:

1) All protective clothing in use at the time was removed and segregated. New protective clothing was issued.

2) Protec tive c lothing used for work in areas with high potential for hot particle contamination, (e.g.

rubber gloves for Shield Tank Cavity work, protective clothing for Steam Generator entries) was discarded after use.

3) Two hour stay times were established for work in the Vapor Container. Personnel were then required to f use the whole body contamination monitors prior to returning to work. Exceptions required by the i nature of the work were handled by establishing alternate controls, (e.g. the issue of new protective clothing).

4) The protective clothing policy was changed to require removal of protective clothing at the Vapor  :

Container exit rather than at the control point.

a

5) One hundred percent of all laundered protective clothing was frisked prior to return to service. A limit of i3OOO cpm was established, protective clothing above this limit was segregated and not returned to service.

6) Paper coveralls were worn over cloth coveralls for work in the Vapor Container.

7) Masslin and sticky tape rolleru were used to survey {

areas to more effectively monitor areas for the presence of ho t par ticles.

O) A portal monitor was set up at the exit from the control point to the radiological control area as an additional check for exceptionally high activity hot particles.

l l

l

l

l. .

t l.

FYR 87-96 I

9) OP-8430, " Personnel Contamination Monitoring and l

Decontamination," was revised to insure that l

regulatory limits would not be exceeded. Actions,

! including transport to an off-site medical facility, I are required prior to approaching a regulatory l limit.

III Corrective Actions Which Will De Taken To Prevent Recurrence l

The corrective actions described above proved to be effective in the prevention of further overexposure due to hot partitles throughout the remainder of the refueling outage.

Specifically, clear procedural guidance is now provided in OP-8430 to both adequately assess dose from hot particles and assure hot particle removal prior to exceeding regulatory limits.

IV The Date That Full Compliance Was Achieved Full compliance was achieved upon institution of adequate dose assessment techniques on June 2, 1987.

The individual's dose of record for the second quarter of 1987 will be changed to reflect a dose to the skin of the whole body of 11.0 rem. In accordance with 10 CFR 20.409(b), the individual will be notified in writing of the change to his dose of record for the second quarter of 1997. This notification will be transmitted prior to September 24, 1987.

Opparent Violation IJ 10 CFR 20.201(b) requires that each licensee make such surveys as may be necessary to comply with all sections of Part 20. As defined in 10 CFR 20.201(a), " survey" means an evaluation of the radiation hazards incident to the pro &:ction, use, release, disposal, or presence of radioactive materials or other sources of radiation under a specific set of conditions.

Contrary to the above, on May 30, 1997, a proper survey (evaluation) of a hot particle located on the skin of the head of an individual was not made to assure compliance with 10 CFR 20.101(a), which limits the radiation exposure to the skin of the whole body in any ca?endar quarter.

Specifically, the initial dose evaluation following identification of the contaminated hot particle was based on

5 FYR 87-96 an inappropriate assumption concerning the size of the contaminant, in that the contaminant was of smaller dimensions than assumed in the initial assessment. As a result, the initial dose rate estimate (8 mrem /hr) was in error. Subsequent calculations indicated a dose rate of greater than 100 mrem /hr.

This is a Severity Level IV violation.

Response

We concur with the Notice of Violation as described above, in that a proper survey was not performed to an'equately assess dose to the skin.

I Evaluation The root cause of this event has been attributed to personnel error. As was described in Apparent Violation A, Radiation Protection management failed to have adequate evaluation techniques in place to conservatively estimate the dose rate to the skin due to the presence of a hot particle. The failure to perform an adequate survey was a direct result of using the guidance in place at the time in OP-8430, " Personnel Contamination Monitoring and Decontamination." Although we were aware that the contaminant was a point source, (hot particle), initial dose calculations effectively treated the hot particle as a plane source by including the active area of the radiation detector, (15 cm"), in the denominator of the equation used. A more thorough technical review of the draft industry document used as the basis for OP-8430 would have identified a misapplication of the equations presented for skin dose calculations.

II Corrective Actions Which Have Deen Taken And Results Achieved Upon removal of the hot particle from the scalp of the individual, a quantitative isotopic analysis was performed. Using this data, the inadequacy of the l initial dose estimate was identified. The following corrective actions were implemented to immediately establish an adequate dose assessment methodology:

l I

, 1) A computer code nimilar to "VARSKIN" was obtained l from another facility and dose conversion factors for all isotopes of interest were tabulated. An i

adequate method for dose assessment of subsequent hot particle contaminations was available as of l l

l 1700, June 2, 1987.

1 1

l l~

I l l l

! I

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

FYR 87-96

2) With the assistance of the YNSD Environmental Laboratory, the computer code "VARSKIN" was obtained and made available for use.

3) OP-8430 was revised to include: >

a) A conservative dose conversion factor for dose assessment of hot particles with G-M Frickers.

b) Action levels requiring isotopic analysis. f '

)

c) Tabulated dose conversion factors for applicable isotopes.

d) The use of the computer code "VARSKIN".

4) Gamma spectroscopy systems were calibrated to measure both hot particles removed from the skin and hot particles still on an individual.

5) The hot particle was transported to the YNSD Environmental Laboratory for confirmatory gamma l isotopic analysis and further evaluation.

6) A hot particle of similar isotopic composition was transported to the YNSD Environmental Laboratory for

"'Sr, 'oSr-Y measurements. Using the data obtained, methods to estimate a'Sr,

• Sr-Y activity based on the activity of various gamma emitting isotopes were developed. This methodology was available for use f on June 15, 1987 and included in a subsequent revision to OP-0430 which was PORC approved July 9, 1907.

III Corrective Actions Which Will De Taken To Prevent Recurrence The corrective actions described above were effective in quickly implementing adequate survey techniques for hot particle contaminations. Although refinements to the program will continue to be made, an adequate hot particle program is currently in place.

IV ]lle Date That Full Compliance Was Achieved Full compliance was achieved on June 2, 1987 when i adequate survey techniques for hot particle contaminations were established.

)  !

l l

I

.____________.J

i FYR 87-96

{

i If you have any questions or desire additional i n fo rma t io n , l please contact us.

Sincerely, H. Heider k Vice President and Mgr. of Operations cc: [1] Region I -

021 Resident Inspector i

l

)

)

)

)

1-

)

l l l

l 1

f

)

}

}

) I

, ATTACHMENT A REFERENCES

1. Memorandum to R.A. Mellor from N. Stanford, " Final Report and Evaluation of Yankee Plant Hot Particle", dated June 10, 1987 EL 355/87.

2. " Design and Operation of an Extrapolation Chamber with Removable Electrodes" by R. Loevinger and W. G. Trott, International Journal of Applied Radiation and Isotopes, 1966, Vol. 17, pp. 103-111.

3. "Use of a Victcrian 500 Electrometer to Determine Ionization Chamber Collection Ef ficiencies", by P. R. Almond, Medical physics Journal, Vol. B. No. 6, Nov./Dec., 1981.

4. Memorandum from R. A. Mellor to C. M. Babineau, " Independent Estimation of Sr-89 and Sr-90 Concentrations in the 5/30/87 Scalp particle", dated June 15, 1987, CH 057/87.

5. " Mathematical Theory of Radiation Dosimetry", by J.J. Fitzgerald, C.L. Brownell and F.J. Mahoney, Gordon and Breach Science Publishers, Inc. New York, NY, copyright 196 7.

6. Radiological Decay Data Tables by David C. Kocher, Technical Information Center, U.S. Department of Energy, 1981.

This report presents further research and testing concerning the use of the Vinten extremity dosimeter for direct measurement of skin dose resulting from irradiation by a hot particle. The two concerns which were raised in NRC inspection report #50-29/87-10 concerning this technique have been addressed.

The areas of concern are: 1) the accuracy of the dosimeter effective area value and 2) the ef fect of the underresponse of the dosimeter to low energy beta radiation. In addition, data is presented from a definitive comparison between this technique, the VARSKIN computer code and extrapolation chamber measurements of a hot particle similar in composition to the originally measured particle. The results of this comparison show the Vinten dosimeter measurement to provide excellent agreement with the extrapolation chamber while the VARSKIN model overestimates the dose significantly.

. Effective Area The standard Vinten extremity dosimeter has a 5 mg/cm2 layer of phosphor over an area of 0.6 cm x 1.0 cm or 0.6 cm2 Due to the limiting dimensions of the plate on which the dosimeter is mounted for processing and the co111mation of the photomultiplier tube (Figure 1), the actual area of phosphor over which thermoluminescence is gathered is calculated to be 0.29 cm2 As the estimated dose to 1 cm2 is very sensitive to this parameter, an experiment was devised to verify this calculated value.

To verify the maximum effective area of the dosimeter, Vinten ins t rument s . LTD was requested to manufacture a batch of dosimeters with phosphor areas of varying lengths. In addition to the normal 1.0 cm length, groups were manuf actured at .8, .7, .6 and .5 cm lengths. All five groups of

dosimeters were manufcetured f rom the s:me batch of phosphor so that c comparison of sensitivities would indicate the depsndence of asnsitivity en length. The smaller dosimeters were expected to produce a reduced signal compared to the full size batch only when their length became the limiting dimension, i.e. when the length of the phosphor approached the diameter of the collimator. Figure 2 contains illustrations of the relative dimensions with the calculated ureas of interest indicated. The ratio of the response of any ,

given dosimeter length to that of the standard 1 cm length was expected to be i equal to the ratio of the esiculated effective areas for equal delivered {

dose.

Ten dosimeters from each of the five groups of test TLDs were exposed to a uniform field of 137 sC gamma radiation. All of the dosimeters were processed following normal protocol, centering the phosphor area over the l opening of the mounting plate. The responses of the dosimeters, corrected for background using unieradiated dosimeters of the same group, were averaged for l each group. The average group responses were then divided by the average l response for the standard I cm batch. The results, which are presented in l i Table 1, indicate excellent agreement between the observed ratio of responses i l and the ratio of the calculated effective areas, supporting the technique for calculating the effective phosphor area.

l l

TABLE 1 OBSERVED VERSUS PREDICTED RATIOS OF THE RESP 0NSE OF DIFFERENT LENGTH DOSIMETERS l

Calculat ed Ratio of Effective Effective Area Observed Ratio of Area to 1.0 cm Dosimeter response to 1.0 cm 2

Length (cm) (cm ) Effective Area Dosimeter Response 1.0 1 .01 .289 - - . --

1 0.8 1 .01 .289 1.00 1.02 1 .09 0.7 1 .01 .289 1.00 1.00 1 .09 0.6 1 .01 .280 0.97 0.97 1 .08 0.5 1 .01 .247 0.85 0.85 1 .07 Re.syonse to Low Energy Beta Radiation An estimate of the underresponse of the dosimeter due to low energy beta i radiation of 20 percent was quoted in the NRC inspection report. This number, l which was felt to be somewhat conservative, was checked by the DSC using e l more rigorous method.

In order to evaluate the expected underresponse of the Vinten dosimeters, l Loevinger's expression (Ref erence 5) was employed to calculate the apparent absorption coefficient based on the end point energies (Reference 6) of the 2

2

_ individual beto partic1cs emittsd by tha isotep) involv26. Th2 me n value  ;

theor n w;s uccd to cvar ss th2 Gxpon:ntial ettsnuaticn of cich bato antrgy over the 5 as em-2 thickness of the Vinten dosimeter. The fractional yield of the beta energy and the fraction of the total dose delivered to 1 cm2 aren under 7 mg em-2, based on the VARSKIN code, was used to weight the i effect of the individual underresponses. The summation of these individual ')

underresponses yicided an estimated overall underresponse of approximately 10 i percent.

Comparison with Extrapolation Chamber i

To further test the accuracy of this method of measurins skin dose from a )

hot particle, a comparison was made with extrapolation chamber measurements, j The Yankee Plant staff obtained a second hot particle (actually three 1 discernable particles, each measuring approximately 200 ym and contained in  !

an area of less than 2mm) of suf ficiently high activity to allow precise extrapolation chamber measurements using a 3 cm2 collecting electrode.

]

The particic/(' logged as "STC UPNH Sump , Speck") was held on a piece of umsking tape which' was in turn covered by a layer of 0.8 mg/cm2 clear mylar j film. This configuration was used for both dosimeter irradiations and j extrapolation chamber measurements. I I

The radionuclides composition of this second hot particle was evaluated using gamma ray spectroscopy and,found to be similar to the origins 1 particle, validating a comparison of dosimeter performance. The relative abundance of the major constituents of each particle, as determined with gamma ray l analysis, are tabulated below.

l I

l l

l l

l l

3 l

l i L___________  !

1 TABLE 2 COMPOSITION OF ORIGINAL AND WEW HOT PARTICLE

% of Total Activity Radionuclides Original Wew 95Z r 16 16 95Nb 21 23 144Ce 11 16 144Pr il 16 141Ce 11 7 103Ru 8 6 140La 5 <1 140Ba 4 <1 l

106Ru 2 1 l l

106Eh 2 1 l

1311 1 ..

l 137Cs 1 1 I 89sr 7 10 90S r 1 2 .

l 90Y 1 2 pos.'eter Measurements l

Vinten dosimeters were used for dose measurements using the same techniques that were outlined for the original particle in June of 1987 ,

(Reference 1). The particle was placed on top of the five dosimeter configuration (Figure 3) snd left in contact with the center dosimeter for a measured time period. The four noncentral dosimeters were then used to calculate the dose delivered to the 1 cm2 area, outside of the 0.29 cm2 effective area of the central dosimeter. The five replicate irradiations using this configuration resulted in an average dose rate estimate of 46.31 6.0 rad /hr.

4 I

t

, Extrapolation Chamber Measurements On July 31, 1987 the YAEL extrapolation chamber was used to determine the absorbed dose rate at 7.0 mg em-2 Six layers of 0.86 mg em-2 aluminized mylar were placed between the previously described particle configuration and the window of the extrapolation chamber. The weight of the plexiglass sheet was allowed to bear down upon the mylar sheets to avoid any air gaps between the sheets. This arrangement yielded a 6.9 mg cm-2 density thickness between the particle and the active volume of the extrapolation chamber.

The extrapolation chamber was equipped with a special electrode having an j l

active area of I cm2, manufactured specifically for these measurements. The  !

l electrode is a one inch thick 4 cm radius lucite cylinder coated with a fine layer of colloidal graphite. A thin (11/1000 inch) groove was etched through the graphite into the lucite to isolate a 0.564 cm (1 cm2) circle from the outer annulus. The machining of the electrode's active area was verified using the capacitance techniques outlined in Reference 2 to be well within ,

1 percent of I cm2 The particle was centered over the active volume by  !

moving the particle across the window until the ionization current reading on the electrometer was maximized.

l This positioning technique was repeated i several times and the maximum current verified. plate spacings of 0.1 mm l increments were used to minimize any geometrical effects associated with a changing active volume. The ionization collection efficiency for each plate l separation was determined to be at least 99.7 percent using the methodologies l of Reference 3. An extrapolation, which was performed between 0.8 and 0.4 nn, yielded an absorbed dose rate of 46.5 1 0.86 rad /hr based on a least squares fit to the measurement data. This measurement was performed in accordance l with YAEL Procedure 950 and the data is maintained in data file E8721113.

I VARSK1W Analysis on "STC UNR Sump Speck" The samma spectroscopy analysis performed at the Yankee plant was used i for input in the VARSKIN computer program. The results of the gamma analysis and the corresponding VARSKIN dose estimales are contained in Table 3.

5 l

l ,

. TABLE 3 GAMMA AED VARSKlW ANALYSIS OF KTC UNR SUMP SPECK JULY 31, 1987 Absorbed Dose Rate (rem /hr) over Isotope pCi yCi*See (3) I cm2 9 7 sag em- 2

'Nb-95 6.04 2.17 E4 5.2 2r-95 4 04 1.45 E4 19.6 Ru-103 1.52 5.47 E4 3.2 Rh-106 .339 1.22 E3 3.2 l

Ru-106 (1) .339 1.22 E3 0.00 Cs-137 .165 5.92 E2 1.1 La-140 (1) .115 4.14 E2 1.0 1

Ba-140 .100 3.59 E2 0.7 Ce-141 1,82 6.54 E3 10.1 Ce-144 4.07 1.46 E4 13.2 Pr-144 (1) 4.07 1.46 EA 38.2 Sr-89 (2) 2.47 8.90 E3 21.5 Br-90 (2) 0.43 1.54 E3 2.9 1

Y-90 (2) 0.43 1,54 E3 3.9 123.8 f 7.5(4) rem /hr (1) Activity for these isotopes based on activity of associated parent isotope.

(2) Calculated based on core inventories as outlined in Reference 4.

(3) The tirne interval used was 3600 seconds in order for the VARSKIW code to calculate rem /hr.

(4) Total uncertainly in the VARSK1W estimate is based on a propagation of the uncertainty in the counting statistics.

6

The racults cf tha three techniqu2s cra listed belcw. For p:rticiss of

. thir cisa cnd r dionuclida mixtura, th2 techniqua using th2 Vinten desimeter

'has been shown to accurately estimate the skin dose, while the VARSKlW model

_substantially overestimates the dose.

TABLE 4 COMPARISON OF THREE TJCHN1 QUES TO EVALUATE SKIN DOSE FROM A HOT PARTICLE (RESULTS IN RAD /HR) i Extrapolation Vinten Chamber Dosimeter VARSKIN ]j l

46.5 1 0.9 46.3 1 6.0 123.8 1 7.5 i

Summary The work presented in this report verifies the accuracy of direct skin dose measurements of submillimeter radioactive particles using the Vinten j extremity thermoluminescent dosimeter. The actual effective area over which  ;

dose is averaged by the dosimeter has been experimentally verified. Most I importantly, the accuracy of this technique has been verified for a particle of mixed fission products using a direct comparison with extrapolation chamber measurements.

i l

l l

l i

1 7

L_________________._ --

FIGURE-1 Relative Dimensions of I cm x 0.6 cm Vinten Extremity Dosimeter, Dosimeter Mounting Plate and Photomultiplier Tube Collimator Photomultiplier Tube Collimator Radius = 0.325 cm

_ /

/ / Dosimeter Plate 7 --- Opening 0.8 cm x 0.5 cm

.__ nosimeter Phosphor r 1 cm x 0.6 cm 2

Area = .289 cm Sc' ale = 10:1 i

l I

w__ _-- -_ _ _ _ _

FIGURE 2 Relativa Dimensions for Tcat Dosin2tsr Crcups Showing Phospbsr l Area Effective Reader Area and Total Effective Dosimeter Area i

/ / > / / ,

/

/ /

/ /

/

\ s i s / /

/ l 0.8 cm x 0.6 cm 2 Effective area = .289 cm

,/ > > s - Scale = 6.5:1 '

\ Area of Collimator

\ and Dosimeter Pla'td

// Area of Dosimeter j ,

[ Phosphor

//s s s _s c 0.7 cm x 0.6 cm Effective area = .289 cm To al Effective

/. / s /

s- s s //

0.6 cm x 0.6 cm 2 Effective Area = .281 cm

// / /

WD . , _

MGURE 3 TOP VIDI - 00$1ETER CONFIGURATION West test Dosimete -

Dosimeter {

Petri Msh i

- ~

- g

' s. 1

/

t

/

i l

-"mi

--- A 7~.--

' ' North l Dosimeter l

-("  ;,p ,,; ::;'I:q ,

I 'a r .!I g < ,

4 l

l n .

• j j '

$ngg \,

J t ^ ,; "-d' n.c ;" '.','

Dosimeter

(  ;;A', ,' , e l: -

$OUrCt

\

\

Central Dosimete

_ _ _ _ _ _ _ _ _ - - - - - - - - -