ML20215M129
| ML20215M129 | |
| Person / Time | |
|---|---|
| Site: | Three Mile Island  |
| Issue date: | 03/20/1987 |
| From: | NATIONAL COUNCIL ON RADIATION PROTECTION & MEASUREMEN |
| To: | |
| Shared Package | |
| ML20215L984 | List: |
| References | |
| PROC-870320, NUDOCS 8705130096 | |
| Download: ML20215M129 (44) | |
Text
C orree < /c
- , . . . b- ',
r.
~
NCRP ComENTARY No. 4 4-GUIDELINES FOR THE RELEASE OF WASTE -
WATER FROM NUCLEAR FACILITIES WITH SPECIAL REFERENCE-TO THE PUBLIC HEALTH SIGNIFICANCE OF THE PROPOSED RELEASE OF TREATED WASTE WATERS AT THREE MILE ISLAND l
l National Council on Radiation Protection and Measurements 7910 Woodmont Avenue, Bethesda, Maryland 20814 8705130096 870507 PDR ADOCK 05000320 P PDR
,y d
,v Pteface In May of 1980, the NCRP issued a report entitled, Krypton-85 in the Atmosphere - With Specific Reference to the Public Health Significance of the Proposed Controlled Release at the Three Mile Island, in response to public concerns over the proposed venting of Krypton from TMI-2. That document was prepared in response to a request from Governor Thornburgh of the Commonwealth of Pennsylvania and as part of NCRP's responsibilities as stated in its congressional charter.- The charter includes, among other things, the responsibility to " collect, analyze, develop and disseminate in the public interest, information and recommendations about (a) protection against-radiation and ...". The study was also recognized as part of the general problem of controlling releases to the atmosphere of which the circumstances at TMI-2 were a special case. ,
The problem of releases of waste water from TMI-2 could raise similar scientific and public issues. Again, TMI-2 is a specific case of a general problem.
In 1980, the Nuclear Regulatory Commission requested the NCRP to examine this issue. Recognizing that it could serve the public interest, the council established a . Task Group to address this problem. The Task Group prepared a draft report which was reviewed by the Council in 1985. The members of the Task Group were:
Frank L. Parker, Chairman Vanderbilt University -
Nashville, Tennessee Members A. Bertrand Brill Bernd Kahn Brookhaven National Laboratory Georgia Institute of Technology Upton, New York Atlanta, Georgia Donald G. Jacobs Edward Watson Roy F. Weston, Inc. Battelle Pacific Northwest Laboratories Oak Ridge, Tennessee Richland, Washington l
However, further proposals to release the waste water at TMI-2 have been developed recently by GPU Nuclear. Therefore NCRP established a new Task Group to review the potential environmental impacts of these proposals.
The U. S. Nuclear Regulatory Commission have reserved to itself the deci-sion on disposal of the waste water. As part of this process, NRC issued a
" Final Programmatic Environmental Impact State ment Related to Decontamination and Disposal of Radioactive Wastes Resulting f rom, March 28, 1979 Accident, l
Three Mile Island Nuclear Station, Unit 2" in 1981. A supplement to the EIS l- is now out for public comment. This Commentary is intended to provide the NCRP's response to the request for public comment.
i l-t
sg
- 9 The System International (SI) units are used La this report but, with the exception of Section 5 are followed by the conventional units in parenthesis in accordance with the procedure set forth in NCRP Report No. 82.
Serving on the Task Group during the preparation of this' Commentary were:
Charles B. Meinhold, Chairman Brookhaven National Laboratory
. Upton, New York Me mbers Leonard Emma Villiam L. Templeton Brookhaven National Laboratory Battelle Pacific Northwest Laboratories Upton, New York Richland,-Washington Donald G. Jacobs John E. Till Roy F. Weston, Inc. Radiological Assessments Corporation Oak Ridge, Tennessee Neeses , South Carolina Consultants Victor P. Bond John W. Healy-Brookhaven National Laboratory Los Alamos, New Mexico Upton, New York Melvin W. Carter Frank L. Parker Georgia Institute of Technology Vanderbilt University Atlanta, Georgia Nashville, Tennessee NCRP Secretariat - E. Ivan White -
The Council wishes to express its appreciation to the Task Group members and consultants for the time and effort devoted to the preparation of this Commentary.
Warren K. Sinclair President, NCRP Bethesda, Maryland February 25, 1987 11
Table of Contents Page Preface 1
- 1. Introduction .................................................
- 2. Status of Accident Generated Waste Waters 2
at Three Mile Island..........................................
- 3. Options for Treatment of Tritiated Waste Waters 4
f ro m Three Mile Is lan d . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4 3.1_ Evaporation / Atmospheric Release..........................
3.2 Evaporation / Surface Water Release........................ 4
- 4. Tritium - Physical and Chemical Properties . . . . . . . . . . . . . . . . . . . . 5
- 5. Tritium - Environmental Transport and Pathways of Exposure.... 6 5.1 Assessment of Releases to the Atmosphere................. 6 5.2 Assessment of Releases to Surf ace Water. . . . . . . . . . . . . . . . . . 8 8
- 6. Dosimetry.....................................................
- 7. Dose Equivalents Resulting f rom Release of Tritiated Waste Water to the Atmosphere and Surface Water 9
at Three Mile Island..........................................
7.1 Assessment of Tritium Releases to the Atmosphere. .. . ... .. 9 7.2 Assessment of Tritium Releases to Surf ace Water. . . . . . . . . . 10 11
- 8. Summary of Health Ef f ects and Conclusions . . . . . . . . . . . . . . . . . . . . .
APPENDIX A - Specific Sources of Was te Water at TMI.. . . . . . . . . . . . . . . 14 APPENDIX B - Decontamination Factors Required to Ensure that the Dose from Radionuclides Other Than Tri tium a re Relatively Insignificant . . . . . . . . . . . . . . . . . . 15 References......................................................... 17 t
iii 1
l
- 1. Introduction The risk associated with the release of waste water f rom nuclear facili-ties, whether generated by accident or produced during." normal" operations, is of interest to the public and the nuclear industry. The presence of waste water at Three Mile Island (TMI) and proposals for its release represent a special case of this problem, the analysis of which is widely applicable.
In many situations, including the case at TMI, the radionuclides other than tritium can be removed f rom the waste water by various processes.
Tritiated water cannot be separated and concentrated from ordinary water by conventional waste treatment techniques (Blomeke, 1964). Isotopic separations are available, but these are impractical for high volume, low concentration operations (IAEA, TRS, 234-1984). Thus, while radionuclides other than tritium can be removed from contaminated waste water, the tritium must be handled by other means. Therefore, the ' focus here is primarily on the disposal of tritiated waste water.
Because of the previous work on this subject by the NCRP, it is possible to apply existing knowledge and procedures to the situation at TMI.
The National Council on Radiation Protection and Measurements (NCRP) in 1979, published NCRP Report No. 62, Tritium in the Environment (NCRP, 1979).
That report dealt with the available inf ormation on tritium in terms of its physical properties, production sources, physical transport, biological behavior, projected future production, waste management and long-term implications in the environment.
The Council, in 1984, published NCRP Report o. 76, Radiological Assessment: Predicting the Transport, Bioaccumulation, and Uptake by Man of Radionuclides Released to the Environment (NCRP, 1984). That report dealt with a review of the current status of the application of radionuclide transport models from the point of discharge to the environment to the point of intake by man. Models are reviewed that describe the transport of radionuclides through the atmosphere, surface and groundwater, deposition on terrestrial surf aces and in sediments and accuculation in food products.
Usage factors are considered that determine the intake of radionuclides by ,
humans due to dietary habits, physiological parameters and living customs.
The information provided in the two cited NCRP reports is used in the analysis here of the public health significance of the disposal of waste water contaminated with tritium. Optional methods for release of tritiated waste water are reviewed, information about tritium provided, and environmental transport and pathways are examined and dosimetry discussed. Finally, doses and health ef fects resulting from waste water releases are given.
- 2. Status of Accident Generated Waste Waters at Three Mile Island The TMI-2 accident resulted in the production of large volumes of con-taminated water. Since the time of the incident, the total inventory of this water has increased to approximately 1.9 million gallons due to continued in-leakage f rom support systems and condensation f rom the Reactor Building air coolers. The specific sources of the waste water and their quantities of radionuclides are given in Appendix A. When the clean up is completed in 1
4 October, 1988, it is estimated that approximately 2.1 million gallons of water
- i will presentrequire are fispo Co, gtionl25 Sr. "b, S $ 's,"an*d C 15ftium, Cs. GPU theNuclear principle has radionuclides noted (GPU, 1986) that prior to ultimate disposition, a considerable amount of this water will require processing to reduce the levels of radioactive contaminants.
This reduction of the radi guelide levels will minimize the total release of activity, particularly of Sr, to the environment and thereby minimize the environmental consequences associated with the various disposal options.
The volume of water requiring processing prior to ultimate disposal depends upon the selection of the final disposal method. Three methods have been considered (see Section 3) - - river discharge, direct solidification and evaporation. For the " river discharge" and " direct solidification" options, essentially all of the water would require initial processing, or reprocessing, through existing ion exchange systems prior to disposal. This is referred to as 100% processing. In light of the decontamination factor achieved by evaporation, it is estimated that only about 40% of the total volume will require reprocessing to reduce the activity levels before final disposal by the release to the atmosphere. This is referred to as 40%
processing. The average characteristics of the water expected after these two degrees of processing are presented in Table 2.1.
The boron concentration in the water is given in Table 2.1. Boron may influence the disposition options either through discharge limits to the envi-ronment (via the federally mandated release limit of 25 ppm boron), increased quantities of concentrates requiring solidification from the evaporation option, or the necessity to add stabilizing agents to ensure proper solidifi-cation. In addition, the water will contain approximately 11 tons of sodium hydroxide.
4
\
l l
2
n l
H m C O u a a i N N d 0 1 0 1 o 0 1 s 0 1 s S 7 A 7 A 4 4 0 0 B B n 0 3 0 0 3 o 0 0 5H 0 51 1 r
o 0, 1 s
0, 1 s
B 3 A 3 A
)
4
- )
s 0 c 1 5 i x 0 t 5 1 )
. s r 1 x 8 i S . )
r 0 1 9 1 0 e ( (
9 0 t (
c 3 4 (
a 0 9 0 r 0 1 9 a
1 x 1 x 0 1
h 3 x 7 c . 3 x 3 3 3 4
r e
t a )
w 5
- )
n 0 6 o 1 i x 0 t 7 1
i s .
s C 3 x )
3 o
p 7 ( 4
( 0 3 )
s 1 3 9 5 (
i 0 2 d 1 0
- x ( 1 x
9 0
e 7 5 1 r 3 9 P 0 1 1 x
1 1 1
2 ) )
1 I
)
e - ) -
0 0 l 0 0 1 2 b 1 2 a x 0 , x 0 T 3 1 3 1 m . (
u ( 1 i 1
( 2 ( 2 t 1 i 1 r 9 0 9 0 T 0 1 0 1 1 x 1 x x 7 x 7 8 7 8 7 . .
3 4 3 4
) y y ) y y n nl t n t g
nl om t
i o t
i g n
om i/
i v i o i t
n i/ v i t n
i ti i t n i ti aC i t a a s ac t a a % s rp t
c r u
% s rp c r u 0 s Q
0 e A) t Q 0 e t( A) t 4 c t(
n i n 1 c n i n g e l o g lC e l o lc r em a( c as r em a( c nm as t n P n/ t nm t n P n/ t oq oq oo oq oq op oo CB TB oP C P TT
' CB Tn C p TT .
O
_ O
a- .
- 3. Options For Treatment Of Tritiated Waste Water trom Three Mile Island In the~ July, 1986, Report, " Disposal Of TM1-2 Water," (GPU, 1986) GPU Nuclear proposed three disposal options for-the processed waste water. These
- options au as follows:
i _1. ' Evaporation - Processing and evaporation of the water would be by an installed evaporation facility followed by controlled atmospheric release.
Shipment to, and disposal of, solidified residues at a licensed, commercial low-level waste dispos,al site would follow.
2; Solidification - Processing and solidification of the water in cement would be followed by burial in an on site industrial landfill. .
- 3. River Discharge - Processing and controlled, monitored discharge to the Susquehanna River would result in significant dilution of the processed water.
A The G?U Report noted that - direct release to the river is the best choice on the basis of overall technical merit .but that political and institutional concerns resulted in their suggesting the evaporation / atmospheric release option as the method of choice. We therefore reviewed the potential impacts of evaporation / atmospheric release option given above slightly modified to ensure that the dose from other radionuclides are appreciably less than those from tritium (see Appendix B). The decision to imploy such a modification in our analysis was based on the unique difficulty of removing tritium from waste water, the comparatively low radiogenic hazard associated with triciated water, and the ease and precision of tritium environmental transport analysis..
One could reasonably argue, however, that in view of = the extremely low doses involved, use of this approach is unnecessarily co.nservative. Since this analysis also suggested that the addition of an evaporation step might improve the viability of the River base option we analyzed this scenario as well.
We have not, however, assessed the solidification option since the GPU Report suggested that it had the lowest overall merit of the three options.
3.1 Evaporation / Atmospheric Release This option employs a standard industrially available evaporation unit which, depending on the available decontamination factor, would require pre-The processing of between 40% and 100% of the waste waters (see Appendix B).
vapor produced by the evaporator, which contains essentially all of the tritium, would be released through a 50 meter stack. If a typical processing rate of 3 gallons per minute is assumed and the operating basis is seven days a week, with two, ten-hour shifts per day of actual processing with an overall availability of 80%, approximately two years would be required to process the total volume of waste water (2.1 million gallons). This option would result
' in a release rate of .93 MBq/sec (25 pCi/sec) or approximately 19.5 TBq/yr i (526 Ci/yr) of tritium to the atmosphere.
i
!! 3.2 Evaporation / Surface Water Release l
This option employs the evaporation step followed by controlled release l
of the condensed vapors to the Susquehanna River. Again, the extent of j required preprocessing of input waste waters would depend on the ef ficiency of
the particular evaporator used. In any case, as shown in Appendix B, the 1
i.
4 I
~s . . -m ,_
U Sr source term is insignificant evaporation step will ensure that the compared to that for tritium.
Af ter evaporation, ' rather than releasing to the 50 meter stack, all the vapor would be condensed and pumped to one of the two 11,000 g'allon condensate test tanks. From these tanks the water would be discharged to the Susquehanna River via blow down from the mechanical draf t cooling tower. The use of the cooling tower blow down in this manner provides a dilutant flow of about 22,000 gallons per minute. Use of this option would also result in a release of 19.5 Bq/yr (526 C1/yr).
For either the " Evaporation / Atmospheric Release" or the " Evaporation /
Surf ace Release" option, disposal of the waste water will require approximately two years. ,
4
- 4. Tritium - Physical and Chemical Properties Tritium is the heaviest and only radioactive isotope of hydrogen. It was discovered in '1939 by Alvarez and Cornog (1939) who determined that it emitted radiation with a very short range and had a long half-life; subsequent work
-established that it decays with a half-life of 12.3 years. It emits a beta partiqle with a' maximum energy of 18 kev and an average energy of 5.7 kev to form 1e.
Tritium is produced naturally by the interaction of cosmic rays with elements in the upper atmosphere. It is also produced by thermal or fast neutron reactions with various light elements utilized in reactors, such as baron, used for reactivity control, and lithium, used for corrosion control.
Tritium formed in this way is circulated in the coolant and from there is released into the environment. Most of the fission product tritium is normally retained within the fuel element cladding, however, an appreciable fraction of the fission product tritium can be released from the fuel elements when the core is damaged such as in the case of TMI-2.
For a detailed review of the physical and chemical properties of tritium, see NCRP Report No. 62 (NCRP, 1979). Tritium closely follows the reactions of ,
ordinary hydrogen, although, the relatively large mass differences among the hydrogen isotopes make isotopic ef fects discernible. Because of the prevalence of water and its importance in the life processes, the isotopic exchange of hydrogen in water with tritium is of special importance.
In the environment, tritiated water behaves generally, though not exclusively like ordinary water. Most of the predicted behavior of tritium is based on existing information regarding the cycling of water, supplemented by observations of the behavior of tritium produced during atmospheric testing of nuclear weapons. Tritium can also become an integral part of any chemical compound containing hydrogen atoms, including the organic compounds that make up living tissue.
5
5.' Tritium - Environmental Transport and Pathways of Expoaure 5.1 Assessment of Releases to the Atmosphere When tritium is released to the atmosphere, it disperses rapidly and mixes with stable hydrogen in the atmosphere, hydrosphere and' biosphere.
Tritium released in forms other than tritiated water (HTO), tends to convert to HTO. Concentrations of tritium in atmospheric water at a given distance and direction from a source are typically estimated using atmospheric dispersion models as described in NCRP Report No. 76 (NCRP, 1984) and the absolute humidity for ,the point of interest. For purposes of assessing dose from tritium released to the atmosphere, it is generally assumed that the chemical form is HTO and that there is uniform mixing between atmospheric water vapor and the bound and unbound hydrogen in biological systems. This approach, referred to as the specific activity method, is based on data reported by Evans (1969) that suggests that body hydrogen is unif ormly labeled with tritium under chronic exposure conditions. Although other approaches, such as the multicompartment model, may lead to more precise estimates of dose, the specific activity method is simpler to apply and generally results in dose estimates that are significantly higher than those which would actually exist. Therefore, if acceptable criteria for exposure are met as determined using this more conservative but simpler approach, use of a more sophisticated model may not be justified. This is especially true since the application of a multicompartment model requires the use of site specific compartment dilution volumes and intercompartment transfer rates. The use of a simple ef fective model in preference to a more complex model is in accordance with previous NCRP Recommendations (NCRP, 1984).
A model based on the specific activity method.and the contribution to total water intake of reference man was proposed by the NCRP (1979, 1984).
The model predicts that the relative importance of different pathways of exposure is determined by the intake of body water derived from each pathway. This model was updated by Killough (1982) to balance total hydrogen intake, accounting for hydrogen in both water and organic products consumed by individuals. The model of hydrogen balance assumed for reference man allows one to convert tritium specific activities in food and fluids and in the individual's ambient air to daily intake rates for steady-state conditions as described below lair = (15 + 9) Aair (TBq/d) (5.1)
-I ingestion = 183 _Awater + Smilk + 120Afood (TBq/d) (5.2) where 131 , and I4 respec-tively. khe specikkestion, areA the intakes via air and ingestion,ir' $ water, etc. a e activities terabequeral per gram of hydrogen (TBq/gH) and the coefficients are grams of hydrogen per day (gH/d) derived from the data on hydrogen intake and losses given for ICRP's Reference Man (ICRP Publication 23) as modified by Killough and others (Killough, 1982). Note that water in food is included in the coefficient 120 associated with Af in Eq. 5.2. It is also important to note that the coefficients 15andOodin Eq. 5.1 (inhalation and absorption through the skin) are sensitive to the ambient absolute humidity, and the~
values given here correspond to an ambient absolute humidity of 6 gH O 2 m The specific activities $ water' $ milk, and flood depend on the environmental sources of the water, milk, and food ingested by the reference individual.
6
Based on information given in ICRP Publication 23 (ICRP, 1975) as updated by Killough (1982), it is assumed that the total hydrogen intake is 360 gH/d. The hydrogen balance model for assessing dose from tritium in the environment can be described as follows:
(15 + 9) Aair inhalation + (1833 water
+ 33Amilk + 120$ food) DCFingestion 365 d/y (Sv/y) (5.3) where,
- b , andetc. = specific activities of tritium in air, water etc.
$ai kTBq/gb)r, DCF DCF inge
= d se conversion factors for tritium by knNakation,naationandinges!kionon which have values of 2.2 Sv/TBq and 23 Sv/TBQ, respective 1y.
The model can be further subdivided to account for different concen-trations of tritium present within a given pathway. For example, it is f requently assumed that individuals drink water from several sources, each containing different concentrations of tritium. If this level of detail is desired in the calculation, then the ingestion of tritium can be determined by calculating the relative intake from each source.
The specific activity methodology assumes that for a given location, the concentration of tritium is the same in atmospheric water and biota. This assumption likely leads to a higher estimate of dose than that which actually occurs because it is unusual for a steady state condition to exist in the environment near a source, considering the intermittent nature of most source terms and the variability of meteorological and climatic conditions. As suming the specific activity of tritium in each component (i.e., air, water, milk, and food) were the same, the model tells us that intake of tritium via ingestion of water and food are the most important pathways of exposure. The contribution to dose from inhalation and skin absorption combined when all pathways of exposure are available, is approximately 7%.
It is likely that this technique significantly overestimates the dose f rom tritium to individuals who do not produce and consume their own food products but import them from regions, outside their area, where tritium .
concentrations in food are substantially lower. Likewise, persons may receive only a f raction of their drinking water supply from a source containing tritium. Nevertheless, this simple model can be easily applied to estimate dose at a given distance f rom the source once the concentration of tritium in atmospheric water at that location is derived using a meteorological model.
One key to applying this model is the determination of the concentration of tritium in drinking water when the only source of release is to the atmosphere. When the release of HTO is to the atmosphere, it is generally assumed that the concentration of tritium in drinking water is 1% of that in air for a given location (NCRP 1984). This assumption is simply an attempt to account for tritium that migrues f rom the atmosphere to drinking water supplies. If a drinking water supply is known to contain tritium f rom another source, then this asst'ption is no longer valid and the concentration in drinking water must be jetermined.
7
5.2 Assess ment of Releases to Surface Water Tritium released to water in' the environment is assumed to be HTO. Com-
~
plete mixing _will ultimately occur. However, the time to achieve complete mixing and the location and steady state concentration where complete mixing occurs depends on the site specific _ characteristics of the body of water receiving the release. Mathematical models describing methods for determining the ' dispersion of tritium in surf ace water are discussed in NCRP Report No. 76 (NCRP, 1984). As with releases of tritium to the atmosphere, the first step .
in. determining.the dose from tritium released to surface water is to calculate the concentration in water at the point of _ interest where water is being consumed.
The pathways of exposure available to humans following a release of tritium' to water are drinking water and foods irrigated by that water. The model described in Eq. 5.3 can be applieH to estimate dose and, for simplicity, the tritium to hydrogen ratio in food due to irrigation is assumed to be equal to that in drinking water. The contribution to dose from tritium in the atmosphere is assumed to be negligible.
- 6. Dosimetry As shown in Eq. 5.3, estimates of dcoe due to tritium are made by multi-plying the activity ingested or inhaled br the Dose Conversion Factor (DCF).
For a given intake mode (ingestion, inhalation, or absorption through the skin), a dose conversion f actor for any radionuclide -is the committed dose equivalent to a specified organ per unit intake of the radionuclide. In lieu of an organ-specific dose conversion factor, one may also consider the commit-ted effective dose equivalent, which is the weighted average of organ-specific DCF's, with weights proportional to risks associated with stochastic fatal health effects, as defined by the International Commission on Radiological -
Protection (ICRP, 1977).
Killough (1982) reviewed the dosimetry for tritium in tissue following intake by ingestion, inhalation, and skin absorption and calculated dose conversion factors. An intake of HTO either by ingestion or inhalation is generally assumed to be completely absorbed and to mix uniformly with the water content of the body. For most organs and tissues, the average emitted beta-ray energy of 5.685 kev is treated as if it were completely absorbed within the organ containing the radionuclide (the source organ). Exceptions
' to this are transfers of energy among skeletal tissues that are treated as discrete targets (er h steal cells, red marrow) and from the contents to the walls of the gastrointestinal tract. A quality factor (Q) of 1 is used in the i derivation of dose conversion factors for tritium.
I Exposure to contaminated atmosphere results in complete uptake of inhaled RTO and its absorption through intact skin at a comparable rate. Pinson and Langham (1957) estimated that the rates were equal, and results of a study
- reported by Osborne (1966) suggest that absorption through the skin accounts j f or 60% of the total uptake rate when inhalation and skin absorption are the only two modes of exposure. The reader is reminded, however, that because of the ubiquitous nature of tritium following a release to the atmosphere, all
! modes of exposure are likely, including ingestion, inhalation, and skin absorption, and that the ingestion pathway likely dominates since that pathway is the primary mode of entry of hydrogen into the body.
8 1
e
- Biological removal of tritium f rom the body occurs through urination, fecal excretion, sweat, exhalation, and insensible water loss through'the skin. Killough (1982) derived organ-specific dose conversion factors using a dynamic compartment model based on hydrogen balance in reference man and equilibrium of specific activities between body water and other tissues.
Killough's data indicate that there is little difference between the dose conversion factors for intake of tritium by ingestion and inhalation or skin absorption.
- 7. Dose Equivalents Resulting from Release of Tritiates aste Water to the Atmosphere and Surface Water at Three Mile Island 7.1 Assessment of Tritium Releases to the Atmosphere In the case where the waste water is decontaminated and tritium is released to the atmosphere as HTO, it is, assumed that the tritium will mix with the water in air and with water in the environmental media in the vicinity of the point of release. The source term for tritium release to atmosphere in this case is estimated to be 19.5 TBq/y (526Ci/y). It is assumed that the release rate is approximately constant throughout the year.
Dispersion parameters (X/Q) for the TMI Nuclear Station have been previously calculated based on the average annual meteorological conditions for the facility (USNRC, 1981). The maximum hypothetical dose from tritium would occur at the point having the highest dispersion parameter and where food products are grown. From the Environmental Impact Statement (NRC, 1981),
we find this point to be 1.7 km (1.05 miles) gast gf the release point and an associated dispersion parameter of 2.13 x 10 s/m,.. Using this information, an upper limit estimate of the dose to a hypothetical individual residing at this point may be determined as follows:
The concentration of tritium in air at the point of interest is calcu-1.3 x latgj by mugtiplying tge source term by the dispersion parameter to give 10 TBq/m (35 pCi/u ). This value represents the average concentration of HTO in atmosphere at the point where maximum exposure occurs. In order to apply the models in Section 5, however, this concentration must be converted into TBq/gH. This is agcomplished by assuming the average specific humidity in the area is 6 gH 20/m (NCRP, 1984) and correcting for the atomic weight of hydrogen {gwater.The result gives an activity concentration ofSince tritiumtritium of 2.0 x 10- TBq/gH (54 pCL/gH) in air at the point of interest.
is not concentrated above ambient levels by biological media, it is reasonable to assume that tritium hydrogen ratios in food products grown at the point approach that in the ambient air. This bounding assumption also yields a higher estimate of dose than what actually occurs because it is most unlikely that all of the individuals food products would be grown in the area where maximum concentrations exist. Further, as stated in Section 5, it is assumed that the concentration of tritium in water being consumed by the individual is 1% of that in air at the point of interest. Applying these assumptions to the model de-scribed in Section 5 f or atmospheric releases , the upper limit of the effective dose equivalent rate to a hypothetical individual is calculated to be 3.0 USv/y (0.3 mrem /y) and the upper limit of the total ef fective dose equivalent commit ment is calculated to be 6 pSv (0.6 mrem) for complete disposal of the waste water over a two year period. A breakdown of contributions to effective dose equivalent by pathway indicates that food ingestion accounts for 67%, milk ingestion 18%, inhalation and skin absorption 132, and drinking water 1%.
9
- 7. 2 Assessment of Tritium Releases to Surf ace Water If the decontaudnation process releases tritium as HTO to river water rather than to air, it is assumed that the HTO mixes completely with the river water. The source term gor tritium relgase is given as approximately 19.5 TBq/y (526 Ci/y) in 84 m / min (2.2 x 10 gal / min). It is assumed that this release rate is constant throughout the year.
of 9.7Dispergiog/s x 10 m (3.4 x 10in the ft /s) Susqgehagna averaged over a year River (USNRC, is estimated1981). by assuming Assuming complete mixing of river water downstream of the discharge point whereconsumptionoccurs,thesteadystateconcentrationggtritigminriver3 TBq /m (17 nCi/m )
water from tgg waste water treatment (This process is 6.4 x 10-or 5.8 x 10- TBq/gli (.15 pCi/gH). assumption is a simplification.
suggested by the statement given in the Environmental Impact Statement (USNRC, 1981) that "below York Haven Dam additional mixing occurs and the full flow of the river may be used in determining dilution f actors.")
The effective dose equivalent can be estimated using the model described in Section 5 and making assumptions to simplify the analysis. It is assumed that the concentration of tritium in drinking water and all food, including aquatic food, and milk equals the concentration of tritium in the river. The only feasible mechanism for tritium in food to have the same concentration as
' that in river water is by assuming all food is derived from a source where irrigation is the only water source for the food crop. Although this assumption is highly unrealistic for the TMI area, it is consistent with other conservative assumptions made in this assessment. The concentration of tritium in air is assumed to be zero when the release is to surface water.
Applying these assumptions to the model in Section 5, the ef fective dose equivalent rate is calculated to be 0.01 pSv/y (1. prem/y) and total effective dose equivalent commituent is calculated to be .02 pSv/y (2 prem) for complete disposal of the waste water over a two year period. These estimates are, as before, for a highly hypothetical individual. The contribution to effective
-dose equivalent commitment from the various pathways is as follows: 54%
drinking water; 36% food; and 10% milk.
10 t
, ,.r---- - . . . - . . .n. - - - - - , , ,--.--..,,-..e.,-.m,.,,vx..n--, n e , --,., ~c-wa ,
4 9
- 8. Summary of Health Ef fects and Conclusions It in emphasized that due to the very conservative assumptions made to simplify this assessment, the estimated ef fective dose equivalents are upper bounds and it is highly unlikely that a person exposed to tritium released during the waste water clean up would receive doses approaching these values.
Table 8.1 summarizes the assumptions made to estimate .the effective dose equivalents and Table 8.2 summarizes the results for _ releases to atmosphere and to surface water. The estimated effective' dose equivalents resulting from each method of release are likely much higher than anyone would realistically receive, and either pa'thway would result in radiation doses that are well within acceptable limits.
In view of the low level of effective dose equivalent calculated for the maximally exposed hypothetical individual, detailed calculations of collective dose equivalents are not warranted. ,
In this Commentary, for the purposes of assessing the health impacts of l
' ingesting and inhaling tritium, it is assumed that a uniform whole body dose J
equivalent of 1 Sv (100 rem) will result in an average lifetime fatal cancer t risk plus severe genetic risk of approximately 2 x 10-2, These risk values reflect current estimates of the ICRP (ICRP, 1977), but .
do not account for potential changes that may result f rom the re evaluation of the Japanese atomic bomb survivor data. In addition, the quality factor of I for tritium beta radiation used in this Commentary is under review. However the net effect of both of these reviews is unlikely to result in an increase
- of the risk values by an order of magnitude. ,
Applying the risk estimates given above to the ef fective dose equivalent values given in Table 8.2 we find that the release to. atmosphere option will result in a lifetime cancer plus severe genetic risk to the most highly exposed hypothetical individual of approximately 1 chance in 10 million.
The release to surface water option will result in a lifetime cancer risk plus severe genetic risk of approximately 1 chance in 1000 million to the most highly exposed hypothetical individual. .
Since these risks are both below the Negligible Individual Risk Level of 10-7/y recommended by the NCRP, and below the risk associated with one day of natural background, the health and safety of the public will be unaffected by the release of the treated waste waters f rom TML-2 and therefore either option is acceptable.
b i
11 f
.a<a- w ---,.----,-y,- ,.m_ , , ,,w--,, -,.,-m-~.n. , - - - - ~ - , .,e- - ,,,. .,c,, e----, . - - . - - , ~ - -
9 Table 8.1 - Assumptions made for calculating tritium releases to atmosphere and surface water and estimating effective dose equivalents for clean up of TML waste water Release to Atmosphere Release to Surface Water Dispersion calculated using highest Uniform mixing in rivgr gater with a y/R for point 1.7 km (1.05 mi) east flow rate gf 9.7 x 10 m /S
'of site 34,000 ft /s averaged over a year Concentrations of tritium in air, All food products including aquatic food, and milk are equal foods have same tritium concentra-
, tion as river water Individual consumes only food All drinking water comes from the grown locally river Concentration of tritium in drinking No inhalation pathway exists water is 1% of that in air l
\
12
.s .
Table 8.2 - Summary of effective dose equivalent commitments resulting from two modes of release Release to Atmosphere Release to Surface Water Effective Dose Equivalerc Commitment Total 6.D pSv (0.6 mrem) 0.02 pSv (2 prem)
Pathway Food (67%) 4 pSv (0.4 mrem) Food (36%) 0.007 pSv (0.7 prem)
- Milk (18%) 1 pSv (0.1 nirem) Milk (10%) 0.002 pSv (0.2 prem)
Inhalation (13%) 0.8 pSv (0.08 mrem) 4 Inhalation (0%) O pSv ( 0 prem)
Drinking Drinking 0.06 pSv (0.006 mrem) Water (54%) 0.011 pSv (1.1 prem)
Water (1%)
e 13
Appendix A. Spscific Sources cE Waste Wster at Three:Hile island TOTAL RADIOACTIVITY VQ.UME H-3 Sr-90 Cs-137 Cs-134 Sb-125 Co-60 GALLONS DATE Ci Ci C Cl C C6 STORAGE DESCRIPTim RCS RE ACTOR COOL ANT SYSTEN . 67,286 3/7/86 . 3.06E+01 4.58E+02 6.62E
- 01 1.88E*00 9.17E+00 2.39E +00 i
PWST-l PROCESSED WATER STORAGE 109,081 2/22/86 1.24E 402 6.61E-03 2.81E-03 PWST-2 PROCESSED WATER STORAGE 480.134 2/24/86 5.09E+02 9.63E-02 8.00E-03 CO-T-I A CONDENSATE STORAGE 101.518 3/3/06 2.lSE + 01 6.92E-02 1.69E-03 5.610 4/12/83 2.76E +00 S ABE-04 2.00E-04 2.00E-05 1.91E-05 WDL-T-9A EVAP.C0re. EST ! ANK
- WDL-T-9B EVAP.COND. TEST TANK 2.231 4/17/03 1.10E400 7A3E-04 4.22E-05 2.%E-06 CC-T-1 EPICOR it off-SPEC 20,500 3/5/86 1.0lE+0i 4 50E-02 IA0E-02 3.88E-04 3.80E-03 EPICOR il CLEAN 16,887 11/15/85 5.62E+00 1.92E-02 9 59E-03 7 03E-03 415E-04 CC-T-2 I
SFP-B SPENT FUEL POOL *B' 241.698 3/2/86 4.12E +01 3.02E-02 3.29E-03 373 3/7/86 1.07E-01 7A8E-03 1.38E-03 9.3?E-04 i SDS-T-1A SDS MONITOR 497 10/10/85 1.37E-01 1.82E-03 1.88E-03 1.77E-03 1.30E-04 SDS-T-lO . SDS MONITOR WDL-T-1 A RC BLEED HOLDUP 3,810 2/24/86 1.23E+00 4.76E-01 1.34E-Ol 3 89E-04 6.63E-03 1.30E-03
" WDL-T-IB RC BLEED HOLDlp 4.420 3/7/86 2.17E*00 2.84E+0i 3.3,5E + 00 9.37E-02 5.69E-0I l .10E-0 l l 57,116 10/31/85 3.68E +0 ! SA0E+02 3:68E*01 1.75E*01 1.41E+00 WDL-T-lC RC BLEED sa Dto
, BORATED WATER STORAGE 458,915 3/4/86 1.15E+02 6.60E-01 2 26E-Ol 1.55E-02 4 69E-03 BWST 6.675 2/28/86 2.96E +00 4.60E+00 6.2 4 ,00 1.74E-O l '
WDL-T-8A MUTRALIZER WDL-T-6B NEUTRALIZER 8.605 3/l/86 2.21E+00 2.5 tE+00 5 86E+00 1.73E-01 HISCELLANEOUS WASTE HOLDUP 3,712 2/28/06 9 69E-01 1.10E+00 2.39E+00 6.88E-02 WDL-T-2 M)L-T-II A CMTANINATED DRAINS 1,931 3/1/86 1.53E-04 1.97E-04 3.07E-04 6.80E-06
)
WDL-T-llB CONTAMINATED DRAINS 820 3/l/86 4.3SE-05 3 A IE-05 1.680 3/2/06 2.86E-O l 6.99E-03 5.60E-03 2.16E-02 5 02E-05 CilEN CLEANING 8tD0 Surf AUXILIARY BLOG SUMP 5,917 10/4/85 2.91E+00 2.69E+00 5.15E-01 REACTOR BLOG BASEMENT 43.082 4/26/85 4.24E+00 2.6IE+02 7.99E*02 SPENT FUEL POOL *A* 205.234 2/27/86 2.02E+02 2A9E+0i 6 84+00 1.86E-01 2.02E + 00 1940-01
, SFP-A DEEP END Of TRANSFER CANAL 50,685 3/12/86 6.66E *01 5.78E+00 1.9IE*00 4.89E-02 6A4E-01 5.55L 02 1
1,908,417 1182.75 1331.17 929A9 j TOTAL AS Cf 1/1/86 l
" This table was taken directly from the CPU Report (GPU 1986) .
l l - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . . _ _ _
, ~ . - . . . ~ _ - . . - . _ - . . _- . .
. w
.o .
Appendix.B - Decontamination Factors Required to Ensure that the Dose from Radionuclides other than Tritium are Relatively Insignificant The source term data shown11n Appendix A were analyzed to' determine the-decontamination factor necessary to ensure .that none of the five radionuclides .
listed in the inventory, would contribute sore than l% of the radiation dose to members of the public from tritium during-the release of the waste water.
B.1 Evaluation of a Process Releasing All-Radionuclides to the Atmosphe're For release'to the atmosphere, it is assumed that no effluent would be released to water and that the radionuclides are' released during processing over a period of one year. The calculation was made using the-AIRDOS-EPA (Moore'et al. 1979) computer code.
l It was also assumed that the radionuclides would be released at a constant rate during the processing of the waste water. Assumptions of the type given in NCRP Report No. 76 (NCRP 1984) were used for atmospheric' dispersion and environmental transport and for the potential pathways of significance. Dose estimates were made for each radionuclide and then normalized to that for tritium. All of the values are rounded to one significant digit. The results are given in Table B-1.
Table B The significance of doses from five radionuclides
. relative to tritium for releases to atmosphere 4
Radionuclide Ratio of Doses to'that from Tritium 3H 1 60Co 20,000
- 90 3r 30,000 123 Sb 3,000 134 Cs -30,000 137 Cs 20,000 i
Although the uncertainty is probably less than a factor of ten, for the l purpose of applying these data to the situation at TMI, an arbitrarily chosen T criteria and doses from other radionuclides should not give rise to radiation
- doses greater than 1% of that from tritium.
90 Sr is 30,000 f The data in Table B-1 indicate that iftheggurcetermfor times less than that for tritium, the dose from Sr would be no more than j
- that f rom tritium assuming the releases of each are occurggng simultaneously Sr is 3,000,000 and are to the atmosphere. Thus, if the source term for times less than that f or tritium, the dose will be no more than 1% of that
- from tritium.
3 Since the source term for ritium is 4 86 GBq/ 3 (1.3 x10-I uCL/cm ), ,
90 Sr oncentration of 1.6 KBq/m (4.3 x 10-b UCi/cm will result in a dose of 10% of that f rom tritium. Using 100% procegg Srvalues from Table concentration 2-5 MBq/m of 0.37 of the 3 July 1986GPURegort (GPU, 1986) we find a (1 x'10-) uCi/cm . This value, taken together with an evapoggtor decontami-nation factor of I for tritium and between 100 and 1000 for Sr results in a 4
15 a
l 4
1
-we -w-r -
,_-,,.-,me, ,._m-,..y,w r.,.--_.__-c _ .
-l PCi/cm ) f # h#itium to 0.37 KBq /m (1 x 10 pCL/cm3) f or gl.3xIO~
release concegtration tt;st
'woulf)be 4.8 GBq/m 0 and 3.7 KBq/3 (1 x 10 pCi/cm Sr.
If it is possible to attain a decontamination factor of between 1000 and 10,000 the 40% process value as 'taken f rom Table 2.1 would meet these criteria.
B.2. Evaluation of a Process Releasing All Radionuclides to Surface Water For releases to surface water, it~ is assumed that tritium and other radionuclides are released to water and that no effluent enters the atmosphere directly during processing. Both drinking water and consumption of aquatic foods are included in the analysis. The results are given in Table B-2.
Table B The significance of doses from five radionuclides relative to tritium for releases to surface water a
Radionuclide Ratio of Doses to that from Tritium 3H 1 60Co 1,100 9 2,000 2pr 100 Sb 3 2,200 Cs f3 Cs 1,500 Again, for the purpose of applying these data,to the situation at TMI, a getoroftwoordersofmagnitude (100) has been used. If wgagain Sr and include this factor of 100, then a source term for consider Sr which is a factor of 200,000 less than the source term for tritium would give rise to a radiation dose of 1% of that from tritium.
3 90 Sr Since the source term for tritium is 4.8 GBq/m (1.3 x10-I pgi/cm),the concentration must be less than 24 KBq/m (6.5 x 10-7 pCi/cm ) for the dose to be lower by a factor of 100 from that due to tritium. Using the 100%ggrocess valuesfromTable2-5ofttjeJuly19g6GPURgport (GPU, 1986) we find a Sr ,
i concentration of .37 MBq/m (1 x 10- pCi/cm ). Processing of ge waste through the evaporator would ensure that the effluent concentration of Sr would be 200,000 times less than the tritium concentration.
i 16
+d
- REFERENCES Alvarez, L. W., and Cornog, R. (1939). " Helium and Hydrogen of Mass 3," Phys. Rev.
569 No. 2, 613.
Blomeke, J. O. (1964). Management of Fission Product Tritium in Fuel Processing Wastes, USAEC Report ORNL-TM-951 (Oak Ridge National Laboratory).
Evans, A.G. (1969). "New Dose Estimates from Chronic Exposure to Tritium," Realth Phys. 16, 57. ,
GPU letter, No. 44-10-86-L-0114, F. R. Standerfer (GPU Nuclear, to W. D. Travers (NRC) dated July 31, 1986. NRC Docket No. 50-320.
lAEA (1984). International Atomic Energy Agency, Management of Tritium at Nuclear Facilities, Technical Reports Series No.,24 (International Atomic Energy Agency, Vienna).
ICRP (1975). International Commission on Radiological Protection, Report of the Task Group on Reference Manual, ICRP Publication 23 (Pergamon Press, New York).
ICRP (1977). International Commission on Radiological Protection, Recommendations Protection, ICRP Publiation 26 (Pergamon Press, New York)
Killough, G. G. (1982). " Derivation of Dose Conversion Factors for Tritium,"
NUREG/CR-2523, ORNL-5853.
P., Hof f man, F.
Moore, R. E., Baes, C. F., III, McDowell-Boyer , L. M. , Watson, A.
O., Pleasant, J. C., and Miller, C. W. (1979).
"AIRDOS-EPA: A Computerized Methodology for Estimating Environmental Concentrations and Dose to Man from Airborne Radionuclides," ORNL-5532.
NCRP (1975). National Council on Radiation Protection and Measurements, Review of the Currect State of Radiation Protection Philosophy, NCRP Report No. 43, National 'h Council on Radiation Protection and Measurements, Washington, D.C.
NCRP (1979). National Council on Radiation Protection and Measurements, Tritium in the Environment, NCRP Report No. 62, National Council on Radiation Protection and Measurements , Washington, D.C.
NCRP (1984). National Council on Radiation Protection and Measurements, Radiological Assessment: Predicting the Transport, Bioaccumulation, and Uptake by Man of Radionuclides Released to the Environment, NCRP Report No. 76, National Council on Radiation Protection and Measurements, Washington, D.C.
Osborne, R.V. (1966). " Absorption of Tritiated Water Vapour by People," Health Phys.
12, 1527.
Pinson, E. A. , and Langham, W.H. (1957). " Physiology and Toxicology of 1087 Tritium in Man," J. Appl. Physiol. 10:108. Reprinted: Health Phys. 38 USNRC (1981) US Nuclear Regulatory Commission (1981). " Final Programatic Environmental Impact Statement Related to the Decontaednation and Disposal of Radioactive Wastes Resulting f rom March 28, 1979, Accident Three Mile Island Nuclear Station, Unit 2," NUREG-0683, Vol. 2, Appendix W, USNRC, Washington, D.C.
17
L. ; .; :r:;; =
- '
- NCRP CDP 9ENTARY No. 4 CUIDELINES FOR THE RELEASE OF WASTE WATER FROM NUCLEAR FACILITIES-WITH SPECIAL REFERENCE TO THE PUBLIC HEALTH
- 1-SIGNIFICANCE OF THE PROPOSED RELEASE OF TREATED WASTE 1
WATERS AT THREE HILE ISLAND s'
i 4
National Council on Radiation Protection and Measurements 7910 Woodmont Avenue, Bethesda, Maryland 20814
- 1 i
i I
4 5
/#
\
e.- --+ , - - ,,. - . . . , ,-.,,~4y, e -e-~-~ . . . - , - oy y,-- , .,--- .-me --
,r-w w w e- - - m yemw -v
Preface In May of 1980, the NCRP issued a report entitled, Krypton-85 la the
' Atmosphere - With Specific Reference to the Public Health Significance of the Proposed Controlled Release at the Three Hile Island, in response to public concerns over the proposed venting of krypton from TMI-2. That document was prepared in response to a request from Governor Thornburgh of the Commonwealth of Pennsylvania and as part of NCRP's responsibilities as stated in its congressional charter. The charter specifies, among other things, the responsibility to " collect, analyze, develop and disseminate in the public interest, information and recommendations about (a) protection against radiation and ...". The study was also recognized as part of the general problem of controlling releases to,the atmosphere of which the circumstances at TMI-2 were a special case.
The problem of releases of waste water from TM1-2 could raise similar scientific and public issues. Again, TMI-2 is a specific case of a general problem.
In 1980, the Nuclear Regulatory Commission (NRC) requested the NCRP to examine this issue. Recognizing that it could serve the public Interest, the Council established a Task Group to address this problem. The Task Group prepared a draft report which was reviewed by the Council in 1985. The members of the Task Group were:
Frank L. Parker, Chairman Vanderbilt University Nashville, Tennessee A. Bertrand Brill Bernd Kahn Brookhaven National Laboratory Georgia Institute of Technology Upton, New York Atlanta, Georgia Donald G. Jacobs Edward Watson Roy F. Weston, Inc. Battelle Pacific Northwest Laboratories Oak Ridge, Tennessee Richland, Washington lloweve r , further proposals to release the waste water at TMI-7 have been developed recently by GPU Nuclear. Therefore NCRP established a new Task Group to review the potential environmental Impacts of these proposals.
The U. S. Nuclear Regulatory Commission has reserved to itself the deci-sion on disposal of the waste water. As part of this process, NRC issued a
" Final Programmatic Environmental Impact Statement Related to Decontamination and Disposal of Radioactive Wastes Resulting from, March 28, 1979 Accident, Three Mile Island Nuclear Station, Unit 2" in 1981. A supplement to the EIS is now out for public comment. This Commentary is intended to provide the NCRP's response to the request for public comment.
i
. .l The System International (SI) units are used in this report but, with the
- exception of Section 5, are followed by the conventional units in parentheses in accordance with the procedure set forth In NCRP Report ?b. 82.
Serving on the Task Group during the preparation of this Commentary were:
Charles B. Meinhold, Chairman Brookhaven National Laboratory .
Upton, New York Leonard Emma William L. Templeton Brookhaven National Laboratory Battelle Pacific Northwest Laboratories Upton, New York Richland, Washington Donald G. Jacobs John E. Till Roy F. Weston, Inc. Radiological Assessments Corporation Oak Ridge, Tennessee Neeses , South Carolina Consultants Victor P. Bond John W. Healy Brookhaven National Laboratory Los Alamos, New Mexico Upton, New York Melvin W. Carter' Frank L. Parker Georgia Institute of Technology Vanderbilt University Atlanta, Georgia Nashville, Tennessee ,
NCRP Secretariat - E, Ivan White The Council wishes to express its appreciation to the Task Group members and consultants for the time and effort devoted to the preparation of this Commentary.
' Warren K. Sinclair President, NCRP Bethesda, Maryland February 25, 1987 l
i I
l 11
s
=g O
CONTENTS Page Preface ............................. * *****************I
- 1. Introduction ...................................................!
- 2. Status of Accident Generated Waste Water at Three Mile Island........................................... 2
- 3. Options for Treatment of Tritiated Waste Waters from Three Mile Island......................................... 4 3.1 Evaporation / Atmospheric Release........................... 4-3.2 Evaporation / Surface Water Release......................... 5
- 4. Tri tium - Physical and Chemical Propert ies. . . . . . . . . . . . . . . . . . . . . 6
- 5. Tritium - Environmental Tranr. port and Pathways of Exposure... . . 7 5.1 Assessment of Releases to the Atmosphere.................. 7 5.2 Assessment of Releases to Surface Water................... 9
- 6. Do s i m e t r y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
- 7. Dose Equivalents Resulting from Release of Tritiated Waste Water to the Atmosphere and Surface Water a t Th r e e Mi l e I s l a n d . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 7.1 Assessment of Tritium Releases to the Atmosphere..........ll 7.2 Assessment of Tri tium Releases to Surface Wa ter. . . . . . . . . . 12 Summary of Health Effects and Conclusions..................... 13 8.
Appendix A - Specific Sources of Waste Water at 15 Three Mile Island........................................
Appendix B - Decontamination Factors Required to Ensare that the Dose from Radionuclides Other Than Tritium are Relatively lasignificant.................. 16 18 References.........................................................
iii
- . iY s.
- 1. Introduction
. The risk associated with the release of . waste . water from' nuclear facili-ties,-whether generated by accident or produced during " normal" operations, is of interest to the public and the nu: lear industry. The presence of waste water at Three Mile Island (TMI) and proposals for its release represent a special case of this problem, the analysis of which is widely applicable. !
In many situations, including the case at TMI, the radionuclides other than tritium can be removed from the waste water by various processes.
Tritiated water cannot be separated and concentrated from ordinary water by conventional waste treatment techniques (Blomeke, 1964). Isotopic separations are available, but these are impractical for high volume, low concentration operations (IAEA, 1984). Thus, while radionuclides other than tritium can be removed from contaminated waste water, the tritium must be handled by other means. Therefore,. the focus here is primarily on the disposal of tritiated waste water.
Because of the previous work on this subject by the NCRP, it is possible to apply existing knowledge and procedures to the situation at TMI.
The National Council on Radiation Protection and Measurements (NCRP) in
.1979, published NCRP Report No. 62, Tritium in the Environment (NCRP, 1979).
That report dealt with the available information on tritium in terms of its physical properties, production sources, physical transport, biological behavior, projected future production, waste management and long-term implications in the environment. I The Council, in 1984, published NCRP Report No. 76, Radiological Assessment: Predicting the Transport, Bioaccumulation, and Uptake by Man of Radionuclides Released to the Environment (NCRP,-1984). That report dealt with a review of the current status of the application of radionuclide transport models from the point of discharge to the environment to the point of Intake by man. Models are reviewed that describe the transport of
-l radionuclides through the atmosphere, surface and groundwater, deposition on l- terrestrial surfaces and in sediments and accumulation in food products.
Usage factors are considered that determine the intake of radionuclides by humans due to dietary habits, physiological parameters and living customs.
I The informati s provided in the two cited NCRP reports is used in the analysis here of the public health significance of the disposal of waste water contaminated with tritium. Optional methods for release of tritiated waste water are reviewed, information about tritium provided, and environmental transport and pathways are examined and dosimetry discussed. Finally, doses and health effects resulting from waste water releases are given.
i l
l l
l 1
c.
,s q:
- 2. -- Status of Accident Generated Waste Waters at Three Mile Island The TM1-2 accident resulted in the production of large volumes,of con-taminated water. Since the - time of the incident , .the total inventory of this water hac: Increased to.approximately 1.9 million gallons due to continued in-leakage from support systems and condensation from the Reactor Building air coolers._ The specific sources of the waste water and the activities of the
- radionuclides in each are given in Appendix A. When the clean-up is completed in October, 1988, it is estimated that approximately 2.1 million gallons of waterwillrequiredisposiggon.ggnadpjgion[gtritium,gpeprinciple Sb, Cs, and Cs. GPU Nuclear radionuclides present are Co , . Sr, has noted (GPU, 1986) that prior to ultimate disposition, a considerable amount of this water will require processing to reduce the levels of radio-
' active contaminants. Thisreductionoftheradggnuclidelevelswillminimize the cotal release of activity, particularly of Sr, to the environment and thereby minimize the environmental consequences associated with the various t
. disposal options.
The volume of water requiring. processing prior to ultimate disposal depends upon the selection of the ' final disposal method. Three methods have-been considered (see Section 3) - - river discharge, direct solidification and evaporation. For the " river discharge" and " direct solidification"' options, essentially all of the water would require initial processing, or reprocessing, through existing lon exchange systems prior to disposal. This is referred to as 100% processing. In light of the decontamination factor achieved by evaporation, it is estimated that only about 40% of the total
. volume will require reprocessing to reduce the activity levels before final disposal by the release to the atmosphere. This is referred to as 40%
processing. The average characteristics of the water expected after these two degrees of processing are presented in Table 2.1.
The boron concentration in the water is given in Table 2.1. Boron may influence the disposition options either through discharge limits to the envi-ronment'(via the federally mandated release limit of 25 ppm boron), increased quantitles of concentrates requiring solidification from the evaporation option, or the necessity to add stabilizing agents to ensure proper solidifi-
. cation. In addition, the water will contain approximately 11 tons of sodium hydroxide.
2 f
,,re . , - , _ + - - - ._, - , ..y, , . , , e. , _ _ . . . , _ . . . . . - _ . - - _ . _ _ . - , _ _ . . . . . - . , - - - . , , _ ~ - nm._ -
= - - .
.,. . - . . .. . . - . . - -.- ~ . . . - . . - .. .. -. -- . , , ,
n' i .;
Table 2.1 - Pre-disposition water characteristics i
40%
137c , 90 Sr Boron Sodium-Processing Tritium l l
Conc neration ( 4.8x109 (1.3x10-I) 1.4x106 (3.7x10-5) 4.3x106 (1.15x10-4)
I Bq/m (uCf/ml) i i
f Total Activity 3.8x10I3 (1020) 1x1010 (0.29) 3.3x1010 (0.9)
Bq (Ci) ,
I Concentration 3,000 700 Ppm
)
J Total Quantity 150 11 Tons As H3B04 As NaOH 100%
Processing 4.8x109 (1.3x10-I) L 1.5x105 (4x10-6) 3.7x105 (1x10-5) i Conegntration Bq/m (uCi/ml)
I j Total Activity 3.8x10I3 (1020) lx109 (0.03) 3x109 (0.08)
Bq (Ci) ,
Concentration 3,000 700 l ppm t
Total Quantity 150 11 l .As NaOH j Tons '
As.H 3E04 ,
i I (
i
- 3 j l
t v
! I
-3. Options For Treatment Of Tritiated Waste Water from Three Mile Island In the July, 1986, Report, ~ Disposal Of TMI-2 Water" (GPU, _1986), GPU Nuclear proposed three disposal options for the processed waste water. These options are as follows:
- 1. Evaporation - Processing and evaporation of the water would be by an Installed evaporation facility followed by controlled atmospheric release.
Shipment to, and disposal of, solidified residues at a licensed, commercial low-level waste disposal site would follow.
- 2. Solidification - Processing and solidification of thu water in cement would be followed by burial in an on-site industrial landfill.
- 3. River Discharge - Processing and controlled, monitored discharge to the Susquehanna River would result in significant dilution of the processed water.
The GPU Report noted that direct release to the river is the best choice on the basis of overall technical merit but that political and institutional concerns resulted in their suggesting the evaporation / atmospheric release option as the method of' choice. We therefore reviewed the potential impacts of evaporation / atmospheric release option given above, slightly modified to ensure that the dose from other radionuclides are appreciably less than those from tritium (see Appendix B). The decision to employ such a modification in our analysis was based on the unique difficulty of removing tritium from waste water, the comparatively low radiogenic hazard associated with tritiated water, and the ease and precision of tritiura environmental transport analysis.
One could reasonably argue, however, that in view of the extremely low doses involved, use of this approach is unnecessarily conservative. Since this analysis also suggested that the addition of an evaporation step might improve the viability of the River discharge option, we analyzed this scenario as well.
We have not, however, assessed the solidification option since the GPU Report suggested that it had the lowest overall merit of the three options.
3.1 Evaporation / Atmospheric Release This option employs a standard industrially available evaporation unit which, depending on the available decontamination factor, would require pre-processing of between 40% and 100% of the waste waters (see Appendix B). The vapor produced by the evaporator, which contains essentially all of the tritium, would be released through a 50-meter stack. If a typical processing rate of 3 gallons per minute is assumed and the operating basis is seven days a week, with two, ten-hour shif ts per day of actual processing with an overall availability of 80%, approximately two years would be required to process the total volume of waste water (2.1 million gallons). This option would result in a release rate of 0.93 MBq/sec (25 pCi/sec) or approximately 19.5 TBq/y (526 Ci/y) of tritium to the atmosphere.
I 4
4 3.2 Evaporation / Surface Water Release This option employs the evaporation step followed by controlled release of the condensed vapors to the -Susquehanna River. Again, the extent of required preprocessing of input waste waters would depend on the efficiency of the particular evaporator used. In ang case, as shown in Appendix B, the evaporation step will ensure that the Sr source term is insignificant compared to that for tritium.
After evaporation, rather than releasing to the 50 meter stack, all the vapor would be condensed and pumped to one of the two ll,000-gallon condensate test tanks. From these tanks the water would be discharged to the Susquehanna River via blow down from the mechanical draft cooling tower. The use of the cooling tower blow down in this manner provides a dilutant flow of about 22,000 gallons per minute. Use of this option would also result in a release-of 19.5 TBq/y (526 Ci/y).
For either the " Evaporation / Atmospheric Release" or the " Evaporation /
Surface Release" option, disposal of the waste water will require approximately two years.
5
f.
- 4. Tritium - Physical and Chemical Properties Tritium is the heaviest and only radioactive isotope of hydrog'en. It was discovered la 1939 by Alvarez:and Cornog (1939) who determined that it had a long half-life and emitted radiation with a very short range. Subsequent work established that it decays with a half-life of 12.3 years. It emits a beta partiglewithamaximumenergyof18keVandanaverageenergyof5.7keVto form lie .
Tritium is produced naturally by the interaction of cosmic rays with elements in the upper atmosphere. It is also produced by thermal or fast neutron reactions with various light elements utilized in reactors, such as boron, used for reactivity control, and lithium, used for corrosion control.
Tritium formed in this way is circulated in the coolant and from there is released into the environment. Most of the fission product tritium is normally retained within the fuel element cladding. However, an appreciable fraction of the fission product tritium can be released from the fuel elements .
when the core is damaged such as in the case of TMI-2.
For a detailed review of the physical and chemical properties of tritium, see NCRP Report No. 62 (NCRP, 1979). Tritium closely follows the reactions of ordinary hydrogen, although, the relatively large mass differences among the hydrogen isotopes make isotopic effects discernible. Because of the prevalence of water and its importance in the life processes, the isotopic exchange of hydrogen in water with tritium is of special importance.
In th'e environment, tritiated water behaves generally, though not exclusively, like ordinary water. Most of the predicted behavior of tritium is based on existing information regarding the cycling of water, supplemented by observations of the behavior of tritium produced during atmospheric testing of nuclear weapons. Tritium can also become an integral part of any chemical compound containing hydrogen atona, including the organic compounds that make up living tissue.
6
1
- 5. Tritium - Environmental Transport and Pathways of Exposure 5.1 Releases to the Atmosphere When tritium is released to the atmosphere, it disperses rapidly and mixes with stable hydrogen in the atmosphere, hydrosphere and biosphere.
Tritium released in forms other than tritiated water (HTO), tends "to convert to HTO. Concentrations of tritium in atmospheric water at a given distance and direction from a source are typically estimated using atmospheric dispersion models as described in NCRP Report No. 76 (NCRP, 1984) and the absolute humidity for the point of interest. For purposes of assessing the dose from tritium released to the atmosphere, it is generally assumed that the chemical form is HTO and that there is uniform mixing between atmospheric water vapor and the bound and unbound hydrogen in biological systems. This approach, referred to as the specific activity method, is based on data reported by Evans (1969) that suggests that body hydrogen is uniformly labeled with tritium under chronic exposure conditions. Although other approaches, such as the multicompartment model, may lead to more precise estimates of dose, the specific activity method is simpler to apply and generally results in dose estimates that are significantly higher than those which would actually exist. Therefore, if acceptable criteria for exposure are met as determined using this more conservative but simpler approach, use of a more sophisticated model may not be justified. This is especially true since the application of a multicompartment model requires the use of site specific compartment dilution volumes and intercompartment transfer rates. The use of a simple effective model in preference to a more complex model is in accordance with previous NCRP recommendations (NCRP, 1984).
A model based on the specific activity method and the contribution to total water intake of reference man was proposed by the NCRP (1979, 1984).
The model predicts that the relative importance of different pathways of exposure is determined by the intake of body water derived from each pathway. This model was updated by K111ough (1982) to balance total hydrogen intake, accounting for hydrogen in both water and organic products consumed by individuals. The model of hydrogen balance assumed for reference man allows one to convert tritium specific activities in food and fluids and in the individual's ambient air to daily intake rates for steady-state conditions as described below (TBq/d) (5.1) lair - (15 + 9) Aair (TBq/d) ( 5. 2) lingestion -
183 $ water + 33$ milk + 120dfood the intakes via air and ingestion, respec-where Iair, and lingestion, are tively. The specific activities Aair' $ water, etc. are in units of terabequeral per gram of hydrogen (TBq/gH) and the coef ficients are grams of 7
hydrogen per day (gH/d) derived from the data on hydrogen intake and losses given for ICRP's Reference Man (ICRP Publication 23) (ICRP, 1975) as modified by Killough and others (Killough, 1982). Note that water in food is included in the coefficient 120 associated with A food in Eq. 5.2. It is als'o important to note that the coefficients 15 and 9 in Eq. 5.1 (inhalation and absorption through the skin) are sensitive to the ambient absolute humidity, and the-3 values given here correspond to an ambient absolute humidity of 6 gH 2O m The specific activities.bwater* $ milk, and Afood depend on the environmental sources of the water, milk, and food lagested by the reference -individual.
Based on information given in ICRP Publication 23 (ICRP, 1975) as updated by K111ough (1982), it is assumed that the total hydrogen intake is 360 gH/d. The hydrogen balance model for assessing dose from tritium in the environment can be described as follows:
[(15 + 9) Aair DCFinhalation + (183dwater
+ 33Amilk + 120Afood) DCFingestion] 365 d/y (Sv/y) (5.3) where, Aair' Awater, etc. = specific activities of tritium in air, water etc.
(TBq/gH), and DCFinhalation, DCF ingestion = d se conversion factors for tritium taken in via inhalation and ingestion and which have values of 22 Sv/TBq and 23 Sv/TBq, respectively.
The model can be further subdivided to account for dif ferent concen-trations of tritium present within a given pathway. For example, it is frequently assumed that individuals drink water from several sources, each containing different concentrations of tritium. If this level of detail is desired in the calculation, then the ingestion of tritium can be determined by calculating the relative intake from each source.
f The specific activity methodology assumes that, for a given location, the concentration of tritium is the same in atmospheric water and blota. This asst;mpr.lon likely leads to a higher estimate of dose than that which actually occurs because it is unusual for a steady state condition to exist in the environment near a source, considering the intermittent nature of most source terms and the variability of meteorological and climatic conditions. Assuning the specific activity of tritium in each component (i.e., air, water, milk, and food) were the same, the model tells us that intake of tritium via 1
8
ingestion of water and food are the most important pathways of exposure. The contribution to dose from Inhalation and skin absorption combined, when all pathways of exposure are available, is approximately 7%. .
It is likely that this technique significantly overestimates the dose
. from tritium to individuals who do not produce and consume.their own food products but import them from regions outside their area, where tritium concentrations in food are substantially lower._ Likewise, persons may receive only a fraction of their drinking water supply from a source containing tritium. Nevertheless, this simple model can be easily applied to estimate the dose at a given distance from the source once the concentration of tritium in atmospheric water at that location is derived using a meteorological model.
One key to applying this model is the determination of the concentration of tritium in drinking water when the only source of release is to the atmosphere. When the release of HTO is to the atmosphere, it is generally assumed that the concentration of tritium in drinking water is 1% of that in air for a given location (NCRP, 1984). This assumption is simply an attempt to account for tritium that migrates from the atmosphere to drinking water supplies. If a drinking water supply is known to contain tritium from another source, then this assumptioa is no longer valid and the concentration in drinking water must be determined.
5.2 Releases to Surface Water Tritium released to water in the environment is assumed to be HTO. Com-plete mixing will ultimately occur. However, the time to achieve complete mixing and the location and steady state concentration where complete mixing occurs depends on the site specific characteristics of the body of water receiving the release. Mathematical models describing methods for determining the dispersion of tritium in surface water are discussed in NCRP Report No. 76 (NCRP, 1984). As with releases of tritium to the atmosphere, the first step in determining the dose from tritium released to surface water is to calculate the concentration in water at the point of interest where water is being consumed.
The pathways of exposure available to humans following a release of 4 tritium to water are drinking water and foods irrigated by that water. The model described in Eq. 5.3 can be applied to estimate dose and, for simplicity, the tritium to hydrogen ratio in food due to irrigation is assumed to be equal to that in drinking water. The contribution to dose from tritium
- in the atmosphere is assumed to be negligible.
4 1
I 9
x; e
W
- 6. Dosimetry As shown in Eq. 5.3, estimates of dose due to tritium are made by multi-plying the activity ingested or Inhaled by the Dose Conversion Fact,or (DCF).
For a given intake mode (ingestion, inhalation, or absorption through the skin), a dose conversion factor for any radionuclide is the committed dose equivalent to a specified organ per unit intake of the radionuclide. In lieu of an organ-specific dose conversion factor, one may also consider the commit-ted ef fective dose equivalent, which is the weighted average of organ-specific-DCF's, with weights proportional to risks associated with stochastic fatal health ef fects, as defined by the International Commission on Radiological Protection (ICRP, 1977).
K111ough (1982) reviewed the dosimetry for tritium in tissue following intake by ingestion, inhalation, and skin absorption and calculated dose conversion factors. An intake of HTO either by ingestion or inhalation is generally assumed to be completely absorbed and to mix uniformly with the water content of the body. For most organs and tissues, the average emitted beta-ray energy of 5.685 kev is treated as if it were completely absorbed within the organ containing the radionuclide (the source organ). Exceptions to this are transfers of energy among skeletal tissues that are treated as discrete targets (endosteal cells, red marrow) and from the contents to the walls of the gastrointestinal tract. A quality factor (Q) of 1 is used in the derivation of dose conversion factors for tritium.
Exposure to contaminated atmosphere results in complete uptake of inhaled HTO and its absorption through intact skin at a comparable rate. Pincon and Langham (1957) estimated that the rates were equal, and results of a study reported by Osborne (1966) suggest that absorption through the skin accounts for 60% of the total uptake rate when inhalation and skin absorption are tite only two modes of exposure. The reader is reminded, however, that because of the ubiquitous nature of tritium following a release to the atmosphere, all modes of exposure are likely, including ingestion, inhalation, and skin absorption, and that the ingestion pathway likely dominates since that pathway is the primary mode of entry of hydrogen into the body.
Biological removal of tritium from the body occurs through urination, fecal excretion, sweat, exhalation, and insensible water loss through the skin. Killough (1982) derived organ-specific dose conversion factors using a dynamic compartment model based on hydrogen balance in reference man and equilibrium of specific activities between body water and other tissues.
K111ough's data indicate that there is little dif ference between the dose conversion factors for intake of tritium by ingestion and inhalation or skin absorption.
10
- '?
, 7. Dose Equivalents Resulting from Release of Tritiated Waste Water to the Atmosphere and Surface Water at three Mile Island 7.1 Dose Equivalents Resulting from Tritium Releases to the Atmosphere In the case where the waste water is decontaminated and tritium is released to the atmosphere as HTO, it is assumed that the tritium will mix with the water in air and with water in the environmental media in ,the
' vicinity of the point of release. The source term for tritium release to atmosphere in this case is estimated to be 19.5 TBq/y (526 Ci/y). It is assumed that the release rate is approximately constant throughout the year.
Dispersion parameters (1/ Q) for the TMI Nuclear Station have been previously calculated based on the average annual meteorological conditions for the facility (USNRC, 1981). The maximum hypothetical dose from tritium would occur at the point having the highest dispersion parameter and where food products are grown. From the Environmental Impact Statement (US NRC ,
1981), we find this point to be 1.7 km (1.05 miles) l i and an associated dispersion parameter of 2.13 x 10~gast gf the s/m . Using this re ease po nt information, an upper limit estimate of the dose to a hypothetical individual residing at this point may be determined as described below.
The concentration of tritium in air at the point of interest is cal-culatedby2multipgylng the sgurceThis termvalue by the dispersion represents theparameter average to give 1.3 x 10- TBq/m (35 pCi/m ).
concentration of HTO in atmosphere at the point where maximum exposure occurs. In order to apply the models in Section 5, however, this concentration must be converted into TBq/gli. This is accomplished by assuming the average specific humidity in the area is 6 gH 0/m3 2 (NCRP, 1984) and correcting for the atomic weight of hydrogen in water. The result gives an activity concentration of tritium of 2.0 x 10-12 TBq/gli (54 pCi/gil) in air at the point of interest. Since tritium is not concentrated above ambient levels by biological media, it is reasonable to assune that tritium hydrogen ratios in food products grown at the point approach that in the ambient air. This bounding assumption also yields a higher estimate of dose than what actually occurs because it is most unlikely that all of the Individual's food products would be grown in the area where maximum concentrations exist. Further, as stated in Section 5, it is assumed that the concentration of tritium in water being consumed by the Individual is 1% of that in air at the point of interest. Applying these assumptions to the model described in Section 5 for atmospheric releases, the upper limit of the ef fective dose equivalent, for each year of the release, to a hypothetical Individual is calculated to be 3.0 p Sv (0.3 mrem) and the upper limit of the total effective dose equivalent is calculated to be 6 pSv (0.6 mrem) for complete disposal of the waste water over a two-year period. A breakdown of contributions to the total effective dose equivalent by pathway indicates that food Ingestion accounts for 67%,
milk ingestion 18%, inhalation and skin absorption 13%, and drinking water 1%.
11 i
i
s c.
" 7.2 Dose Equivalents Resulting from Tritium Releases to Surface Water If the decontamination process releases tritium as HTO to river water rather than to air, it Is assumed that the HTO mixes completely with-the river water. The source term TBq/y(526C1/y)in84m{ortritiumrelgaseisgivenasapproximately19.5
/ min (2.2 x 10 gal / min). It is assumed that this release rate is constant throughout the year.
flow rate in the Susqgehanna Disperglog/s(3.4x10 River is estimated by assuming a of 9.7 x 10 m f t 3/s) averaged over a year (USNRC,1981).
Assuming complete mixing of river water downstream of the discharge point where consumption occurs, the steady state concentration10of tritigm in river 3 TBq/m (17 nC1/m )
waterfromthgwastewatertreatment process is 6.4 x 10 or L8 x 10- TBq/gH (.15 pCi/gH). (This assumption is a simplification suggested by the statement given in the Environmental Impact Statement-(USKRC, 1981) that "below York Haven Dam additional mixing ~ occurs and the full flow of the river may be used in determining dilution factors.")
The effective dose equivalent can be estimated using the model described in Section 5 and making assumptions to simplify the analysis. It is assumed that the concentration of tritium in drinking water and all food, including aquatic food, and milk equals the concentration of tritium in the river. The only feasible mechanism for tritium in food to have the same concentration as that in river water is by assuming all food is derived from a source where irrigation is the only water source for the food crop. Although this assumption is highly unrealistic for the TMI area, it is consistent-with other conservative assumptions made in this assessment. The concentration of tritium in air is assumed to be zero when the release is to surface water.
Applying these assumptions to the model in Section 5, the effective dose equivalent, for each year of the release, is calculated to be 0.01 USv (1 p rem) and total ef fective dose equivalent is calculated to be 0.02 pSv/y (2 p rem) for complete disposal of the waste water over a two-year period. These estimates are, as before, for a highly hypothetical individual. The contribution to the total effective dose equivalent from the various pathways is as follows: 54% drinking water; 36% food; and 10% milk.
i 1
I l
12 l
I
'f
+;
- 8. Summary of Health Ef fects and Conclusions It is emphasized that due to the very conservative assumptions made to simplify this assessment, the estimated effective dose equivalents are upper bounds and it is highly unlikely that a person exposed to tritium released during the waste water clean up would receive doses approaching these values.
Table 8.1 summarizes the assumptions made to estimate the effective dose equivalents and Table 8.2 summarizes the results for releases to atmosphere and to surface water. The estimated ef fective dose equivalents resulting from each method of. release are likely much higher than anyone would realistically receive, and either pathway would result in radiation doses that are well within acceptable limits.
In view of the low level of ef fective dose equivalent calculated for the maximally exposed hypothetical Individual, detailed calculations of collective dose equivalents are not warranted.
In this Commentary, for the purposes of assessing the health impacts of ingesting and Inhaling tritium, it is assumed that a uniform whole body dose equivalent of I Sv (100 rem) will result in an average lifetige fatal cancer risk plus severe genetic risk totalling approximately 2 x 10-This risk value reflects current estimates of the ICRP (ICRP, 1977), but does not account for potential changes that may result from the re-evaluation of 'he Japanese atomic bomb survivor data. In addition, the quality factor of I for tritium beta radiation used in this Commentary is under review. Ilowever the net effect of both of these reviews is unlikely to result in an increase of the risk values by an order of magnitude.
Applying the risk estimates given above to the effective dose equivalent values given in Table 8.2, we find that the release to atmosphere option will result in a lifetime cancer risk plus severe genetic risk to the most highly exposed hypothetical Individual of approximately I chance in 10 million.
The release to surface water option will result in a lifetime cancer risk j plus severe genetic risk of approximately I chance in 1000 million to the most highly exposed hypothetical individual.
i
^
Since these risks are both below the Negligible Individual Risk Level of f 10-7/y recommended by the NCRP, and below the risk associated with one day of
. natural background, the health and safety of the public will be unaf fected by the release of the treated waste waters from TMI-2 and therefore either option
- is acceptable.
13
., .. 7 Table 8.1 - Assumptions made for calculating tritium releases to atmosphere and surface water and estimating effective dose equivalents for clean up of TMI waste water Release to Atmosphere Release to Surface Water Dispersion calculated using highest Uniformmixinginrivgrwaterwitha 3 g/(1 for point 1.7 km (1.05 mi) east flow rate gf 9.7 x 10 m /s of site (34,000 ft /s) averaged over.a year Concentrations of tritium in air, All food products including aquatic food, and milk are equal foods have same tritium concentra-tion as river water Individual consumes only food All drinking water comes from the grown locally river Concentration of tritium in drinking No inhalation pathway exists water is 1% of that in air Table 8.2 - Summary of the total ef fective dose equivalents resulting from two modes of release 4
Release to Atmosphere Release to Surface Water Total 6.0 p Sv (0.6 mrem) 0.02 pSv (2 prem)
Pathway Food (67%) 4 pSv (0.4 mrem) Food (36%) 0.007 pSv (0.7 prem)
I pSv (0.1 mrem) Milk (10%) 0.002 pSv (0.2 prem)
Milk (18%) 0 pSv ( 0 prem)
Inhalation (13%) 0.8 p Sv (0.08 mrem) Inhalation (0%)
- Drinking Drinking 0.06 pSv (0.006 mrem) Water (54%) 0.011 pSv (1.1 prem)
Water (1%)
14
~
- O Appendix A. Specific Sources cf Waste Water ct Three Mile Isicnd" ,
TOTAL RADIOACTIVITY 3 90 137 134 12 60 Co VOLUME H Sr Cs Cs Sb STORAGE DESCRIPTION GALLONS DATE Ci Ci Ci Ci Ci C1 RCS REACTOR COOLANT SYS1Et 67,286 3/7/86 3.06E+01 4.58E+02 6.62E+01 1.88E+00 9.17E+00 2.39E+00 PWST-1 PROCESSED WATER STORAGE 109,081 2/22/86 1.24E+02 6.61E-03 2.81E-03 PWST-2 PROCESSED WATER STORAGE 480,134 2/24/86 5.09E+02 9.63E-02 8.00E-03 CO-T-1A CONDENSATE STORAGE 101,518 3/3/86 2.15E+01 6.92E-02 1.69E-03 WDL-T-9A EVAP. COND. TEST TANK 5,610 4/12/83 2.76E+00 5.48E-04 2.00E-04 2.00E-05 1.91E-0 5 WDL-T-9B EVAP. COND. TEST TANK 2,231 4/17/83 1.10E+00 7.43E-04 4.22E-05 2.96E-06 CC-T-1 EPICOR II OFF-SPEC 20,500 3/5/86 1. ole +01 4.50E-02 1.40E-02 3.88E-04 3.80E-03 CC-T-2 EPICOR II CLEAN 16,887 11/15/85 5.62E+00 1.92E-02 9.59E-03 7.03E-03 4.15E-04 SFP-B SPENT FUEL POOL ~B" 241,698 3/2/86 4.12E+01 3.02E-02 3.29E-03 SDS-T-1A SDS MONITOR 373 3/7/86 1.07E-01 7.48E-03 1.38E-03 9.32E-04 SDS-T-1B SDS MONITOR 497 10/10/85 1.37E-01 1.82E-03 1.88E-03 1.77E-03 1.30E-04 UDL-T-1A RC BLEED HOLDUP 3,810 2/24/86 1.23E+00 4.76E-01 1.34E-01 3.89E-04 6.63E-03 1.30E-03 WDL-T-1B RC BLEED HOLDUP 4,420 3/7/86 2.17E+00 2.84E+01 3.35E+00 9.37E-02 5.69E-01 1.10E-01 WDL-T-lC RC BLEED HOLDUP 57,116 10/31/85 3.68E+01 5.40E+02 3.68E+01 1.75E+01 1.41E+00 BWST BORATED WATER STORAGE 458,915 3/4/86 1.15E+02 6.60E-01 2.26E-01 1.55E-02 4.69E-03 WDL-T-8A NEUTRALIZER 8,675 2/28/86 2.96E+00 4.60E+00 6.24E+00 1.74E-01 WDL-T-8B NEUTRALIZER 8,605 3/1/86 2.21E+00 2.51E+00 5.86E+00 1.73E-01 WDL-T-2 MISCELLANEOUS WASTE HOLDUP 3,712 2/28/86 9.69E-01 1.10E+00 2.39E+00 6.88E-02 WDL-T-IIA CONTAMINATED DRAINS 1,931 3/1/86 1.53E-04 1.97E-04 3.07E-04 6.80E-06 WDL-T-IIB CONTAMINATED DRAINS 820 3/1/86 4.35E-05 3.41E-05 CHEM CLEANING BLDG SUMP 1,680 3/2/86 2.86E-01 6.99E-03 5.60E-03 2.16E-02 5.02E-05 AUXILIARY BLDG SUMP 5,917 10/4/85 2.91E+00 2.69E+00 5.15E-01 REACTOR BLDG BASEMENT 43,082 4/26/85 4.24E+00 2.61E+02 7.99E+02 SFP-A SPENT FUEL POOL "A" 205,234 2/27/86 2.02E+02 2.49E+01 6.84E+00 1.86E-01 2.02E+00 1.94E-01 DEEP END OF TRANSFER CANAL 58,685 3/12/86 6.66E+01 5.78E+00 1.91E+00 4.89E-02 6.44E-01 5.55E-02 TOTAL AS OF 1/1/86 1,908,417 1182.75 1331.17 929.49 8 This table was taken directly from the GPU Report (GPU, 1986) 15
- /
Appendix B - Decontamination Factors Required to Ensure that the Dose from Radionuclides other than Tritium are Relatively Insignificant The source term data shown in Appendix A were analyzed to. determine the decontamination factor necessary to ensure that none of the five radionuclides listed in the inventory would contribute more than 1% of the radiation dose to members of the public from tritium during the release of the waste water.
B.1 Evaluation of a Process Releasing All Radionuclides to the Atm'osphere For release to the atmosphere, it is assumed that no effluent would be released to water and that the radionuclides are released during processing over a period of one year. The calculation was made using the AIRDOS-EPA (Moore et al. 1979) computer code.
It was also assumed that the radionuclides would be released at a constant rate during the processing of the waste water. Assumptions of the type given in NCRP Report No. 76 (NCRP, 1984) were used for atmospheric dispersion and enviroamental transport and for the potential pathways of significance. Dose estimates were made for each radionuclide and then normalized to that for tritium. All of the values are rounded to one significant digit. The results are given in Table B-1.
Table B The significance of doses from five radionuclides relative to tritium for releases to atmosphere Radionuclide Ratio of Doses to that from Tritium 3
3 i 60,o C 20,000 90 30,000 fr 12aSb 3,000 134 Cs 30,000 137 Cs 20,000 Although the uncertainty is probably less than a factor of ten, for the purpose of applying these data to the situation at TMI, an arbitrarily chosen criterion that doses from other radionuclides should not give rise to radiation doses greater than 1% of that from tritium was employed.
90 Sr is 30,000 The data in Table B-1 indicate that if the gurce term for times less than that for tritium, the dose from Sr would be no more than that fromtritiumassumingthereleasesofeachareoccurgngsimultaneously and are to the atmosphere. Thus, if the source term for Sr is 3,000,000 times less than that for tritium, the dose will be no more than 1% of that from tritium.
3 Since the source term for gritium is 4-g 8 GBq/m33 (1.3 x10-I pCi/cm ), a 90Sr concentration of 1.6 KBq/m (4.3 x 10 pC1/cm ) will result in a dose of 1% of that from tritium. Using100%pgcessvaluesfromTable2-5ofjheJuly 1986 CPU Report 1986) we find a Sr concentration of 0.37 MBq/m (1 x 10-5 p C1/cm{ ). CPU, This value, taken together with an evapgator decontami-nation factor of I for tritium and between 100 and 1000 for Sr results in a si l 16 l
.h
, . . ._ .- _ _. ._. . . __.m . _
! .g id' .f 3 I 3 release and concegtration 3.7 KBq/m (1 x 10- tpat pC1/cmwoulg)be 4.8 GBq/m to 0.37 KBq/m gl.3x108 (1 x 10- -pC1/cm-) pCi/cm )for for 651'i""
Sr.
If it is- possible to attain a decontamination factor of between 1000 and 10,000 the-40% process value as taken from Table 2.1 would meet these criteria.
B.2 Evaluation of a Process Releasing All Radionuclides to Surface Water For releases to surface water, it is assumed that tritium and other s '- radionuclides are released to water and that no effluent enters the, atmosphere directly during processing. Both drinking water and consumption of aquatic 4 foods are included in the analysis. The results are given in Table B-2.
1 Table B The significance of doses from five radionuclides I
relative to tritium for releases to surface water Radionuclide Ratio of Doses to that from Tritium I-3 I H
60 Co 1,100 l 90 2,000 12$srSb 100 l '
- 134 Cs 2,200 137 1,500
, Cs i
f Again, for the purpose of applying these data to the situation at TM1, a (getoroftwoordersofmagnitude(100)hasbeenused.
Sr and include this factor of 100, then a source term for Ifwh0"rwhichisa S
factor of 200,000 less than the source term for tritium would give rise to a radiation dose of 1% of that from tritium.
3 Since the source term for tritium is 4.8 GBq/m3 (1.3 x10-I pgi/cm),the j 90Sr concentration must be less than 24 KBq/m3 (6.5 x 10-7 pCi/cm ) for the dose
- to be lower by a factor of 100 from that due to tritium. Usingthe100%ggrocess Sr j values from Table 2-5 of thg July 198g GPU Regort (GPU,1986) we find a
- concentration of 0.37 MBq/m (1 x 10- pC1/cm ). Processingofthewas55 through the evaporator would ensure that the effluent concentration of Sr would be 200,000 times less than the tritium concentration.
~
i 4
1 r
i
['
17
{
- g .* /
REFERENCES Alvarez, L. W., and Cornog, R. (1939). " Helium and Hydrogen of Mass 3," Phys. Rev.
56, No. 2, 613.
Blomeke, J. O. (1964). " Management of Fission Product Tritium in. Fuel Processing Wastes," USAEC Report ORNL-TM-951 (National Techaical Information Service, Spriagfield , Virginia).
Evans, A.G. (1969). "New Dose Estimates from Chronic Exposure to Tritium," Health Phys. 16, 57.
GPU (1986). General Public Utility let ter (with at tached report, Disposal of TMI Water), No. 44-10-86-L-Oll4, F. R. Standerfer (GPU Nuclear) to W. D. Travers (NRC) dated July 31, 1986. NRC Docket No. 50-320 (U.S. Nuclear Regulatory Commission, Washington).
IAEA (1984). International Atomic Energy Agency. Management of Tritium at Nuclear Facilities, Technical Reports Series No. 24 (International Atomic Energy Agency, Vienna).
ICRP (1975). International Commission on Radiological Protection. Report of the Task Group on Reference Man, ICRP Publication 23 (Pergamon Press, New York).
ICRP (1977). International Commission on Radiological Protection. Recommendations of the International Commission on Radiological Protection, ICRP Pub 11ation 26 (Pergamon Press, New York)
K111ough, G. G. (1982). " Derivation of Dose Conversion Factors for Tritium,"
NUREC/CR-2523, ORNL-5853 (National Technical Information Service, Springfield, Virginia).
Moore , R. E. , Baes , C. F., III, McDowell-Boyer, L. M., Watson, A. P., Hoffman, F.
O., Pleasant, J. C. , and Miller, C. W. (1979). "AIRDOS-EPA: A Computerized Methodology for Estimating Environmental Concentrations and Dose to Ibn from Airborne Radionuclides," ORNL-5532 (National Technical Information Service, Springfield , Virginia).
NCRP (1979). National Council on Radiation Protection and Measurements. Tritium in the Environment, NCRP Report No. 62 (National Council on Radiation Protection and Measurements , Bethesda , Maryland).
NCRP (1984). National Council on Radiation Protection and Measurements, Radiological Assessment: Predicting the Transport, Bioaccumulation, and Uptake by Man of Radionuclides Released to the Environment, NCRP Report No. 76 (National Council on Radiation Protection and Measurements, Bethesda, Maryland).
Osborne, R.V. (1966). "Absorpt ion of Trit tated Water Vapour by People," Health Phys.
12, 1527.
18 I
.g . 2 Pinson, E. A. , and Langham, W.H. (1957). " Physiology and Toxicology of Tritium in Man," J. Appl. Physiol. 10:108. Reprinted: Health Phys. 38, 1087.
USNRC (1981). U. S. Nuclear Regulatory Commission, Final Programatic Environmental Impact Statement Related to the Decontamination and Disposal of Radioactive Wastes Resulting from tiarch 28, 1979, Accident, Three Mile Island Nuclear Station, Unit 2, NUREG-0683, Vol. 2, Appendix W (National Technical Information Service, Springfield, Virginia).
19
+ -_
oy- 9 4
P.O. Box 983 Pierre, South Dakota 57501 February 21, 1987 Dr. Michael T. Masnik TMI Project Directorate ;
Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Re: Comments on Draft Supplement #2, NUREG-0683, Programmatic Environmental Impact Statement related to Decontamination and nisposal of Radiorctive Wastec reculting from March 28, 1979 Accident, Three Mile Island Nuclear Station, Unit 2, Docket
- 50-320 (December 1986)
Dear Dr. Masnik:
I have reviewed this draft supplement and wish to make the following comments.
In the Summary on page v, I note that the disposal volume of accident-generated water was " expected to be 40,000 to 80,000 cubic feet (11,000 to 13,000 cubic meters)". I believe that cubic feet have been converted here into square meters and not cubic meters.
In the second paragraph of the Summary, I see that the final proces-sing will involve about 2.1 million gallons, or 7.9 million liters with about 1,000 curies of tritium and smaller amounts of cesium 137 and strontium 90. There is no niention here of uranium, plutonium or other transuranics,nor of other of the 500 different radionuc-S. lides of potential importance in the assessment of contamination (sjpO around nuclear facilities. This is a very serious oversight. I
,gg believe that the concentration of all of these should be determined.
M8' O@ The Summary estimates that the considered disposal alternatives will have an impact of only 0 to .003 radiation-induced cancer deaths in dx
@g the worker population and only 0 to 0.0003 for radiation-induced
@ cancer fatalities in the offsite population. If this water is really og that innocuous, should the plant save it to be used in drinking water fountains for the employees at the plant? Or should it be carbonated, h@a bottled, and sold in stores as spring water? 3-
2.
As I recall, the reactor core in this plant was partially melted down and this water has been in and around the 100 tons of partially melted uranium (with plutonium and other activation and fission products) for nearly seven years. Many of these metals and compounds are quite water-soluble, especially uranium. The Schwariwalder Uranium Mine, for example, near Golden, Colorado, at times pumps out more than a million gallons of water each day, and in the past (perhaps today also) this has been discharged-into public water sup-plies. The water at times contains more than 10,000 picocuries per liter of alpha radiation from the urarium. The contact thore between water and uranium ore has been at rather cool temperatures, not in a super-heated environment such as has occurred at TMI-2. I can't believe that there is not a large amount of uranium and its progeny and other transuranics dissolved in this water in TMI-2. And yet, in reading this report I didn't see any mention of alpha radiation levels per liter of water nor of the concentration of uranium and other transuranics in the water.
In the manuscript there was a discussion of background radiation levels in surface waters downstream and these discuss the levels of alpha radiation and radium in the water amounting to several picocu-ries per liter. The lack ot intormation in this dratt report on tne concentration ot uranium and transuranics in the waste water is very puzzling.
Tne range and concentrations or radionuclides in tne water snould be determined by number 01 agencies and independent laboratories, and the radiation protection guides snould be those developed by tne EPA or by moretonservative inoependent researchers. For example, the EPA nas advisea a .?mit or 10 picocuries per 11ter or uranium in water,in contrast to a limit of 6000 supported by the Dept. of Energy.
I rurtner, the units in tne DooK snould De Consistent with present EPA practice. Atter all, tnis is an e::vironmental impact statement.
Nadiation activities snould ne expresseo in terms or picocuries per alter or water ano picocuries per cuole meter or air. Tne use ot'
3.
awkward units like microcuries per milliliter and the use of large negative exponents should be avoided,since these are c.onfusing even to experts and especially confusing to the public.
In several places the text reads as if the tritium in th'e water is there as the gas. In fact, tritium (which is hydrogen) oxidizes with oxygen and ozone over time to form tritiated water or heavy water.
The evaporation process will simply evaporate off all the tritium as tritiated vapor which is much more toxic on inhalation or ingestion than is tritiun. gas.
I think that we do not have enough information to make a decision about thedispositionofdhiswater. I recommend against any of the methods of disposal at this time, until there has been exhaustive analyses of the water by a number of agencies and independent laboratories at universities, including one or two in Canada. The water should be analyzed alsc, for example, by the EPA and by the U.S. Geological Survey, which does get involved in what happens to water in the environment. I have attached a figure from an EPA report en liquid emissions from a nuclear power plant in normal operation to show the range of radionuclides released in such normal operations.
I think we need to know more about the assumptionshade in cal-culatinq doses to persons around the plant from the radionuclides which mi.ght be relenced by the various altarnativen pronosed. Those dose estimates should also include exposure to every one of the 500 radionuclides of potential importance in this water, and should also 1
i consider concentrations of radionuclides by marine plants and animals in the food chain.
Sincerely,
/
/. i d-Carl J. Johnson, M.D., M.P.H.
1
1 1 ~ . . _ _.
. . . . . . . . . . . . , . . . . - . . .. . . . . . . ~ . . . . . . . . . . . . - . . . . . . . . . . . . . . - - . . . . . . . . . -. .. . . . . . . . . . . - . . . . - . . . . . . . . . . . . . . . . . . ~ . . . -
M ,a6 * ..
. '- $b.
- * *
- g,
...- 1 g
- Til w &
y- I i
hC *2O t
'Il IE !
- t- c e *{3 bd 1f!
.~ '! ! '
4$ }'
i i ,1 t p d j
gi g
y' ,!
ife,I
.. ,!*} 3,
['g
~
n g I
'\, f & 4'
.$$$l$4 9 %
e
> =( 4"ikk
,8
<. 1 .
TII i E i ti, :
/i i
i [ ,,
r8 i il 'I(*!i
< t l E g =
i '
h d.[,,a U
%? f i !,!
st ,
1 s!
I-
}'! I. .
i 11 % v.y' r l'!'r{q 11'\ ' I) '\
I .... g't 1
Jh )
flh% .lli m See ##
. T y* g g ' ,u .
g I ITi ggii 3T i
"" 1 '. 5 111 I
lI i (l! I t
' i 3 t ! 'i i li
' i I Ii 6
i gt i l i fi l j '
... \M, g$ t T / ,'! 1 -! -
M.I371. li , i,1 ,$ 4.' i '
t {'
I D hJdJ[ . 4 i
g j< ,.
{
11 1 il i-
,,. rt t,i l !!
',i1.I 1 , . ',
r
,i li't 111 4 * .
a *O*
..e
. _ . ._. . . _ - - . _ _ _ . _ _ _ . - _ _ _ _ _ _ , _ - - _ _ . _ _ . _ _ _ _ _ . ~ _ _ _ . . . _ _ . . . - _ _ _ _ _ _ _ _ _ . _ . _ . . _ . _ . - _ . _ _ _ .
- - _