ML20055A481
| ML20055A481 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 05/31/1982 |
| From: | Rich B EG&G, INC. |
| To: | Barrett L Office of Nuclear Reactor Regulation |
| References | |
| CON-FIN-A-6465 EGG-SSDC-5883, NUDOCS 8207190062 | |
| Download: ML20055A481 (13) | |
Text
.
EGG-SSDC-5883 May 1982 INDEPENDENT EVALUATION OF TMI WORKER EXPOSURE AND PERSONNEL DOSIMETRY 4
A)l0 ftSf6te$ a etke{sstak, sst$ banct f
l Bryce L. Rich l
U.S. Department of Energy Idaho Operations Office
- Idaho National Engineering Laboratory 44 T
- y,-
f
.4.,
- **mt W m m l %, ff}
,s~p.
mmmm 4[,.
"~'
f
- m.._ :
_ m..Li:aw i
- g % w c u -... - n,
.W'
~ ~ %s.K
=3r%
3 g
L -
T This is an informal report intended for use as a preliminary or working document Prepared for the U.S. Nuclear Regulatory Comission y
Under DOE Contract No. DE-AC07-761ID01570 66E6imo FIN No. A6465 8207190062 820531 PDR RES 8207190062 PDR
h EGzG,.,~.. -
FORM EG40-396 e.+ u n INTERIM REPORT Accession No.
Report No.
EGG-SSDC-5883 Contract Program or Project
Title:
Independent Evaluation of TMI Worker Exposure and Personnel Dosimetry Subject of this Document:
Independent Evaluation of TMI Worker Exposure and Personnel Dosimetry Type of Document:
Final Report Author (s):
Bryce L. Rich Date of Document:
May 1982 Responsible NRC Individual and NRC Office or Division:
Lake Barrett Deputy Program Director, TMI Program Office 7
This document was prepared primarily for preliminary or internal use. it has not received full review and approval. Since there may be substantive changes, this document should not be considered finai.
EG&G Idaho, Inc.
Idaho Falls, Idaho 83415 Prepared for the U.S. Nuclear Regulatory Commission l
Washington, D.C.
Under DOE Contract No. DE-AC07 76lD01570 l
NRC FIN No. A6465 INTERIM REPORT r
i
ABSTRACT A self-initiated (GPU) review of personnel dosimetry records col-lected d' ring the TMI accident period revealed probable exposures of u
uncovered skin areas' of at least one operator above the permissible a
guides which occurred on the day of the accident. A formal investi-gation was conducted by the GPU Technical Staff and a report sub-mitted to DOE on January 15, 1981. An independent evaluation of this exposure and the personnel dosimetry system upgrade was contracted by the NRC and is the subject of this report.
2 This evaluation verified that the reported over-exposure was technically justified and probably was not as conservative as originally estimated.
It was further decided that the new dosimetry system offered significant potential for state-of-the-art service, but should be carefully calibrated and evaluated.
4 1
e i
iii
- _=..
i CONTENTS i
AB ST RACT - - - - - -- - -- - - - - - - - -- ------ -- - --- -- - -- - - - - -- - -- - -- - - - - -- - - - i i i.
e
SUMMARY
1 INTRODUCTION -----------------------------------------------------
3 CONCLUSIONS ------------------------------------------------------
4 pt i '
Conclusion A --------------------------------------------------
4 Conclusion B --------------------------------------------------
5~
4 Conclusion C --------------------------------------------------
6 C o n cl u s i o n D - - -- -- -- - - -- - - -- - - - - - - --- - - -- -- - - - - -- - - - - -- -- - - - - -
6 Conclusion E --------------------------------------------------
7 Co n cl u s i o n F - - -- - - - - -- - - - - --- - - - - - -- - -- - -- - - -- - - - -- - - - - -- -- -- -
7 RECOMMENDATIONS --------------------------------------------------
8 REFERENCES -------------------------------------------------------
9 e
(
l ka e
4 r
I 1
.,,,_,-,,,..---n
--~,-,.m.,
---m..
- - --- - - +, - -,
SUMMARY
o During the turmoil incident to the accident at TMI, a considerable amount of personnel dosimetry data was processed under trying conditions.
6 Recognizing the increased probability of errors during those circum-stances, GPU initiated a retrospective review of the dosimetry records during the period of the accident. This review revealed an unrecorded beta skin exposure above permissible limits to an operator who entered the auxiliary building during periods when high concentrations of fission gases were present in the work places.
The GPU technical staff performed a detailed evaluation and concluded that the personnel dosi-meter results should be recorded as the ' legal' exposure, though those values were believed to be grossly conservative.
NRC contracted for an independent technical review / evaluation of the exposure (s).
That review and evaluation reported herein resulted in agreement that the personnel dosimeter results should be recorded as the correct exposure, but concluded that the results were not con-servative.
1 Previous reviews of the GPU personnel dosimetry system,5 concluded o
that the system in use (2-chip Harshaw) was inadequate for the post accident TMI conditions.
Considerable study and technical evaluation was expended by GPU and technical consultants and a Panasonic dosimeter and reader system was ordered by GPU to precise performance specifica-tions.
In addition, a Hewlett-Packard computer system was designed and ordered to interface with both the Panasonic reader system and the IBM computer records system in GPU headquarters. The second objective of the NRC contract for technical support was to evaluate the perform-8 ance of this upgraded personnel dosimetry system. This evaluation is reported in this document and concludes that the proposed system is
.1 technically sound and has significant potential of providing state-of-the-art capability in detecting and evaluating personnel dose in the TMI work places.
However, the system has not been completed nor independently evaluated to date.
Calibration of the GPU Panasonic 1
1
badge _ has been initiated by exposing a set.of badges to calibrated fields at the INEL calibration laboratory.
The results have not been received or evaluated as of the date of this writing, but will-be reported separately as they become available, e
1 I
i i
i i
o 2
INTRODUCTION The TMI accident resulted in the release of mixed-fission products from the primary containment system such that high-level beta / gamma radiation fields existed in some occupational work locations in the plant during and following the emergency.
Personnel exposures in excess of regulatory limits resulted and were evaluated and reported as they were detected.I Dosimetry deficiencies were identified in these investigations which centered around the inability of the personnel dosimeter in use at TMI to measure and record the gamma or beta exposure in a mixed beta / gamma field accurately. These evaluations highlighted general nuclear industry deficiencies in beta dosi-metry and instrumentation areas as well as specific nuclear power needs.
A self-initiated retrospective review of the accident period (March 28,-
1979) dosimetry records was conducted by the GPU TMI Dosimetry Group and detected an apparent high beta exposure on April 24, 1980, which was overlooked in the original routine dosimetry reports. An investigation was conducted, and a GPU report was submitted to the NRC on January 15, 1981.2 An "Inde-pendent Evaluation of TMI Worker Exposure and Personnel Dosimetry" INEL tech-nical assistance contract with the NRC (FIN A6465) was issued on December 1981.
This report summarizes the results of the evaluations which were begun on 4
January ll, 1982 in response to this contract.
Preliminary results of the in-vestigation of the worker exposure were reported verbally to the NRC TMI staff at a close out meeting of the site visit on January 13, 1982.
The GPU TLD Beta Dosimetry system was nearing completion at the time of the site visit.
GPU plans and design consideration were reviewed and procedures i
for an independent calibration discussed.
Panasonic dosimeters of a design to meet GPU performance specifications were received at the INEL for irradiation to a variety of calibrated beta and gama sources in April 1982.
In order for O
this report to be submitted on schedule, it provides only a qualitative evalua-i tion of the new dosimetry system, since the results of the calibration test l
have not been completed. The test results will be tabulated, evaluated, and discussed verbally and by formal letter as soon as they are received.
3
CONCLUSIONS The referenced GPU incident report to the NRC was reviewed.
In addition worker interview reports (including NUREG-0600, " Investigation into the March 28, 1979 Three Mile Island Accident by Office of Inspection and Enforcement"),
dosimetry records, personnel dosimetry system (PDS) justification and perform-ance specifications, PDS calibration reports from the University of Michigan, and interviews with NRC and GPU Technical staff members were used to form the bases of the following conclusions and recommendations.
Conclusion A The overall reconstruction of the related circumstances and evaluation of the technical aspects of the exposure were done in a technically competent manner. However, reconstruction of circumstances 1 to 1-1/2 years after the fact assure large uncertainties in the estimated dose.
Basis Known transfer of primary water containing mixed fission products to 3
open areas of the auxiliary building undoubtedly produced the majority of the radiation levels in the form of fission gases and their daughters, near the Radwaste Panel (305 ft el.) and particularly during the period in ques-tion when the ventilation system was off.
Documented short-lived contamina-tion on workers as they left the auxiliary building areas provide further evi-dence of a high level of fission gas / daughters in these areas. The computer codes and basic assumptions used to estimate the fission gas / daughter concen-trations were appropriate and would result in estimates as reasonable as could be obtained under the circumstances. Time and motion studies were realistically constructed and applied, although attemptsto reconstruct the exposure circumstances at times in the one year range reduce the confidence in details. Agreement in the gama dose estimates through time-and-motion studies and the actual dosimeter reading provide a degree of confidence in that portion of the ex-posure and provide the base upon which to estimate the skin exposure (the overexposure).
4
There are several areas of uncertainty which probably cannot be further refined and which provide the greatest uncertainty in the final results.
The two-chip TLD personnel dosimeter has been shown to produce large inaccura-cies when exposed to mixed beta-gama fields; however, response to a mixed 3,4 fission gas / daughter cloud has not been evaluated. Plato's experiments with 133 Xe and the panasonic environmental dosimeter provides some useful qualitative information but are not specific to this situation.
Conclusion B Though the analyses and assumptions discussed in the text of the incident report would indicate that the exposure recorded by the personnel dosimeter was conservative, there are convincing evidences to indicate that the recorded exposure is probably not conservative as estimated.
Basis A review of the personnel dosimetry records indicates that several other workers in areas at TMI during the time of the accident / emergency received recorded beta exposures--some with beta / gamma ratios in the 3 to 5 range.
In addition, the Unit II HP control point area TLD badge during the period o
March 1-31,1979, read 1.08 rem ga.uma and 7.3 rad beta for a ratio of 6.8.
The HP control point area was known to have been exposed to fission gases during the accident and the ratio is consistent with that recorded by operator "T" who entered areas of fission gases.
Further evidence is seen in Plato's experiments exposing panasonic dosimeters immersed in 133Xe clouds 3 ',4 which recorded ratios in the 7 to 10 range.
Another point is worthy of note regarding the position of operator "T's" badge.
After 14 months, the operator ' thought' his badge was under the plastic wet suit, while the evidence would indicate direct exposure to the gases.
However, Plato's experiments indicate rapid diffusion of the noble gases through plastic bags, such that the badge (and worker's skin) could have been directly exposed even if covered.
5
Conclusion C The analyses and review of circumstances associated with the exposure o
of operator "T" indicate that other plant personnel could have received un-recorded beta exposures which were 3 to 10 times the recorded gama dose
)
assigned to those personnel.
Basis A review of the dosimetry records indicate lack of awareness on the part of the TLD System Technicians to the beta-dose problem. There were a few times when the TLD reader system malfunctioned at the time of the incident and several manual readings were necessary for workers who had been in the plant areas. Only the gama chip was read and/or recorded. There is a high probability that beta dose was significant but not recorded.
In addition, the variation in where the personnel dosimeters were worn (how much clothing was covering the dosimeter) could have resulted in an unrealistically low recorded value if an area of uncovered skin was exposed and the badge covered by multiple layers of protective clothing in such a manner that even gas diffusion was minimized.
Conclusion D The personnel dosimetry system chosen by GPU is not in service but re-i presents a significant effort to upgrade to state-of-the-art.
Basis During the plant visit, the new personnel dosimetry system was discussed l
and the associated facilities were visited.
The decision had been made to procure a personnel dosimetry system which offered simplicity and reliability l
of badge processing by technicians of moderate training and experience.
' Clean room' facilities have been allocated to house the associated and sophisticated e
data processing computers and the delicate dosimeter readers. The dosimeters and processors have been procured from Panasonic on a performance specification l
l contract, which will require careful evaluation by knowledgeable professionals.
6 l
Conclusion E The personnel dosimetry system is fundamentally sound from a technical standpoint and will provide TMI with capability to measure and record per-sonnel dose to beta-gamma radiation equivalent or superior to any in the nuclear industry, if the system functions as designed.
Basis 2
The panasonic dosimeter utilizes ' thin' (10-15 mg/cm ) TLD powder detectors in a multi-filter badge. Technically, this detector represents the best dosimeter connercially available for mixed-field use due to the thinness of the dosimeters and should provide high quality results provided the energy response, badge filter system, calibration and calculational algorithm are understood and performed in a technically competent manner.
Routine processing and data analyses must also be performed under close supervision and control.
A Hewlett-Packard 85 and 9845B system has been purchased and installed to interface with the Panasonic Reader System and process the data to dose form. This local on-site computer will then interface with the IBM business computer in GPU headquarters by ' modem' for detailed, cumulative records.
This system is excellent.
Conclusion F 4
Since the personnel dosimetry system represents an advance in the state-of-the-art, and has not been completely calibrated and/or verified or field tested, it remains questionable until fully implemented and evaluated.
Basis Beta dosimetry capability within the nuclear industry is in a vigorous i
period of development. Though the GPU-TMI dosimeter system represents a significant advance in capabilities to deal with the TMI radiation field problems by combining demonstrated and/or proven elements of current i
a
N l
j technology, there are aspects of the system which have not been demonstrated in addition to the fact that the entire system as assembled at TMI has not been proven. Dr. Phil Plato, at the University of Michigan (a recognized expert) under contract to Panasonic, has been providing calibration and calculational algorithm design services in order to assure that the dosi-meter system meets the GpU performance specifications. Through discussion with Dr. Plato and a review of his progress reports, it appears that the system will be able to perform at a level consistent with the TMI needs and stated specifications.
However, careful and independent evaluation should be performed to verify the performance.
RECOMMENDATIONS 1.
The incident report should be modified slightly to include: a) recog-nition of other data from the personnel dosimetry records which indicate significant beta exposures, b) specific recognition of the comparable results from the area dosimeter at the HP control point, and c) the qualitative evidence of Plato's 133Xe experiments and omit the statements that the recorded exposure is conservative.
It would be well to recog-nize that diffusion of gases into the badge (and clothing) was probable and would make clothing or dosimeter shields protection fac-tors of questionable application. The main focus of the report should j
recognize the probability of high beta exposure to the skin and state that the large uncertainties in the report values cannot be reduced without an inordinate amount of experimental work. The reported values probably represent a reasonable upper limit.
i 2.
The report should recognize that there were other workers who could have j
received unrecorded high beta exposures.
Dosimeters from several j
operators were processed manually and recorded the gamma dose only.
In l
addition, there are other questions concerning unique shielding and p
modifying covers during the exposure of other workers.
It probably is l
not necessary (based on the lack of sufficient data) to try and recon-struct dose estimates on each individual; however, a statement in the incident report and selected personnel folders that unrecorded beta exposures could have occurred would be advisable to complete the record.
If deemed necessary, a 7 to 1 beta-gamma ratio could be applied to the recorded gamma dose during the time of the incident.
8
3.
An independent evaluation of the capabilities of the new personnel dosi-metry system should be performed by a professional group with capability to evaluate mixed field response to a variety of energies for betas and i
photons (x and gamma).
1 REFERENCES 1.
B. L. Rich and S. R. Adams, Technical Assessment of Radiation Overexposures at Three Mile Island from the August 28, 1979 Entry Into the Unit 2 Fuel Handling Building Makeup Valve Room, EGG-SD-5159, June 1980 2.
Investigation and Assessment of the Beta and Gamma Radiation Exposure of Auxiliary Operator
'T' on March 28, 1979, PCI Technical Report-186, December 19, 1980, transmitted to V. Stello by G. K. Hovey letter LL2 0003, January 15, 1981.
3.
P. Plato and G. Hudson, " Calibration of Environmental and Personal Dosi-meters by Submersion in and Distant Exposure to Xe-133", Health Physics Vol. 38, No. 4, pg. 523, April 1980.
4.
D. Oatley, G. Hudson and P. Plato, " Projected Response of Panasonic Dosimeters to Submersion in and Distance Exposure by Xe-133", Health Physics, Vol. 41, No. 3, pg. 513.
5.
B. L. Rich, Joseph L. Alvarez and Steven R. Adams, Interim Status Report of the TMI Personnel Dosimetry Project, GEND 004, June 1981 1
9