P/EPA
United States
Environmental Protection
Agency
Office of Health and Ecological
Effects
Washington DC 20460
EPA-600/5-78-013
June 1978
Research and Development
Scoping Assessment
of the Environmental
Health Risk
Associated with
Accidents in the
LWR Supporting
Fuel Cycle
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U S Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields
The nine series are,
1 Environmental Health Effects Research
2 Environmental Protection Technology
3 Ecological Research
4 Environmental Monitoring
5 Socioeconomic Environmental Studies
6 Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8 ' Special" Reports
9 Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL HEALTH EFFECTS RE-
SEARCH series This series describes projects and studies relating to the toler-
ances of man for unhealthful substances or conditions This work is generally
assessed from a medical viewpoint, including physiological or psychological
studies In addition to toxicology and other medical specialities, study areas in-
clude biomedical instrumentation and health research techniques utilizing ani-
mals — but always with intended application to human health measures
This document is available to tne public througn the National Technical Informa-
tion Service, Springfield Virginia 22161
-------
EPA-600/5-78-013
February 1977
SCOPING ASSESSMENT OF THE
ENVIRONMENTAL HEALTH RISK
ASSOCIATED WITH ACCIDENTS IN
THE LWR SUPPORTING FUEL CYCLE
S.C. Cohen
K.D. Dance
EPA Contract No. 69-01-2237
Project Officers
Alan P. Carl in
Paul H. Gerhardt
Office of Health and Ecological Effects
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C. 20460
Ti-" — ' -' Projection
-------
DISCLAIMER
This report has been reviewed by the Office of Research and Development
U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views
and policies of the Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or recommendation
for use.
-------
ACKNOWLEDGMENTS
Teknekron, Inc. wishes to acknowledge the Office of Research and
Devleopment, U.S. Environmental Protection Agency, for support of this
study. In particular, we are grateful to Mr. Paul H. Gerhardt, the Project
Officer, for his guidance in the performance of this research.
Several members of the Office of Radiation Programs offered helpful suggestions
and comments. Special thanks are extended to Messrs. Stephen Goldberg and
James Hardin for reviewing the draft report.
We are also grateful to Professor George Yadigaroglu of the University
of California for reviewing the manuscript, and to Messrs. Ray Scott
and Wayne Smalley of the Oak Ridge National Laboratory for supplying
r*
"^ pertinent data.
m
-------
ABSTRACT
A framework is developed for the assessment of the health risk associated
with postulated accidents in the fuel cycle supporting the annual opera-
tion of a 1000 Mwe light water-cooled reactor. The methodology applied
consists of a synthesis of nominal radiological source terms and corres-
ponding likelihoods for postulated accidents at generic fuel cycle facili-
ties (with the exception of the reactor and waste management) considered
to comprise the nuclear industry toward the end of the current decade.
The accident definitions, source terms, and likelihoods are compiled
from a number of diverse studies and subjected to interpretation, renor-
malization, and, in some cases, revision. Risk is defined in this study
as the product of the expected consequences of an accident and its likeli-
hood of occurrence. The source terms are converted to 50-year dose com-
mitments following, in general, the recommendations of the International
Commission on Radiological Protection, and considering inhalation,dietary,
plume submersion and ground shine pathways to man for generic representa-
tions of population density. Dose commitments are converted to health
effects using the linear, non-threshold dose-response relationship de-
veloped in the BEIR report (Report of the Advisory Committee on the Bio-
logical Effects of Ionizing Radiation) and interpreted by EPA.
An initial quantitative estimate of the aggregated somatic health risk
from accidents in the supporting fuel cycle, normalized to the annual
o
operation of a 1000 Mwe LWR, is of the order of roughly 10 excess can-
cers. This estimate is subject to uncertainties associated with the
dose conversion model, the dose-response relationship, and the conse-
quence expectation value estimate. The results, however, are broken
iv
-------
down by accident category in each component of the fuel cycle, so that
the framework thus established can be expanded or revised as more data
or the results of more refined analyses become available.
Transportation accidents dominate the total accident risk, whereas the
risks from accidents in mining, milling, and plutonium storage are rela-
tively insignificant. The risks from the remaining components of the
fuel cycle, with the exception of spent fuel reprocessing, are roughly
of equal orders of magnitude, at least at the upper limits of the range
of estimates. Comparisons with the risks from normal operations of
the supporting fuel cycle and with occupational risks indicate that,
on the basis of the annual operation of a 1000 Mwe LWR, the risk from
accidents in the supporting fuel cycle is orders of magnitude lower.
On the same basis, the risk from accidents in the supporting fuel cycle
is also slightly lower than that from reactor accidents and comparable
to that from normal reactor operation.
A more comprehensive scoping analysis would include the risks associated
with site-induced and other high consequence, low probability ("class 9")
accidents, which were, in general, neglected here. For example, the like-
lihoods and consequences of accidents at the reprocessing facility invol-
ving the interim storage of high level wastes should be evaluated. Nor
have the risks associated with proposed and postulated waste management
alternatives been addressed, although this area is the subject of an ex-
tensive effort currently sponsored by the Energy Research and Development
Administration. Moreover, the accidental releases of chemicals at fuel
cycle facilities have the potential of producing environmental health
-------
effects. Finally, more confidence could be placed in these results if
a more detailed assessment of HEPA filter failure probabilities were
available, and if accident scenarios associated with spent fuel repro-
cessing and transportation were examined in considerably more detail.
-------
TABLE OF CONTENTS
1. Introduction 1
1.1 Background 1
1.2 Objectives and Scope 3
1.3 Methodology 4
1.4 Limitations 6
1.5 Organization 9
2. The LWR Fuel Cycle 10
/
2.1 Description of the Fuel Cycle 10
2.2 Normalization to the Annual Fuel Requirement of
a Generic LWR 20
3. Generic Models for Demography, Dispersion, Dose
Conversion, and Health Effects 26
3.1 Demography 2o
3.2 Dispersion and Dose Conversion ^°
3.3 Health Effects 37
4. Source Terms and Likelihoods 41
4.1 Mining 50
4.2 Milling 51
4.3 UFC Conversion 63
o
4.4 Enrichment 77
4.5 Uranium Fuel Fabrication 39
Vll
-------
4.6 Reprocessing 1°8
4.7 Mixed Oxide Fuel Fabrication 144
4.8 Plutonium Storage 169
4.9 Transportation 174
5. Risk Assessment 197
5.1 Milling 198
5.2 UF6 Conversion 200
5.3 Enrichment 202
5.4 Uranium Fuel Fabrication 204
5.5 Reprocessing 206
5.6 Mixed Oxide Fuel Fabrication 211
5.7 Plutonium Storage 214.
5.8 Transportation 21^
5.9 Overall Fuel Cycle Risks and Comparisons 218
6. Conclusions and Recommendations for Future Work 226
2?9
7. References
Appendix A Dose Commitment Factors 233
Appendix B Source Terms from Normal Operations
-------
List of Tables
Table: Page
2-1 Capabilities of the Generic Facilities Comprising ,?
the LWR Supporting Fuel Cycle
2-2 Transportation Data for the Fuel Cycle 19
2-3 Fuel Inventories for Generic LWR 21
2-4 Annual Mass Flows in the LWR Supporting Fuel Cycle 22
2-5 Number of 1000 MWe LWR's Serviced Annually by Each
Generic Facility in the LWR Supporting Fuel Cycle 23
2-6 Number of Shipments and Total Travel Miles in Each
Fuel Cycle Transportation Step Normalized to the 25
Annual Operation of a 1000 MWe LWR
3-1 Population Densities in the Vicinity of the Generic 27
Fuel Cycle Facilities
3-2 Average Distances to Unrestricted Area for Generic
Fuel Cycle Facilities 28
3-3 Radiation Dose to an Indiv-idual as a Function of
Distance Resulting from the Direct "Shine" of a 36
Criticality Incident (10lb Fissions)
3-4 Effects of a Single Exposure of afiPopulation of One
Million to One Rem Per Person (10 Man-Rem) 38
4-1 Categories of Accidents 42
4-2 Explosion Incidents in Fuel Cycle Facilities 44
4-3 Fire Incidents in Fuel Cycle Facilities 45
4-4 Loss of Containment Incidents in Fuel Cycle Facilities 46
4-5 Filter Failure Incidents in Fuel Gycle Facilities 49
4-6 Milling Accidents 52
4-7 Uranium Mills in Operation as of March, 1975 53
4-8 Uranium Mills No Longer in Operation 55
4-9 Summary of Accidental Tailings Slurry Releases 59
ix
-------
List of Tables
(continued)
Table: Page
4-10 Concentrations of Radioactive Efflupnts in Waste
Liquor from the Model Uranium Mill 60
4-11 UFg Conversion Accidents 64
4-12 UFg Releases (> 5Kg) Associated with Filling or Feeding
Cylinders at the Gaseous Diffusion Plants in the Period ''
1969-1973
4-13 Radionuclide Concentrations in UFfi Conversion Waste
Retention Pond ° 76
4-14 Enrichment Plant Accidents 78
4-15 Existing Enrichment Plant Data 79
4-16 Isotopic Activities of the Enrichment Plant Feed 81
4-17 Specific Activities by Isotope of Product and Tails
Cylinders at the Enrichment Plant 84
4-18 Radionuclide Release Resulting From a Criticality
Incident at the Enrichment Plant 87
4-19 Uranium Fuel Fabrication Accidents 90
4-20 Estimate of Plant-Years of Production Since 1942
Involving Fuel Fabrication 91
4-21 Radionuclide Release Resulting From a Criticality
Incident at the Uranium Fuel Fabrication Facility 103
4-22 Estimated Concentrations of Radionuclides in Waste
Retention Pond 106
4-23 Spent Fuel Reprocessing Accidents 109
4-24 Radionuclide Release Resulting From an HAW Concentrator
Explosion at the Reprocessing Facility H2
4-25 Radionuclide Release Resulting From a LAW Concentrator
Explosion at the Reprocessing Facility 115
4-26 Radionuclide Release Resulting From a HAF Tank Explosion
at the Reprocessing Facility 118
4-27 Radionuclide Release Resulting From an Explosion in the
Waste Calciner at the Reprocessing Facility 120
-------
List of Tables
(continued)
Table: Page
4-28 Radionuclide Release Resulting From a Fire in the
Codecontamination Cycle at the Reprocessing Facility 124
4-29 Radionuclide Release Resulting From a Fire in the
Plutonium Extraction Cycle at the Reprocessing Facility 127
4-30 Radionuclide Release Resulting From an Ion-Exchange
Resin Fire at the Reprocessing Facility 129
4-31 Radionuclide Release Resulting From a Fuel Assembly
Rupture and Release in Fuel Receiving and Storage 132
Area at the Reprocessing Facility
4-32 Radionuclide Release Resulting From a Dissolver Seal
Failure at the Reprocessing Facility 135
4-33 Isotopic Activities of Uranium with 33,000 MWD/MT
Burnup 139
4-34 Radionuclide Release Resulting From a Criticality
Incident at the Fuel Reprocessing Facility 141
4-35 Mixed Oxide Fuel Fabrication Accidents 145
4-36 Radionuclide Release Resulting From an Explosion in the
Oxidation-Reduction Scrap Furnace at the Mixed Oxide
Fabrication Plant 148
4-37 Radionuclide Release Resulting From a Major Fire at the
Mixed Oxide Fabrication Plant 151
4-38 Radionuclide Release Resulting From a Fire in the Waste
Compaction Glove Box at the Mixed Oxide Fabrication Plant I53
4-39 Radionuclide Release Resulting From a Fire in an Ion
Exchange Resin Column at the Mixed Oxide Fabrication 156
Plant
4-40 Radionuclide Release Resulting From a Dissolver Fire in
Scrap Recovery at the Mixed Oxide Fabrication Plant 159
4-41 Radionuclide Release Resulting From a Glove Failure at
the Mixed Oxide Fabrication Plant 161
4-42 Radionuclide Release Resulting From Severe Glove Box
Damage at the Mixed Oxide Fabrication Plant !64
xi
-------
List of Tables
(continued)
Table: Page
4-43 Radionuclide Release Resulting From a Criticality
Incident at the Mixed Oxide Fuel Fabrication Facility 166
4-44 Radionuclide Release Resulting From a Criticality
Incident at the Plutonium Storage Facility 171
4-45 Transportation Accidents 175
4-46 Radionuclide Release Resulting From an Improperly
Closed Plutonium Oxide Container 179
4-47 Accident Probabilities for Truck, Rail, and Barge
Accidents of Various Severities 182
4-48 Radionuclide Release Resulting From an Extra Severe
Collision Involving Irradiated Fuel 184
4-49 Radionuclide Release Resulting From a Collision
Involving Plutonium Oxide 188
4-50 Radionuclide Release Resulting From a Criticality
Incident Involving Enriched UOp in Transport I93
4-51 Radionuclide Release Resulting From a Criticality
Incident Involving Pu02 in Transport 196
5-1 Environmental Risks From Accidents in Uranium Milling 1"
5-2 Environmental Risks From Accidents in Uranium
Hexaflouride Conversion
5-3 Environmental Risks From Accidents in Enrichment ^03
5-4 Environmental Risks From Accidents in Uranium Fabrication 205
5-5 Environmental Risks From Accidents in Fuel Reprocessing 207
5-6 Environmental Risks From Accidents in Mixed Oxide Fuel 212
Fabrication
5-7 Environmental Risks From Accidents in Plutonium Storage 215
5-8 Environmental Risks From Accidents in Transportation 217
5-9 Total Environmental Health Risks From Accidents in the
LWR Supporting Fuel Cycle 219
5-10 Comparison Between Environmental Health Risks From Accidents
and From Normal Operations of the LWR Fuel Cycle 222
xii
-------
1. INTRODUCTION
1.1 Background
1
Since the advent of the National Environmental Policy Act of 1970 (NEPA),
the perspective in the assessment of environmental health effects associ-
ated with nuclear power production has shifted significantly. Historically,
the radiological releases from normal operations of nuclear reactors were
evaluated against maximum individual dose limitations at the site boundary.
Accidental releases were evaluated using conservative assumptions regarding
*
release magnitudes and environmental dispersion. No serious attempts
were made to estimate the likelihood of accidental releases. Moreover,
the evaluation of the environmental health effects associated with the
nuclear fuel cycle in support of reactor operation received little atten-
tion, either during normal operations or in the accident mode.
The mandate of NEPA resulted in a number of studies which assess the environ-
mental health effects associated with nuclear power generation on a realistic
basis. These assessments are now performed routinely in the form of De-
tailed Environmental Impact Statements as part of the licensing process
associated with the construction and operation of each reactor or indi-
vidual fuel cycle facility. Moreover, a number of generic studies have been
performed in support of impending legislation or regulatory rule-makings.
For example, the environmental statement in support of Appendix I to 10 CFR
4
50, which established numerical guides to meet the "as low as practicable"
2
* Examples of studies utilizing this approach are WASH-740 for nuclear
reactors and ORNL-3441J for fuel fabrication .and processing plants.
-------
criterion,evaluated population doses and potential health effects resulting from
the normal operation of reactors assuming various levels of control tech-
nology. For the supporting nuclear fuel cycle, the GESMO (Generic Envi-
5
ronmental Statement on Mixed Oxide Fuel) Report evaluated the magnitude
of effluents, population doses, and potential health effects for each
component of the fuel cycle both with and without plutonium recycle. The
consequences and likelihoods of major reactor accidents were predicted on
6
a realistic basis in the recent Rasmussen Report.
Potential accidents in the supporting nuclear fuel cycle have received only
cursory evaluation. A survey of fuel cycle accidents and their potential
consequences was incorporated in the Environmental Survey of the Uranium
7
Fuel Cycle, but this study did not attempt an evaluation of accident
likelihoods or health risks. Much of the accident information incorpor-
ated in this study was based upon data contained within detailed environ-
mental reports and statements for individual fuel cycle facilities, each
of which contains an accident assessment based upon realistic assumptions
Accident likelihoods have been assessed on a generic basis for the mixed
8 9 10
oxide fuel fabrication, spent fuel reprocessing, and transportation
components of the fuel cycle. Different methodologies were used in these
assessments, and the results are not readily translatable to health risks.
None of the previous accident evaluations-has normalized the predicted
consequences to the operation of the reactor supported by the fuel cycle.
-------
1.2 Objectives and Scope
The objective of the current study is to provide a framework for the risk
evaluation of postulated accidents in the nuclear fuel cycle, and to
establish a preliminary quantitative estimate of this risk. The esti-
mated risk will be normalized to the annual operation of a generic re-
actor so that the impact of potential fuel cycle accidents can be in-
corporated in the overall cost/benefit balance for nuclear power gen-
eration. The results will be broken down by accident category in each
component of the fuel cycle, so that the framework thus established can
be expanded or revised as more data or the results of more refined anal-
yses become available. The expanded format will also be useful in iden-
tifying relatively high risk operations in the fuel cycle. Finally, the
overall risk from accidental releases will be compared with the risk
from normal operation of the reactor and the supporting fuel cycle.
The analysis will be confined to the light-water-cooled reactor (LWR) fuel
cycle incorporating the recycle of both uranium and plutonium. All compon-
ents of the conventional LWR fuel cycle will be included with the exception
of waste management, which is omitted because of uncertainties in process
definition. Although accidents in the fuel cycle may involve the release of
toxic chemicals to the environment, only the effects of radiation on human
health (somatic effects) will be considered. Similarly, occupational risks
and the risks to the public from deliberate acts of sabotage or diversion
of nuclear materials are considered outside of the scope of the current
study.
* The most efficacious way to express the social benefits of nuclear
fuel cycle activities in support of reactor operation is in terms of
the electricity delivered by the reactor.
-------
1.3 Methodology
Risk is defined in this study as the product of the expected consequences
of an accident and its likelihood of occurrence. The accident risk in a
component of the fuel cycle is obtained by summing the risks from all
postulated accidents for a generic facility representing that component.
The risk from each component is normalized to the annual operation of a
1000 MWe generating station using mass flow data derived for a generic
LWR. The risk from accidents in the entire fuel cycle is then obtained
by summing the normalized risks from the individual components of the
fuel cycle.
The heart of this study lies in the definition of postulated accidents,
source terms (compositions and magnitudes of radiological releases to
the environment) associated with these accidents, and likelihoods of
their occurrence. In most cases, this information has been synthesized
from existing studies of considerably less ambitious scope. Individual
items of data, however, have been subjected to considerable manipulation,
interpretation, and renormalization. Accident likelihoods, in particu-
lar, are based upon a diversity of original methodologies, including
actual incidents on record (lacking, unfortunately, in comprehensiveness),
analogous chemical industry statistics, and fault tree analysis. Ranges
are Quoted in a number of instances, and judgment enters in nearly all
instances. The lack of rigor is hopefully compensated by a thorough
documentation of original sources.
The source terms are converted to doses generally following the recom-
mendations of the International Commission on Radiological Protection
-------
(ICRP). Inhalation, dietary, plume submersion, and ground shine path-
ways are considered in deriving 50 year dose commitments. Then, assum-
/
ing uniform population distributions surrounding each generic fuel cycle
facility, a deposition model is used to determine the total population
dose, in man-rem, to critical organs of humans in the environment. Simi-
larly, a generic river model is defined to estimate the dilution of liq-
uid sources in the watercourse. Population dose from liquid sources is
11
then estimated from the drinking water pathway for the generic river.
The population dose expresses the integral of the radiation
12
doses received by individuals over the entire population exposed. Coup-
/
led with the linear non-threshold hypothesis, this approach permits the
direct conversion of population dose to health effects. According to
this hypothesis, there exists a linear relationship between the total
accumulated dose and the number of health effects from zero exposure
to the highest exposure which does not cause acute mortality. This lin-
12
ear relationship is defined from estimates of somatic effects origin-
13
ally obtained from the BEIR Report.
-------
1.4 Limitations
The rather simple concept of risk employed in this study is not intended
to minimize the philosophical difficulties in defining risk or the com-
14
plexities of applying the concept to decision-making. The linear non-
threshold hypothesis for converting population dose to health effects is
15
not universally accepted, but has been generally adopted as a conserv-
12
ative approach for the establishment of radiation standards. Moreover,
the application of this linear model permits the normalization of risk
to the annual operation of a generic LWR and generally simplifies the
analysis.
As discussed in the previous section, the weakest link in the accident an-
alysis lies in the establishment of source terms and likelihoods for the
accidents postulated in each component of the fuel cycle. First of all,
the comprehensiveness of the list of accidents postulated cannot be as-
sured. Indeed, it will become apparent in Section 4 that this scoping
assessment is not altogether complete. For example, the consequences
of natural disasters and other "class 9" accidents have not, in general,
been examined here. An analysis of the nature attempted here is a con-
tinuing exercise, thus explaining the early identification of this study
within the context of a "framework."
The use of incidents on record to establish an accident data base is de-
ficient in at least two respects. The existing data base is not compre-
hensive and does not, in general, contain estimates of environmental
releases. Moreover, processes and safeguards have been improved in a
-------
number of specific cases, so that the historical record is not necessar-
ily indicative of current or future events.
The use of analogous chemical industry statistics may be misleading be-
cause of differences in process constraints and the generally higher con-
cern for safeguards in handling nuclear materials. The fault tree ap-
proach depends upon the ability to conceptualize all potential release
events and suffers from the unavailability of a complete set of quanti-
tative failure data.
In reality, a complete spectrum of source terms and corresponding prob-
abilities is associated with each potential accident class. The choice
in this assessment of "nominaT'source term and a single likelihood
associated with this release is a simplification. However, the nominal
source terms and associated likelihoods have been chosen to be as repre-
sentative of the actual continuum of sources as possible. The selections
of source terms are based upon past analyses or actual data representative
of process variables. In some cases, though, past experience is not com-
pletely representative of the processes and plant capacities selected for
this study. Also, in some cases, it is not possible to tightly couple the
accident likelihoods with the source terms.
Despite the foregoing limitations, it is felt that the synthesis pre-
sented here is a necessary first step in consolidating and placing in
perspective previous estimates of the impact of accidents in the fuel
-------
cycle. The nature of the study and the many assumptions employed render
the quantitative estimates of risk highly tentative. A framework is
developed, however, which can be used to update these estimates as more
complete and accurate data become available.
-------
1.5 Organization
Section 2 of this report provides a brief description of the LWR support-
ing fuel cycle conceptualized for this study, including the factors used
in the normalization of the results to the annual operation of a generic
1000 MWe LWR. Section 3 describes the models used for the dispersion
of radioactive sources, population distributions, dose conversions, and
health effects.
The heart of the study is contained in Section 4, which provides esti-
mates of the source terms and corresponding likelihoods, including docu-
mented rationale, for the accidents postulated in each component of the
fuel cycle. The predicted population doses and normalized health risks
corresponding to each postulated accident are compiled in Section 5. The
results are consolidated to provide an estimate of the normalized health
risk associated with the entire supporting fuel cycle, and compared with
the risk associated with normal operations of the supporting fuel cycle
and the reactor. Conclusions and recommendations for future work are
contained in Section 6.
-------
2. THE LIGHT-WATER-COOLED REACTOR (LWR) FUEL CYCLE
2.1 Description of the Fuel Cycle
The conventional LWR fuel cycle is shown schematically in Figure 2-1.
The reactor and high-level waste management are included in the figure,
although these components are not treated in the current study. The
capacities of the generic facilities comprising the remaining elements,
denoted as the LWR supporting fuel cycle, are given in Table 2-1.
The fuel cycle shown in Figure 2-1 assumes the complete recycle of the
uranium and plutonium contained in the spent fuel. In the absence of
plutonium recycle, the dotted elements in the figure are eliminated.
Should spent fuel reprocessing also be eliminated from the fuel cycle,
the spent fuel itself would be stored or disposed of, and plutonium
storage, reprocessing, and the recycle of uranium to the enrichment plant
would be eliminated in addition to the dotted elements.
The specific components comprising the LWR supporting fuel cycle are de-
scribed in References 5 and 7. Summary descriptions are given below:
• Mining - The generic uranium mine is considered to be a sur-
face mine in the Western United States. The nominal capacity
is 480,000 MT ore/yr, a larger than average surface mine, situ-
ated on 3000 acres of land. The U-^Og content is assumed to
be 0.2%, and the ore body lies at various levels from 100 to
450 feet below the surface. The ratio of overburden to ore
body is estimated to average about 30 to 1. About 33 surface
mines and 122 open pit mines (supplying 36% of the uranium)
10
-------
Fuel
Assemblies
Spent Fuel
n
Plutonium
Mixed Oxide
Fuel Fabrication
Enriched UF6
u
Plutonium
Storage
Enrichment
Recovered*
Uranium
Natural UFg
Wastes Storage
High-Level Waste
Uranium Mines ft Mills
Ore
Figure 2-1. The LWR Fuel Cycle
11
-------
TABLE 2-1
CAPABILITIES OF THE GENERIC
FACILITIES COMPRISING THE
LWR SUPPORTING FUEL CYCLE
FACILITY ANNUAL CAPACITY
Mine 480,000 MT ore
Mill 960 MT U30g
UFg Conversion Plant 5,000 MTU
Enrichment Plant 8.75 xlO6 kg SWU
Uranium Fuel Fabrication 900 MTU
Reprocessing 1,500 MTHM
Mixed Oxide Fuel Fabrication 300 MTHM
Plutonium Storage 40 MT Pu-
MT= metric ton = 2205 Ibs.
MTU = metric tons uranium
kg SWU = kilograms of separative work units
MTHM = metric tons heavy metal
MTPu.r = metric tons fissile plutonium
12
-------
serviced the industry in 1973, and the output of uranium mines
in 1990 is estimated to be roughly a factor of seven higher than
16
current levels.
• Milling - The milling step extracts the uranium in the ore and
produces a refined UoOR product. The generic uranium mill is
located adjacent to the mine and produces approximately 960 MT
U30g annually, which is roughly the average nominal capacity of
mills currently in operation. It uses a mechanical crushing and
acid leach, solvent-extraction process, which is the predomin-
ant chemical processing method. Approximately 17 uranium mills
are currently in operation, and industry forecasts anticipate
5
roughly 80 mills by 1990.
• UFg Conversion - The U30g extracted from the uranium ore must
be converted to a volatile compound, uranium hexafluoride, for
enrichment by the gaseous diffusion process. There are currently
two facilities in the U.S., each producing UFg by a different
process. The dry hydrofluor process consists of continuous suc-
cessive reduction, hydrofluorination and fluorination followed
by fractional distillation for purification of the product. The
wet process uses solvent extraction at the head end to prepare
a high purity feed followed by reduction, hydrofluorination,
and fluorination steps. The bulk of the impurities in the ore
are rejected from the hydrofluor process as solids, whereas in
the wet process, the impurities are contained in the raffinate
stream. The generic UFg conversion plant produces 5000 MTU as
UFg annually, and consists of a synthesis of the two existing
13
-------
-------
processes. By the year 1990, it is anticipated that approxi-
mately 5-6 plants of 15,000 MTU/yr capacity will be in opera-
5
tion.
Enrichment - The uranium used by the LWR must be enriched from
poc 2^5
the 0.7% U content in the ore (and the-0.8% U content in
the uranium recycled from the reprocessing step) to approximate-
ooc
ly 2-4% U. The generic enrichment plant uses the gaseous
diffusion technology which is the process adopted by the three
existing government-owned and-operated enrichment plants. In
the gaseous diffusion process, volatile UFg feed is compressed
and pumped through hundreds of stages of porous barriers through
which the gas molecules diffuse. In addition to the enriched
UFg product, the gaseous diffusion process produces UFg depleted
p-JC
in U, called tails. The generic enrichment plant has a capa-
city of approximately 8.75 x 10 Kg separative work units (SWU's)
annually, larger than any of the three individual enrichment
plants currently in existence, but typical of the new plants
planned for the future. It is estimated that by the year 1990,
approximately eight plants of this capacity will be required by
5
the industry.
Uranium Fuel Fabrication - The enriched UFg is converted into fuel
assemblies at the fabrication plant for use in the LWR. The pre-
dominant current method for uranium fuel fabrication is the ammoniuir
diuranate (ADU) process for conversion of UFg to U02 powder, fol-
lowed by pelletizing and sintering in a reducing atmosphere to a-
chieve the required density. The finished pellets are loaded into
15
-------
Zircaloy or stainless steel tubes, and the completed fuel rods
are'assembled in fixed arrays. Scrap uranium is recycled into
the main process through a scrap recovery cycle. The prepara-
tion of the UCL powder or pellets may be carried out at a separ-
ate location from the final steps of the fabrication process.
The generic fabrication plant is assumed to produce annually
fuel assemblies containing 900 MTU , a large plant by today's
standards. It is anticipated that roughly nine plants of somewhat
5
larger capacity vn'll be in existence by the year 1990.
• Spent Fuel Reprocessing - Following a cycle of exposure in the
LWR, the fuel i>s chemically processed to recover the unburned
uranium and plutonium and separate the fission products for
storage or ultimate disposal. The generic spent fuel reproces-
sing plant is located on a relatively large site and utilizes
the Purex process to separate fissile material from the fission
products. In this process, the fuel elements are chopped into
short pieces, and the metal oxides are leached by hot nitric
acid, leaving behind the chopped tubing. The nitric acid solu-
tion, containing uranium, transuranics and fission products, is
adjusted chemically and processed through solvent extraction and
ion exchange systems. These process steps separate the fission
products, uranium, and plutonium from each other. The purified
uranium product is converted to uranium hexafluoride and is
shipped to the gaseous diffusion plant for reenrichment. The
purified plutonium product is converted to Pu02 for recycle to
the mixed oxide fabrication plant. The high level liquid wastes
16
-------
are stored temporarily on-site in a water-cooled storage basin
and ultimately converted to solid form for shipment to the waste-
respository. The generic fuel reprocessing plant is assumed to
have a capacity of 1500 MTHM/yr. Although there are currently
no commercial reprocessing plants in operation in the U.S.,
approximately seven plants of the generic plant capacity are
5
predicted to be in operation by the year 1990.
Mixed Oxide Fuel Fabrication - The extracted plutonium from the
reprocessing step is combined with natural uranium at the mixed
oxide fuel fabrication plant to form mixed uranium dioxide-
plutonium dioxide fuel pins for recycle to the LWR. , In the con-
ventional process, plutonium dioxide powder is blended with cer-
amic grade uranium dioxide powder, followed by the pelletizing,
sintering, and mechanical steps employed in the uranium fuel
fabrication steps. Scrap plutonium is recycled into the main
process through a scrap recovery cycle. The fuel assemblies
themselves may be fabricated from the loaded fuel rods at the
uranium fuel fabrication plant. The generic mixed oxide plant
is assumed to produce annually fuel assemblies containing 300
MTHM. Although the existing mixed oxide facilities are only
pilot plant scale, it is anticipated that by the year 1990, ap-
proximately eight facilities of the generic plant capacity will
5
be in existence.
Plutonium Storage - Plutonium recovered from the reprocessing
step must be stored on an interim basis prior to use in mixed
oxide fuel fabrication because of the difference between rates
17
-------
of production and use. Should plutonium not be recycled in
LWR's, all of the plutonium recovered from LWR spent fuel would
have to be stored awaiting the introduction of the breeder re-
actor or some alternative destiny. It is anticipated that the
plutonium will be stored in the oxide form under quiescent con-
ditions. The generic storage facility is assumed to hold ap-
proximately 40 MT of fissile plutonium.
Transportation - Transportation implies tiie shipment of materials
between each component of the fuel cycle described above, in-
cluding the shipment of unirradiated fuel from fabrication to
the reactor and irradiated fuel from the reactor to the repro-
cessing plant. It is assumed that all of these shipments, with
the exception of the transportation of irradiated fuel, are
made by truck. Irradiated fuel is brought a short distance by
truck to the rail head, from which it is transported by rail to
the reprocessing plant. The assumed quantities of materials
and transportation distances for each shipment are presented
in Table 2-2 from data given in References 5 and 7. Note that
the conversion of enriched UFg to U02 powder and the subsequent
steps in fuel fabrication are considered,for the purposes of
transportation analysis, to be conducted at separate locations.
The same assumption is made regarding the production
of mixed oxide fuel rods and the fabrication of fuel assemblies
containing mixed oxide fuel.
18
-------
TABLE 2-2
TRANSPORTATION DATA FOR
THE FUEL CYCLE
Transportation
Step
Mine-Mill
Mill-UFc Conversion
0
UFC Conversion-Enrichment
0
Enrichment-UOp Plant
U00 PI ant- Fuel Fabri-
Material
Transported
Uranium Ore
U3°8
Natural UFC
o
Enriched UFg
Enriched U00
Quantity
per Shipment
(MT)
27.2
15.2
12.7
11.0
4.5
Miles
per Trip
5
1000
500
750
750
cation
Fuel Fabrication-LWR Unirradiated Fuel
LWR-Fuel Reprocessing Irradiated Fuel
Recycle UFg-Enrichment Depleted UFg
PuO
Natural U0?-M0 Plant
£ /\
Stored Pu09-M0 Plant
C~ J\
MO Fuel Rods-Fuel
Fabrication
Natural UO,
i
PuOo
M0v Fuel Rods
X
5.8
1000
*
3.7
11.0
**
0.26
15.2
**
0.26
5.8
Truck-20
Rail -1000
1000
300
200
300
200
* 4.6 MT for no recycle case
** 0.32 MT for no recycle case
19
-------
2.2 Normalization to the Annual Fuel Requirement of a Generic LWR
7 5
The fuel inventories in the initial core, annual reloads, and spent dis-
\j
charges for the generic LWR are given in Table 2-3. Inventories are
given for the case of no plutonium recycle as well as that of recycle.
Both cases assume complete recycle of the uranium recovered from the
spent fuel. The fuel management scheme assumed for the recycle case is
5
the so-called 1.15 SGR model, which recycles plutonium equivalent to
115% of that which could be self-generated within the reactor. The gen-
eric LWR is based upon a 1000 MWe boiling water reactor (BWR) design with
an annual reload of about 172 fuel assemblies, or approximately 25% of
the core.
Using the data given in Table 2-3, the flow of materials in the fuel cy-
cle normalized to the annual operation of the generic LWR is developed
for both the recycle and no recycle cases. The results, given in Table
2-4, are based upon attributing 1/30 of the mass flows to the initial
core and 29/30 of the mass flows to the annual reloads (30-year reactor
lifetime).
Combining the information in Tables 2-1 and 2-4 gives the number of 1000
MWe LWR's serviced annually by each generic facility in the LWR support-
ing fuel cycle. The results, presented in Table 2-5, are used in Sec-
tion 5 to normalize the estimated risks from each component of the fuel
cycle to the annual operation of a generic LWR.
Similarly, the transportation data given in Table 2-2 are combined with
the annual mass flows in Table 2-4 to provide estimates of the number
20
-------
TABLE 2-3
FUEL INVENTORIES FOR
GENERIC LWR
Initial Core
Uranium (MT)
235U Enrichment (35)
134
2.6
Annual Reload
Uranium (MT)
235U Enrichment (%)
**
Plutonium (MT)
Pu Recycle
U Rods MO Rods
19.0
2.6
12.7
0.71
0.59
No Pu Recycle
32.3
2.6
Uranium (MT)
235U Enrichment (%)
Plutonium (MT)
Annual Discharge
Pu Recycle
30.6
0.63
0.57
No Pu Recycle
31.0
0.82
0.28
* MO rods, enriched to —3.3% fissile, comprise 40% of the fuel rods
A
** 57.5% fissile plutonium
21
-------
TABLE 2-4
ANNUAL MASS FLOWS IN
THE LWR .SUPPORTING FUEL CYCLE
Material Annual Requirements
Pu Recycle No Pu Recycle
Uranium Ore (MT) 60,200 82,900
U30g (MT) 120 166
Natural UFg (MTU) 89.9 141
Recycle UFg (MTU) 29.3 29.7
Separative Work (MTSWU) 69.5 102
Uranium Fuel Rods (MTU) 22.8 35.7
Spent Fuel.(MTHM) 34.5 34.6
Pu to Storage (MTPuf) 0.07 0.20
MOY Fuel Rods (MTHM) 12.8
Assuming:
1% losses in reprocessing and fabrication
Enrichment tails assay = 0.25%
Ore contains 0.2% U30g
22
-------
TABLE 2-5
NUMBER OF 1000 MWe LWR's
SERVICED ANNUALLY BY EACH GENERIC
FACILITY IN THE LWR SUPPORTING FUEL CYCLE
Facility
Number of 1000 MWE
Mine
Mill
UFg Conversion Facility
Enrichment Plant
Uranium Fuel Fabrication Plant
Reprocessing Facility
Mixed Oxide Fabrication Plant
Pu Storage Facility
LWR's Serviced
Pu Recycle No
8.0
8.0
42
126
25
43
23
570
Annually
Pu Recycle
5.8
5.8
29
86
25
43
-
200
2-3
-------
of shipments and resulting travel miles for each transportation step
in support of the annual operation of a 1000 MWe LWR. These results,
given in Table 2-6, are used as required in Section 5 to normalize the
estimated risks from transportation to the annual operation of a generic
LWR.
24
-------
TABLE 2-6
ro
en
Transportation
Step
Mine-Mill
Mill-UF,. Conversion
0
UFg Conversion-Enrichment
Enrichment-U02 Plant
U02 Plant-Fuel Fabrication
Fuel Fabrication-LWR
LWR-Fuel Reprocessing
Recycle UFg-Enrichment
Pu02-Storage
Natural U02-MOX Plant
Stored Pu09-M0 Plant
£ /\
MO Fuel Rods-Fuel Fabrication
NUMBER OF SHIPMENTS AND TOTAL
TRAVEL MILES IN EACH FUEL
CYCLE TRANSPORTATION STEP NORMALIZED
TO THE ANNUAL OPERATION OF A
1000 MWe LWR
Pu Recycle
No Pu Recycle
Truck
Rail
Number of
Shipments
2,210
7.9
10.5
3.1
5.7
7.0
9.3
9.3
3.9
0.27
0.91
2.5
2.5
Total
Number of
Travel Miles Shipments
11,000
7,900
5,300
2,300
4,300
7,000
190
9,300
3,900
80
180
750
500
3,050
10.9
16.4
4.8
9.0
7.0
Truck 8.5
Rail 8.5
4.0
0.63
-
-
_
Total
Travel Miles
15,300
10,900
8,200
3,600
6,800
7,000
170
8,500
4,000
190
-
-
_
-------
3. GENERIC MODELS FOR DEMOGRAPHY, DISPERSION,
DOSE CONVERSION AND HEALTH EFFECTS
3.1 Demography
The distribution of human receptors in the vicinity of the generic fuel
cycle facilities considered in this study is assumed to be uniform. A
tabulation of these uniform population densities is given in Table 3-1.
With the exception of mixed oxide fuel fabrication, plutonium storage,
and transportation, the tabulated numbers are the averages for the popu-
lation density ranges estimated in Reference 7. The population density
in the vicinity o^ the mixed oxide fabrication plant is taken from esti-
mates given in Reference 5, and the plutonium storage population density
is assumed to be the same as that of mixed oxide fuel fabrication.
The population density for transportation is obtained by integrating over
the population density distributions given in Reference 17 for highways
and railroads.
Most of the fuel cycle facilities considered in this study are surrounded
by areas of land which are restricted from use by the general public.
This "restricted area" is tabulated for each of the generic fuel cycle
facilities in Table 3-2, accompanied by the closest distance to the un-
restricted area, obtained by assuming a circular plot. With the excep-
tion of the numbers for mixed oxide fuel fabrication and plutonium stor-
age, which are obtained from Reference 5, these data are obtained from
Reference 7. Although the distance to the unrestricted area may be
ignored in estimating population doses from airborne radiological
26
-------
TABLE 3-1
POPULATION DENSITIES IN THE VICINITY OF
THE GENERIC FUEL CYCLE FACILITIES
2
Facility Population Density (persons/ml )
Mine 7.5
Mill 7.5
UFC Conversion Plant 47.5
o
Enrichment Plant 35
Uranium Fuel Fabrication 240
Reprocessing 90
Mixed Oxide Fuel Fabrication 100
Plutonium Storage 100
Transportation 290
27
-------
TABLE 3-2
AVERAGE DISTANCES TO UNRESTRICTED AREA
FOR GENERIC FUEL CYCLE FACILITIES
Facility
Mine
Mill
Restricted Area
(acres)
3000
(co-located
with mine)
1400
1500
UFg Conversion Plant
Enrichment Plant
Uranium Fuel
Fabrication 500
Reprocessing 2000
Mixed Oxide Fuel
Fabrication 1000
Plutonium Storage 250
Transportation 0
Approximate Distance to Unrestricted
Area (meters)
1000
1000
1000
1250
750
1500
1000
500
10*
* Assumed distance to populated area.
28
-------
effluents, it is important in assessing the population dose from the
prompt gamma and neutron radiation resulting from a postulated criti-
cal ity incident (see Section 3-2).
For the estimation of population dose from liquid effluents, the approach
outlined in Reference 11 is adopted,. It is assumed that 2000 persons/km
of a generic river for 300 km downstream from the point of release drink
water from the river. This places the population at risk at approximately
5
6 x 10 individuals. For mines and mills, the population at risk is taken
to be approximately 44,000 individuals, obtained by reducing the popula-
tion density by a factor of 0.037 and multiplying the result by a factor
of two, in order to account for the clustering of people around water
supplies in arid regions.
29
-------
3.2 Dispersion and Dose Conversion
All doses estimated in this report were calculated to represent a 50-year
*
dose commitment from a given accident release (the routine operational
release doses given for comparison were calculated for a single year's re-
lease). The estimates incorporate long-term persistence in the environment
as well as residence of the material in the body after uptake has stopped.
In general, the dose conversion calculations have been made following the
recommendations of the International Commission on Radiological Protection
(ICRP). Specifically, internal dose conversion calculations have been made
20
using the maximum permissible concentrations given by ICRP Committee 2, as
21 22
updated in ICRP publications 6 and 10. For isotopes appearing in the
accident source terms which are not given in ICRP publications, dose con-
version factors were taken from the latest Oak Ridge data as contained in
23
the INREM code. For the few short-lived isotopes which are neither in
ICRP nor INREM listings, dose conversion factors were derived using half-
24
life and decay energy data from the Radiological Health Handbook, along
with biological data from ICRP II on isotopes of the same element. Isotopes
with radiological half-lives of less than 10 minutes were neglected, as
their contribution to the 50-year population dose commitment is negligible.
* This time period includes the bulk of the dose delivered by most of the
isotopes released in the fuel cycle. For the very long-lived radionuclides,
such as plutonium-239 and iodine-129, much longer time periods are required.
However, quantitative estimates of dose over thousands to millions of years
are highly uncertain. Earlier studies have arbitrarily chosen 100 years
and 70 years'9 as cutoffs. Considering the scoping nature of this study,
the use of a 100, rather than a 50-year integration period for estimating
dose commitment would not significantly affect the results.
30
-------
Dose conversion factors for airborne insoluble participates followed the ap-
11
proach of EPA, converting from the 1959 lung model of ICRP II to the newer
lung model given in the ICRP Task Force report, "Deposition and Retention
25
Models for Internal Dosimetry of the Human Respiratory Tract." Typically,
this provides an increase in dose for a given airborne concentration by a
factor of eight. Dose conversion calculations for airborne radon-222 were
3 1]
made using the EPA suggested value of 1 pCi/m = 4 mrem/year, rather than
the 1.5 mrem/year value of ICRP II.
Plume submersion doses were calculated using a semi-infinite cloud model.
Total body and organ doses were calculated on the basis of total energy de-
posited with 5 cm of tissue shielding. Lung doses include both the external,
26
5 cm dose and dose from inhaled material, following the treatment of Snyder.
Ground plane irradiation dose calculations were made using Oak Ridge EXREM
23
III dose for gamma irradiation 100 cm above an infinite ground plane.
All dose calculations included resuspension of deposited materials and build-
up of daughter products during the 50 years following the initial release.
The resuspension coefficient was taken as 10 m initially, and was assumed
-9 -1
to decay to 10 m with a 50 day half-life. For long-lived isotopes, this
model leads to a resuspension inhalation dose contribution equivalent to 64%
of initial inhalation.
Dietary doses were calculated from intake of vegetation, milk and meat. The
assumed intake of each type of food used in the calculation was:
Vegetation - 400 grams/day
Milk - 350 grams/day
Meat - 250 grams/day
31
-------
Isotopes were assumed to be deposited directly on vegetation as well as de-
posited on soil and taken up by plant roots. Transfers of isotopes from
soil to plants and from plants to animals were based on relative concentra-
tion data of stable elements in the pathway of concern as given by Livermore
27-29
and Oak Ridge data.
Since this study deals with a number of different existing and planned
facilities, the dispersion modeling was done on a generic rather than site-
specific basis. For generic facilities,the actual spatially-dependent
population densities and dispersion factors are not known, but reasonable
average population doses can be determined assuming uniform, average
population densities. In general the total population dose is given
by:
Population dose = K / space T+ P? Jr .
where V4 is the pointwise concentration time integral and P+ is the
pointwise population density. For a general population distribution the
spatial integral will be a function of both Vp and fpr , but for a uniform
population density the population dose may be written as:
Population dose = K*FP ^
where f* is the spatially averaged concentration time integral appro-
priate for the population of P individuals. At any point, the concen-
tration time integral, V^, will be related to the ground concentration,
W*, and the deposition velocity, Vg, by:
Vg = W*/^+ .
Thus, the population dose can be expressed as:
Population dose = K W P
Vg '
o o
3 C-
-------
where W is the average ground concentration. In the above equation only
the average ground concentration, ft, is needed. Noting that whatever is
released will eventually settle, we can define the average W over a large
arbitrary areas as:
W = Q/A ,
where Q is the total source released. This gives:
Population Dose = K .TT*-. jr ,
where P/A is the average population density (people per square meter),*
Q is the total source released (Curies), Vg is the deposition velocity
(meters per second) and K is the dose conversion factor (rem per Ci-sec/
n,3).
The above equation was used to determine population doses for all parti-
culate emissions using the average population densities given in Table
3-1. A deposition velocity of 0.01 m/sec was used in all the calculations.
The above equation does not take into account radioactive decay during
dispersion. Assuming mixing in a plume limited to a height, L, it can
30
be shown that the population dose is given now by the formula:
K Q P
(Vg +*L)~A '
where V +AL is the effective plume depletion rate per unit height. The
* It should be noted that a time-invariant population density has been
assumed in this study. In actuality, the population will be increasing
over the time period used in calculating dose commitment, leading to
somewhat higher estimates. Considering the other approximations em-
ployed, however, the neglect of this effect does not significantly
affect the results.
33^
-------
mixing height boundary, L, was taken as 1000 meters. For short-lived
radioactive gases, the same formula as above was used with Vg = 0. For
long-lived gases, namely H-3, C-14 and Kr-85, dilution of the gases into
the earth's reservoirs and exposure of the total world's population was
used to calculate doses. Kr-85 was diluted in the world's atmosphere
TOO Q
(3.8 x 10 m ) and doses calculated for exposure of 4 x 10 people.
15 3
Tritium was diluted in the circulating water volume of 7 x 10 m , and
C-14 releases were assumed to reach isotopic equilibrium with the (XL
in the earth's atmosphere.
11
For releases to waterways, the EPA water pathway model was used to cal-
culate doses. Basically this model assumes a river with a flow of about
1000 cfs, with a density of people of 2000 persons/km for 300 km down-
^f
stream who drink water from the river. The effective for this river
model is 4 x 10 .' ,when all dilution factors are taken into account.
pCi/sec
For mines and mills, which are located in relatively remote parts of the
country, special models were used in calculating doses. For waterborne ef-
fluents, the EPA model for mills was used which takes a river flow of 1/20
of that given above, and the population at risk was taken as 44, 000, as opposed
to the 600,000 used for the other facilities. For radon releases to air
from the mines and mills, the effective travel time before decay is
sufficiently large that populations much beyond the mill proper may be
exposed. Thus, three different generic population densities were used
in assessing dose. Within 50 miles of the facility, the generic population
5
density of 7.5 people/mile given in Table 3-1 was used; for distances
•D
between 50 and 500 miles a population density of 50 people/mile, typical
34
-------
of the Western United States, was used; and for distances from 500 to 2000
5
miles a population density of 160 people/mile, typical of the Eastern
United States, was used. Interestingly, the largest dose commitment from
radon releases, using this model, is that received by the far high density
region corresponding to the Eastern United States. Note also that when build-
up of Pb-210 from the decay of Rn-222 is taken into account, uptake of lead
in the diet (which persists over 50 years) yields a larger dose to man than
from direct inhalation of the Rn-222 in the initial plume passage.
The final population dose commitment values used in the calculations are
given in Appendix A. It is worth noting here that the organ receiving the
largest dose is selected in each case as the critical organ, regardless of
the actual chemical characteristics of the material at the time of release.
This approach, which provides a conservative estimate of dose commitment,
is necessitated by uncertainties in the long-term behavior of these materi-
als in the environment.
For criticality incidents, the population dose from the prompt gamma and neu-
tron radiation emitted by the fission burst must be added to the dose from
radiological effluents released to the atmosphere. The dose received by in-
18
dividuals from the direct radiation "shine" of a 10 fission burst is given
as a function of distance from the burst in Table 3-3. The dose at 100 meters
was obtained from the value given in Reference 8. However, in addition to the
geometrical inverse square attenuation, the dose was further attenuated by an
air removal cross section of 0.027 cm" for both neutrons and gamma rays, as-
suming dry air. It is also important to note that the dose at 100 meters has
been corrected for facility shielding equivalent to 8 inches of ordinary con-
crete (fractional transmission of 0.17 for the neutron radiation and 0.22
for gamma radiation).
35
-------
TABLE 3-3
RADIATION DOSE TO AN INDIVIDUAL AS A FUNCTION OF DISTANCE
RESULTING FROM THE DIRECT "SHINE" OF A
CRITICALITY INCIDENT (1018 FISSIONS)
Distance (meters) Radiation Dose (rem)
100 1.6
500 1.7 x 10"2
600 8.4 x 10"3
700 4.6 x 10"3
800 2.4 x 10"3
900 1.3 x 10"3
1000 5.9 x 10"4
1250 2.2 x 10"4
1500 6.8 x 10"5
2000 7.6 x 10~6
3000 1.2 x 10"7
4000 2.5 x 10"9
5000 5.9 x 10"11
10000 7.5 x 10"19
36
-------
3.3 Health Effects
Most of the radiation exposures to the general public estimated by this
study are well below the range in which effects have been directly observed.
Accordingly, theoretical estimates of the relationship between absorbed
dose and health effects must be relied upon to provide quantitative esti-
mates. There exists considerable disagreement within the scientific
community as to the appropriate relationship for such estimates. However,
most advisory and standards setting groups suggest that it is correct
and prudent to use the linear, non-threshold hypothesis in standard setting.
According to this hypotheses, there is a linear relationship between
the total accumulated dose and the number of effects (for all types
of effects observed) from zero exposure to the highest exposure which
does not cause acute mortality before expression of the delayed effect.
Using this hypothesis, if 1 rem produces 10 effects,then 2 rems produce
20 effects and 0.1 rem produces 1 effect. Differences in magnitude of
dose or dose rate do not have any influence on the relationship if the
linear, non-threshold hypothesis is adopted.
Because of the widespread use of this model in standards setting, and
since no other model has achieved widespread acceptance for quantitative
estimates of health effects at low levels of exposure, the linear, non-
threshold dose-response relationship is adopted for use in this
study. Table 3-4, developed by EPA, lists the number and types of
effects expected after the exposure of a population of one million to
12
a total of one million man-rem. This table was developed by inte-
grating forward over time the estimates of annual effects from the BEIR
13
report.
37
-------
TABLE 3-4
EFFECTS OF A SINGLE EXPOSURE OF A POPULATION OF ONE MILLION
fi 12
TO ONE REM PER PERSON (10° MAN-REM)
Target Organ
Total Body
Distributed as follows:
Breast
Lung
GI Tract
Bone
Leukemia
Other
Thyroid
Somatic Effects
Cexcess cancers)
200 fatal
200 non-fatal
90
40
62
12
49
147
62
400
Genetic Effects
(congenital defects)
300
38
-------
All of the effects to populations of exposures at low dose levels of
radiation are delayed. These delayed effects show up after a relatively
long latency period and fall into two broad categories - somatic and
genetic. Somatic effects are generally some form of cancer, and are
the effects which are quantified in this study. Depending on the dis-
tribution of exposure in the body, these cancers develop at specific
locations in the body or in specific organ systems. Approximately 50%
of the cancers are fatal. Delayed genetic effects apparently arise
from mutations in the sperm and ova, and usually result in children
with hereditary congenital defects.
In this study, Table 3-4 is used by determining the population dose
to the "critical organ" (the body organ receiving the maximum absorbed
dose), and converting this dose to health effects using the conversion
factor in the table. For example, if the population dose to the GI tract
were x man-rem, the number of estimated health effects would be given
by 62x/10 . Since estimated effects are assumed to be cumulative across
the total body and tissues, the estimated effects to the remainder of
the total body must be added to the effects to the critical organ. For
example, if the corresponding dose to the total body in the above example
were y man-rem, the number of additional health effects would be given
by (400-62)y/106. The total estimated health effects, then, would be
given by the sum of the effects to the critical organ and the total
body.
39
-------
For a few of the postulated accidents, acute effects as well as long-term
delayed effects would be expected. The number of acute effects are expec-
ted to occur within a period of two to three months after the exposure.
For purposes of this study, only possible deaths were considered, and a
very simple relation was used to estimate the number of fatalities.
Namely, for exposures in the range of 200 to 600 rem, 50% of the exposed
population is assumed to receive a lethal dose, and above 600 rem, 100%
of the exposed population is assumed to die. The population dose commit-
ment to the surviving individuals in the 200 to 600 rem range, and to
those individuals receiving less than 200 rem, is added to the low-level
dose commitment for the purposes of estimating long-term somatic health
risks.
40
-------
4. SOURCE TERMS AND LIKELIHOODS
For the purpose of ease in documentation and subsequent expansion or re-
vision, the accidents examined in this study are organized within the
broad categories given in Table 4-1. Containment failures, Category C,
include vessel, pipe, valve, and ventilation system failures of all types.
A loss-of-coolant incident, for example, would be incorporated within
this category. External events, including acts-of-God and inadvertent
human-induced events, would comprise a separate category were they in-
cluded within the scope of this study.
The source terms,and likelihoods estimated for each postulated accident
are generally synthesized from the existing literature. The major sources
of information are generic environmental statements and/or safety assess-
ments.
;
These documents generally derived their data from individual detailed en-
vironmental statements, safety analysis reports, incidents abstracts, or
compliance files. As discussed in Section 1.3, the original data are
subject to considerable manipulation in this study. This includes con-
solidation, renormalization, and reinterpretation. Little attempt has
been made in this preliminary risk assessment to utilize a consistent
methodology for deriving source terms or likelihoods.
Accident likelihoods have been the most difficult data to obtain and
are generally the softest numbers quoted in this study. A cursory sur-
31 32
vey was made of incidents on record in government and commercial
facilities to augment the information discussed above. The accidents
41
-------
TABLE 4-1
CATEGORIES OF ACCIDENTS
A. Explosions
B. Fires
C. Containment Failures
D. Criticality
E. Retention Pond Releases
42
-------
identified in this survey* which are pertinent to this study are tabu-
lated in Tables 4-2 through 4-5. Note that, in general, the magnitude
of the release has not been documented. Criticality incidents are well
31,33,34
documented in a number of sources, and are therefore not tabula-
ted here.
* This survey was by no means comprehensive.
43
-------
Info. Source Date
TABLE 4-2
EXPLOSION INCIDENTS IN FUEL CYCLE FACILITIES
Where Process Type Release Magn.
NSIC
NSIC
NSIC
WASH-1192
WASH-1192
WASH-1192
WASH-1192
WASH-1192
1/67
12/72
12/68
7/59
1/60
4/60
12/60
8/65
NUMEC
GUNF,
Elmsf.
UNC,
Wood
River
Hanford
ORNL
Hanford
Hanford
Battelle-
M0x Fab.
MO Fab.
U Fab.
U Fab.
U Fab.
Repro.
U. Fab.
MOV Fab.
0.1 gm Pu
?
7
Probably zero
?
7
Probably zero
?
WASH-1192
Northwest
4/68
Mound
Lab.
MO Fab.
Details
Glovebox breach from projectiles
resulting from H-O- decomp.
Glovebox containing sintered
U-Pu02
Explosion 1n scrap recovery
Autoclave explosion
Sintering furnace explosion
Dlssolver explosion
Autoclave explosion
Cleaning fluid in glovebox
ignited
Glovebox drying oven
explosion
-------
TABLE 4-3
FIRE INCIDENTS IN FUEL CYCLE FACILITIES
•en
Info. Source
NSIC
NSIC
NSIC
NSIC
NSIC
NSIC
NSIC
NSIC
WASH- 11 92
WASH- 11 92
Date
1/74
9/72
4/71
3/72
12/68
7/63
'62- '68
3/66
11/56
12/62
Where
NUMEC
UNC
UNC-
Hematite
NFS,
Erwin
Petro-
tomics
Rocky
Flats
Several
B&W
Paducah
Paducah
Process Type
U Fab.
U Fab.
U Fab.
U Fab.
Mill
MOX Fab.
General -Ion
Exchange
U Fab.
Enrich.
Enrich.
WASH-1192 10/64 Savannah Repro.
River
Release Magn. Details
? Hydrogen ignited, burned filter
? Fire in scrap recovery burned
through stack
? Leak in off-gas system of UFc-UO?
converter. Hydrogen release ignited
? Flash fire in tray dissolver
Fire in solvent extraction circuit
Ion-exchange recovery fire and
explosion
Seven ion-exchange resin fires
from self-ignition of nitrate-form
resins
Dissolver fire
Fire spread thru roof ($2,100,000)
Major fire in gaseous diffusion cell
($2,900,000)
Fire occurred around an ion exchange
column in hot canyon
-------
TABLE 4-4
LOSS OF CONTAINMENT INCIDENTS IN FUEL CYCLE FACILITIES
Info. Source
NSIC
NSIC
NSIC
NSIC
NSIC
NSIC
NSIC
NSIC
NSIC
Date
1/73
4/71
4/69
3/69
7/68
5/68
11/67
12/68
11/67
Where
NUMEC-
Apollo
NFS- West
Valley
B&W
NUMEC
Kerr-
McGee-Okl
NUMEC,
Apollo
NUMEC,
Apollo
Allied
Chem. ,
Metr.
NUMEC,
Apollo
Process Type Release Magn.
U Fab. 2.5gms235U to
Repro . ?
MOxFab. ?
MOxFab. ?
Fab. 100 Ibs. UF,
O
U Fab. ?
M0x Fab. ?
UF, Conv. 90 Ibs. UF,
6 6
U Fab. 600gms235U to
River
Details
Extraction column rupture
Valve left open in Pu Prod.
Storage Tank
Leak in Pu calcining furnace
Leak in bag containing glovebox
sweepings
Inadvertent valve open on vapori;
equipment
Leaking gasket on blender
Pump leak
Ruptured valve in dist. process
Inadvertently opened valve to
dump tank
-------
Info. Source
NSIC
NSIC
NSIC
WASH- 11 92
WASH- 11 92
WASH- 11 92
WASH- 11 92
WASH- 11 92
WASH- 11 92
WASH- 11 92
Date
2/69
5/72
6/71
8/59
9/60
11/60
11/60
5/61
11/61
2/66
Where
NFS
NFS,
W. Valley
NFS,
W. Valley
Savannah
River
Savannah
River
ORNL
Natl.
Lead
Mound
Lab.
Han ford
Natl.
Lead
TABLE 4-4
(continued)
Process Type Release Magn.
U Fab. ?
Repro . ?
Repro. ?
Repro. ?
Repro . ?
Enrich. 3077 kg U
U Fab. ?
U Fab. ?
U Fab. 1089 Ibs. depl U to sewer
U Fab. 3844 Lbs. UFg
(to environ,? }
Details
5 Kg U02 spilled when blender
discharge valve opened acci-
dentally
Acid and water released into
extraction aisle
Release of contaminated water
in sewage
Leakage from waste evaporator
Contamin. cooling water dis-
charged from canyon onto floor
Ten-ton UF, cylinder rupture
SI. enr. UF* lost thru stack
of dust collector
Caustic scrubber radiation
release
Operator inadvertently unscrewed
valve from head of a 10 ton
cylinder
-------
TABLE 4-4
(continued)
Info. Source Date Where Process Type Release Magn. Details
WASH-1192 11/69 Savannah Repro. ? Acidic waste soln. inadvertently
River transf. to underground waste
system due to leak, valve, and
lost
Health Physics 9/63 Hanford Repro. 60 Ci 1-131 ?
Vol. 11, pp.
1009-1015
CO
-------
Info. Source
NSIC
NSIC
NSIC
NSIC
NSIC
£ NSIC
Date
1/74
7/73
1/73
6/70
3/68
9/68
Where
NFS-
Erwin
GE.N.C.
UNC, New
Haven
NUMEC,
Apollo
NFS
NFS,
W. Valley
Process Type
U Fab.
U Fab.
U Fab.
MOX Fab.
U Fab.
Repro.
NSIC
NSIC
TABLE 4-5
FILTER FAILURE INCIDENTS IN FUEL CYCLE FACILITIES
Release Magn.
8/67 UNC- U Fab.
Wood River
11/72 UNC- U Fab.
New Haven
67 C1 Pu to air
Details
Filter in vacuum cleaner failed
Roughing filter plugged
Holes in HEPA from calciner
Dissolver exhaust line filter
failed
Monthly allowance (?) Filter failure
11% monthly allow. Filter failure
1 gm U-235
Ruptured exhaust filter
Acid fumes caused failure on
non-acid resistant HEPA filter
-------
4.1 Mining
No accidents have been identified in the mining of uranium ore which would
result in significantly higher releases to the environment than incurred
during normal operations. Fire or earth collapse could lead to occupa-
tional injuries, but would not result in airborne radioactive effluents
(uranium-bearing dusts and radon and its daughters) in excess of those
released when the ore body is exposed and broken up during mining operations.
Power failure to the ventilation system in an underground mine could
lead to a buildup of radon, but the integrated release over time would
be unchanged.
A failure of the mine drainage system to dewater the mine area could lead
to flooding, or flooding of the mine area could result from natural causes.
However, the total release of activity in mine drainage water would not
be substantially higher than that released during normal mine drainage.
At some mines, the drainage water is held in retention before being re-
leased. Failure of a retention pond dike could result in the release to
the environment of the contaminated drainage water together with sus-
pended solids from the mines. Since the fate of the suspended solids
is site-specific, and the activity in the mine drainage water is gener-
ally not significantly greater than that of natural mineral springs in
the vicinity of the mine, this accident was not analyzed in the current
study.
50
-------
4.2 Milling
The accidents considered in uranium milling facilities, keyed to the ac-
cident categories given in Table 4-l^are listed in Table 4-6.
Other incidents may be postulated or have occurred in mills from which
the environmental risks are deemed insignificant from the existing in-
formation. These include local fires, overflows from process tanks,
failures of process lines, failures of offgas filtration or scrubbing'
systems, or storage tank spills.
Natural or man-induced disasters, such as tornados, earthquakes, floods,
or missile impacts are highly unlikely, although their occurrence could
result in varying releases of activity. The risks associated with
these events were not analyzed in the current study.
Table 4-7 provides historical data on uranium mills currently in opera-
35,36
tion. Historical data on mills no longer in operation are provided
37
in Table 4-8.
B.I Fire in Solvent Extraction Circuit
In the solvent-extraction step of the milling process, the uranium is
purified and concentrated. This is accomplished by contacting the
gaseous phase from the leaching step, which contains the uranium and
impurities, with an organic solvent. The solvent extraction circuit is
likely to be in a separate building containing several thousand gallons
of solvent (mostly kerosene) and several thousand pounds of natural
51
-------
TABLE 4-6
MILLING ACCIDENTS
B.I Fire in solvent extraction circuit
E.I Release of tailings slurry from
tailing pond
E.2 Release of tailing slurry from
tailings distribution pipelines
52
-------
TABLE 4-7
URANIUM MILLS IN OPERATION AS OF MARCH, 1975
35,36
COMPANY
Anaconda Company
Atlas Corporation
Conoco & Pioneer
Nuclear, Inc.
Cotter Corporation
Dawn Mining Company
Exxon, U.S.A.
Federal-American
Partners
Kerr-McGee Nuclear
Petrotomics Company
Rio Algom Corp.
Union Carbide Corp.
Union Carbide Corp.
LOCATION
Grants, New Mexico
Moab, Utah
Falls City, Texas
YEAR OPERATIONS
INITIATED
1953
1956
1961
Canon, City, Colorado 1958
Ford, Washington 1957
Powder River Basin, Wyoming --1971
Gas Hills, Wyoming 1959
Grants, New Mexico 1958
Shirley Basin, Wyoming 1962
La Sal, Utah ^1972
Uravan, Colorado -'1950
Natrona County, Wyoming 1960
NOMINAL CAPACITY
(Tons of Ore/Day)
3000
800-1500
220-1750
150-450
0-400
2000
500-950
3600-7000
525-1500
500
0-1300
1000
-------
TABLE 4-7
(continued)
YEAR OPERATIONS NOMINAL CAPACITY
COMPANY LOCATION INITIATED(Tons of Ore/Day)
United Nuclear- Grants, New Mexico 1958 1650-3500
Homestake Partners
Utah International, Gas Hills, Wyoming 1958 750-1200
Inc.
Utah International, Shirley Basin, Wyoming 1971 1200
Inc.
£ Western Nuclear,Inc. Jeffrey City, Wyoming 1957 400-1200
TVA (Mines Develop- Edgemont, South Dakota 1956 250-500
ment, Inc.)
-------
TABLE 4-8
COMPANY
Foote Mineral Co.
El Paso Natural Gas
Vanadium Corp.
Climax Uranium Co.
Colorado Ventures
Union Carbide Corp.
Vanadium Corp.
Union Carbide Corp.
Union Carbide Corp.
North Continent
Union Carbide
Michigan Chemical
Corp.
United Nuclear Corp.
Foote Mineral Co.
Atlantic Richfield
Susquehanna Western
Union Carbide Corp.
A-Z Minerals
Vitro u>rp.
Western Nuclear
URANIUM MILLS NO LONGER IN OPERATION
37
LOCATION
Monument, Arizona
Tuba City, Arizona
Durango, Colorado
Grand Junction, Colorado
Gunnison, Colorado
Maybe!1, Colorado
Naturita, Colorado
New Rifle, Colorado
Old Rifle, Colorado
Slick Rock, Colorado
Slick Rock, Colorado
Lowman, Idaho
Ambrosia Lake, New Mexico
Shiprock, New Mexico
Lakeview, Oregon
Ray Point, Texas
Green River, Utah
Mexican Hat, Utah
Salt Lake City, Utah
converse County, Wyoming
YEARS OPERATED
1955-
1956-
1943-
1951-
1958-
1957-
1939-
1958-
1924-
1931-
1957-
1955-
1967
1966
1963
1970
1962
1964
1963
1972
1958
1943
1961
1960
1958-1963
1954-1968
1958-1960
1970-1973
1958-1961
1957-1965
1951-1968
1962-1965
TONS OF TAILINGS
1,200,000
800,000
1,555,000
1,900,000
540,000
2,600,000
704,000
2,700,000
350,000
37,000
350,000
90,000
2,
1,
600,000
500,000
130,000
490,000
123,000
2,200,000
1,700,000
187,QUO
-------
7
uranium. The flammability of the solvent provides the potential 'for
a serious fire.
Source Term
Two to three thousand pounds of uranium were present in the solvent ex-
7
traction circuits involved in major fires in 1968. This is a reason-
able inventory for a 960 MT U-jOg/yr mill which produces approximately
6000 Ibs/day of uranium.
If 1% of the uranium were dispersed, as assumed in estimates of the
9
amount of plutonium dispersed in a solvent extraction fire, 20 to 30
pounds of uranium could be released to the atmosphere. If 0.5-0.7%
were dispersed, as assumed in estimates of the amount of plutonium re-
8*
leased from a fire near nitrate blending tanks, 10 to 20 pounds of
uranium could be released to the atmosphere.
It is assumed that 99% of the radium had been removed in the leaching
step, but that the thorium in equilibrium with uranium is present in
the solvent extraction step. Thus the estimated concentrations of radio-
— 7
isotopes in the material released are: 3.3 x 10" Ci/gm of U, 1.54 x
p OOC _7 9QA _7
10"tt Ci/gm of J U, 3.52 x 10 ' Ci/gm of "HU, 3.31 x 10 Ci/gm of
234Th, 3.31 x 10"7 Ci/gm of 230Th, and 3.31 x 10"9Ci/gm of 226Ra.
* Experiments have demonstrated that approximately 80% of the uranium
aerosol generated from a nitrate solution involved in a gasoline
fire is the respirable size range.8
56
-------
Likelihood
From chemical industry data, the probability of major fires per plant-year
-4 8
is estimated to be 4 x 10 . However, at least two major solvent ex-
7
traction circuit fires are documented in the literature. From the data
in Tables 4-7 and 4-8, there have been 515 plant-years of mill operation,
or the equivalent of 282 plant-years handling 480,000 MT ore/yr. Thus,
from the historical incidents, the likelihood of a major solvent extrac-
tion fire is in the range of 3 to 7 x 10 /plant-year.
— ^ o 7 o — A
Estimate: Source term & 3. 3 x 10~ Ci Us 1.5 x 10~ Ci U,
1 ? Id 1 914
3.5 x 10 Ci U> 3.3 x 10 Ci Th,
3.3 x 10~3 Ci 2S°Th3 3.3 x 10~5 Ci 226Ra to air. Likelihood
-3 -4
3 x 10 to 4 x 10 /plant-year.
E.I Release of Tailings Slurry from Tailings Pond
The solid residues (tailings) from the leaching circuit of the milling
process are suspended in a wash solution and pumped to the tailings re-
tention pond. The aqueous phase from the solvent extraction circuit,
which is called the raffinate and contains most of the impurities in the
ore, is also pumped to the tailings retention pond. The tailings are
composed mostly of sandstone and clay particles, and contain about 85%
of the activity originally in the ore. The tailings pond is constructed
57
-------
by erecting an earth fill, clay core dam across a natural basin. The
system may hold on the order of 1,500,000 MT of solid tailings. Tail-
ings dams have been known to fail due to flooding, earthquake, or inat-
tention.
Source Term
Accidental tailings slurry releases have been documented and the result-
7
ing release estimates compiled. Table 4-9 contains a summary of re-
corded incidents in the period from 1959 to 1971. From these data, the
average releases from tailings dam failure or flooding are 3000 m3
of liquids and 7.5 x 106 Ibs. of solids.
The estimated concentrations of radioactive effluents in the waste liquor
from a generic uranium mill are given in Table 4-10. The estimated
total specific activity of solids is 3.7 /*iCi/lb. The solids, however,
are assumed to deposit in the vicinity of the entry point to the water-
course.
Likelihood
Eight out of eleven of tne releases documented in Table 4-9 reached the
watercourse, ana of the total of 12 recorded incidents, eight involved
dam failure or flooding. Referring to Tables 4-7 a-nd 4-8, there were
270 plant-years of operation in the period 1956 through 1970, or the
equivalent of 153 plant-years handling 480,000 MT ore/yr. Using these
historical data, the likelihood of release from the tailings pond itself
_2
to the watercourse is in the range of 2 to 4 xlO /plant-year.
58
-------
TABLE 4-9
1
SUMMARY OF ACCIDENTAL TAILINGS SLURRY RELEASES
Cause
Flash flood
Dam failure
Dam failure
Dam failure
Pipeline failure
Flooding
Pipeline failure
Flooding
Dam failure
Pipeline failure
Dam failure
Pipeline failure
Solids Released Liquids Released Reached
(Ibs.)(gallons)Watercourse
30 x 10l
2 x 10
1 x 10(
4 x 10J
6 x
,6*
3.3 x 10
2.4 x 10{
1 x 10
5 x 10
6 x
,6*
,5*
.4*
No quantitative information
1.4 x 10
4 x 10
3-30 x 10
3 x 10
2 x 10
,5*
,6*
6*
5*
,4*
1.6 x 10*
4.4 x 105
3-30 x 105
3.5 x 104
2 x 103
No quantitative information
yes
yes
no
yes
yes
?
small amount
yes
yes
yes
no
no
* Assuming equal weights of solids and liquids released and density of
liquids approximately 9 Ibs/gallon.
59
-------
TABLE 4-10
CONCENTRATIONS OF RADIOACTIVE EFFLUENTS
IN WASTE LIQUOR FROM THE MODEL URANIUM MILL
Contaminant Concentration
Uranium - natural 8.0 x 10"
Radium 226 3.5 x 10
Thorium 230 2.2 x 10 ju.Ci/ml
60
-------
It should be pointed out that these historical incident data may be mis-
leading, since the evaluation of early dike construction led to a deter-
mination that dikes need strengthening. Mills having dikes similar in
construction to those that failed were required to strengthen the dikes
and new mills were required to use new construction standards.
r? f) "7Q C O "2 C
Estimate: Source term £? 1. 1 x 10 Ci U3 5.3 x 10 Ci
t f\ -7 /•»" *5 ~« & O~r .. ~j ~j — -— O -. , £1 0~r _.-r -* -7 -/- — O -^ • £1 &) _
1.2 x 10 C^ U, l.l x 10 C^ Th, 1. 1 x 10 C^ Ra,
_o 270 —2
6.6 x 10 Ci Th to watercourse; Likelihood <* '4 x 10 /plant year.
E.2 Release of Tailings Slurry from Tailings Distribution Pipelines
Tailings distribution pipelines have been known to fctil, resulting in
the accidental release of tailings slurry to the watercourse.
Source Term
From the data given in Table 4-9, the average release from incidents on
record involving the failure of pipelines in the tailings distribution
3 5
system is 130 m of liquids and 3.5 x 10 Ibs. of solids.
Likelihood
Of the total of 12 recorded incidents documented in Table 4-9, four in-
volved pipeline failures. Using the same approach as that employed in
deriving the likelihood of dam failures, the likelihood of release from
_2
tailings distribution pipelines is in the range of 1 to 2 x 10 /plant/yr.
-------
Estimate: Source term & 4.8 x I0~u Ci "^U, 2.3 x W~6 Ci 235U3
5.2 x 10~5 Ci 2S4V, 4.8 x lO~5 Ci 234Th, 4.8 x 10~5 Ci 226 Pa,
—3 230 —2
2.9 x 10 Ci Th to watercourse. Likelihood^ I x 10~ /plant-year.
62
-------
4.3 UFC Conversion
b
The accidents considered in UFg Conversion facilities, keyed to the acci-
dent categories given in Table 4-1, are listed in Table 4-11.
Other incidents have occurred or are possible for which the associated
environmental risks are judged to be insignificant in comparison with
those of the accidents listed in Table 4-11. These include yellowcake
spills at the head end of the process, local fires, loss of refrigeration
to cold traps, or small releases of UFg from valve or process line fail-
ures. There are also credible accidents which would release toxic chemi-
cals to the environment, such as the rupture of a hydrogen fluoride tank
or a leak in the fluorine production plant, but these are considered out-
side of the scope of the current study. Also, the risks associated with
the release of radioactivity from disasters, such as tornados, or missile
impact, were not analyzed.
There are currently two commercial facilities for the production of uran-
ium hexafluoride, operated by Allied Chemical Corporation and Kerr-McGee
Corporation, respectively. The former plant has operated a total of 14
years since 1959 at production capacities up to 5000 MTU/yr. The latter
initiated operations in 1970 and has accumulated 5' to 6 years at produc-
tion capacities close to 5000 MTU/yr.
A.I Uranyl Nitrate Evaporator Explosion
In the wet chemical solvent extraction process, for producing uranium hexa-
fluoride, the uranium concentrate is dissolved in nitric acid and sent to
63
-------
TABLE 4-11
UFC CONVERSION ACCIDENTS
0
A.I Uranyl nitrate evaporator explosion.
A.2 Hydrogen explosion in the reduction step of the
process.
B.I Fire in the solvent extraction operation.
C.I Release from a hot UFg cylinder.
C.2 Valve rupture in the distillation step.
E.I Release of raffinate from the waste retention pond.
64
-------
the uranium-extraction column, where the aqueous solution of uranyl ni-
trate is extracted countercurrently withtributyl phosphate in hexane.
The uranium is then reextracted as uranyl nitrate solution into a large
volume of water, which is concentrated by evaporation. An explosion could
occur in the evaporator from a "red-oil" reaction. "Red-oil" is a materi-
al that is formed from a heavy metal nitrate and nitric acid solution
9
mixed with tributyl phosphate solvent at temperatures exceeding 135°C.
Under optimum conditions, the reaction becomes explosive,thus the evap-
orator temperature must be limited to avoid explosive conditions.
Source Term
The evaporator would typically contain about 2000 gallons of uranyl ni-
7 3
trate. At a uranyl nitrate density of 2.8 gms/cm , the total mass of
uranium in the evaporator would be about 10,000 kg. Adjacent to the
evaporator, an equal amount of product might be stored in a surge tank.
The consequences of explosion accidents are limited by the material that
can be maintained in the air rather than by the total volume or mass of
material involved in the explosion. The airborne concentration of heavy
particles in the respirable range appears to be limited to approximately
3 8
100 mg/m . The material splattered on the walls and floors which subse-
quently becomes airborne as it dries is expected to constitute an insig-
nificant fraction of the original source term. Thus for a room volume
4 3
assumed to be approximately 10m, the quantity of uranium released to
the environment is estimated to be approximately 1000 kg.
66-
-------
It is assumed that the milling operation has removed most of the uranium
daughter products in the ore, so that nearly all of the activity released
in the evaporator explosion is from the isotopes of natural uranium. Their
specific activities are 3.31 x 10"7 Ci/gm of 238U, 1.54 x 10"8 Ci/gm of
-7 P34 -7
, 3.52 x 10 ' Ci/gm of "\l, and 3.31 x 10 'Ci/gm of
Likelihood
Historically, one explosion associated with evaporator operation has
7
occurred due to a "red-oil" reaction, but this was not in a commercial
uranium hexafluoride conversion facility.
The probability of a red-oil explosion in the low activity waste concen-
tration of a commercial fuels reprocessing plant has been estimated to be
. g
-v 10" /plant-year. The likelihood of a chemical explosion in a fuel
-3 8
fabrication plant has been estimated to be "10 plarit-year.
— 7 238 —2 235
Estimate : Source term "Z> 3. 3 x W Ci Uf 1.5 x 10 Ci U3
3.5 x I0~l Ci 2MV, 3.3 x I0~l Ci 2S4Th to air. Likelihood
-------
Source Term
It is assumed that the reductor contains a uranium inventory of approxi-
mately 10,000 kg. As discussed in the case of the uranyl nitrate evap-
orator explosion, the consequences of explosion accidents are limited by
the concentration of heavy material that can be maintained in the air,
38
shown to be approximately 100 mg/m . Then for a room volume assumed to
4 3
be of the order of 10 m , the quantity of uranium released to the environ-
ment is estimated to be approximately 1000 kg. The isotopic composition
of the material released is taken to be that of freshly separated natural
uranium.
Likelihood
No incidents have been recorded involving hydrogen explosions in the re-
16
duction step of uranium hexafluoride conversion. Hydrogen fires or ex-
plosions have been documented, however, in association with fuel proces-
31,32
sing activities. An estimate of the probability of a hydrogen ex-
plosion in the sintering step of fuel fabrication is
-------
B.I Fire in the Solvent Extraction Operation
In the wet chemical solvent extraction process for producing uranium
hexafluoride, the uranium concentrate is dissolved in nitric acid and sent
to the uranium-extraction column, where the aqueous solution of uranyl
nitrate is extracted countercurrently with tributyl phosphate in hexane.
The solvent extraction operation is likely to be carried out in a separate
building consisting of two operating parts. The first is a solvent rework
section where most of the organic solvent used in the process, hexane, is
stored and prepared for use. The other is the solvent extraction section
where the uranium purification operation is carried out. The flammability
of the solvent provides the potential for a serious fire in either section,
however, only a fire in the solvent extraction circuit would involve the
release of radioactivity.
Source Term
It is estimated that the amount of loaded solvent which might typically
be involved in a fire in the solvent extraction section is approximately
2500 gallons containing 800 kg of uranium.
If 1% of the uranium were dispersed, as assumed in estimates of the amount
$
of plutonium dispersed in a solvent extraction fire, 8 kg of uranium
could be released to the atmosphere. If 0^.5-0.7% were dispersed, as
assumed in estimates of the amount of plutonium released from a fire near
68'
-------
8*
nitrate blending tanks, 4 to 6 kilograms of uranium could be released
to the atmosphere.
It is assumed that the thorium and 1% of the radium in equilibrium with
natural uranium is present in this initial purification step. Thus the
estimated concentrations of radioisotopes in the material released are:
3.31 x 10"7Ci/gm of 238U, 1.54 x 10~® Ci/gm of 235U, 3.52 x 10~7 Ci/gm
of 234U, 3.31 x 10"7 Ci/gm of 234Th, 3.31 x 10~7 Ci/gm of 230Th, and
-Q
3.31 x 10 y Ci/gm of Ra.
Likelihood
There have been no solvent extraction fires on record associated with
31 ,32
uranium hexafluoride production. However, from chemical industry
data, the probability of major fires per plant-year is estimated to be
4 x 10"4.
Estimate: Souroe term ^2.0 x 10~3 Ci 238V3 9.2 x 10~5 Ci 235U,
2.1 x 10~3 Ci 234U3 2.0 x 10~3 Ci 234Fh3 2.0 x 10~3 Ci 23°Th3 2.0
x 10~5 Ci 26Ra, to air; Likelihood ^ 4 x I0~4/plant-yr.
C.I Release from a Hot UF6 Cyl i nder
At the tail end of the uranium hexafluoride production process, the liq-
uid uranium hexafluoride is transferred under pressure to a large cylinder,
* Experiments have demonstrated that approximately 80% of the uranium
aerosol generated from a nitrate-solution involved in a gasoline fire
is in the respirable size range.
69
-------
nominally either 14 or 10 tons. The cylinder is placed in a steam-heated
chest for about 12 hours at 200°F to homogenize the contents. A valve is
installed and a sample taken. The cylinder is then removed by fork lift
to an outdoor storage area where the UFg cools and solidifies in approxi-
mately 72 hours. At any time during this sequence of events, while the
UFg is in the liquid state, the failure of a valve, an operator error,
or a cylinder rupture could release significant quantities of UFg to the
environment.
Source Term
If a large cylinder containing 12.7 MT of UFg were to rupture or lose a
valve, and if the release were to go unchecked until the contents of the
cylinder solidified, analyses result in estimates of approximately 2800
7
kg of uranium .released to the environment. In fact, the rupture of a
ten ton cylinder at an enrichment plant,resulting in the release of 3077
31
kg of uranium, is a matter of record. However, the most probable re-
lease in the event of a breach in containment is not likely to coincide
with the maximum release. A summary of releases associated with hookup
and disconnect operations on UFg cylinders at the three existing gaseous
38,39
diffusion plants has been compiled and is reproduced in Table 4-12.
The range of release falls between approximately 10 and 400 kg uranium
in the period 1969 through 1973. The average release in six recorded in-
cidents resulting in more than 5 kg uranium loss was 108 kg of uranium.
The uranium at the tail end of the uranium hexafluoride conversion pro-
cess is freshly separated from daughter products.
70
-------
TABLE 4-12
UFg RELEASES (> 5Kg) ASSOCIATED
WITH FILLING OR FEEDING CYLINDERS AT
THE GASEOUS DIFFUSION PLANTS
IN THE PERIOD 1969-1973
Year Plant Quantity Reason
(KgU)
1969 Portsmouth 400 Tails cylinder valve
would not close
1969 Portsmouth 9.9 Leaking pigtail
1970 Oak Ridge 153.3 Pigtail connection leak
in normal assay feed
autoclave
1970 Portsmouth 12 Feed position pigtail
gasket failure
1971 Paducah 14.7 Pigtail rupture during
sampling of cylinder
1973 Portsmouth 59.4 Undetected plugged and
open valve on hot cyl-
i nder
-------
Likelihood
There are no UF,. cylinder releases on record in the <~> 20 plant-years of
b 40,41
operation of commercial uranium hexafluoride conversion plants.
6
From failure data on valve ruptures, the likelihood of a UFg cylinder
valve rupture while the UFg is in the liquid state is estimated to lie
-5 -3
in the range, 5xlOto5xlO /plant-yr for a 5000 MTU/yr plant ca-
pacity. However, none of the releases associated with the filling or
feeding of cylinders at an enrichment plant, as shown in Table 4-12, can
be attributed to valve ruptures.
From the incidents on record at the three enrichment facilities, a rough
estimate of the likelihood of UFg cylinder releases can be made which may
be extrapolated to filling operations at a UFg conversion facility. In
1973, roughly 7800 UFg cylinder hookup and disconnect operations were per
formed at the Paducah plant, 6700 at the Oak Ridge plant, and 10,000 at
the Portsmouth plant. Then for the six releases on record documented in
Table 4-12, and assuming the same level of operations over the entire 5-
year period, the likelihood of a release in handling a cylinder is approx
_c
imately 5 x 10 /cylinder.
The generic UFg conversion plant operating at a nominal capacity of 5000
MTU/yr would require the filling of about 580 large product cylinders an-
nually. Using the release probabilities developed in the previous dis-
cussion from the enrichment plant data, the likelihood of a release is
_2
approximately 3 x 10 /plant-year.
7;2
-------
_ — ^?
Estimate: Source term ^S.6 x 10 Ci U, 1.7 x 10 Ci U,
_9 - — —
3.9 x 10 Ci U3 3.6 x 10 Ci Th to air. Likelihood ~ 3 x 10 /
plant-year.
C.2 Valve Failure in the Distillation Step
In the dry hydrofluor process for the conversion of uranium hexafluoride,
the final step of process involves fractional distillation. The distil-
lation step removes volatile fluorides generated in the fluorination step.
The failure of a valve or piping in this step could result in the release
of UF,. to the environment.
D
Source Term
In an incident of record, a valve bonnet failed in a line from a vaporizer
tank in the distillation area of a plant and released approximately 40 kg
7
of uranium as UFg to the building.
Small releases of UFg also occur in the enrichment plants from time to
time due to defective valves or tubing, and the parallel to the distilla-
tion step of a conversion plant can be drawn. Thus it is interesting to
note that in 13 such incidents recorded at the diffusion plants over the
past 20 years, an average of about 44 kg of UF,., or approximately 30 kg
42 b
of uranium, was released.
Likelihood
At least one accident on record involved the failure of a valve bonnet
7
in the distillation area of a UF,. plant. Using the 20 plant-years of
73
-------
commercial conversion plant operation as a base, the one incident results
_o
in a likelihood of /-' 5 x 10 /plant-year.
Another approach to estimating the likelihood of such an accident is to
a 7 6
use the failure data on valve ruptures, 10 to 10 /hr. Assuming 300
day around-the-clock operation, and the order of 10 to 100 valves in
the process, one arrives at a range of 10 to 10 valve failures/plant
yr. Pipe failure likelihoods are an order of magnitude lower on a per
6
section basis, and thus can be neglected in comparison with valve fail-
ures.
Estimate: 1.3 x 10~2 Ci 238U3 6.2 x lO~4 Ci 235U3 3 1.4 x lO~2 Ci 2S4U3
—2 234 —9
1.3 x 10 Ci Th to air; Likelihood /& 5 x 10 /plant-yr.
E.I Release of Raffinate from the Waste Retention Pond
In the wet chemical solvent extraction process for producing uranium
hexafluoride, the raffinate stream from the solvent extraction step must
be permanently disposed of. Currently, the raffinate stream is combined
with miscellaneous liquid streams from the process, such as spent scrub-
ber solutions, is neutralized by ammonia, and permanently impounded in
earthen-walled retention basins. The impounded liquid is composed of
ammonium nitrate, nitric acids, and metallic salts, as well as soluble
radionuclides. If a dike failure were to occur, as much as eight million
gallons of contaminated water could be released to the watercourse.
Source Term
The measured concentrations of radionuclides in the waste retention pond
74-
-------
are given in Table 4-13.
There have been no recorded major releases in the four to five years of
operation of the UF,. conversion waste retention pond, although minor
b 41
seepage has been noted. Accordingly, the release d.ata related to urani-
um mill tailings ponds will be cited (see Section 4.2, Accident E.I).
From the tailings pond data, the average releases to the watercourse from
dike failure or flooding are 3000 m of liquids and 7.5 x 10 Ibs of
solids. The solids, however, are assumed to deposit in the vicinity of
the entry point to the watercourse.
Likelihood
There has been no major release recorded during the four to five years of
41
operation of the UFg conversion waste retention pond. From the mill
tailings pond data (see Section 4.2, accident E.I), the likelihood of a
release involving the source term described in the previous section is
_2
approximately 2 x 10 /plant-yr of mill operation. As pointed out in
Section 4.2, however, this estimate may be high, since the evaluation of
early dike construction led to strengthening of existing dikes and im-
proved construction standards.
_A f) ?Q s* o -7 c
Estimate: Source term
-------
TABLE 4-13
RADIONUCLIDE CONCENTRATIONS IN
7
UFC CONVERSION WASTE RETENTION POND
b
Radionuclide Concentration
Ra-226 1 x 10"6
Th-230 1 x 10"8
Uranium 1.2 x 10
76
-------
4.4 Enrichment
The accidents considered in enrichment facilities, keyed to the accident
categories given in Table 4-1; are listed in Table 4-14.
Small releases resulting from occasional leaks in piping or valves are
considered to be credited to the normal operation of the plant. The
potentially toxic effects resulting from the postulated release of an-
hydrous hydrogen fluoride is considered outside of the scope of the cur-
rent study. Natural or man-induced disasters, such as tornados, earth-
quakes, floods, or missile impacts are highly unlikely, although their
occurrence could result in varying releases of activity. The risks
associated with these events were not analyzed in the current study.
Table 4-15 provides historical data on the three gaseous diffusion en-
43
richment complexes currently in operation.
B.I Catastrophic Fire
Gaseous UFg from leaks in the process stages can combine with oil vapors
from pumps or other machinery if the ventilation system allows these vapors
to accumulate in parts of the process building. The reaction is explo-
sive, following an equation of the type:
[CH2]3 + UFg + 402 — * 6HF + U02
or others in which UCLF^ and H20 can be generated. The heat generated
can melt the roof and floor structures, and release the uranium in pro
cess in the stages involved in the fire.
77
-------
TABLE 4-14
ENRICHMENT PLANT ACCIDENTS
B.I Catastrophic fire
C.I Release from a hot UFg cylinder
C.2 Significant leaks or failure of
valves or piping within the plant
D.I Criticality
78
-------
Plant
Oak Ridge
Paducah
Portsmouth
TABLE 4-15
43
EXISTING ENRICHMENT PLANT DATA
Capacity
(millions
of kg SWU
in 1970)
4.73
7.31
5.19
Enrichment
Range
(% U-235 in
1970)
0.7-4
0.7-1.1
0.7-97.7
Completion
Dates
1945-1954
1953-1954
1955-1956
Years of
Operation
~n
-23
~21
Total —75
79
-------
Source Term
Although catastrophic fires have occurred in enrichment plants, resulting
31
in significant property damage, estimates of the resulting quantity of
uranium released to the environment are not available.
At least one analytical estimate has been made of the consequences of a
16
catastrophic fire in an enrichment plant. It was assumed that the con-
tents of the 16 stages comprising a cell are released to the environment.
The resulting release of uranium was estimated to be 1550 kg, with a
composition characteristic of the feed.
In determining the isotopic composition of the feed, it was assumed that
in the generic enrichment plant, 90% of the feed is natural uranium and
10% recycled uranium from reprocessing LWR fuel. Traces of fission pro-
ducts are also carried along with the recycled uranium, but their activity
is sufficiently low to be neglected. Thus the estimated isotopic activi-
ties of the feed are given in Table 4-16.
Likelihood
From chemical industry data, the probability of major fires is estimated
-4 8
to be 4 x 10 /plant-year. However, at least two major fires, each re-
sulting in property damage in excess of $2 million, have occurred at
31
the Paducah enrichment plant. Referring to Table 4-15, there have been
/•-75 plant-years of enrichment plant operating experience. Thus, from
the historical incidents, the likelihood of a major fire in the enrich-
_2
ment plant is approximately 3 x 10 /plant-year.
ao
-------
TABLE 4-16
ISOTOPIC ACTIVITIES OF THE ENRICHMENT
PLANT FEED
Isotope Activity per Gram of Uranium (Ci)
238 7
"°U 3.31 x 10 '
237U 2.43 x 10"7
236U 3.02 x 10'8
235U 1.57X10'8
234U 4.06 x 10"7
234Th 3.31 x 10"7
81
-------
Estimate: Source term ~ 5.1 x 10~1 Ci 2Z8U3 3.8 x I0~l Ci 23?U,
4.7 x 10~2 Ci 236U, 2.4 x 10~2 Ci 235US 6.3 x lO~l Ci 2S4U, 5.1 X
— 1 234 —2 —4
10 Ci Th to air; Likelihood '3J 3 x 10 to 4 x 10 /plant-year.
C.I Release from a Hot UFg Cylinder
D -*
Natural uranium hexafluoride from the conversion facility and recycle
uranium hexafluoride from the reprocessing plant arrive at the enrichment
facility in cylinders which must be heated in autoclaves in order to ef-
fect the transfer of material. This takes place at the feed vaporization
operation. Valve rupture at this stage would have no other consequences
than a negligible release inside the building because the UFg is in the
solid phase. Once the cylinder has been heated in the sealed autoclave,
significant releases of UFg outside of the process containment system
are not possible. Thus accidents involving feed cylinders are not cred-
ible sources of significant environmental releases. In contrast, product
cylinders, containing 2.2 MT of UFg, and tails cylinders, containing up
to 12.7 MT of UFg, are carried out of the autoclave to a cylinder storage
area while they are still hot. The failure of a valve, an operator error,
or rupture of the cylinder itself while the contents are still in the
liquid phase could release significant quantities of UFg to the envi-
ronment.
Source Term
A summary of the releases over 5 kg associated with the filling or feed-
ing of UFg cylinders at the diffusion plants over the period 1969-1973
38,39
was presented in Table 4-12. The range of releases falls between
82
-------
approximately 10 and 400 kg uranium. The average release in six recorded
incidents was 108 kg of uranium.
The predicted specific activities by isotope of product and waste cylinders,
assuming that 90% of the feed is natural uranium and 10% is recycled
235
uranium, and that the product U enrichment is 2.6% and the tails compo-
235
sition is 0.25% U, are given in Table 4-17. Although the type of cyl-
inder involved in the release is not specified in the historical release
data given in Table 4-12, it is estimated that an enrichment facility
operating at a capacity of 8.75 x 10 kg SWU/yr would require the filling
of approximately 2000 product cylinders and 1300 waste cylinders. Assum-
ing that the likelihood of a release from a particular cylinder is pro-
portional to the number of operations involving that type of cylinder,
the weighted averages of the specific activities given in Table 4-17 are
as follows: 3.28 x 10"7 Ci/gm of 238U, 2.92 x 10"7 Ci/gm of 237U, 4.86
x 10"8 Ci/gm of 236U, 3.55 x 10"8 Ci/gm of 235U, and 1.40 x 10"6 Ci/gm
of 234U.
Likelihood
From the historical data on releases at the gaseous diffusion facilities
summarized in Table 4-12, a release probability of 0.4/plant-year is
derived for the six releases on record.
Estimate: Source term ^ 3.5 x 10~2 Ci 238U^3.2 x lO~2 Ci 23?U,
5.3 x 10~3 Ci 236U3 3.8 x 10~3 Ci 235U, 1.5 x lO~l Ci 234U to air;
Likelihood & 4 x 10~ /plant-year.
83
-------
TABLE 4-17
SPECIFIC ACTIVITIES BY ISOTOPE OF PRODUCT
AND TAILS CYLINDERS AT THE
ENRICHMENT PLANT
Isotope Activity (Ci/gram Uranium)
Product Tails
(2.2 MT Cylinder) (12.7 MT Cylinder)
238U 3.24 x 10"7 3.33 x 10"7
237U 3.73 x 10"7 1.71 x 10"7
236U 7.11 x 10"8 1.48 x 10"8
235U 5.56 x 10"8 5.39 x 10"9
234U 2.26 x 10"6 9.67 x 10"8
84
-------
C.2 Significant Leaks or Failure of Valves or Piping Within the Plant
Approximately 1100 stages are required for the enrichment of
ooc
natural uranium to 2.6% U for a tails assay of 0.25%. Each stage
requires extensive piping and valves to route the compressed gas through
the cascade. Releases of UFg occur from time to time due to valve or
piping failures, or significant leaks.
Source Term
An average of about 2.4 Kg of uranium was released in the plant from
significant leakage or the failure of valves or piping in the three
38,39
gaseous diffusion facilities over the period 1969 through 1973.
It is assumed that the isotopic composition of uranium released in such
incidents was characteristic of the feed (see Table 4-16).
Likelihood
There were 27 recorded incidents of releases from the three gaseous
diffusion facilities in the period 1969 through 1973 resulting in the
release of greater than 100 grams of uranium. On this basis, assuming
that the release probability is independent of the plant capacity, the
likelihood of such releases is approximately 1.8/plant-year.
Estimate: Source term~7.9 x 10~4 Ci 238U3 5.8 x 10~4 Ci 2S7U, 7.3 x
10~5 Ci 236U, 3.8 x 10~5 Ci 235U, 9.7 x lO~4 Ci 234U3 7.9 x 10~4 Ci
224
Th to air; Likelihood22 1. 8/plant-year.
D.I Criticality
85
-------
Although process configurations and administrative procedures are care-
fully designed and monitored with criticality control in mind, the
possibility of the formation of a critical mass, accompanied by the
evolution of ionizing radiation and fission products, is always a
possibility when handling enriched uranium. For the low enrichments
involved in LWR fuel processing, and the low densities of uranium handled
in the enrichment operation, inadvertant formation of a critical mass is,
indeed, a remote possibility. Water moderation would be required
accompanied by a rearrangement in configuration, both of which are pre-
vented by deliberate design and administrative procedures.
Source Term
31
For the 26 criticality incidents on record, the total number of fissions
15 19
range from —- 3 x 10 to 4 x 10 . For the 11 non-solution critical-
;
ities, considered more representative of a postulated criticality
incident for enrichment, the average number of fissions is ^-10 .
The radionuclides released ten minutes after an incident involving 10
44
fissions, based upon ORIGEN calculations, are given in Table 4-18.
These estimates further assume 100% of the noble gases, 50% of the
halogens, and 0.2% of the actinides released to the environment.
Additionally, the neutron and gamma radiation associated with a burst
of 10 fissions would result in a dose to the population. The method-
ology for evaluating this direct "shine" dose is discussed in Section
3.2.
86
-------
TABLE 4-18
RADIONUCLIDE RELEASE RESULTING FROM A
CRITICALITY INCIDENT AT THE ENRICHMENT PLANT
Nucllde Activity Released Nuclide Activity Released
(CO (CO
Br 80
Br 80m
Br 82
Br 82m
Br ,83
Br 84
Br 84m
Br 85
Br 86
Br 87
Kr 83m
Kr 85
Kr 85m
Kr 87
Kr 88
Kr 89
Kr 90
I 128
I 130
I 131
I 132
I 133
I 134
I 135
1.2 x 10"2
5.7 x 10"6
1.9 x 10"6
3.2 x 10"4
3.5 x 10"1
3.5
1.5 x 10"1
7.2
1.5 x 10"1
2.1 x 10"1
3.4 x 10"2
3.2 x 10"6
1.3
9.4
6.3
5.2 x 101
9.2 x 10'3
1.4 x 10"4
1.0 x 10"4
5.5 x 10"3
1.6 x 10"1
1.5 x 10"1
5.2
2.3
I 136
I 137
Xe 133
Xe 133m
Xe 135
Xe 135m
Xe 137
Xe 138
Xe 139
Xe 140
Th 231
Th 234
Pa 234m
U 233
U 234
U 235
U 236
U 237
U 238
U 239
Np 237
Np 239
Np 240
Pu 239
2.3
3.3 x 10"5
1.7 x 10"4
9.9 x 10"6
2.2 x 10"1
4.7 x 10"1
9.0 x 101
7.0 x 101
1.5 x 10"1
2.2 x 10"8
2.0 x 10~9
2.2 x 10"10
1.8 x 10"10
1.7 x 10'16
3.4 x 10'14
4.5 x 10"7
1.3 x 10"12
7.1 x 10"6
1.1 x 10"6
6.1 x 10"1
4.4 x 10"17
1.5 x 10"3
3.5 x 10"12
4.1 x 10"13
87
-------
Likelihood
There have been no criticality incidents in the 75 plant-years of
uranium enrichment. Moreover, there has not been a criticality inci-
dent recorded involving low enrichment uranium. The only criticality
incidents on record involved the handling of plutonium, highly enriched
uranium (> 83% enriched in 235U), or 233U. J
An estimate has been made of the probability of criticality in fuel
8
fabrication plants. From the four recorded incidents, and the estimated
432 (increased to^490 through 1975) plant-years of production involving
_3
uranium and plutonium fuel fabrication, a probability of A) 8 x 10
criticality accidents/plant-year has been derived. As discussed in
Reference 8, an improved basis for such an estimate would consider the
total fuel throughput, the fuel forms during processing, and the fuel
reactivities involved.
The only critical!ties on record in fuel fabrication or reprocessing
facilities occurred in solution. Dry criticalities have only occurred
in reactor experiments. Since enrichment operations are dry, the like-
lihood of criticality is estimated to be at least an order of magnitude
lower than the above estimate. Moreover, the low enrichment associated
with LWR fuel is estimated to reduce the likelihood of criticality by
at least another order of magnitude.
Estimate: Source Term - Radionuolides given in Table 4-18 to air plus
ionizing radiations (see Section 3.2 for dose methodology); Likelihood
^ 8 x 10~ /plant-year.
88
-------
4.5 Uranium Fuel Fabrication
The accidents considered in uranium fuel fabrication facilities, keyed to
the accident categories given in Table 4-1, are listed in Table 4-19.
The risk associated with other incidents, some of which have occurred in
the past, is judged to be insignificant in comparison with the accidents
considered in Table 4-19. These include spills of U02 powder, sintering
furnace explosions, autoclave explosions, inadvertent release of liquid
wastes, and ventilation problems from, for example, loss of electrical
power. These have resulted in excessive airborne concentrations in work
areas, but the release to the environment is generally inconsequential.
Filter failures in the process ventilation stream have occurred from time
32
to time, but the filters are generally replaced before the time inte-
grated release becomes a significant fraction of the annual release from
normal operations.
A tornado which is capable of demolishing the building structures could
disperse significantly large quantities of respirable uranium to the
environment,but the risk associated with such an accident has not been
evaluated due to lack of pertinent data.
A list of government contractors and commercial firms engaged in fuel
Q
fabrication activities since 1942 is given in Table 4-20. The levels of
production and the nature of the processes vary substantially. Also,
some of the commercial firms have conducted activities at more than one
plant, although they are listed only once in the Table. Nevertheless,
89
-------
TABLE 4-19
URANIUM FUEL FABRICATION ACCIDENTS
A.I Hydrogen explosion in reduction furnace
B.I Major facility fire
B.2 Fire in a roughing filter
C.I Release from a hot UFg cylinder
C.2 Failure of valves or piping within the
plant
D.I Criticality
E.I Waste retention pond failure
90
-------
TABLE 4-20
ESTIMATE OF PLANT-YEARS OF PRODUCTION SINCE 1942
8
INVOLVING FUEL FABRICATION
Plant Estimated
Dates
Hanford 1944-1975
Savannah River Laboratory 1954-1975
Los Alamos Scientific Laboratory 1943-1975
National Lead Company of Ohio 1944-1975
Oak Ridge National Laboratory 1943-1975
Lawrence Radiation Laboratory 1949-1975
Argonne National Laboratory 1949-1975
Aerojet General Nuclear 1955-1970
Atomics International 1955-1975
Babcock & Wilcox 1957-1975
Clevite Research Corporation 1957-1969
Combustion Engineering 1955-1975
Curtiss-Wright Davison 1955-1975
Gulf General Atomic 1958-1975
General Electric 1955-1975
Gulf United Nuclear 1971-1975
M & C Nuclear, Incorporated 1961-1965
Exxon 1971-1975
Mallinckrodt Nuclear Corporation 1960-1970
Martin Company 1960-1370
Kerr-McGee 1969-1975
National Carbon Company 1960-1965
National Lead Company 1962-1975
Engelhard Industries, Incorporated 1957-1970
Nuclear Development Corporation of
America ' 1957-1968
Nuclear Materials and Equipment
Corporation 1960-1971
Sylvania-Corning Nuclear Corporation 1960-1968
Westinghouse Electric Corporation 1955-1975
United Nuclear 1957-1975
U.S. Nuclear 1972-1975
Nuclear Fuel Services 1966-1975
Estimated
Plant-Years
31
21
32
31
32
26
?6
15
20
18
12
20
20
17
20
4
4
4
10
10
6
5
13
13
11
11
8
20
18
3
9
Total
490
91
-------
the total number of plant-years may be used for rough estimates of
accident probabilities from historical records on incidents.
A.I Hydrogen Explosion in Reduction Furnace
In the process for producing uranium dioxide from uranium hexafluoride,
the gaseous UFg is hydrolyzed to uranyl fluoride and reacted with ammonia
to precipitate ammonium diuranate (ADD). The ADU slurries are concen-
trated and then calcined to form tUCL. The U-0Q is reduced to uranium
o o o o
dioxide at a temperature of approximately 1000°F in a reducing atmosphere
of hydrogen. The hydrogen concentration is controlled to prevent the
buildup of an explosive atmosphere. However, should these controls fail,
the hydrogen could ignite and explosively blow the uranium out of the
furnace.
Source Term
The rotary kiln reduction furnace would typically contain in excess of
100 kg uranium dioxide. The consequences of explosion accidents are
limited by the material that can be maintained in the air rather than by
the total volume or mass of material involved in the explosion. The air-
borne concentration of heavy particles in the respirable range appears to
o 8
be limited to approximately 100 mg/nr . Thus for a room volume assumed
4 3
to be approximately 10 m , the quantity of uranium released
* A reducing atmosphere is also used later in the fabrication process
during the sintering of pellets. However, by virtue of the integral
form of the uranium dioxide at this stage of the process, the conse-
quences of a postulated explosion in the sintering furnace is judged
to be insignificant in comparison with a similar event in the reduc-
tion furnace.
92
-------
to the building ventilation system would be approximately 1 kg of
uranium.
Uranium fuel fabrication plants have at least one high efficiency par-
ticulate air (HEPA) filter in the process ventilation system. However,
the existence of HEPA filters in the building air ventilation system is
not assured. Some of the existing plants have one HEPA filter in the
building ventilation system and others have none. It is not clear at
the present time whether new uranium fuel fabrication plants will be
required to incorporate a HEPA filter in the building ventilation system.
The efficiency of a single HEPA filter against particles in the respir-
8
able size range is taken to be 99.9%. However, because of the variability
in the design of building ventilation systems, a source term range is
adopted for this accident of 1 to 1000 gms uranium. The isotopic com-
position of the uranium is taken to be that of the product from the enrich-
ment plant, given in Table 4-17.
Likelihood
An estimate of 3:5 x 10" /plant-year has been made for the likelihood of
8 **
a hydrogen explosion in a fuel fabrication sintering furnace. The same
study derived an estimate of AJ 10" /plant-year for chemical explosions in
general.
* Only one-jplant has two HEPA filters in the building air ventilation
system. ^
** The facility associated with this estimate was a 300 MT HM/yr mixed
oxide plant which possesses a significantly lower annual furnace
throughput than the generic uranium fuel fabrication plant selected
for this study.
93
-------
At least one accident has occurred Cat a sintering furnace) resulting from
8,31
detonation of an explosive mixture of hydrogen and oxygen. On the
basis of the ^490 plant-years of fuel fabrication experience, this would
result in a likelihood of 7 2 x 10"3/plant-year.
45
In another accident evaluation, the probability of a hydrogen explosion
in the sintering furnace has been crudely estimated to be lower than 10
to 10~ /year and higher than 10" to 10 /year.
A 79?P 4
Estimate: Source term % 3.2 x 10 to 3.2 x 10~ Ci V, 3.7 x 10 to
3.7 x 10~? Ci 23?U3 7.1 x 10~5 to 7.1 x 10~8 Ci 236U, 5.6 x 10~5 to 5.6 x
10~8 Ci 235U> 2.3 x 10~3 to 2.3 x 10~6 Ci 234U to air; Likelihood & 5 x
—2 -3
10 to 2
-------
It is estimated that, in the event of a major facility fire, as much as
8,42
U of the dispersible U(L powder could be rendered airborne. The
exhaust fans would continue to operate, sweeping the powder out through
the ventilation system. However, the final HEPA filter barrier, if
available, would continue to function with an estimated efficiency of
8
99.9%. As discussed in the previous section, the variability in building
ventilation system designs requires the consideration of a range for the
source term. Thus the amount of uranium released to the environment from
this accident is estimated to be in the range of 1 to 1000 kg. The
isotopic composition of the uranium is taken to be that of the product
from the enrichment plant, given in Table 4-17.
Likelihood
The likelihood of a major fire in a mixed oxide fabrication plant, which
is similar in design to the uranium fabrication facility considered here,
is estimated to be^ 2 x 10 /plant-year. Although a catastrophic facility
31
fire has occurred at the Rocky Flats facility in 1969, this incident is
not considered germane to the current estimate, since metallic plutonium
42
was the probable source of ignition.
_7 A 9 7Q 7
Estimate: Source term $ 3.2 x 10 to 3.2 x 10 Ci U3 3.7 x 10 to
—A 9?7 P 5 ? ^fi 9
3.7 x 10 Ci ° U3 7.1 x 10 to 7.1 x 10 Ci V3 5.6 x 10~ to 5.6
x 10~S Ci 225UJ 2.3 to 2.3 x 10~3 Ci 234U to air; Likelihood? 2 x ld~4/plant-yr.
B.2 Fire in a Roughing Filter
Roughing filters are installed in the exhaust plenums from dusty oper-
ations, such as the oxide milling station, in the fuel fabrication plant.
95
-------
In addition to the uranium oxide powder, small amounts of lint, lubri-
cating materials, or other combustible materials may be trapped in these
filters. A fire could be started by an electrical spark or static elec-
tricity, resulting in the destruction of the filter and the release of
its entire uranium inventory.
Source Term
It is assumed that the inventory in the enclosure surrounding the ball
mill served by the roughing filter is approximately 25 kilogram^ of
uranium dioxide. Although the maximum filter loading may be as high as
16
5% of the total enclosure inventory, it is assumed that the nominal
filter loading'is 1% of the inventory, all in the respirable range. The
fire is assumed to release to the building air all of the powder trapped
within the filter.
As discussed earlier, if a HEPA filter is included within the building
ventilation system, a reduction in the amount of uranium released to the
8
environment by 99.9% is estimated. However, since many fabrication
plants do not incorporate HEPA filters in the building air exhaust
systems, a range of values is selected for the source term. Thus the
quantity of uranium released to the environment from this accident is
estimated to fall within the range of 0.25 to 250 gms. The isotopic
composition of the uranium is taken to be that of the product from the
enrichment plant, given in Table 4-17.
96
-------
Likelihood
The likelihood of a local fire in a mixed oxide fuel fabrication facility
has been estimated to be*£ 10"2/plant-year. According to the incidents
on record, at least five local fires (exclusive of ion-exchange resin fires)
32
have occurred in the 490 plant-years of fuel fabrication activities.
C Q f) 70 _ C
Estimate: Source term%8.l x 10~ to 8.1 x 10~ Ci Us 9.3 x 10 to
9.3 x 10~8 Ci 227U3 1.8 x 10~5 to 1.8 x 10~8 Ci 226V3 1.4 x 10~5 to
Dp^r A 7 914
1.4 x 10 ° Ci V3 5.7 x 10 to 5.7 x 10 Ci U to air; Likelihood
-2
*£ 10 /plant-year.
C.I Release from a Hot UF6 Cylinder
Enriched uranium is received at the fabrication facility in a cylinder
containing 2.2 MT of uranium hexafluoride. The cylinder is placed in
a steam-heated chest, where the uranium hexafluoride is vaporized and
passed directly to the first step of the conversion process. Should the
cylinder be overpressurized, a rupture could cause the entire contents
of the cylinder to be released into the vaporization room. A more prob-
able accident would be the development of a leak, or an operator error
resulting in an inadvertently backed out valve, in which case a portion
of the liquid would evaporate to replace the escaping gas. Evaporation
would continue until enough heat is removed by the escaping gas, or
by forced cooling through a water-spray protective system, to cause the
remaining UFg to solidify.
97
-------
Source Term
If a cylinder containing 2.2 MT of UFg were to rupture or lose a valve,
and if the release were to go unchecked until the contents of the cylinder
solidified, analyses result in estimates of approximately 700 kg of
7
uranium released to the vaporization room. However, the most probable
release is not likely to coincide with the maximum release. A summary
of releases associated with hookup and disconnect operations at the
gaseous diffusion plants in the period 1969 through 1973 is given in
Table 4-12. The average release in six recorded incidents resulting in
more than 5 kg uranium loss was 108 kg of uranium. Since the data
base associated with enrichment is more extensive than that associated
with fabrication operations, and since the UFg cylinder hookup and dis-
connect operations are similar, the average source term from enrichment
will be adapted here.
It is further assumed that the HEPA filter in the process ventilation
system becomes plugged with hydrolized UFg, allowing the UFg and its
reaction products, UO^Fp and HF, to seep from the vaporization room.
If a HEPA filter is incorporated within the building ventilation system,
a reduction in the amount of uranium released to the environment by
8
99.9% is estimated. However, since many fabrication plants do not have
HEPA filters in the building air exhaust systems, a range of releases,
from .108 to 108 kg is estimated. The isotopic composition of the
uranium is taken to be that of the product from the enrichment plant,
given in Table 4-17.
98
-------
Likelihood
At least two incidents involving inadvertent releases of UFg from
31 32
cylinder operations have been recorded in fabrication facilities.
On the basis of theA^490 plant-years of fuel fabrication experience,
the two recorded incidents would result in a UFC release likelihood of
o
Sf 8 x 10"3/plant-year.
From the incidents on record at the three enrichment facilities, the
likelihood of a release in handling a cylinder was estimated to be
approximately 5 x 10" /cylinder (see Section 4.3). The generic uranium
fabrication plant operating at a nominal capacity of 900 MTU/yr would
require the emptying of approximately 600 feed cylinders annually.
Then, on the basis of the release likelihood developed from enrichment
operations, the likelihood of a release at the fabrication plant is
_2
approximately 3 x 10 /plant-year.
45
In another accident evaluation, the probability of a UFg cylinder
release has been crudely estimated to lie in the range of 10"1 to 10"3/
year.
* Assuming, however, that only about one-half of the facilities utilize
UF as the feed material.
99
-------
p r P^fl 9
Estimate: Sourae term t& 3. 5 x 10~ to 3.5 x 10 Ci U, 4.0 x 10~
*> 9?7 ? — # P^fl 7
to 4.0 x 10 Ci Uf 7.7 x 10 to 7.7 x 10 Ci Uf 6.0 x 10~ to
6.0 x 10~6 Ci 235U3 2.4 x 10'** to 2.4 x IQ~* Ci 2Z4U to air; Likelihood
—2
g 3 x 10 /plant-year.
C.2 Failure of Valves or Piping Within the Plant
In the head end of the fuel fabrication process, the vaporized UFg is
hydrolized to uranyl fluoride and then reacted with ammonia to form
ammonium diuranate. Releases of UFC occur from time to time due to
o
valve or piping failures, or significant leaks.
Source Term
In an incident on record, a valve on a newly installed unit was inadver-
tently left open, resulting in approximately 30 kg of uranium as UFfi
32 D
released from the process equipment. The magnitude of this release is
comparable to the recorded incidents in other components of the fuel
cycle. For example, in the recorded incidents at the AEC diffusion plants
over the past 20 years, an average of about 44 kg of UFfi was released
42
within the building.
If a HEPA filter is incorporated within the building ventilation system, a
reduction in the amount of uranium released to the environment by 99.9% is
estimated. Since, however, the existence of a HEPA filter is not assured,
a range of releases, from 440 to 44000 gms» is estimated. The isotopic
composition of the uranium is taken to be that of the product from the
enrichment plant, given in Table 4-17.
100
-------
Likelihood
At least one incident involving an inadvertent release of UFfi from the
b 32
valves or piping within a fabrication plant has been recorded. On the
*
basis of the ^ 490 plant-years of fuel fabrication experience, the one
o
incident on record would result in a UFg release likelihood of<> 4 x 10 /
plant-year.
Estimate: Source term -Z 9.7 x 10~3 to 9.7 x 10~6 Ci 238U3 1.1 x 10~2 to
l.l x 10~5 Ci 237U> 2.1 x 10~3 to 2.1 x 10~6 Ci 236U3 1. 7 x 10~3 to
1.7 x 10~6 Ci 235U3 6.8 x 10~2 to 6.8 x 10~5 Ci 234U to air; Likelihood
> 4 x 10 /plant-year.
/
D.I Criticality
Nuclear criticality safety is a consideration in the design, operation
and licensing of fuel fabrication plants. Equipment is designed to be
maintained in a safe geometry or to contain fixed nuclear poisons to
prevent criticality. For the low enrichments involved in LWR fuel pro-
cessing, the inadvertent formation of a critical mass is a remote
possibility. Nevertheless, an incident in which interlocks have failed
and a double batch of material has come together can be conceived. The
most likely point in the plant for this to occur is at the head end con-
version of UFg to U02, or in scrap recovery, where low enriched uranium
is processed in a water solution.
* Assuming, however that only about one-half of the facilities utilize
UF as the feed material.
101
-------
Source Term
31
For the ten solution criticalities on record, the average number of
18
fissions was i^ 4 x 10 . However, one of the incidents Involved a
19
record */ 4 x 10 fissions at the Chemical Processing Plant of the
Idaho Reactor Testing Area, considered unrepresentative of the fuel
fabrication operations. Neglecting this incident, the average number
17 18
of fissions is
-------
TABLE 4-21
RADIONUCLIDE RELEASE RESULTING FROM A CRITICALITY
INCIDENT AT THE URANIUM FUEL FABRICATION FACILITY
Activity Released Activity Released
Nuclide (Ci) Nuclide (Ci)
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Kr
Kr
Kr
Kr
Kr
Kr
Kr
I
I
I
I
I
I
I
80
80m
82
82m
83
84
84m
85
86
87
83m
85
85m
87
88
89
90
128
130
131
132
133
134
135
1.2
5.7
1.9
3.2
3.5
3.5
1.5
7.2
1.5
2.1
3.4
3.2
1.3
9.4
6.3
5.2
9.2
1.4
1.0
5.5
1.6
1.5
5.2
2.3
x 10
x 10
x 10
x 10
x 10
x 10
x 10
x 10
x 10
x 10
x 10
-1
-5
-5
-3
1
1
-1
-5
1
1
1
x IO2
x 10
x 10
x 10
x 10
x 10
x 10
-2
-3
-3
~2
1
1
I
I
Xe
Xe
Xe
Xe
Xe
Xe
Xe
Xe
Th
Th
Pa
U
U
U
U
U
U
U
Np
Np
Np
Pu
136
137
133
133m
135
135m
137
138
139
140
231
234
234m
233
234
235
236
237
238
239
237
239
240
239
2.3
3.3
1.7
9.9
2.2
4.7
9.0
7.0
1.5
2.2
2.0
2.2
1.8
1.7
3.4
4.5
1.3
7.1
1.1
6.1
4.4
1.5
3.5
4.1
x
x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
to
X
X
X
X
IO1
10"4
io-3
IO-5
io2
io2
io-7
ID'8
io-9
io-9
io-'5
ID'13
io-6
io-'1
io-5
io-5
6.1
io-16
ID'2
io-11
io-12
to 2.0 x
to 2.2 x
to 1.8 x
to 1.7 x
to 3.4 x
to 4.5 x
to 1.3 x
to 7.1 x
to 1.1 x
x IO"3
to 4.4 x
to 1.5 x
to 3.5 x
to 4.1 x
io-11
io-12
io-12
io-18
io-16
io-9
io-14
ID'8
io-8
io-19
io-5
10"14
ID'15
103
-------
historical data base, a probability of criticality of^ 8 x 10 /plant-
year is derived for fuel fabrication operations. As discussed in
Reference 8, an improved basis for such an estimate would consider the
total fuel throughput, the fuel forms during processing, and the fuel
reactivities involved.
It is important to note, however, that these four incidents in fabri-
cation operations involved plutonium or fully enriched uranium (> 83%
poc
enriched in U), and that there has not been a criticality incident
recorded in any type of operation which involved uranium of low enrich-
ment. Thus the low uranium enrichment associated with LWR fuel is esti-
mated to reduce the likelihood of criticality by at least an order of
magnitude.
45
In another accident evaluation, the probability of a criticality
incident in a low enrichment fabrication plant is crudely estimated to
-1 -3
lie in the range of 10 to 10 /year.
Estimate: Source term - Eadionuclides given in Table 4-21 to air
plus ionizing radiations (see Section 3.2 for dose methodology);
rJ -4
Likelihood ^8 x 10 /plant-year.
E.I Waste Retention Pond Failure
The liquid effluents from the UFg to U02 conversion process and from
the scrap recovery operation, which contain strongly acidic or basic
chemical wastes and traces of radioactivity, are treated with lime to
form a calcium fluoride precipitate, and discharged to a waste retention
104
-------
pond.- The pond is sealed to prevent seepage to the ground water.
The surface of the pond is exposed to the atmosphere, allowing evapor-
ation to take place. If a dike failure were to occur, or through a
natural process, such as earthquake or flooding, several million gallons
of contaminated water could be released to the watercourse.
Source Term
At least one incident on record involving a leak in the retention dam
of a liquid waste lagoon resulted in a loss of approximately 1.4 million
gallons. From the mill tailings pond data (see Section 4.2, Accident
E.I), the average release to the watercourse from dike failure or flooding
is 3000 m3 (800,000 gallons).
The concentration of uranium in the liquid waste lagoon is estimated to
-5 45*
be approximately 2.5 x 10 gm/ml. Then from the isotopic composition
of uranium given in Table 4-17, the estimated concentrations of radio-
nuclides in the waste retention pond are given in Table 4-22.
Likelihood
At least one retention pond leak associated with fuel fabrication
activities has been documented, resulting in the loss of approximately
* This is obtained by scaling both the size of the waste lagoon and
the U02 production rate from the estimates in Reference 45, and
assuming a nominal 4-ft. water depth in the lagoon.
105
-------
TABLE 4-22
ESTIMATED CONCENTRATIONS OF -RADION'ICLIDES
IN WASTE RETENTION POND
Isotope Activity (Ci/ml)
U-238 8.1 x 10~12
U-237 9.3 x 10"12
U-236 1.8 x 10"12
U-235 1.4 x 10"12
U-234 5.7 x 10"11
Th-234 8.1 x 10"12
Pa-234 8.1 x 10"12
106
-------
1.4 million gallons of contaminated liquids. On the basis of 490
plant-years of fuel fabrication activities, this would result in a
_3
release likelihood ofrvZ x 10 /plant-year.
From the mill tailings pond data (see Section 4.2, Accident E.I), the
likelihood of a release involving the source term described in the
2
previous section is approximately 2 x 10 /plant-year of mill operation.
As pointed out in Section 4.2, however, this estimate may be high,
since the evaluation of early dike construction led to strengthening
of existing dikes and improved construction standards.
n f> 7O 9 ??7
Estimate: Source term ^2. 4 x 10~ Ci U3 2.8 x 10~ Ci U, 5.4
x 10~3 Ci 236U3 4.2 x 10~3 Ci 225V3 1.7 x I0~l Ci 23\ 2.4 x 10~2
91 A 9 9 Id -/>,
Ci Th, 2.4 x 10~ Ci Pa to watercourse; Likelihood^. 2 x 10
to 2 x 10~ ./plant-year.
107
-------
4.6 Reprocessing
The accidents considered in spent fuel reprocessing facilities, keyed
to the accident categories given in table 4-1, are listed in
Table 4-23.
Other accidents can be postulated to occur in reprocessing operations.
For example, fission gases could be released from the fuel pins should
the fuel cask be dropped during unloading or if the fuel element becomes
overheated during transfer to the shear operation. However, the risk
associated with these events is judged to be small in comparison with
the accident examined in the fuel receiving and storage area. Fires
involving leached zirconium hulls have occurred, but the duration, in
general, is short, and the risk is judged to be insignificant in com-
parison with the fires considered in Table 4-23. Small leaks in vessels
or pipes containing radioactive material, or operator errors resulting
in inadvertent discharges of radioactive solutions may occur relatively
32
frequently, but the source term to the environment is generally
negligible in comparison with the accidents analyzed in this section.
Similarly, filter failures or loss of ventilation zone differential
pressure may occur, but the small releases which result during the
abnormal condition are generally an insignificant fraction of the normal
annual release.
High level wastes are stored in solution for up to five years prior to
solidification. The radiolytic heat generated by the fission products
requires water cooling of the high level waste tanks. A loss of coolant
108
-------
TABLE 4-23
SPENT FUEL REPROCESSING ACCIDENTS
A.I Explosion in the high aqueous waste concentrator
A.2 Explosion in the low aqueous waste concentrator
A.3 Explosion in the high aqueous feed tank
A.4 Explosion in the waste calciner
A.5 Explosion in the iodine adsorber
B.I Solvent fire in the codecontamination cycle
B.2 Solvent fire in the plutonium extraction cycle
B.3 Ion-exchange resin fire
C.I Fuel assembly rupture and release in fuel receiving
and storage area
C.2 Dissolver seal failure
C.3 Release from a hot UFg cylinder
D.I Criticality
109
-------
to these tanks could result in the release of a significant amount of
,-i
activity, but is only remotely possible because of the defenses in depth
which would be operative. The risk associated with this accident was not
considered in this study due to a lack of pertinent data.
The potentially toxic effects resulting from the release of
acids or hydrogen fluoride to the environment is considered outside of
the scope of the current study.
Current regulatory criteria related to the design of a reprocessing
facility require that the structures, systems and components withstand
the effects of natural phenomena. These include all floods, tornados,
earthquakes, or missiles of intensity more severe than experienced
historically in the locality of the plant. Although events with inten-
sities outside of this range are conceivable, the exceedingly low prob-
abilities associated with their occurrence were not evaluated in the
current study.
Only one commercial reprocessing facility has operated in the United
States. This plant, which is located in West Valley, New York, was
operated by Nuclear Fuel Services in the period from 1966 to 1972.
Federally supported facilities which employ processes similar to current
designs for LWR spent fuel reprocessing plants, are in operation at
Hanford, Savannah River, Oak Ridge, and Idaho. If these facilities are
included, approximately 100 plant-years of experience have been
accumulated in the reprocessing of spent fuel.
no
-------
Most of the data presented in this section were developed in an earlier
9
hazards analysis of a generic fuel reprocessing facility. The Safety
46
Analysis, Report for the Barnwell Nuclear Fuel Plant was the source of
most of the release data presented ,in that study.
A.I Explosion in the High Aqueous Waste (HAW).Concentrator
The high aqueous waste concentrator, located in the remote process cell,
concentrates the high-level radioactive waste streams from all the
solvent extraction cycles to recover nitric acid and water for reuse
in the process, while reducing the waste volumes for storage in waste
facilities. An explosion in the HAW waste concentrator could con-
ceivably be caused by ignition of an explosive mixture of hydrogen in
the air above the liquid in the evaporator or a "red-oil" explosion.
Hydrogen and oxygen are generated by radiolysis of aqueous solutions.
To avoid reaching a combustible hydrogen concentration, dilution of
the off-gases with continously flowing air is used in the processing
operations. However, a failure of the air purge system through failure
of blowers or their power supply, filter blockages, or ventilation
control failure could result in a hydrogen explosion. To reduce the
likelihood of air flow failure, the plant is designed with redundant
air flow features.
Source Term
The estimated quantities of radionuclides released to the environment
9
in the event of a HAW concentrator explosion are given in Table 4-24.
These values are based upon an estimated concentrator volume of 600
111
-------
TABLE 4-24
RADIONUCLIDE RELEASE RESULTING FROM AN HAW CONCENTRATOR
9
EXPLOSION AT THE REPROCESSING FACILITY
Nuclide Activity 1n Fuel Activity Released*
(C1/MTHM) (C1)
Sr 89
Sr 90
Y 90
Y 91
Zr 95
Nb 95
Ru 103
Ru 106
I 129
I 131
Cs 134
Cs 137
Ce 141
Ce 144
Pm 147
Pu 238
Pu 239
Pu 240
Pu 241
Pu 242
Am 241
Am 242
Cm 242
Cm 243
Cm 244
9.0 x 104
8.4 x 104
8.4 x 104
1.9 x TO5
3.5 x 105
6.5 x 105
1.2 x 105
6.1 x 105
3.6 x 10"2
1.6
2.4 x 105
1.2 x 105
7.9 x 104
8.8 x 105
1.4 x 105
•3
4.3 x 10J
3.2 x 102
6.3 x 102
1.7 x 105
3.6
2.5 x 102
4.0
4.4 x 104
3.4 x 101
5.7 x 103
3.1 x 10"4
2.9 x 10"4
2.9 x 10'4
6.6 x 10"4
1.2 x 10"3
2.3 x 10"3
1.3 x 102
6.4 x 102
3.1 x 10"3
1.4 x 10"1
8.3 x 10"4
4.2 x 10"4
2.7 x 10"4
3.1 x 10"3
4.8 x 10"4
c
1.5 x 10 s
1.1 x 10"6
2.2 x 10"6
5.9 x 10"4
1.3 x 10"8
8.7 x 10"7
0
1.4 x 10 8
1.5 x 10"4
1.2 x 10"8
2.0 x 10"6
The values tabulated are for normal operation of the two series HEPA
filters. In the event of simultaneous failure of the filters, the
values in this column, with^the exception of iodine and ruthenium, are
increased by a factor of 10 . Since iodine is assumed to be a vapor,
the activity released is assumed to be unchanged. Ruthenium is part
112
-------
liters with an equivalence of 1.76 kg of heavy metal per liter. The
estimated fraction of radionuclides in the concentrate which were
2 7
originally present in the fuel are 8.3 x 10 for iodine, 10 for
Plutonium, and unity for the remainder. The estimated volatile fraction
o
of ruthenium is 10 and of iodine is unity. The amount of concen-
trate released to the cell is limited by the allowable concentration
, 8
of respirable heavy particles in the air, approximately 100 mg/m .
The remote process cell volume is taken to be 2850 m3. The estimated
fraction of non-volatile material passing through two HEPA filters in
5 8
series under normal conditions is 10 . In the event of simultaneous
failure of the two HEPA filters, all of the particulate activity dis-
charged to the cell is assumed to be released to the atmosphere.
Likelihood
The likelihood of an HAW concentrator explosion in a generic spent
fuel reprocessing facility with normal HEPA filter operation has been
c 9
estimated to be approximately 10 /yr. The likelihood of a simultaneous
failure of the two series HEPA filters has been further estimated to
-3 9
be roughly 10 /demand.
* vapor and part particulate. In the event of simultaneous filter
failure, thencuthenium activity released is assumed to be the
following: IUJRu - 1.7 x 10* Ci, T06Ru - 8.5 x 10Z Ci.
113
-------
Estimate: Source term: Radionuolide given in Table 4-24 to air for
normal filter operation and the event of filter failure; Likelihood
&10 /year for normal filter operation, 10~ /year for simultaneous
failure of two series HEPA filters.
A. 2 Explosion in the Low Activity Waste (LAM) Concentrator
The low activity waste concentrator,located in the high intermediate
level cell, concentrates the low-level radioactive waste streams from
all the solvent extraction cycles. An explosion in the LAW concen-
trator could also be caused by a hydrogen explosion or a "red-oil"
explosion. "Red-oil" is a material that can be formed from a heavy
metal nitrate and/or nitric acid solutions mixed with tributyl
phosphate solvent at temperatures exceeding 135°C. Under optimum
conditions, the reaction is explosive and oxides of nitrogen are
evolved. In order for a "red-oil" explosion to occur, several inde-
pendent instrument control failures which are designed to keep the
temperature in the waste concentrators below 135°C and to keep the
solvent out of the aqueous stream would have to occur.
Source Term
The estimated quantities of radionuclides released to the environment
9
in the event of an LAW concentrator explosion are given in Table 4-25.
These values are based upon an estimated concentrator volume of 1500
liters with an equivalence of 2.31 kg of heavy metal per liter. The
estimated fractions of radionuclides in the concentrate which were
114
-------
TABLE 4-25
RADIONUCLIDE RELEASE RESULTING FROM A LAW
9
CONCENTRATOR EXPLOSION AT THE REPROCESSING FACILITY
Nuclide Activity Released (Ci)*
Sr 89 5.0 x 10"'
Sr 90 4.7 x 10~7
Y 90 4.7 x TO"7
Y 91 1.1 x 10~6
Zr 95 2.0 x 10"5
Nb 95 3.6 x 10"5
Ru 103 8.4
Ru 106 4.3 x 101
I 129 4.0 x 10"3
I 131 1.8 x 10"1
Cs 134 1.3 x 10"6
Cs 137 6.7 x 10~7
Ce 141 4.4 x 10"7
Ce 144 4.8 x 10"6
Pm 147 7.8 x 10"7
Pu 238 1.1 x 10"8
Pu 239 8.1 x 10"10
Pu 240 8.1 x 10~10
Pu 241 4.3 x 10"7
Pu 242 9.1 x 10~12
Am 241 1.4 x 10"9
Am 242 2.2 x 10'11
Cm 242 2.5 x 10"7
Cm 243 1.9 x 10"10
Cm 244 3.2 x 10"8
The values tabulated are for normal operation of the two series HEPA
filters. In the event of simultaneous failure of the filters, the
values in this column, with thegexception of iodine and ruthenium,
are increased by a factor of 10 . Since iodine is assumed to be a
vapor, the activity released is assumed to be unchanged. Ruthenium
is part vapor and part particulate. In the event of simultaneous
filter failure,^the ruthenium activity re1eased,is assumed to be
the following: tuJRu - 9.1 Ci, Ru - 4.6 x 10 Ci.
115
-------
-2 -4
originally present in the fuel are 3.2 x 10 for iodine, 9 x 10 for
2 -3
plutonium, 2 x 10 for ruthenium, zirconium and niobium, and 2x10
for the remainder. The estimated volatile fraction of ruthenium is •
_o
10 and of iodine in unity. The amount of concentrate released to the
cell is limited by the allowable concentration of respirable heavy
38
particles in the air, approximately 100 mg/m . The high intermediate
3
level cell volume is taken to be 1510 m . The estimated fraction of
non-volatile material passing through the two HEPA filters in series
-5 8
under normal conditions is 10 . In the event of a simultaneous failure
of the two HEPA filters, all of the particulate activity discharged to
the cell is assumed to be released to the atmosphere.
Likelihood
The likelihood of a LAW concentrator explosion in a generic spent.fuel
reprocessing facility with normal filter operation has been estimated
-4 9
to be approximately 10 /year. The likelihood of a simultaneous
failure of the two series HEPA filters has been further estimated to
-3 9
be roughly 10 /demand.
Estimate: Source term: Radionuolides g-iven in Table 4-25 to air for
normal filter operation and in the event of filter failure; Likelihood
A *7
% 10~ /year for normal filter operation^ 10~ /year for simultaneous
failure of two series HEPA filters.
116
-------
A.3 Explosion in the High Aqueous Feed (HAF) Tank
The high aqueous feed tank, located in the high level cell, receives
the solution from the dissolution step and feeds the extraction steps
of the process. An explosion in the HAF tank could be caused by ignition
of an explosive mixture of radiolytically generated hydrogen in the air
above the liquid. The occurrence of an explosive concentration of
hydrogen would require a failure in the air purge system.
Source Term
The estimated quantities of radionuclides released to the environment
9
in the event of a HAF tank explosion are given in Table 4-26. These
values are based upon estimated fractions of radionuclides in the
solution which were originally present in the fuel of 0.5 for iodine
and unity for all the remaining radionuclides. The estimated volatile
fraction of ruthenium is 10" and of iodine is unity. The estimated
fraction of the non-volatile radionuclides which is dispersed to the
cell from the explosion is 5.9 x 10. The non-volatile radionuclides
are assumed to pass through one HEPA filter which is estimated to trap
8
99.9% of the respirable material. In the event of a filter failure,
all of the particulate activity discharged to the cell is assumed to
be released to the atmosphere.
Likelihood
The likelihood of a HAF tank explosion in a generic fuel reprocessing
facility with normal HEPA filter operation has been estimated to be
117
-------
TABLE 4-26
RADIONUCLIDE RELEASE RESULTING FROM A HAF TANK EXPLOSION
9
AT THE REPROCESSING FACILITY
Nuclide Activity Released (Ci)*
Sr 89 2.1 x 10"3
Sr 90 2.0 x 10"3
Y 90 2.0 x 10~3
Y 91 4.5 x 10"3
Zr 95 8.2 x 10"3
Nb 95 1.5 x 10"2
Ru 103 4.8 x 102
Ru 106 2.4 x 103
I 129 7.2 x 10~2
I 131 3.2
Cs 134 5.7 x 10"3
Cs 137 2.8 x 10"3
Ce 141 1.9 x 10"3
Ce 144 2.1 x 10"2
Pm 147 3.3 x 10"3
Pu 238 1.0 x 10"4
Pu 239 7.6 x 10"6
Pu 240 1.5 x 10"5
Pu 241 4.0 x 10"3
Pu 242 8.5 x 10"8
Am 241 5.9 x 10"3
Am 242 9.5 x 10"8
Cm 242 1.0 x 10"3
Cm 243 8.0 x 10"7
Cm 244 1-4 x 10"4
The values tabulated are for normal operation of the single HEPA filter.
In the event of failure of the filter, the values in this column, with,
the exception of iodine and ruthenium, are increased by a factor of 10 .
Since iodine is assumed to be a vapor, the activity released is assumed
to be unchanged. Ruthenium is part vapor and part particulate. In
the event of filter failure, however, the ruthenium activity released
is unchanged.
118
-------
-5 y
approximately 10 /year. The likelihood of a filter failure has been
p
estimated to be roughly 10 /demand.
Estimate: Source term: Radionuclides given in Table 4-26 to air for
normal filter operation and in the event of filter failure; Likelihood
•£ 10 /year for normal filter operation^ I0~7/year for filter failure.
A. 4 Explosion in the Waste Calciner
The concentrated wastes are fed from the HAW concentrator to the waste
calciner, where the high level wastes are calcined at a temperature in
excess of 450°C. The waste calciner is located in the remote process
cell. An explosion in the waste calciner could be initiated by ignition
of an explosive mixture of hydrogen in air or by an excess pressure
buildup in the steam supply. In either case, several independent
instrument control failures would have to occur.
Source Term
The estimated quantities of radionuclides released to the environment
9
in the event of a waste calciner explosion are given in Table 4-27.
The source term for this accident is identical to that of the HAW
concentrator explosion, with one exception. Since the calciner operates
at several hundred degrees centigrade, the amount of ruthenium volatil-
ized in the course of the accident is estimated to be approximately a
factor of ten higher than that volatilized in the course of the HAW
concentrator explosion.
119
-------
TABLE 4-27
RADIONUCLIDE RELtASE RESULTING FROM AN EXPLOSION
IN THE WASTE CALCINER AT THE REPROCESSING FACILITY
Nuclide Activity Released (Ci)*
Sr 89 3.1 x 10"4
Sr 90 2.9 x 10"4
Y 90 2.9 x 10"4
Y 91 6.6 x 10'4
Zr 95 1.2 x 10"3
Nb 95 2.3 x 10'3
Ru 103 1.3 x 103
Ru 106 6.4 x 103
I 129 3.1 x 10'3
I 131 1.4 x 10"1
Cs 134 8.3 x 10"4
Cs 137 4.2 x 10"4
Ce 141 2.7 x 10"4
Ce 144 3.1 x 10"3
Pm 147 4.8 x 10"4
Pu 238 1.5 x 10'5
Pu 239 1.1 x 10"6
Pu 240 2.2 x 10"6
Pu 241 5.9 x 10"4
Pu 242 1.3 x 10"8
Am 241 8.7 x 10"7
Am 242 1.4 x 10'8
Cm 242 1.5 x 10"4
Cm 243 1.2 x 10"8
Cm 244 2.0 x 10"6
The values tabulated are for normal operation of the two series HEPA
filters. In the event of simultaneous failure of the filters, the
values in this column, with thegexception of iodine and ruthenium,
are increased by a factor of 10 . Since iodine is assumed to be a
vapor, the activity released is assumed to be unchanged. Ruthenium
is part vapor and part particulate. In the event of simultaneous
filter failura, the ruthenium activity released is-assumed to be the
following: ia3Ru - 1.4 x 103 Ci, 106Ru - 6.6 x 10J Ci.
120
-------
Likelihood
The likelihood of a waste calciner explosion in the generic fuel
reprocessing facility with normal HEPA filter operation has been
estimated to be approximately 10 /year. The likelihood of a
simultaneous failure of the two series HEPA filters has been further
-3 9
estimated to be roughly 10 /demand.
Estimate: Source term: Radionuolides given in Table 4-27 to air for
normal filter operation and in the event of filter failure; likelihood
—6 —9
J5* 10 /year for normal filter operation, 10 /year for simultaneous
failure of two series HEPA filters.
A.5 Explosfon in the Iodine Adsorber
Silver zeolite sorbents are incorporated in the process off-gas treat-
ment to limit the release of radioiodine to the environment. A silver
reactor explosion could result from the introduction of ammonia to the
sorbent with resultant formation of an explosive azide compound.
Although ammonia is not used in the process, the inadvertent use of
plant reagent chemicals such as hydrazine or hydroxylamine nitrate,
which are used in the uram'um-plutonium partitioning and extraction
steps, could result in the formation of ammonia vapors.
Source Term
Assuming that the radioiodine present in 4 MT of partially dissolved
fuel were available for release to the atmosphere, and that one-half
121
-------
of the volatilized radioiodine is removed by the exhaust filter, the
9
estimated release to the environment of 1-131 is 6 Ci. Ratioing the
activities of 1-129 and 1-131 in ^a fuel, the estimated release of
1-129 is approximately 1.4 x 10 Ci.
Likelihood
The likelihood of an iodine adsorber explosion in the generic fuel
reprocessing facility has been estimated to be approximately 2 x 10" '
9
year.
7?7 _7 7pq
Estimate: Source term: 6 Ci of I and 1.4 x 10 Ci of I to
air; Likelihood $} 2 x 10~ /year.
B.I Solvent Fire in the Codecontanri nation Cycle
Codecontamination is the operation which removes most of the fission
products and other undesirable impurities from solutions of uranium
and plutonium without separating the uranium and plutonium components.
The solvent extraction cycle, which takes place in the high intermediate
level cell, employs 30 volume percent tributyl phosphate in a normal
parafinic hydrocarbon (dodecane) solvent. Because of the relatively
low flash point ( /^70°C) of the organic solvent, there exists the
potential for a solvent fire during this processing step. The operating
temperature is maintained below 70°C by temperature controls and the
flow rates are monitored to avoid spills and to maintain the desired
compositions in all feed and discharge streams of the equipment used.
A solvent fire could result from the failure of temperature control
which would allow the flash point to be reached.
122
-------
Source Term
The estimated quantities of radionuclides released to the environment
in the event of a solvent fire in the codecontamination cycle are given
9
in Table 4-28. These releases are based upon an estimated 378 liters
of solvent burned with an equivalence of 0.23 kg heavy metal per liter.
The estimated fraction of the radionuclides in the solvent which were
-2 ?
originally present in the fuel are 4 x 10 for iodine, 10 for
_3
Plutonium, ruthenium, zirconium and niobium, and 1 x 10 for the
remainder. The estimated volatile fraction of ruthenium is 10 and
of iodine is unity. It is assumed that 1% of the non-volatile radio-
nuclides in the burned solvent are dispersed by the fire. The estimated
fraction of the dispersed, non-volatile material passing through the
-5 8
two HEPA filters under normal conditions in series is 10 . In the
event of simultaneous failure of the two HEPA filters, all of the
particulate activity discharged to the cell is assumed to be released
to the atmosphere.
Likelihood
The likelihood of a solvent fire in the codecontamination cycle of a
generic fuel reprocessing facility has been estimated to fall in the
"a
-5 -6 9
range of 10 to 10 /year. However, from chemical industry data, the
probability of major fires per plant-year is estimated to be 4 x 10 .
This probability is reduced to 2 x 10 /plant-year in Reference 8 for a
generic mixed oxide fuel fabrication plant, in which defenses are
employed uncharacteristic of the typical chemical plant. The likelihood
123
-------
TABLE 4-28
RADIONUCLIDE RELEASE RESULTING FROM A FIRE IN THE
9
CODECONTAMINATION CYCLE AT THE REPROCESSING FACILITY
Nuclide Activity Released (Ci)
Sr 89 1.5 x 10~7
Sr 90 7.1 x 10"7
Y 90 7.1 x 10"7
Y 91 3.7 x 10"7
Zr 95 4.2 x 10"6
Nb 95 7.7 x 10"6
Ru 103 1.6
Ru 106 3.8 x 101
I 129 1.2 x 10'4
I 13T 1.7 x 10"4
Cs 134 1.6 x 10'6
Cs 137 9.8 x 10"7
Ce 141 8.7 x 10"8
Ce 144 4.9 x 10"6
Pm 147 1.1 x 10"6
Pu 238 3.7 x 10"7
Pu 239 2.7 x 10"8
Pu 240 5.4 x 10"8
Pu 241 1.4 x 10"5
Pu 242 3.1 x 10"10
Am 241 2.1 x 10"9
Am 242 3.4 x TO"11
Cm 242 1.6 x 10"7
Cm 243 2.9 x 10"10
Cm 244 4.8 x 10"8
The values tabulated are for normal operation of the two series HEPA
filters. In the event of simultaneous failure of the filters, the
values in this column, with the exception of iodine and ruthenium, are
increased by a factor of 105. Since iodine is assumed to be a vapor,
the activity released is assumed to be unchanged. Ruthenium is part
vapor and part particulate. In the event of simultaneous filter
failure, the ruthenium activity released is assumed to be the following:
103Ru - 1.8 Ci, 106Ru - 4.2 x 101 Ci.
124
-------
of a simultaneous failure of the two series HERA filters has been
-3 - 9
estimated to be roughly 10 /demand.
Estimate:' Source term: Radionuolides given in Table 4-28 to air for
normal filter operation and in the event of filter failure; Likelihood
fit 10~ to 10~ /plant-year for normal filter operation, 10~ to 10~ /
plant-year for simultaneous failure of the two series HEPA filters.
B.2 Solvent Fire in the Plutonium Extraction Cycle
The organic solution from the codecontamination step may be passed through
a partitioning column located in the plutonium product cell, where tetra-
valent plutonium is electrochemically reduced to the less extractable
trivalent state, and subsequently stripped into another aqueous nitric
*
acid stream containing hydrazine. Solvent extraction cycles are also
used for the partitioning operation in conjunction with various chemical
adjustments. As in the codecontamination step, a solvent fire is possible
in the partitioning process, and is avoided by limiting the operating
temperature of the process.
Source Term
The estimated quantities of radionuclides released to the environment in
the event of a solvent fire in the plutonium extraction cycle are given
* Alternatively, anion exchange could be used for partitioning plutonium
and uranium into separate streams.
125
-------
9
in Table 4-29. Radionuclides other than plutonium are considered neg-
ligible. These releases are based upon an estimated 14 liters of solvent
burned with an equivalence of 2.44 kg heavy metal per liter. The esti-
mated fraction of plutonium in the solvent which was originally present
in the fuel is unity. It is assumed that 1% of the plutonium in the
burned solvent is dispersed by the fire. The estimated fraction of the
dispersed plutonium passing through the three HEPA filters in series
-78
under normal conditions is 6 x 10 . In the event of a simultaneous
failure of the three HEPA filters, all of the particulate activity
discharged to the cell is assumed to be released to the atmosphere.
Likelihood
The likelihood of a solvent fire in the plutonium extraction cycle of
a generic fuel reprocessing facility has been estimated to fall in the:
5 -6 9
range of 10 to 10 /year. However, from chemical industry data, the
-4 8
probability of major fires per plant-year is estimated to be 4 x 10 .
This probability is reduced to 2 x 10" /plant-year in Reference 8 for a
generic mixed oxide fuel fabrication plant, in which defenses are
employed uncharacteristic of the typical chemical plant.
_2
The failure of a single HEPA filter has been estimated to be 10 /event,
whereas the failure of two HEPA filters in series has been estimated to
-3 9
be 10 /event. Since the cell exhaust filter and the exit filters
from the ventilation system are assumed to be independent, the combined
probability of all three filters failing simultaneously is further
_5
estimated to be 10 /demand.
126
-------
TABLE 4-29
RADIONUCLIDE RELEASE RESULTING FROM A FIRE IN THE
9
PLUTONIUM EXTRACTION CYCLE AT THE REPROCESSING FACILITY
Nuclide Activity Released (C1)*
Pu 238 8.8 x 10"7
Pu 239 6.5 x 10"8
Pu 240 1.3 x 10~7
Pu 241 3.5 x 10"5
Pu 242 7.3 x 10"10
* The values tabulated are for normal operation of the three series
HEPA filters. In the event of simultaneous failure of the filters,
the values in this column are increased by a factor of 1.6 x 10^.
127
-------
Estimate: Source term: Eadionuolides given in Table 4-29 to air for
normal filter operation and in the event of filter failure; Likelihood
AX 10 to 10 /plant-year for normal operation, 10~ to 10~ /plant-
year for simultaneous failure of three series HEPA filters.
B.3 Ion-Exchange Resin Fire
Ion exchange resin columns, contained in the plutonium product cell, are
used to partition plutonium, uranium and/or neptunium into separate streams,
while providing for additional fission product decontamination following
the initial codecontamination step. A resin-nitrate reaction in these col-
umns could lead to an ion exchange resin fire. A fire is prevented by pro-
cess control which limits the acidity and temperature of the resin to less
than 135°C. If the resin beds are highly loaded with plutonium, radiolytic
heating could augment the temperature rise of the bed.
Source Term
The estimated quantities of radionuclides released to the environment
9
in the event of an ion-exchange resin fire are given in Table 4-30.
The estimated fractions of radionuclides in the resin which were originally
present in the fuel are 3.0 x 10" for ruthenium, 5.0 x 10 for plutonium,
0.95 for neptunium, 6.6 x 10 for zirconium and niobium, 3.1 x 10"
Q
for iodine, and 1.0 x 10" for the remainder. In this estimate, the
_2
volatile fraction of ruthenium is taken to-be 5 x 10 and of iodine,
0.5. All of the activity contained within the resin from a full day's
processing is assumed to be released to the cell in the fire. The
128
-------
TABLE 4-30
RADIONUCLIDE RELEASE RESULTING FROM AN ION-EXCHANGE
9
RESIN FIRE AT THE REPROCESSING FACILITY
Nuclide Activity Released (Ci)*
Sr 89 1.2 x 1Q"9
Sr 90 1.2 x ICf9
Zr 95 3.5 X 10"6
Nb 95 6.8 x 10"6
Ru 103 9.3 x 10"2
Ru 106 5.4 x 10"1
I 129 2.8 x 10"8
I 131 1.2 x 10"6
Cs 134 2.7 x 10"9
Cs 137 1.9 x 10"9
Ba 137m 1.7 x 10~9
Ce 144 8.4 x 10"9
Np 238 9.1 x 10"6
Pu 238 3.3 x 10""6
Pu 239 2.4 x 10"7
Pu 240 4.4 x 10~7
Pu 241 1.2 x 10'4
Cm 242 6.0 x 10"10
Cm 244 7.1 x 10"11
* The values tabulated are for normal operation of the three series
HEPA filters. In the event of simultaneous failure of the filters,
the values in this column, with the exception of iodine and
ruthenium, are increased by a factor of 1.6 x 106. Since iodine
is assumed to be a vapor, the activity released is assumed to be
unchanged. Ruthenium is part vapor "and part particulate. In the
event of simultaneous filter failure, the ruthenium activity
released is assumed to be the following: 103RU _ 9.3 x IQ-J ci,106Ru
5.4 Ci.
129
-------
estimated fraction of non-volatile material passed through the three
7 8
HEPA filters in series under normal conditions is 6 x 10" . In the
event of simultaneous failure of the three HEPA filters, all of the
particulate activity discharged to the cell is assumed to' be released
to the atmosphere.
vikelihood
The likelihood of an ion exchange resin fire in a generic fuel repro-
-4 9
cessing facility has been estimated to be approximately 10 /year.
However, an estimate of the likelihood of an ion exchange fire in a
-1 8
mixed oxide fuel fabrication plant has been given as < 10 /year.
Four incidents have been reported resulting in a release of radio-
activity as a result of thermochemical instabilities in ion-exchange
8
processing. On the basis of 490 plant-years of fuel fabrication
o
activities, this history would result in a likelihood of«*» 8 x 10 /
plant-year.
At least one incident has been reported involving a fire around an
31
anion exchange column in fuel reprocessing activities (at Savannah River).
On the basis of 100 plant-years of reprocessing activities, this history
_2
would result in a likelihood of A»IO /plant year.
The failure of a single HEPA filter has been estimated to be 10"2/
event, whereas the failure of two HEPA filters in series has been
3 9
estimated to be 10 /event. Since the cell exhaust filter and the
exit filters from the ventilation system are assumed to be independent,
the combined probability of all three filters failing simultaneously is
further estimated to be 10" /demand.
130
-------
Estimate: Source term: Eadionuolides given in Table 4-30 to air
for normal filter operation and in the event of filter failure; Like-
-l -4 -6
lihood^lO to 10 /plant-year for normal filter operation, 10
—9
to 10 /plant-year for simultaneous failure of three series EEPA filters.
C.I Fuel Assembly Rupture and Release In Fuel Receiving and Storage
Area
Irradiated fuel assemblies arrive at the reprocessing plant in shielded
casks, where they are removed from the carriers and submerged in a pool
of water for unloading the fuel assemblies. The cask is opened and the
fuel assemblies Removed and placed in storage canisters. If the cask-
were to lose its heat removal capability during shipment, the self-
heating of the spent fuel rods from fission products could elevate the
cladding temperature beyond the failure point. On /opening the cask,
mobile radionuclides would be expelled from the cask cavity as a
stream of bubbles which rise to the pool surface.
Source Term
The estimated quantities of radionuclides released to the environment
in the event of a fuel assembly rupture and release in the fuel receiving
9
and storage area are given in Table 4-31. This estimate is based upon
the assumption that all of the noble gases, tritium and radioiodine;
10% of the ruthenium; and 1% of the cesium and strontium in the breached
elements are released to the pool water. The airborne release of noble
gases and tritium are neglected in the accident evaluation since they
131
-------
TABLE 4-31
RADIONUCLIDE RELEASE RESULTING FROM A FUEL ASSEMBLY
RUPTURE AND RELEASE IN FUEL RECEIVING AND
9
STORAGE AREA AT THE REPROCESSING FACILITY
Nucllde Activity Release (C1)
Ru 103 3.8 x 10"3
Ru 106 1.9 x 10"2
I 129 1.6 x 10"11
I 131 7.2 x 10'10
Cs 134 1.1 x 10"7
Cs 137 5.4 x 10"8
* The values given assume normal operation of the two series HEPA
filters. In the event of simultaneous failure of the filters,
the values in this column are increased by a factor of 105.
132
-------
are normally released in the dissolution step of the process. Ninety
percent of the radioiodine and ruthenium; 99.9% of the cesium; and all
of the strontium are assumed to remain in the pool water. The released
gases subsequently pass through a scrubber, which removes 93% of the
ruthenium, 99.9% of the cesium, and 99.99% of the iodine, and two HEPA
filters in series, which under normal conditions pass 10 of the
materials. In the event of simultaneous failure of the two HEPA filters,
all of the particulate activity discharged to the fuel receiving and
storage area is assumed to be released to the atmosphere.
Likelihood
The likelihood of a fuel assembly rupture and release in the fuel receiving
and storage area has been estimated to be in the range of 10"1 to 10"2/
9
year. The likelihood of a simultaneous failure of the two series HEPA
? 9
filters has been further estimated to be roughly 10"°/demand. The
potential failure of the scrubber has not been considered, since the
failure probability would have to be in excess of A/10 /event before a
significant contribution to risk would result.
* It is assumed in Reference 9 that the materials released from the
cask as vapors are converted to the particulate form in passage
through the pool.
133
-------
Estimate: Source term - Radionuolides given in Table 4-31 to air for
normal filter operation and in the event of filter failure; Likelihood 5&
10~ to 10 /plant-year for normal filter operation^ 10~ to 10~ /plant-
year for simultaneous failure of two series HEPA filters.
C.2 Dissolver Seal Failure
The segmented fuel containing the unspent uranium and radionuclides
formed during irradiation is dissolved out of the cladding hulls with nitric
acid to form the feed for subsequent extraction steps. It is assumed
that a leak in the seam of the dissolver releases the solution onto a
hot surface, producing a powder of all the contained non-volatile materials
and evolving the volatile materials.
Source Term
The estimated quantities of radionuclides released to the environment in
42
the event of a dissolver seal failure is given in Table 4-32. These
estimates are based upon the assumption that .all materials associated
with one kg of heavy metal are dispersed to the cell atmosphere, and
subsequently pass through the cell ventilation system.
Likelihood
An estimate in the range of 10 to 10" /year has been made of the
likelihood of plumbing failure due to corrosion in the high-level and
9
intermediate-level cells of a generic fuel reprocessing plant. In
the same source, an estimate of 10" /year has been made for the like-
lihood of failure in the primary boundary. The likelihood of a
134
-------
TABLE 4-32
RADIONUCLIDE RELEASE RESULTING FROM A DISSOL^ER SEAL
FAILURE AT THE REPROCESSING FACILITY
Nuclide Activity Released (Ci)*
Sr 89 6.52 x 10"6
Sr 90 4.67 x TO"5
Y 90 4.67 x 10"5
Y 91 1.64 x 10"5
Zr 95 4.57 x 10"5
Nb 95 9.71 x 10"5
Ru 103 4.16 x 10"6
Ru 106 6.88 x 10"4
Ag 110 3.93 x 10"7
Sb 125 2.03 x TO"5
Te 127 3.54 x TO"6
Te 129 4.20 x 10"8
I 129 3.92 x 10"6
I 131 3.10 x 10"12
Cs 134 2.11 x TO"5
Cs 137 1.24 x 10"4
Ce 141 7.46 x 10"7
Ce 144 5.18 x 10"4
Pm 147 2.71 x 10"4
Eu 154 1.44 x 10"6
Eu 155 3.78 x 10"5
U 234 1.66 x 10"10
U 235 2.64 x 10"12
U 236 8.24 x 10"12
U 238 2.87 x 10"10
Pu 238 1.89 x 10"5
Pu 239 3.85 x 10"6
135
-------
TABLE 4-32
(continued)
NucMde Activity Released (Ci)*
Pu 240 5.24x 10"6
Pu 241 5.71 x TO"4
Am 241 2.95x 10"6
Am 243 5.41 x 10"8
Cm 242 1.73 x 10"5
Cm 244 1.44x 10"6
The values calculated are for normal operation of the two series
HEPA filters. In the event of simultaneous failure of the filters,
the values in this column, with the exception of iodine, are increased
by a factor of 10$. Since iodine is assumed to be a vapor, the
activity released is assumed to be unchanged.
136
-------
simultaneous failure of the two series HEPA filters has been further estl-
-3 9
mated to be roughly 10 /demand.
Estimate: Source Term: Radionuelides given in Table 4-32 to air for
normal filter operation and in the event of filter failure; Likelihood
8 10~ /plant-year for normal filter operation3 10~ /plant-year for
simultaneous failure of two series HEPA filters.
C.3 Release from a Failure of Hot UF£ Cylinder
The recovered uranium from fuel reprocessing is converted to UFg for
shipment to the enrichment plant. The transfer of UP. from the surge tanks
6
to the cylinders for shipment is accomplished by melting the UFg into a
pressurized liquid. The cylinder is removed from the loadout area by fork
lift to an outdoor storage area where the UFg cools and solidifies. At
any time during this sequence of events, while the UFg is in the liquid
state, the failure of a valve, an operator error, or a cylinder rupture
could release significant quantities of UFg to the environment.
Source Term
A summary of releases associated with hookup and disconnect operations
on UFg cylinders at the three existing gaseous diffusion plants has been
compiled and was summarized in Table 4-12. The average release in six
recorded incidents resulting in more than 5 kg loss was 108 kg of
uranium. Since there exists no data base associated with UFg releases
at reprocessing plants, the average source term from enrichment will be
137
-------
adopted here. The specific activities by isotope of uranium having
47
undergone 33,000 MWD/MT burnup are given in Table 4-33.
Likelihood
From the incidents on record at the three enrichment facilities, the
likelihood of a release in handling a cylinder was estimated to be
approximately 5 x 10" /cylinder (see Section 4.3). The generic repro-
cessing fabrication plant operating at a nominal capacity of 1500
MT HM/year would require the filling of approximately 1000 cylinders
annually of 2.2 MT UFg capacity. Thus, on the basis of the release
likelihood developed from enrichment operations, the likelihood of a
2
release at the reprocessing facility is approximately 5 x 10 /plant-
year.
Estimate: Source term^f 3.6 x 10~2 Ci 2S8U> 3.0 x I0~l Ci 23?U, 3.2
x 10~2 Ci 236U, 1.9 x Id"2 Ci 235U, 8.5 x 10~2 Ci 234U, 3.6 x 10~2 Ci
234 . —2
Th to air; Likelihood ^f5xlO /plant-year.
D.I Critical ity
Accidental criticality in fuel receiving and storage operations is un-
likely because the areas where these operations are performed are
designed to be subcritical with unirradiated fuel of 5%'enrichment.
Light water reactor fuel is normally enriched to less than 4%, and
after burnup the enrichment is significantly reduced and fission pro-
duct poisons are present in the fuel.
138
-------
TABLE 4-33
ISOTOPIC ACTIVITIES OF URANIUM WITH
33,000 MWD/MT BURNUP47
Isotope Activity per Gram of Uranium (C1)
238U 3.3 x 10"7
237U 2.8 x 10"6
236U 3.0 x 10"7
235U 1.8 x 10"8
234U 7.9 x 10"7
234Th 3.3 x 10~7
139
-------
Criticality could accidentally occur by overfilling a dissolver, or by
accidental transfer of plutonium fines to the dissolver transfer tank and
to the accountability tank. Following the concentration of uranium and
plutonium, criticality is inhibited by controlling the concentrations of
fissile materials in the solutions. A criticality accident could occur
in this part of the process from a failure of process control that results
in higher-than-normal fissile material concentrations in solution concur-
rently with multiple monitoring failures or it could result from admin-
istrative error by processing higher enrichment fuel under specifications
normally used for lower uranium enrichment.
Criticality in product loadout is also possible, particularly in the
plutonium loadout area, if failure of both concentration control and
monitor failure were to occur, or if the plutonium product cell were to
be flooded.
Source Term
18
As discussed in Section 4.5 (accident D.I), 10 fissions are selected
as representative of the energy release in a solution criticality for
purposes of this study.
The radionuclides released to the environment ten minutes after an
18
inadvertent criticality involving plutonium and resulting in 10 fissions,
235 44
based upon ORIGEN calculations (for a U criticality incident) are
given in Table 4-34. These estimates assume that 100% of the noble gases
are released to the environment, but that prior to release, 99.99% of
the halogens are removed by scrubbers. Approximately 0.2% of the
140
-------
TABLE 4-34
RADIONUCLIDE RELEASE RESULTING FROM A CRITICALITY INCIDENT
AT THE FUEL REPROCESSING FACILITY
Nucllde Activity Released (Ci) Nucllde Activity Released (C1)
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Kr
Kr
Kr
Kr
Kr
Kr
Kr
80
80m
82
82m
83
84
84m
85
86
87
83m
85
85m
87
88
89
90
2.
1.
3.
6.
7.
7.
3.
1.
3.
4.
3.
3.
1.
9.
6.
5.
9.
4 x
1 x
8 x
4 x
0 x
0 x
0 x
4 x
0 x
2 x
4 x
2 x
3 x
4 x
3 x
2 x
2 x
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
-5
-8
-9
-7
-4
-3
-4
~2
-4
-4
-1
-5
1
1
1
2
-2
I
I
Xe
Xe
Xe
Xe
Xe
Xe
Xe
Xe
Th
Th
Pa
U
U
U
U
136
137
133
133m
135
135m
137
138
139
140
*
231
234*
234m*
*
233
234*
*
235
*
236
4.
6.
1.
9.
2.
4.
9.
7.
1.
2.
2.
2.
1.
1.
3.
4.
1.
6
6
7
9
2
7
0
0
5
2
0
2
8
7
4
5
3
x
x
X
X
X
X
X
X
X
X
X
X
X
X
io-3
io-8
io-3
10'5
io2
io2
ID'7
io-13
io-14
ID'14
ID'20
io-18
ID'11
io-16
14T
-------
TABLE 4-34
(continued)
Nucllde Activity Released (Ci) Nuclide Activity Released (C1)
I 128
I 130
I 131
I 132
I 133
I 134
I 135
2.8 x 10"7
2.0 x 10"7
1.1 x 10'5
3.2 x 10"4
3.0 x 10"4
1.0 x 10"2
4.6 x 10"3
U 237*
U 238*
U 239*
Np 237*
Np 239*
Np 240*
Pu 239*
7.1 x ID'10
1.1 x lO'10
6.1 x 10"5
4.4 x 10"21
1.5 x 10"7
3.5 x 10"16
4.1 x 10"17
* The values tabulated are for normal operation of the two HEPA filters.
In the event of simultaneous failure of the filters-, these values are
increased by a factor of 105.
142
-------
activities generated in the criticality are assumed released to the building,
and these are attenuated by a factor of 10~ as they pass through two HEPA
filters in series during normal operation. In the event of simultaneous
failure of the two HEPA filters, all of the particulate activity discharged
to the cell is assumed to be released to the atmosphere. The neutron and
gamma "shine" from the event is neglected as a source of dose to the popu-
lation, since these radiations are attenuated by at least five ft. of
concrete before emerging from the facility.
Likelihood
The likelihood of a criticality event in a generic fuel reprocessing plant
It
9
18 -5
resulting in 10 fissions has been estimated to be approximately 3 x 10~ /
year.
However, an analysis of the incidents on record in fuel fabrication facil-
ities results in a probability for accidental criticality of ^8 x 10" /plant-
8
year (four solution criticality incidents in 490 plant-years of operation).
Although an analogy can be drawn between scrap recovery operations in fab-
rication and spent fuel reprocessing, it should be noted that there have
been no criticality incidents on record since 1968.
The likelihood of a simultaneous failure of the two series HEPA filters
-3 9
has been further estimated to be roughly 10 /demand.
Estimate: Source term - Radionuclides given in Table 4-34 to air for
normal filter operation and in the event of filter failure; Likelihood ?£
•y r /*
8 x 10 to 3 x 10 /plant-year for normal filter operation^ 8 x 10 to
_ Q
3 x 10 for simultaneous failure of two series HEPA filters.
143
-------
4.7 Mixed Oxide Fuel Fabrication
The accidents considered in mixed oxide fuel fabrication, keyed to the
accident categories given in Table 4-1, are listed in Table 4-35.
The risk associated with other incidents, some of which have occurred in
the past, is judged to be insignificant in comparison with the accidents
considered in Table 4-35. These include sintering furnace explosions,
metallographic glove box explosions, autoclave explosions, and ventilation
problems from, for example, loss of electrical power. These have resulted
in excessive airborne concentrations in work areas, but the release to the
environment is generally inconsequential. Filter failures in the venti-
32
lation streams have occurred from time to time, but the fillers are
generally replaced before the time integrated release becomes a significant
fraction of the annual release from normal operations.
Current regulatory criteria for the design of new mixed oxide fabrication
facilities require that the structures, systems, and other components with-
stand the effects of natural phenomena. These include all floods, tornados,
earthquakes, or missiles of intensity more severe than experienced histor-
ically in the location of the plant. Although events with intensities
outside of this range are conceivable, the exceedingly low probabilities
associated with their occurrence were not evaluated in the current study.
Nine commercial facilities are currently licensed for the production of
fuels containing plutonium. However, since the operations employed in
mixed oxide fuel fabrication are similar to those utilized in oxide fuel
144
-------
TABLE 4-35
MIXED OXIDE FUEL FABRICATION ACCIDENTS
A.I Explosion 1n oxidation-reduction scrap furnace
B.I Major facility fire
B.2 Fire 1n waste compaction glove box
B.3 Ion exchange resin fire
B.4 Dissolver fire in scrap recovery
C.I Glove failure
C.2 Severe glove box damage
D.I Criticality
145
-------
fabrication in general, the fabrication facilities listed in Section 4.5
are deemed germane to the consideration of accident likelihoods in mixed
oxide fuel fabrication. A total of 490 plant-years of fuel fabrication
experience is derived from Table 4-20.
"ost of the data presented in this section were developed in an earlier
8
ssessment of effluents from a generic mixed oxide fabrication facility.
A.I Explosion in Oxidation-Reduction Scrap Furnace
Scrap mixed oxide pellets are conditioned for recycle by heating success-
ively in air and a reducing atmosphere containing hydrogen. The hydrogen
may be mixed with nitrogen or another inert gas, and the hydrogen concen-
tration is controlled to prevent the buildup of an explosive mixture.
However, a failure of controls on the mixture could allow pure hydrogen
to reach the hot, airfilled furnace after the oxidation step. An explosion
in the furnace might damage the glove box and release mixed oxide pellets
and dust into the room.
Source Term
It is assumed that approximately 150 kg of pellets and 1.5 kg of dust
42
are released to the room. However, the airborne concentration of heavy
* A reducing atmosphere is also used during the sintering of pellets.
However, by virtue of the integral form-of the pellets at this stage
of the process, the consequences of a postulated explosion in the
sintering furnace is judged to be insignificant in comparison with
a similar event in the reduction furnace.
146
-------
particles in the respirable range appears to be limited to approximately
38 3
100 mg/m . For a room volume assumed to be approximately 5000 m , the
quantity of mixed oxide released to the building ventilation system would
be approximately 500 gm.
The building ventilation system in a generic mixed oxide fabrication plant
is assumed to incorporate two HEPA filters in series providing,under normal
operations, an attenuation of the source to the environment of approximately
-5 8
10 . In the event of simultaneous failure of the two HEPA filters, all
of the activity discharged to the room is assumed to be released to the
atmosphere. The mixed oxide is assumed to contain 4.4% plutonium oxide
and 95.6% natural uranium oxide. The estimated activities of radionuclides
released to the environment are given in column 4, Table 4-36, based upon
the specific activities of the components of the mixed oxide in column
5
3. Uranium isotopes are neglected since they contribute negligibly to
dose.
Likelihood
i n
An estimate of < 5 x 10" /plant-year has been made for the likelihood of
8
a hydrogen explosion in a fuel fabrication sintering furnace. The same
study derived an estimate of /o 10"3/plant-year for chemical explosions
in general.
At least one accident has occurred (at a sintering furnace) resulting
8,31
from detonation of an explosive mixture of hydrogen and oxygen. On
the basis of the ^490 plant-years of fuel fabrication experience, this
would result in a likelihood of 3» 2 x 10"3/plant-year.
!47
-------
TABLE 4-36
RADIONUCLIDE RELEASE RESULTING FROM AN EXPLOSION IN THE
OXIDATION-REDUCTION SCRAP FURNACE AT THE
MIXED OXIDE FABRICATION PLANT
Nuclide wt. % Specific Activity (Ci/gm MO) Activity Released (Ci)
Pu 236
Pu 238
Pu 239
Pu 240
Pu 241
Pu 242
3.
8.
2.
1.
5.
3.
08 x
36 x
38
10
28 x
08 x
10
10
ID'1
10"
1.
1.
1.
2.
5.
1.
66 x
43 x
48 x
50 x
99 x
32 x
10
10
10
10
10
10
-3
-2
-3
-3
-1
-5
8.
7.
7.
1.
3.
6.
3
2
4
3
0
6
x
x
X
X
X
X
io-6
ID'5
10-6
ID"5
io-3
ID'8
* The values tabulated are for normal operation of the two HEPA filters.
In the event of simultaneous failure of the filters, these values
are increased by a factor of 10^.
148
-------
45
In another accident evaluation, the probability of a hydrogen explosion
in the sintering furnace has been crudely estimated to be lower than 10
to 10" /year and higher than 10"1 to 10~3/year.
The likelihood of a simultaneous failure of the two series HEPA filters
3 9
has been further estimated to be roughly 10 /demand.
Estimate: Source term - Radionuclides given in Table 4-36 to air for
normal filter operation and in the event of filter failure; Likelihood
_ 9 _ 7 £-
X 5 x 10 to 2 x 10~ /plant-year for normal filter operation, 5 x 10~
n.
to 2 x 10 /plant-year for simultaneous failure of two series HEPA filters.
B.I Major Facility Fire
Since the combustible inventory at a mixed oxide fabrication plant is
limited, the chance of a catastrophic fire is remote. Also, the use of
glass windows in the glove boxes and of fire-suppression equipment both
42
on the inside and outside of the glove boxes is expected. The glove
box lines shall also have suitable barriers to isolate various portions
of the lines. Nonetheless, a general facility fire is assumed to be
initiated by the ignition of a solvent container which is improperly
handled. It is assumed that the fire spreads to a large fraction of the
glove boxes in several parallel fabrication lines.
Source Term
The bulk of the plutonium in the facility would be in storage in a fire-
proof vault. It is estimated that approximately five times the daily
149
-------
output, about 5000 kg mixed oxide would be in process. It is estimated
that approximately one-half of the inprocess inventory would be in the
dispersible form, and 1% of the dispersible material would be released
to the ventilation system during a major facility fire.
The final barrier, composed of two HEPA filters in series and providing
an attenuation in source strength of approximately 10" , would remain
8
intact. In the event of a simultaneous failure of the two HEPA filters,
all of the activity discharged to the room is assumed to be released to
the environment. From the specific activities of mixed oxide radio-
nuclides given in Table 4-36, the estimated qualities of radionuclides
released to the environment are given in Table 4-37.
Likelihood
The likelihood of a major facility fire in a generic mixed oxide fuel
-4 8
fabrication plant has been estimated to be approximately 2 x 10 /year.
The likelihood of a simultaneous failure of the two series HEPA filters
-3 9
has been further estimated to be roughly 10 /demand.
Estimate: Sourae term - Radionuolides given in Table 4-Z7 to air for
normal filter operation and in the event of filter failure; Likelihood
-4 -7
~ 2 x 10 /plant-year for normal filter operation, 2 x 10 /plant-year
for simultaneous failure of two series HEPA filters.
150
-------
TABLE 4-37
RADIONUCLIDE RELEASE RESULTING FROM A MAJOR FIRE
AT THE MIXED OXIDE FABRICATION PLANT
Nuclide Activity Released (Ci)
Pu 236 4.2 x 10~4
Pu 238 3.6 x 10"3
Pu 239 3.7 x 10~4
Pu 240 6.3 x 10~4
Pu 241 1.5 x 10"1
Pu 242 3.3 x 10~6
* The values tabulated are for normal operation of the two HEPA filters.
In the event of simultaneous failure of the filters, these values are
increased by a factor of 105.
151
-------
B.2 Fire in Waste Compaction Glove Box
Burnable wastes are collected at a compaction station until enough has
been collected to fill a 55-gallon drum. Waste is composed of poly-
ethylene bagging material, rubber gloves, and cellulose wipes. A solvent-
damp wipe in the waste could ignite due to the discharge of static elec-
tricity. The fire could breach the containment of the glove box and
spread mixed oxide in the dispersible form to the room.
Source Term
It is assumed that the box is at roughly one-half full capacity at the
time of the fire, or roughly 50 Ibs. of waste. It is further assumed
42
that the waste is contaminated to the extent of 0.01% plutonium by weight.
On this basis, approximately 50 gms of mixed oxide would be released to
the room. Taking into account the factor of 10 attenuation of the two
series HEPA filters operating normally in the building ventilation system,
-4
approximately 5x10 gms of mixed oxide would be released to the
environment. In the event of simultaneous failure of the two HEPA filters,
all of the activity discharged to the room is assumed to be released to
the environment. From the specific activities of mixed oxide radionuclides
given in Table 4-36, the estimated quantities of radionuclides released to
the environment are given in Table 4-38.
Likelihood
The likelihood of a local fire in a generic mixed oxide fuel fabrication
-2 8
plant has been estimated to be approximately 10 /year. Moreover, from
152
-------
TABLE 4-38
RADIONUCLIDE RELEASE RESULTING FROM A FIRE IN THE WASTE COMPACTION
GLOVE BOX AT THE MIXED OXIDE FABRICATION PLANT
Nuclide Activity Released (Ci)
Pu 236 8.3 x 10"7
Pu 238 7.2 x 10"6
Pu 239 7.4 x 10"7
Pu 240 1.3 x 10"6
Pu 241 3.0 x 10"4
Pu 242 6.6 x 10"9
* The values tabulated are for normal operation of the two HEPA filters.
In the event of simultaneous failure of the filters, these values are
increased by a factor of 105.
153
-------
32
the Incidents on record, there have been at least five fires in fuel
fabrication facilities. On the basis of 490 plant-years of fuel fab-
rication activities, this results in a probability of about 10"2/year.
The likelihood of simultaneous failure of the two series HEPA filters
has been further estimated to be roughly 10 /demand.
Estimate: Source term - Radionuolides given in Table 4-38 to air for
normal filter operation and in the event of filter failure; Likelihood
4^4 ~»P C
"* 10 /plant-year for normal filter operation, 10~ /year for simultaneous
failure of two series HEPA filters.
B.3 Ion Exchange Resin Fire
Anion exchange may be used in scrap recovery to recover chemicalTy con-
taminated plutonium. Thermal transients may occur in the resin from
radiolytic heating, excessive external heating, or a resin-nitrate
reaction. A fire is prevented by process and administrative controls
which limits the acidity and temperature of the resin to less than 135°C.
However, should these controls fail, and an excursion go unchecked, the
ion exchange column could become pressurized and rupture to discharge
the resin and contained solution.
Source Term
It is assumed that the column inventory at the time of the rupture is
8
1400 gms of plutonium, equivalent to approximately 32 kg. of mixed oxide.
The fire is assumed to render 0.5 to 0.7% of the mixed oxide airborne,
8
approximately 80% of which is in the respirable range. This results in
154
-------
approximately 150 gms of mixed oxide made airborne within the facility,
or 1.5 x 10 gms released to the environment, taking Into account the
factor of 10"5 attenuation of the two series HEPA filters operating
8
normally 1n the building ventilation system. In the event of simul-
taneous failure of the two HEPA filters, all of the activity discharged
to the room 1s assumed to be released to the atmosphere. From the specific
activities of mixed oxide radionuclides given in Table 4-36, the estimated
quantities of radionuclides released to the environment are given 1n Table
4-39.
Likelihood
The likelihood of an ion exchange resin fire in a generic mixed oxide
- -1 ' 8
fuel fabrication plant has been estimated to be <10 /year. However,
the likelihood of an ion-exchange resin fire in a generic fuel repro-
-4 9
cessing facility has been estimated to be approximately 10 /year.
Four incidents have been reported,resulting in a release of radioac-
tivity as a result of thermochemical instabilities in ion-exchange pro-
8
cessing. On the basis of 490 plant-years of fuel fabrication activities,
_3
this history would result in a likelihood of*«8 x 10 /plant-year.
The likelihood of simultaneous failure of the two series HEPA filters
-3 9
has been further estimated to be roughly 10 /demand.
155
-------
TABLE 4-39
RADIONUCLIDE RELEASE RESULTING FROM A FIRE IN AN ION EXCHANGE
RESIN COLUMN AT THE MIXED OXIDE
FABRICATION PLANT
Nuclide Activity Released (Cp*
Pu 236 2.5 x 10"6
Pu 238 2.1 x 10"5
Pu 239 2.2 x 10"6
Pu 240 3.8 x 10"6
Pu 241 9.0 x 10"4
Pu 242 2.0 x 10"8
* The values tabulated are for normal.operation of the two HEPA
filters. In the event of simultaneous failure of the filters,
these values are Increased by a factor of 105.
156
-------
Estimate: Source term - Radionuclid.es given in Table 4-39 to air for
normal filter operation and in the event of filter failure; Likelihood
ft? 10 to 10 /plant-year for normal filter operation, 10~ to 10~ /
plant-year for simultaneous failure of two series HEPA filters.
B.4 Dissolver Fire In Scrap Recovery
After calcining the combustible materials, the scrap and waste from
the process are dissolved in nitric acid. Fire around the dissolution
tank could heat the liquid and cause it to boil. If the dissolution
vessel is closed, it could be ruptured by the internal pressure, and
the nitrate would spill on the floor and be dried by the fire.
Source Term
The scrap recovery operation might incorporate an inventory of 25 kg
8
of plutonium, or roughly 500 kg of mixed oxide. The fractional release
of mixed oxide into the ventilation system might range from 0.2% to
0.7%, depending upon the duration of the fire and the location of the
8
liquid during the fire. Taking the midrange of these estimates,it is
assumed that the fire burns the solution dry, and releases approximately
2.5 kg of mixed oxide to the room air. The source to the environment
_2
would be roughly 2.5 x 10 gms of mixed oxide, taking into account
the 10" attenuation afforded by the two exit HEPA filters in series
operating normally, in the event of simultaneous failure of the two
HEPA filters, all of the activity discharged to the room is assumed to
be released to the atmosphere. - From the specific activities of mixed
157
-------
oxide radionuclides given in Table 4-36, the estimated quantities of
radionuclides released to the environment are given in Table 4-40.
Likelihood
The likelihood of a local fire in a generic mixed oxide fuel fabrication
8
plant has been estimated to be ** 10 /year. At least three fires in
scrap recovery (not involving ion exchange resin columns) have been
32
recorded in fuel fabrication operations. On the basis of 490 plant-
years of fuel fabrication activities, this history would result in a
likelihood of approximately 6 x 10" /plant-year. The likelihood of a
simultaneous failure of the two series HEPA filters has been further
-3 9
estimated to be roughly 10 /demand.
Estimate: Source term - Radionuclides given in Table 4-40 to air for
normal filter operation and in the event of filter failurej likelihood
£» W~ /plant-year for normal filter operation3 10 /plant-year for
simultaneous failure of two series HEPA filters.
C.I Glove Failure
Glove failures of various types are a frequent occurrence. They gener-
ally involve pinholes or tears, which may result in general contamination
of work areas, but a negligible environmental release. Occasionally,
a more significant failure, such as an operator inadvertently pulling
a glove off of the box, occurs. The glove box might contain the mill
used to condition the mixed oxide powder, which is a dusty operation.
158
-------
TABLE 4-40
/•
RADIONUCLIDE RELEASE RESULTING FROM A DISSOLVER FIRE IN
SCRAP RECOVERY AT THE MIXED OXIDE
FABRICATION PLANT
Nuclide Activity Released (Ci)*
Pu 236 4.2 x 10"5
Pu 238 3.6 x 10~4
Pu 239 3.7 x 10"5
Pu 240 6.3 x 10~5
Pu 241 1.5 x 10"2
Pu 242 3.3 x 10"7
* The values tabulated are for normal operation of the two HEPA filters.
In the event of simultaneous failure of the filters, these values are
increased by a factor of 10$.
159
-------
Source Term
It is assumed that the glove which 1s involved in the Incident contains
a gram of mixed oxide powder on the surface, and that approximately 20%
42
of this powder becomes suspended in the room atmosphere. The source
to the environment would be approximately 2 x 10" gms taking Into
-5
account the 10 attenuation of the two series HEPA filters operating
8
normally. In the event of simultaneous failure of the two HEPA filters,
all of the activity discharged to the room is assumed to be released to
the atmosphere. From the specific activities of mixed oxide radio-
nuclides given in Table 4-36, the estimated quantities of radlonuclldes
released to the environment are given in Table 4-41.
Likelihood
A large glove failure such as described here might be expected to occur
8
once a year. Small failures, such as pin-hole leaks, etc., might be
expected to occur more frequently. The likelihood of a simultaneous
failure of the two series HEPA filters has been further estimated to be
-3
roughly 10 /demand.
Estimate: Source term - Radionuolides given in Table 4-41 to air for
normal filter operation and in the event of filter failure; Likelihood
—3
ftJ I/plant-year for normal filter operation, 10 /plant-year for simul-
taneous failure of two series HEPA filters.
160
-------
TABLE 4-41
RADIONUCLIDE RELEASE RESULTING FROM A GLOVE FAILURE
AT THE MIXED OXIDE FABRICATION PLANT
Nuclide Activity Released (Ci)*
Pu 236 3.3 x TO"9
Pu 238 2.9 x TO"8
Pu 239 3.0 x 10"9
Pu 240 5.0 x 10"9
Pu 241 1.2 x 10"6
Pu 242 2.6 x 10"11
* The values tabulated are for normal operation of the two HEPA
filters. In the event of simultaneous failure of the filters,
these values are increased by a factor of 105,
161
-------
C-2 Severe Glove Box Damage
Severe mechanical damage to a glove box could rupture the glove box
and if the damage occurs in the powder line, mixed oxide powder could
be released to the room air. Such an accident might be initiated by
a falling beam or crane, or a runaway fork lift truck. Should the
incident also breach a compressed air line, the jet of air could provide
8
an additional dispersal mechanism. However, such a sequence of events
is not likely to constitute a nominal glove box rupture accident.
Source Term
It is assumed that the glove box involved contains mixed oxide powder, with
8
a batch limit of 11.3 kg of plutonium, or roughly 250 kg of mixed oxide
powder. It is further assumed that the initiating event which breaches the
glove box contains sufficient energy to disperse a small fraction of the
powder. In any case, the airborne concentration of heavy particles in the re-
38
spirable range would be limited to approximately 100 mg/m . For a room vol-
m n
ume assumed to be roughly 10 m ,the quantity of mixed oxide released to the
building ventilation system would be approximately 1000 gm.
The building ventilation system in the generic plant is assumed to
incorporate two HEPA filters in series, providing an attenuation of
-5 8
approximately 10 in normal operation, resulting in the release to
_2
the environment of 1.0 x 10 gms >f mixed oxide. In the event of
simultaneous failure of the two HEPA filters, all of the activity
discharged to the room is assumed to be released to the atmosphere.
162
-------
From the specific activities of mixed oxide radionuclides given in
Table 4-36, the estimated quantities of radionuclides released to the
environment are given in Table 4-42.
Likelihood
The likelihood of severe glove box damage in a generic mixed oxide fuel
? 8
fabrication plant has been estimated to be roughly < 10 /year. The
likelihood of a simultaneous failure of the two series HEPA filters has
1 9
been further estimated to be roughly 10 /demand.
Estimate: Source term - Radionuclides given in Table 4-42 to air for
normal filter operation and in the event of filter failure; Likelihood
_ p r
3* 10 /plant-year for normal filter operation^ ld~ /plant-year for
simultaneous failure of two series HEPA filters.
D.I Criticality
Criticality could occur at several locations within the mixed oxide
plant. Criticality in the dry operations is most likely to occur prior
to blending, where undiluted plutonium oxide powder is handled. Scrap
recovery, however, is a more probable location for a criticality
incident, since the fissile material is moderated by water and trans-
ferred as a solution. These high risk operations, however, are
recognized in the design and operation of the plant, and equipment
is sized to be maintained in a safe geometry or administrative limits
are placed on batch sizes.
163
-------
TABLE 4-42
RADIONUCLIDE RELEASE RESULTING FROM SEVERE GLOVE BOX
DAMAGE AT THE MIXED OXIDE
FABRICATION PLANT
Nuelide Activity Released (C1)*
Pu 236 1.7 x 10"5
Pu 2$8 1.4 x 10"4
Pu 239 1.5 x 10"5
Pu 240 2.5 x 10"5
Pu 241 6.0'X 10"3
Pu 242 1.3 x 10"7
* The values tabulated are for normal operation of the two HEPA
filters. In the event of simultaneous failure of the filters,
these values are increased by a factor of 105.
164
-------
Source Term
Criticality is assumed to occur in the dissolution step of scrap recovery,
8
which may contain as much as 2.5 kg of plutonium. For the ten solution
31
critical1t1es on record, the average number of fissions on record was
18
*v»4 x 10 . However, one of the incidents involved a record ~ 4 x
19
10 fissions at the Chemical Processing Plant of the Idaho Reactor
Testing Area, considered unrepresentative of the fuel fabrication oper-
ations. Neglecting this incident, the average number of fissions 1s
17 18
ft/4 x 10 .Other evaluations have selected rtflO fissions as represen-
8,16
tatlve for criticality 1n a fuel fabrication plant, and this estimate
will be adopted here as well.
18
The radionuclides released ten minutes after an incident Involving 10
?T5 44
fissions, based upon ORI6EN calculations (for a U criticality Incident)
are given in Table 4-43. These estimates further assume 100% of the noble
gases and 50% of the halogens released to the environment. Approximately
0.2% of the activities created in the excursion are released to the
building air. Additionally, some release of plutonium in the solution
can be expected. For an excursion terminated following the evaporation
of 10 liters of excess solution, containing 150 gms of Pu per liter,
approximately 0.2%, or 3 gms of plutonium are assumed to be rendered
8
airborne. This contribution to the source term is included in Table
-5
4-43, following an attenuation by a factor of 10 from two stages of
8
HEPA filtration operating normally in series. Additionally, the
18
neutron and gamma radiation associated with a burst of 10 fissions
165
-------
TABLE 4-43
RADIONUCLIDE RELEASE RESULTING FROM A CRITICALITY INCIDENT
AT THE MIXED OXIDE FUEL
FABRICATION FACILITY
Nuclide Activity Released CCi) Nuclide Activity Released (C1)
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Kr
Kr
Kr
Kr
Kr
Kr
Kr
I
80
80m
82
82m
83
84
84m
85
86
87
83m
85
85m
87
88
89
90
128
1.
5.
1.
3.
3.
3.
1.
7.
1.
2.
3.
3.
1.
9.
6.
5.
9.
1.
2
7
9
2
5
5
5
2
5
1
4
2
3
4
3
2
2
4
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10"1
io-5
io-5
io-3
IO1
IO1
io-1
io-5
101
IO1
IO1
IO2
io-2
io-3
I
I
Xe
Xe
Xe
Xe
Xe
Xe
Xe
Xe
Th
Th
Pa
U
U
U
U
U
136
137
133
133m
135
135m
137
138
139
140
*
231
*
234
234m*
*
233
*
234
*
235
- *
236
*
237
2
3
1
9
2
4
9
7
1
2
2
2
1
1
3
4
1
7
.3
.3
.7
.9
.2
.7
.0
.0
.5
.2
.0
.2
.8
.7
.4
.5
.3
.1
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
IO1
ID'4
l(f3
ID'5
IO2
io2
io-7
io-13
io-14
io-14
ID'20
io-18
10'11
io-16
!0-10
166
-------
TABLE 4-43
(continued)
Nuclide Activity Released (C1) Nucllde Activity Released (Ci)
I 130
I 131
I 132
I 133
I 134
I 135
1.0 x 10"3
5.5 x 10~2
1.6
1.5
5.2 x 101
2.3 x 101
U 238*
U 239*
Np 237*
*
Np 239
Np 240*
Pu 236*
*
Pu 238
Pu 239*
Pu 240*
*
Pu 241
*
Pu 242
1.1 x ID'10
6.1 x 10"5
4.4 x 10"21
_7
1.5 x 10 '
3.5 x 10"16
1.1 x 10"6
-fi
9.7 x 10 b
1.0 x 10"6
1.7 x 10"6
-4
4.1 x 10 *
_Q
9.0 x 10 y
* The values tabulated are for normal operation of the two HEPA
filters. In the event of simultaneous failure of the filters,
these values are increased by a factor of 105.
167
-------
would result in a dose to the population. The methodology for evaluating
this direct "shine" dose is discussed in Section 3.2.
Likelihood
The likelihood of a criticality incident in a mixed oxide fuel fabrication
-3 8*
plant has been estimated to be approximately 8x10 /plant-year. As
discussed in Reference 8. am improved basis for such an estimate would con-
sider the total fuel throughput, the fuel forms during processing and the
fuel reactivities involved.
In a separate study for a generic fuel reprocessing facility, the likelihood
_5 °
of criticality has been estimated to be approximately 3 x 10 /year.
The likelihood of a simultaneous failure of the two series HEPA filters has
-3 9
been further estimated to be roughly 10 /demand.
Estimate: Souroe term - Badionuolides given in Table 4-43 to air for
normal filter operation and in the event of filter failure plus ionizing
—3
radiations (see Section 3.2 for dose methodology); Likelihood+H 8 x 10
r _/j _ O
to 2 x 10~ /plant-year for normal filter operation, 8 x 10 to 3 x 10 /
plant-year for simultaneous failure of two series HEPA filters.
* The estimate contained in Reference 8 has been revised to reflect the
increase in experience to 490 plant-years.
168
-------
4.8 Plutonium Storage
The only accident considered in the generic plutonium storage facility
is criticality. The solid plutonium oxide is neither flammable nor
explosive, and no flammable or explosive compounds are expected in signif-
icant amounts in the storage area. It is not possible for overheating of
the plutonium to occur, since the material is cooled by natural convec-
tion air. Spills are unlikely when the material is in the solid state,
and if they were to occur, the contamination would be contaiped largely
within the building with negligible risk to the environment. The building
structures will be engineered to resist earthquakes and tornadoes, and
the storage/site will be selected to withstand the maximum credible flood.
Natural events outside of the range of design basis tornadoes, earth-
quakes, or floods are not considered within the scope of the current
study.
There are no existing storage facilities for plutonium of the capacity of
the generic plutonium storage facility. A few small storage facilities
located at or in the vicinity of fuel fabrication or reprocessing plants
handle plutonium in the liquid form as a plutonium nitrate solution.
Thus there does not exist an adequate data base to assess the likelihood
of accidents at the generic plutonium storage facility postulated in this
study.
0.1 Criticality
The generic plutonium storage facility is envisioned to store plutonium
in carefully analyzed, predetermined arrays. Because of the quiescent,
169
-------
uniform nature of storage, with no other operations planned, the likeli-
hood of critical ity is extremely remote. A number of storage containers
would have to be forcefully ruptured or crushed simultaneously and their
contents would have to be collected in a favorable geometry for criticality
to occur.
Source Term
31
For the 26 criticality incidents on record, the total number of fissions
15 19
range from *s3 x 10 to 4 x 10 . For the 11 non-solution cr1ticalit1es,
considered more representative of a postulated critical ity incident
involving solid plutonium storage, the average number of fissions is
The radionuclides released ten minutes after an Incident involving
17 44
10 fissions, based upon ORIGEN calculations, are given in Table 4-44.
These estimates further assume 100% of the noble gases and 50% of the
halogens released to the environment. Approximately 0.2% of the actlnldes
created in the excursion are released to the building air. Additionally,
some release of the plutonium powder in the containers can be expected
from the excursion. It is assumed that the energy release in the crit-
icality would be sufficient to rupture one of the storage containers and
release 1% of the 4.3 kg of plutonium in the container to the building
atmosphere. This contribution to the source term is included 1n Table
4-44, following an attenuation by a factor of 10" from two stages of HEPA
8
filtration in series during normal operation. In the event of simul-
taneous failure of the two HEPA filters, all of the particulate activity
discharged to the building is assumed to be released to the atmosphere.
170
-------
TABLE 4-44
RADIONUCLIDE RELEASE RESULTING FROM A CRITICALITY INCIDENT
AT THE PLUTONIUM STORAGE FACILITY
Nucltde Activity Released (Ci) Nuclide Activity Released (Ci)
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Kr
Kr
Kr
Kr
Kr
Kr
Kr
I
I
80
80m
82
82m
83
84
84m
85
86
87
83m
85
85m
87
88
89
90
128
130
1.
5.
1.
3.
3.
3.
1.
7.
1.
2.
3.
3.
1.
9.
6.
5.
9.
1.
1.
2 x
7 x
9 x
2 x
5 x
5
5 x
2
5 x
1 x
4 x
2 x
3
4
3
2 x
2 x
4 x
0 x
io-2
10
10
10
10
10
10
10
10
10
10
10
10
-6
-6
-4
•]
-1
1
-1
-1
-2
-6
1
-3
-4
io-4
I
I
Xe
Xe
Xe
Xe
Xe
Xe
Xe
Xe
Th
Th
Pa
U
U
U
U
U
U
136
137
133
133m
135
135m
137
138
139
140
*
231
234*
234m*
*
233
*
234
*
235
*~
236
*
237
238*
2.
3.
1.
9.
2.
4.
9.
7.
1.
2.
2.
2.
1.
1.
3.
4.
1.
7.
1.
3
3
7
9
2
7
0
0
5
2
0
2
8
7
4
5
3
1
1
x
x
X
X
X
X
X
X
X
X
X
X
X
X
x'
X
X
X
io-5
ID'4
io-6
10"'
10"'
i
io1
10'
io-1
io-8
io-14
io-15
io-15
91
10""
_1Q
10 IS
ID'12
io-17
io-11
io-1'
171
-------
TABLE 4-44
(continued)
NuclIde Activity Released (Ci) Nuclide Activity Released (C1)
I 131 5.5 x 10~3
_1
I 132 1.6 x 10 '
I 133 1.5 x 10"1
I 134 5.2
I 135 2.3
U 239*
*
Np 237
Np 239*
Np 240*
*
Pu 236
*
Pu 238
*
Pu 239
Pu 240*
Pu 241*
n.. o/io
6.1 x 10"6
22
4.4 x 10""
1.5 x 10"8
3.5 x 10"17
/-
1.6 x 10"°
.
1.4 x 10"^
j.
1.4 x 10"5
2.4 x 10"5
5.9 x 10"3
T O \.* 1 f\mm *
* The values tabulated are for normal operation of the two HEPA filters.
In the event of simultaneous failure of the filters, these values are
increased by a factor of 105.
172
-------
Additionally, the neutron and gamroa radiation associated with a burst of
1017 fissions would result in a dose to the population. The methodology
for evaluating thts direct "shine" dose is discussed in Section 3.2.
Likelihood
The likelihood of a criticality incident in a mixed oxide fuel fabrication
-3 8*
plant has been estimated to be approximately 8 x 10 /plant year . The
extrapolation of a probability estimated for fuel fabrication operations
to a storage situation is questionable, but no other data exist. The
only criticalities on record in fuel fabrication facilities occurred
in solution. Dry criticalities have only occurred in reactor experiments.
Since plutonium storage operations are dry, the likelihood of criticality
is estimated to be at least an order of magnitude lower than the above
estimate. Moreover, fuel fabrication operations are active, whereas
plutonium storage is a quiescent operation. This factor is assumed to
reduce the likelihood of criticality by at least another order of magni-
tude. The likelihood of a simultaneous failure of the two series HEPA
-3 9
filters has been further estimated to be roughly 10 / demand.
Estimate: Source term - Radionuolides given in Table 4-44 to air for normal
filter operation and in the event of filter failure plus ionizing radiations
(see Section 3.2 for dose methodology); Likelihood < 8 x 10~ /plant-year
— 8
for normal filter operation, 8 x 10~ /plant-year for simultaneous failure
of two series HEPA filters.
* The estimate contained in Reference 8 has been revised to reflect the
increase in experience to 490 plant/years.
173
-------
4.9 Transportation
The accidents considered in transportation of fuel cycle materials are
given in Table 4-45. The accident category format given in Table
4-1 is not considered appropriate to the consideration of transportation
accidents.
Only improperly closed plutonium oxide packages have been considered at
the low consequence end of the accident spectrum since the risk associated
with the release of ore or ore concentrate is insignificant, and the
remaining fuel cycle materials are not dispersible in the absence of a
driving force
-------
TABLE 4-45
TRANSPORTATION ACCIDENTS
1. Leakage of coolant from Irradiated fuel cask
2. Improperly closed plutonlum oxide container
3. Release from a collision Involving natural uranium
hexafluorlde
4. Release from a collision Involving enriched uranium
hexafluorlde'
5. Release from a collision Involving Irradiated fuel
6. Release from a collision involving irradiated fuel
followed by release of fuel from the cask
7. Release from a collision involving plutonium oxide
8. Criticality of unirradlated fuel
9. Criticality of enriched U02
10. Criticality of plutonium oxide
175
-------
The reliance for safety in transport of radioactive materials is placed
on packaging. The packaging must conform with regulatory standards
established by the Department of Transportation, the Nuclear Regulatory
Commission, and some of the States. The standards require the packaging
to prevent the loss or dispersal of the radioactive contents, retain the
shielding efficiency, assure criticality safety, and provide adequate heat
dissipation. Type A packaging, which may be used to transport uranium
concentrate, natural uranium hexafluoride, and low level wastes, is unlikely
to be breached in a minor accident, and some fraction would not be breached
in very severe accidents. Type B packaging, used for the remaining
materials transported in the fuel cycle, is likely to withstand all but
very severe accidents.
In the past 25 years, approximately 2600 packages of irradiated fuel have
42
been transported in routine commerce. The number of shipments of the
other materials in the fuel cycle have not been well documented.
Much of the data presented in this section was developed in two earlier
studies. The first was a survey of the transportation of radioactive mate-
10
rials to and from nuclear power plants, and the second was an assessment of
48
transportation risks in the nuclear power industry.
1. Leakage of Coolant from Irradiated Fuel Cask
A rail cask carries up to 18 LWR fuel assemblies. The cask must provide
the means to dissipate the heat produced by radioactive decay. In the
usual configuration, water coolant is sealed in the cask and the heat is
dissipated through natural convection to the surface, which is cooled in
turn by natural or forced air convection over fins. The rugged, leaktight
176
-------
design of the cask, coupled with rigid procedures imposed on the shipper,
prevent leaks from occurring. Also, each cask is held at the origin
until checks have been made on pressure, temperature and leakage. However,
a small leak which is undetectable by visual observation could go unchecked
for some time.
Source Term
A previous analysis of the consequences of a leak in a spent fuel shipping
cask indicated that a leakage rate of approximately 0.001 cm /sec. is the
10
largest that can go undetected by visual observation. Based upon
0.25% of the fuel rods being perforated, it was estimated that approximately
2
l^Ci/cm of gross fission product activity might be in the cask coolant,
and that this activity is primarily Cs-137. In five days, approximately
400/4. Ci of activity would be released to the surface of the cask, and
approximately 1% of this activity is assumed to be released to the environ-
ment as an aerosol.
Likelihood
Of the 3600 shipments of spent fuel, there have been no incidents recorded
involving a leak of coolant from the cask. Accordingly, lacking any other
data to estimate the probability of a small leak, it is estimated that
the likelihood of a small leak of irradiated fuel coolant is 3f 3 x 10~4/
shipment.
177
-------
137 t* -4
Estimate: Source term SV 4/4.0, Cs to air; Likelihood ^ 3 x W /
shipment.
2. Improperly Closed Plutonium Oxide Container
Current package designs for plutonlum oxide incorporate sealed metal cans
within an inner, gasketed steel container supported within an outer steel
drum of from 10 to 100 gallon capacity. Current package designs are
expected to be modified to dissipate up to a maximum of 100 watts of heat,
5
allowing container loadings of up to 6.4 kg Pu. It is envisioned that
4D of these containers would comprise a truck shipment. The generic
shipping distance is estimated to be 300 miles from the reprocessing
facility to the plutonium storage facility and an equal distance to the
mixed oxide fabrication plant. A small amount of dispersible plutonium
might be released.to the environment from an improperly closed package.
Source Term
An improperly closed package containing plutonium oxide would release only
a small amount of its contents because of the absence of a driving force
and the several barriers between the powder and the environment. It is
assumed that in the event of a complete breach of containment, 0,1% of the
plutonium oxide would be released from the package. According to estimates
49
made by Mishima, the fractional airborne release of plutonium dioxide
particles less than 10/xm from either a stainless steel or asphalt-gravel
aggregate surface at a nominal wind speed of 2.5 mph is approximately
8 x 10 in 24 hours. Thus for a 6.4 kg container, approximately 5.1 mg
of plutonium might be released to the environment. The estimated activities
of the plutonium isotopes released are given in Table 4-46.
178
-------
TABLE 4-46
RADIONUCLIDE RELEASE RESULTING FROM AN IMPROPERLY
CLOSED PLUTONIUM OXIDE CONTAINER
Nuclide Activity Released (Ci)
Pu 236 1.9 x 10~4
Pu 238 1.7 x 10~3
Pu 239 1.8 x 10"4
Pu 240 2.9 x 10"4
Pu 241 7.0 x 10"2
Pu 242 1.5 x 10"6
179
-------
Likelihood
It has been estimated that for Type B packages, approximately 1 1n 100,000
10
might be improperly closed when shipped. Then, on the basis of 40 con-
-4
tainers per shipment, the likelihood of improper closure is 4 x 10 /
50
shipment. Also in a recent survey of package closure faults, while
several container packaging errors were observed, there was no instance of
complete loss of container integrity in 775 shipments.
Estimate: Source term - Radionuolides given in Table 4-46 to air. Like-
lihood $4 I x W~ /shipment to 4 x 10~ /shipment.
3. Release from a Collision Involving Natural Uranium Hexafluouride
Natural UFg is typically shipped as a solid from the UFg conversion plant
to the enrichment facility in cylinders containing 12.7 MT of UFg. The
generic shipping distance is estimated to be 500 miles and only one
cylinder would be shipped in a truck. A collision resulting in a fire
could volatilize and release the UFC to the environment.
o
Source Term
For a fire of long duration following a collision, the entire contents of
the cylinder could be volatilized and released to the environment. This
would result in approximately 8600 kg of freshly separated uranium dis-
charged to the atmosphere. The isotopic composition of this uranium is
given in Section 4.3.
Likelihood
Although the cylinder could conceivably be breached by a less than severe
accident, it would take a severe accident to produce a fire of sufficient
180
-------
duration to vaporize the entire contents of the cylinder. The accident
probabilities for truck, rail, and barge accidents of various severities
10
are given in Table 4-47. For a shipping distance of 500 miles, then,
the likelihood of an accident resulting in a fire which volatilizes the
UFg is approximately 4 x 10~ .
9 ? o 7 P?^ P^4
Estimate: Source term % 2.8 Ci V, 1.3 x 10~ Ci U3 3.0 Ci V,
2.8 Ci Th to air; Likelihood & 4 x 10~ /shipment.
4. Release from a Collision Involving Enriched Uranium Hexafluoride
Enriched UFg is shipped as a solid from the enrichment plant to the fuel
fabrication facility in cylinders containing 2.2 MT of UFg. The generic
shipping distance is estimated to be 750 miles and typically five cylinders
would comprise one truck shipment. A collision resulting in a fire could
volatilize and release the UFC to the environment.
o
Source Term
For a fire of long duration following a collision, the entire contents of
the cylinders could be volatilized and released to the environment. This
would result in approximately 7400 kg of enriched uranium discharged to
the atmosphere. The isotopic composition of this uranium is given in
Table 4-17.
likelihood
Although the cylinders could conceivably be breached by a less than severe
accident, it would take a severe accident to produced fire of sufficient
duration to vaporize the entire contents of the cylinders. The accident
181
-------
TABLE 4-47
ACCIDENT PROBABILITIES FOR TRUCK, RAIL, AND BARGE ACCIDENTS
10
OF VARIOUS SEVERITIES
Severity Accident Probability
(per vehicle mile)
Minor 2 x 10"6
Moderate 3 x 10"7
Severe 8 x 10"9
Extra Severe 2 x 10
Extreme 1 x 10"13
182
-------
probabilities for truck, rail, and barge accidents of various severities
10
are given in Table 4-47. For a shipping distance of 750 miles, then,
the likelihood of an accident resulting in a fire which volatilizes the
UF- is approximately € x 10" .
o
970 ? 37 -1
Estimate: Source term % 2.4 Ci ^aU, 2.8 Ci U, 5.3 x 10 Ci
226U3 4.1 x I0~l Ci 235U, 17 Ci 234U to air; Likelihood S* 6 x 10~6/
shipment.
5. Release from a Collision Involving Irradiated Fuel
Approximately 3.7 MT of irradiated fuel, comprising from 7 to 18 LWR fuel
assemblies, are transported in a cask from the reactor to the reprocessing
plant after a minimum of 150 days cooling. The spent fuel cask is de-
signed to provide shielding, criticality safety, heat dissipation, and pro-
tection against severe accidents. The cask is assumed to be transported
for 20 miles by truck to the rail head, where it is shipped 1000 miles
to the reprocessing facility by rail. An extra severe collision coupled
with a fire could cause a breach in the cask, releasing a fraction of
the contained fission products to the environment.
Source Term
The estimated quantities of fission products released to the environment in
the event of a collision involving irradiated fuel are given in Table 4-48.
The values given are based upon release fraction estimates from References
10 and 48. These sources are in agreement in their predictions of the
release fractions for Kr-85 (*/3%) and iodine (/^0.2%). Their estimates for
the release fractions of the remaining fission products, however, differ by
183
-------
TABLE 4-48
RADIONUCLIDE RELEASE RESULTING FROM AN EXTRA SEVERE
COLLISION INVOLVING IRRADIATED FUEL
Nucllde Activity Released (Ci)
Kr 85
I 129
I 131
Ru 103
Ru 106
Zr 95
Nb 95
Sr 89
Sr 90
Y 90
Y 91
Cs 134
Cs 137
Ce 141
Ce 144
Pm 147
1200
1.5 x 10"5
1.6 x 10"2
27
140
80
150
20
19
19
42
55
27
18
200
32
184
-------
an order of magnitude. Reference 10 assumes that A/10" % of the remaining
fission products are released to the environment, whereas Reference 48
o
adopts the more conservative value of 10 %. Accordingly, the values 1n
Table 4-48 assume a release fraction of 5 x 10" . Moreover, the 1sotop1c
breakdown of fission products other than krypton and Iodine given 1n
Table 4-48 1s based upon the tabulated activities 1n the fuel given 1n
Table 4-24.
Likelihood
According to Reference 10, an "extra severe" accident would be required
for a break.to occur 1n a spent fuel shipping cask. The probability of ar
extra severe accident, as tabulated in Table 4-47, is 2 x 10" /vehicle-
mile. However, according to a more recent analysis involving fault tree
48
analysis, the average accident involving a typical release of radlo-
_g
activity is expected to have a probability of 4.6 x 10 /mile for a truck
shipment, and 9.6 x 10"9/mile for a rail shipment. The typical shipment
of spent fuel travels a distance of 20 miles by truck and 1000 miles by
rail.
Estimate: Source term - Radionuolides given in Table 4-48 to air;
—fi —8
Likelihood j& 9 x 10~ /shipment to 2 a 10 /shipment.
6. Release from a Collision Involving Irradiated Fuel Followed by
Release of Fuel from the Cask
It is possible, although highly improbable, in the event of a collision,
that the irradiated fuel cask could be damaged to the extent that one or
more of the fuel elements would be released from the cask.
185
-------
Source Term
The release magnitude of radlonuclldes to the environment is assumed to
be similar to that given in Table 4-48. However, if irradiated fuel
elements are released from the cask, a significant dose to the population
results from the direct radiation "shine" emitted by the decaying fission
products in the fuel. Assuming that seven irradiated fuel assemblies are
released from the cask in an extremely severe collision, the radiation
4 10
level at 100 feet has been estimated to be 10 r/hr. This dose is
attenuated by the inverse square relationship and a dry air removal
cross section, as described in Section 3.2, to arrive at a population
dose from direct shine. It is further assumed that the fuel elements
10
remain unshielded for 10 hours, and that the uniform population density
2
beyond 10 meters is 290 people/mi , as given in Table 3-1. It is unreal-
istic to expect that the population close to the accident would remain
exposed to the radiation for the full 10 hours. Accordingly, it 1s
assumed that people within 100 meters of the fuel would be evacuated within
1/2 hour and those within one mile would be evacuated within 2 hours after
the accident (a total of /v800 people). The population dose from direct
shine is added to the total body dose accruing from exposure to the
radlonuclldes given in Table 4-48.
Likelihood
According to Reference 10, the release of irradiated fuel elements from
the cask could only occur under extremely severe accident conditions. Refer-
ring to Table 4-47, the likelihood of an extremely severe accident is roughly
a factor of 10 less likely than that of an accident of sufficient
186
-------
severity to breach the shipping container. Applying this factor to the
breach likelihoods given in the previous section, the likelihood of a
release of fuel elements from the cask for a generic shipment of approxi-
-8
mately 1000 miles is estimated to range from 9 x 10 / shipment to
2 x 10~10/shipment.
Estimate: Source term - Radionuolides given in Table 4-48 to air plus
ionizing radiations (see Section S.2 for dose methodology); Likelihood &
_o —TO
9 x 10 /shipment to 2 x 10~ /shipment.
7. Release from a Collision Involving Plutonium Oxide
In the event of a very severe collision, a container holding 6.4 kg of
dispersible plutonium oxide might be breached, releasing a fraction of
its contents to the environment. The generic shipping distance is
estimated to be 300 miles from the reprocessing facility to the plutonium
storage facility and an equal distance to the mixed oxide fabrication
plant.
Source Term
The estimated quantities of plutonium isotopes released to the environment
in the event of a severe accident are given in Table 4-49. These
estimates are based upon 0.1% of the material in a container holding
6.4 kg of plutonium oxide being rendered immediately airborne and released
to the environment when a container fails in an accident, A fractional
release to the environment of 0.1% has been estimated as a nominal value
in Reference 51, and as a large release in Reference 48,
187
-------
TABLE 4-49
RADIONUCLIDE RELEASE RESULTING FROM
A COLLISION INVOLVING PLUTONIUM OXIDE
Nuclide Activity Released (Ci)
Pu 236 2.4 x TO"1
Pu 238 2.1
Pu 239 2.3 x 10"1
Pu 240 3,6 x 10"1
Pu 241 8.8 x 101
Pu 242 1.9 x 10"3
188
-------
Likelihood
According to Reference 42, even under extra severe collision conditions,
plutonium'oxide containment is not considered to be breached. However,
two recent studies have derived, using fault tree analysis, finite
probabilities of release in the event of an accident. Reference 51
estimates one accident involving a release for every 220 accidents during
transport. Coupling this estimate with the estimated truck accident
probability of 2.5 x 10 /mile, a release likelihood of 1,1 x, 10 /mile
IP
is derived. A considerably lower estimate of approximately 7 x 10 /
mile is obtained in Reference 48 corresponding to the release fraction of
0.1%. Using/an average shipment distance of 300 miles, the likelihood of
6 -9
a release, then, lies in the range of 3 x 10 /shipment to 2 x 10 /ship-
ment.
Estimate: Source term - Radionuolides given in Table 4-49 to air.
-6 -9
Likelihood^ 3 x 10 /shipment to 2 x 10 /shipment.
8- Criticality of Unirradiated Fuel
Unirradiated fuel, both fuel assemblies and mixed oxide fuel pins, are
shipped by truck in metal containers which support the fuel along its
entire length during transport. A single shipment might consist of from
six to sixteen packages, each containing the equivalent of two LWR
assemblies. The shipping containers are designed for criticality safety
under all credible accident conditions, including submersion in water.
139
-------
Mixed oxide fuel pins are assumed to be transported approximately 200
miles by truck to the fuel fabrication plant and completed assemblies
approximately 1000 miles by truck to the reactor. Only under extra severe
accident conditions, which could compromise the safe geometry, coupled with
the presence of water moderation, would a criticality incident be possible.
Source Term
31
For the 26 criticality incidents on record, the total number of fissions
15 19
range from ^3 x 10 to 4 x 10 . For the 11 non-solution incidents,
considered more representative of a postulated criticality involving
unirradiated fuel in transport, the average number of fissions is -^10 .
However, criticality in unirradiated fuel is not expected to cause any
release of radioactive materials from the fuel pins, since the fuel
10
cladding would retain the fission products created in the excursion.
The only anticipated source of exposure to the general population beyond
10 meters from the excursion would be from the prompt neutron and gamma
radiation associated with the burst. The methodology for evaluating the
population dose under these conditions is discussed in Section 3.2.
Likelihood
A criticality has never occurred in the transportation of fissile mate-
rials. For a criticality to occur, a rearrangement of the fuel into a
favorable geometry must take place. It is assumed that such an event
* The shielding factor associated with this methodology is retained, since
the water necessary for moderation/reflection is assumed to provide
shielding equivalent to 8 inches of concrete.
190
-------
could only take place under extremely severe accident conditions. Refer-
ring to Table 4-47, the likelihood of an extremely severe accident is
_2
roughly a factor of 10 less likely than that of an accident of suffi-
cient severity to breach the shipping container. Lacking estimates for -.
a fresh fuel container, it is assumed that the likelihood of an accident
of sufficient severity to breach the container is comparable to
_Q
previous estimates for spent fuel containers, namely 4..6 x 10 /mile to
2 x 10" /mile for shipments by truck. Thus for the assumed 1000 mile
distance from the fabrication facility to the reactor,* the likelihood of
-8
a criticality incident for fresh fuel is in the range of 5 x 10 /shipment
to 2 x 10"10/shipment.
Estimate: Source term - Ionizing radiations associated with a criticality
27
of 10 fissions (see Section 3.2 for dose methodology); Likelihood ^
—8 —JO
5 x 10~ /shipment to 2 x 10~ /shipment.
9. Criticality of Enriched UP
Enriched IKL powder is shipped by truck in containers designed to prevent
criticality under all credible normal transport conditions as well as
severe accidents. A generic shipment of uranium dioxide contains approx-
imately 4.5 MT of powder for a distance of 750 miles. Only under extra
severe accident conditions, which could compromise the safe geometry,
coupled with the presence of water moderation, would a crit1cal1ty
incident be possible.
* The shipments of mixed oxide fuel pins from the mixed oxide plant to
the uranium fabrication plant has the effect of Increasing the total
travel miles by roughly 7%.
191
-------
Source Term
Approximately 10 fissions, representative of non-solution criticality
incidents, are considered appropriate for a postulated criticality
involving enriched U(L in transport. The radionuclides released ten
minutes after an incident involving 10 fissions, based upon ORIGEN
44
calculations, are given in Table 4-50. These estimates assume that
100% of the noble gases and halogens, and 0.2% of the actinides created
in the excursion are released to the environment. Additionally, the neutron
and gamma radiation associated with a burst of 10 fissions would result
in a dose to the population assumed to be uniformly distributed beyond 10
meters from the excursion. The methodology for evaluating this direct
"shine" dose is discussed in Section 3.2.
Likelihood
A criticality has never occurred in the transportation of fissile mate-
rials. For a criticality to occur, the containers of IKL would have to
depart from the safe geometry provided by the shipping containers. It
is assumed that such an event could only occur under extremely severe
accident conditions. Referring to Table 4-47, the likelihood of an
_2
extremely severe accident is roughly a factor of 10 less likely than an
accident of sufficient severity to breach the shipping container. Lack-
ing estimates for a U02 container.it is assumed that the likelihood of
an accident of sufficient severity to breach the container is comparable
192
-------
TABLE 4-50
RADIONUCLIDE RELEASE RESULTING FROM A CRITICALITY INCIDENT
INVOLVING ENRICHED U02 IN TRANSPORT
Nuclide Activity Released (C1) Nuclide Activity Released (Ci)
Br 80
Br 80m
Br 82
Br 82m
Br 83
Br 84
Br 84m
Br 85
Br 86
Br 87
Kr 83m
Kr 85
Kr 85m
Kr 87
Kr 88
Kr 89
Kr 90
I 128
I 130
I 131
I 132
I 133
I 134
I 135
2.4 x 10"2
1.1 x 10"5
3.8 x 10"6
6.4 x 10"4
7.0 x 10"1
7.0
3.0 x 10'1
1.4 x 10'1
3.0 x 10"1
4.2 x 10"1
3.4 x 10"2
3.2 x 10"6
1.3
9.4
6.3
5.2 x 101
9.2 x 10"3
2.8 x 10"4
2.0 x 10"4
1.1 x 10"2
3.2 x 10"1
3.0 x 10"1
1.0 x 101
4.6
I 136
I 137
Xe 133
Xe 133m
Xe 135
Xe 135m
Xe 137
Xe 138
Xe 139
Xe 140
Th 231
Th 234
Pa 234m
U 233
U 234
U 235
U 236
U 237
U 238
U 239
Np 237
Np 239
Np 240
Pu 239
4.6
6.6 x 10"5
1.7 x 10"4
9.9 x 10"6
2.2 x 10"1
4.7 x 10"1
9.0 x 101
7.0 x 101
1.5 x 10"1
2.2 x 10"8
2.0 x 10"9
2.2 x 10"10
-in
1.8 x 10 IU
1.7 x 10"16
3.4 x 10"14
4.5 x 10"7
1.3 x 10"12
7.1 x 10"6
1.1 x 10"6
6.1 x 10"1
4.4 x 10"16
1.5 x 10"3
3.5 x 10"11
4.1 x 10"13
193
-------
-8
to previous estimates for a Pu02 container, namely 1.1 x 10" /mile to
7 x 10"12/mile. Thus for the assumed 750 mile distance from the U02
plant to fhe fabrication facility,the likelihood of a criticality inci-
O IT
dent for U02 is in the range of 8 x 10 / shipment to 5 x 10 '/shipment.
E, timate: Source term - Radionuclides given in Table 4-50 to air plus
i nizing radiations (see Section 3.2 for dose methodology); Likelihood
a _77
8 x 10~ /shipment to 5 x 10 /shipment.
10. Criticality of Pu00
WT"~L-'m-- *—"-""£,
Plutonium dioxide is shipped by truck within metal cans contained in steel
containers supported within an outer steel drum. Approximately 40 containers,
each containing up to 6.4 kg Pu, are contained in a generic shipment of
300 miles from the reprocessing plant to the mixed oxide fabrication plant.
The shipping containers are designed to prevent criticality under all
conceivable conditions. Only under extra severe accident conditions
could the safe geometry be compromised, leading to a potential criticality
incident.
Source Term
Approximately 10 fissions, representative of non-solution criticality
incidents, is considered appropriate for a postulated criticality involv-
ing Pu02 in transport. The radionuclides released ten minutes after an
194
-------
incident involving 10 fissions, based upon ORIGEN calculations (for a
235 44
U crlticality), are given in Table 4-5). These estimates assume
that 100% of the noble gases and halogens, and 0.2% of the actinides
created in the excursion are released to the environment. Additionally,
it is assumed that the energy release would be sufficient to expel
approximately 1% of the plutonium in one of the containers. This con-
tribution to the source term is included in Table 4-51. Finally, the
neutron and gamma radiation associated with a burst of 10 fissions would
result in a dose to the population. The methodology for evaluating this
direct "shine" dose is discussed in Section 3.2.
Likelihood
A criticality has never occurred in the transportation of fissile mate-
rials. For a criticality to occur, the containers of PuO« would have to
depart from the safe geometry provided by the shipping containers. It is
assumed that such an event would only occur under extremely severe
accident conditions. Referring to Table 4-47, the likelihood of an
_2
extremely severe accident is roughly a factor of 10 less likely than an
accident of sufficient severity to breach the shipping container.
According to estimates given earlier, the likelihood of an accident of
o
sufficient severity to breach the container is in the range of 1.1 x 10. /
12
mile to 7 x 10 /mile. Thus for the assumed 300 mile distance of the
shipment, the likelihood of a criticality incident for Pu02 is in the
range of 3 x 10"8/shipment to 2 x 10-11/shipment.
Estimate: Source term - Radionuclid.es given in Table 4-51 to air plus
ionizing radiations (see Section 3.2 for dose methodology); Likelihood
n 77
£* 3 x 10 /shipment to 2 x 10~ /shipment.
195
-------
TABLE 4-51
RADIONUCLIDE RELEASE RESULTING FROM A CRITICALITY
INCIDENT INVOLVING PuOgIN TRANSPORT
Nuclide Activity Released (Ci) Nuclide Activity Released (Ci)
Br 80
Br 80m
Br 82
Br 82m
Br 83
Br 84
Br 84m
Br 85
Br 86
Br 87
Kr 83m
Kr 85
Kr 85m
Kr 87
Kr 88
Kr 89
Kr 90
I 128
I 130
I 131
I 132
I 133
I 134
I 135
2.4 x IO"2
1.1 x 10"5
3.8 x IO"6
6.4 x IO"4
7.0 x IO"1
7.0
3.0 x IO"1
1.4 x IO1
3.0 x IO"1
4.2 x IO"1
3.4 x IO"2
3.2 x IO"6
1.3
9.4
6.3
i
5.2 x IO1
9.2 x IO"3
2.8 x IO"4
2.0 x IO"4
1.1 x IO"2
3.2 x IO"1
3.0 x IO"1
1
1.0 x 101
4.6
I 136
I 137
Xe 133
Xe 133m
Xe 135
Xe 135m
Xe 137
Xe 138
Xe 139
Xe 140
Th 231
Th 234
Pa 234m
U 233
U 234
U 235
U 236
U 237
U 238
U 239
Np 237
Np 239
Np 240
Pu 236
Pu 238
Pu 239
Pu 240
Pu 241
Pu 242
4.6
6.6 x IO"5
1.7 x IO"4
9.9 x IO"6
2.2 x IO"1
4.7 x IO"1
9.0 x 101
7.0 x IO1
1.5 x IO"1
2.2 x IO"8
2.0 x IO"9
2.2 x IO"10
1.8 x IO"10
-16
1.7 x 10 lb
3.4 x IO"14
_7
4.5 x 10 '
1.3 x IO"12
7.1 x IO"6
1.1 x 10"6
6.1 x IO"1
4.4 x IO"16
1.5 x IO"3
• -11
3.5 x 10 "
2.4
21
2.1
3.6
870
1.9 x 10"2
196
-------
5. RISK ASSESSMENT
For each component of the fuel cycle, and for the source terms associated
with the accidents discussed in Section 4, the population dose commitment
has been evaluated using the methodology discussed in Section 3.2. For
each accident, the critical organ (organ receiving maximum dose) popula-
tion dose is given together with the population dose to the total body
(T.B.). Combining these results with the accident likelihoods also given in
Section 4, the expectation value of the population dose commitment is
derived and normalized to the annual operation of the generic 1000 MWe
LWR using the mass flow factors given in Section 2.2. The normalized
population dose commitments in man-rem are then converted to normalized
health risks (somatic effects) using the methodology discussed in Section
3.3. All of these results are presented in Tables 5-1 through 5-8 for
each component of the supporting LWR fuel cycle.
197
-------
5.1 Milling
The results for the accidents considered in milling are given in Table
5-1. The consequences and risks associated with the tailings slurry
release dominate the accidents from this component of the fuel cycle.
As pointed out in Section 4.2, however, since the likelihood of mill
tailings dike failure was obtained from historical data, and since the
construction techniques for these dikes have been improved, these results
probably provide an overestimate of the current risk.
193
-------
TABtE 5-1
ENVIRONMENTAL RISKS FROM ACCIDENTS IN URANIUM MILLING
Accident
B.I Fire In Solvent
Extraction Circuit
E.I Release of Tailings Slurry
from Tailings Pond
E.2 Release of Tailings Slurry
from Tailings Distribution
Pipeline
Totals
(bone
(T.B.
Population Dose
ror Generic Plant
(man-rem)
1.6 (lung)
1.0 x 10 1 (T.B.)
2.9 (bone)
1.9 x 10-1 (T.B.)
1.3 x 10"! (bone)
8.3 x 10"3 (T.B.)
Accident Population Dose
Likelihood Expectation Value
(plant-year)-' (man-rem)
3 x 10"3 to 4 x 10"* 4.8 x
3.0 x
— 4 x 10"2 1.2 x
7.6 x
~ 1 X 10"2 1.3 x
8.3 x
1.? x
8.0 x
10"3 to 6.4 x 10"* *
10" to 4.0 x 10
10"'
10"J
10"3
10"5
_]
10"3 to 7.7 x 10"3
Population Dose
per 1000 MWe-ye»r
(man-rem)
6.0 x 10"? to 8.0 x 10"5
3.8 x 10" to 5.0 x 10
1.5 x 10'2
9.5 x 10"*
1.6 x 10"?
1.0 x 10"b
1.5 x 10"2
1.0 x 10"° to 9.7 x 10
Health Risk
per 1000 MWe-vear
(1 of
3.8 x
5.5 x
5.8 x
5.9 x
excess cancers)
10"8to 5.0 x 10"9
10"7
10-*
10'7 to 5.6X 10'7
-------
5.2 UFg Conversion
The results for accidents considered 1n uranium hexafluorlde conversion
are given 1n Table 5-2. The highest consequence accidents are explosions
1n the uranyl nitrate evaporator or In the hydrogen reduction step of the
operation. The risk from the latter accident predominates, because of the
higher estimated probability range. If the high end of the estimate 1s
appropriate, the risk from a hydrogen explosion in the reduction step
dominates the risk from accidents 1n uranium hexafluorlde conversion. If
the low end of the probability range is more correct, the risk from
accidents in uranium hexafluoride conversion is dominated by accidents
involving the release of UFg from cylinders or piping/valve failures 1n
distillation. The probabilities of these events are based upon historical
data, and considerable attention has been given 1n recent years to re-
ducing such releases to a practical minimum.
200
-------
TABLE 5-2
ENVIRONMENTAL RISKS FROM ACCIDENTS IN URANIUM HEXAFLOURIDE CONVERSION
Accident
ro
0
A. '
A. 2
B.I
C.I
C.Z
E.I
Uraryl Nitrate
Evaforator Explosion
Hydrogen Explosion
In (eduction
Flrt 1n Solvent
Extraction Operation
Release from a
Hot UF, Cylinder
D
Valve Rupture 1n
Distillation Step
Release of Rafflnate
from Waste Retention
Population Dose
for Generic Plant
(man-rem)
720 (lung)
4.0 (T.B.)
720 (lung)
4.0 (T.B.)
6.2 (lung)
3.9 x 10-1 (T.B.)
79 (lung)
4.3 x 10-' (T.B.)
29 (lung)
1.6 x ID-' (T.B.)
3.7 (bone)
Pond 3.1 x 10-1 (T.B.)
Accident
Likelihood
(plant-yr. )-'
10"3 to 10"4
5 x 10"2 to 10"3
^_ 4 x 10"4
^ 3 x 10'2
*s 5 x 10"2
~ 2 x 10'2
Population Dose
Expectation Value
(man-rem)
7.2 x 10"' to 7.2 x 10~2
4.0 x 10"J to 4.0 x 10
36 to 7.2.x 10"' ,
2.0 x 10" to 4.0 x 10"J
2.5 x 10"?
1.6 x 10"
2.4 -
1.3 x 10
1.5 ,
8.0 x 10"J
7.4 x 10"?
6.2 x 10"3
Population Dose
per 1000 MHe-yr.
(man-rem)
1.7 x 10"? to 1.7 x 10"3
9.5 x 10" to 9.5 x 10"°
8.6 x 10"' to 1.7 x tO"2
4.3 x 10" to 9.5 x 10"
6.0 x 10"*
3.3 x 10"
5.7 x 10"2
3.1 x 10
3.6 x 10"?
1.9 x 10"*
1.8 X 10"?
1.5 x 10"*
Health Risk
per 7000 MWe-yr.
(f of excess cancers)
7.1 x 10"7to 7.1 x 10'8'
3.6 x 10"5to 7.1 x 10'7
3.8 x 10"9
2.4 x 10"6
1.5 x 10'6
8.0 x 10"8
Totals (lung)
(T.B.)
41 to 4.7 .
2.3 to 10"' to 3.2 x 10
,-2
9.7 x 10"! to V.I x 10"!
5.6 x 10"J to 7.6 x 10"
4.1 x 10"5 to 4.8 X 10"6
-------
5.3 Enrichment
The results for accidents considered in enrichment are given in Table
5-3. The highest consequence accident postulated in enrichment is the
catastrophic fire, for which the upper range of the estimated likelihood
is based upon incidents on record (the source term associated with this
accident, however, is a rough estimate). However, should the lower range,
based upon general chemical industry data, be a more appropriate esti-
mate for the future, the risk associated with the catastrophic fire is
relatively insignificant. In this case the release from a hot UFg
cylinder dominates the risk, and the data used in the assessment of
this accident is based upon incidents on record. However, an examination
of the historical data reveals that the magnitude of the release associated
with this category of accidents has been decreasing over the years.
202
-------
TABLE 5-3
ENVIRONMENTAL RISKS FROM ACCIDENTS IN ENRICHMENT
rsi
O
GO
Acc1d«nt
B.I Catastrophic Fire
C.I Re lease from t Hot
UF(. Cylinder
C.-2 Le.iks or Failure
of Valves or Piping
0.1 Cr1t1cal1ty
Total (lung)
(T.B.)
Population Pose
for Generic Plant
(man-rent)
930 (lung)
4.9 (T.B.)
150 (lung)
7.5 x 10-'
1.4 (lung)
7.7 x 10-J (T.B.)
4.6 x 10"! (thyroid)
1.2 x 10 (T.B.)
Accident
UTlceTTnobd
(plant-yr.)'1
3 x in"2 to 4 x 10"4
~ 4 x 10"1
~- 1.8
^ 8 x 10"5
Population Dose
Expectation Value
(man-ran)
28 to 3.7.x 10"' ,
1.5 x 10 to 2.0 x 10"-1
64 -1
3.0 x 10
2.5 ,
1.4 x \0~i
3.7 x 10",
9.6 x 10
Population Dose
per 1000 MWe-yr.
(raan-rera)
2.2 x 10", to 2.9 x \0'l
1.2 x 10"J to 1.6 x 10"3
5.1 x 10"'
2.4 x 10
2.0 x 10"?
1.1 x 10"'
2.9 x 10"J
7.6 x 10
Health Risk
per lOOd MUe-yr.
(1 of excess cancers)
9.2 x 10"6to 1.2 x ID"7
2.1 x 10~5
8.4 x )0"7
2.1 x 10""
95 to 67 ,
4.0 x 10 to 3.2 x 10"
. 7.5 x 10"!. to 5.3 X 10"'
1 3.7 x 10"J to 2.5 x 10"J
3.1 x 10"5 to 2.2 x 10"5
-------
5.4 Uranium Fuel Fabrication
The results for accidents considered in uranium fuel fabrication are
given in Table 5-4. The range of a factor of 10 in most of the source
terms reflects the variability in the design of building ventilation sys-
tems in uranium fuel fabrication plants. The highest consequence accident
in plants with no building exhaust HEPA filter is the postulated major
facility fire, whereas criticality is the highest consequence accident in
plants equipped with a building exhaust HEPA filter. Because of the cop-
siderably higher probability associated with the release from a hot UFg
cylinder, the risk associated with this accident dominates the total risk
from uranium fuel fabrication. It should be pointed out that the con-
sequences of this particular accident in uranium fuel fabrication (in the
absence of a building exhaust HEPA filter) is significantly higher than
in other components of the fuel cycle, largely because of the higher
population density in the vicinity of the generic uranium fuel fabrication
plant.
204
-------
TABLE 5-4
ENVIRONMENTAL RISKS FROM ACCIDENTS IN URANIUM FABRICATION
Accident
A.I
B.I
1X>
o
01 B.2
C.I
C.2
D.I
E.L
Hydrogen Explosion
in Reduction Furnace
Major Facility
Fire
Fir» In a
Routing Filter
Release fron a Hot
UF, Cylinder
O
Failure of Valves
or Piping
Crltlcallty
Haste Retention
Ponl Failure
Population Dose
for Generic Plant
(man-rem)
16 to 1.6 x 10"2 (lung)
7.4 x 10"2to 7.4 x 10-'
(T.B.)
1.6 x 104to 16 (lung)
74 to 7.4 x 10'z
3.8 to 3.8 x 10"3(lung)
1.8 x 10-2to 1.8 x 10-5
(T.B.)
1600 to 1.6 (lung)
7.8 to 7.8 x 1Q-3
(T.B.)
460 to 4.6 x lO'lOung)
2.2 to 2.2 x 10"J(T.B.)
32 (thyroid)
1.1 (T.B.)
5.7 x 10"! (bone)
3.5 x 10 (T.B.)
Totals (lung)
(thyroid)
(F.B.)
Accident Population
Dose
Population Dose
Likelihood , Expectation Value
(plant-yr.)
5 x 10"2to 2 x 10"3 8.0
3.7
~-2 x 10'4 3.2
1.5
~10'2 3.8
1.8
-.3 x IO"Z 48
2.3
~4 x 10'3 1.8
8.8
-< x 10"4 2.6
8.8
2 x 10"2to 2 x 10" 3 1.1
7.0
54
2.6
2.6
per 1000 MWe-vr.
(man-rem)
x
x
10'lto
10~3to
to 3.2 x
x 10-Zto
x
X
to
X
10'2to
10'4to
4.8 x
10- 'to
to 1.8 x
X
X
X
X
X
to
X
X
10-3tO
.o-2
lO'4
10"!to
10'4tO
5.3 x
10-2
10 to
3.2 x
1.5 x
ID'3
1.5 x
3.8 x
1.8 x
ID'2
2.3 x
ID'3
8.8 x
1.1 x
7.0 x
ID'2
1.2 x
io-5
10 '
ID'5
10'-7
10 '
io-4
10'6
ID'3
lO'5
10- 3
3.
1.
(man-rem)
2 x 10-f.to 1.3 x
5 x 10 to 6.0 x
1.3 x 10'lto 1.3 x
6.0 x 10"*to 6.0 x
1.
7.
1.
9.
7.
3.
1.
3.
4.
2.
2.
1.
I.
5 x 10"3to 1.5 x
2 x 10"°to 7.2 x
9 to 1.9 x IO'3
2 x 10-3to 9.2 x
2 x 10"?to 7.2 x
5 x 10 to 3.5 x
0 x ID"?
5 x 10°
4 x 10'f.tO 4.4 x
8 x 10 to 2.8 x
1 to 2.1 x 10'3
0 x 10-J
0 x IQ-'to 4.8 x
'°:t
10 9
ID'4
10 7
'»:!
10 *
Itf6
ID-*
10'7
10-*
lO'6
e
10'5
Health Risk
per 1000 Mwe-yr.
(* of excess cancers)
1.3 x 10"6to 5.4 x 10'"
5.4 x 10'6to 5.4 x 10"'
6.3 x 10"8to 6.3 x 10"11
7.9 x IO"Sto 7.9 x 10"*
3.0 x 10'6to 3.0 x 10~*
7.4 x 10"8
1.6 x 10"8to 1.6 x 10*'
8.9 x 10"5to 1.6 x 10'7
-------
5.5 Reprocessing
The results for accidents considered in reprocessing facilities are given
in Table 5-5. The accident of highest consequence is the postulated
explosion in the waste calciner, for which a volatile fraction of ruthenium
_2
of 10 has been postulated because of the high operating temperatures.
The consequences of several of the accidents compounded by postulated
failures of the exhaust HEPA filters are nearly of the same order of
magnitude, however the corresponding likelihoods are exceedingly low. It
is of interest to note, in fact, that with the exception of three accidents,
the estimated risk associated with multiple HEPA filter failure is lower
than that associated with normal filter operation. This result derives
from the fact that, in most cases, the volatile fraction of ruthenium
dominates the resulting population dose commitment.
One of the three accidents for which the postulated multiple filter failure
contributes significantly to the dose is the release in the fuel receiving
and storage area, for which the vapors released to the cask were assumed to
be converted to the particulate form during passage through the pool. In
fact, this accident dominates the risk at the high end of the estimate of
likelihood and constitutes about 25% of the total risk from reprocessing
at the low end of the range of likelihoods. The ion exchange resin fire
contributes substantially to the total risk at the high end of the range
of likelihoods, and is insignificant at the low end. Because of the
potential importance of both of these accidents, the probabilities of
their occurrence should be placed on a firmer foundation. The risk from
206
-------
TABLE 5-5
ENVIRONMENTAL RISKS FROM ACCIDENTS IN FUEL REPROCESSING
Accident
ro
O
Population Dose
for Generic Plant
(man-rem)
A.I Explosion 1n High Aqueous
Haste Concentration
a. Normal HEPA Filtration
b. HEPA Filter Failure
A.2 Explosion In Low Aqueous
Haste Concentrator
•a. Normal HEPA Filtration
b. HEPA Filter Failure
A.3 Explosion In High Aqueous
Feed Tank
a. Normal HEPA Filtration
b. HEPA Filter Failure
230.000 (G.I.)
430 (T.B )
3.3 x 10» (G.I.)
9.5 x 103 (T.B.)
15.000 (G.I.)
28 (T.B.)
1.6 x 10* (G.I.)
4.8 x 10' (T.B.)
840.000 (G.I.)
1,600 (T.B.)
8.4 x 105 '
1.7 x 10*
[S.I.j
[T.B.j
A 4 Explosion In Waste Caldner
a. Normal HEPA Filtration 2.3 x 106 (G.I.)
b. HEFA Filter Failure
4.300 (T.B.)
2.4 x ID* (G.I.)
1.3 x 10* (T.B.)
Accident
Likelihood
(plant-yr.) -1
A. 10
A, 10-'
'5
M 10
At 10'
-5
A. 10
A» 10"7
10
'6
Population Dose
Expectation Value
(man-rera)
2.3
4.3 x 10
3.3 x 10
9.5 x 10
1.5
2.8 x
1.6 x
4.8 x
8.4
1.6 x
8.4 x
1.7 x
2.3
4.3 x
2.4 x
1.3 x
Population Dose
per 1000 Hfle^yr.
(man-ren)
5.3 x 10
1.0 x
7.7 x
2.2 2
3.5 x
6.5 x
3.7 x
1.1 x
2.0 x
3.7 x
2.0 x
3.8 x
5.3 x
1.0 x
5.6 x
3.0 x
Health Risk
per 1000 ffife-yr.
(I of excess cancers)
3.3 x 10
5.5 x 10
2.2 x 10
2.3 x 10
-9
1.2 x 10'
1.2 x 10
,-5
,-7
3.3 x 10'
3.6 x 10
,-9
-------
Accident
A.5 Explosion In Iodine
Adsorber
B.I Solvent Fire In Codecon-
tamlnatlon Cycle
a. Normal HEPA Filtration
b. HEPA Filter Failure
B.2 Solvent Fire In Plutonium
Extraction Cycle
a. Nornal HEPA Filtration
b. HEPA Filter Failure
B.3 Ion-Exchange Resin Fire
a. Nonnal HEPA Filtration
b. HEPA Filter Failure
Population Dose
for Generic Plant
(man-rem)
1.9 x 103 (thyroid)
4.8 (T.B.)
Accident
Likelihood
(plant-yr. H
/o2 x 10"4
TABLE 5-5
(continued)
Population Opse
Expectation Value
(man-rem)
3.8 x 10"!
9.6 x 10"*
Population Dose
per 1000 MWe-yr.
(man-rem)
8.8 x IO"?
2.2 x 10°
Health Risk
per lOOti HUe-yr.
(* of excess cancers)
5.6 x 10"7
14.000 (G.I.) 10"4 to 10"6
23 (T.B.) .
1.5 x 104 (6.1.) 10"7 to 10"9
5.6 x 10' (T.B.)
1.5 x
3.1 x
2.6 x
5.2 x
190 (6
3.6 x
8.3 x
1.8 x
10"? (bone) 10"4 to 10"6
io:4 (T.B.) , ..
10? (bone) 10"9 to 10"11
10* (T.B.)
.1.) 10"1 to 10"4
10-1 (T.B.)
104 (bone) 10"6 to 10"9
10J (T.B.)
1.4 to 1.4
2.3 x 10-3
1.5 x 10";?
5.6 x 10"°
1.5
3.1
2.6
5.2
19
3.6
8.3
1.8
x
x
X
X
to
X
X
X
101
10 5
10 7
'0 '
xlO-2
to 2.3 x
to 1.5 x
to 5.6 x
to 1.
to 3.
to 2.
to 5.
5 x
1 x
6 x
2 x
ID"*
10 ~l
10"8
o o o o
i i i i
10 ^J — *OD
O
).9xlO"2
10-2
10 3
10"3
to 3.
to 8.
to 1.
6 x
3 x
8 x
'°~6
IO"6
3.3 x
5.3 x
3.5 x
1.3 x
3.5 x
7.2 x
6.0 x
1.2 x
4.4 x
8.4 x
1.9 x
4.2 x
10I2
IO"7
to 3.3 x 10"i
to 5.3 x 10"'
to 3.5 x 10".
to 1.3 x 10""
10"8nto 3.5
lO'J0 to 7.
10"a to 6.0
10"" to 1.2
w:j
"1
10 b
to 4.4
to K9
to 4.2
,V5»
: !?'«
_4
x 10 Z
* 10~«
x 10"B
2.0 x 10"6 to 2.0 x 10'8
2.2 x IO"9 to 2.2 x 10""
x 10'13 to 6.6 x 10"1S
-13
x 10"" to 4.1 X 10
2.7 x 10"5 to 2.7 x 10"8
1.3 x 10"7 to 1.3 x 10"'°
-------
ro
O
Accident
C.I Fuel Assembly Rupturt and
Release 1n Fuel Receiving
and Storage
a. Normal HEPA Filtration
b. HEPA Filter Failure
C.2 dssolver Seal Failure
a. Normal HEPA Filtration
b. HEPA Filter Failure
C.3 Release from a Hot UF,
Cylinder 6
0.1 Crltlcallty
a. Normal HEPA Filtration
b. HEPA Filter Failure
Totals (G.I.)
(lungj
(T.B.)
Population Dose Accident
for Generic Plant Likelihood
(man-rem) (plant-yr.) -1
6.0 (G.I.) 10"1 to 10"2
1.3 x 10-2 (T.B.) . s
6.8 x 105 (G.I.) 10 to 10
1.3 x 103 (T.B.)
1.6 x 10"' (lung) „. 10"5
2.1 x 10"2 (T.B.) fi
1.6 x 10* (lung) *, 10"8
2.3 x 103 (T.B.)
3.2 x 102 (lung) /w 5 x 10~2
1.5 (T.B.)
3.0 x 10"2 (T.B.) 8 x 10"3 to c
, 3 x 10"*
2.5 x 10 , (lung) 8 x 10"6 to „
3.5 x 10"' (total 3 x 10
body)
TABLE 5-5
(continued)
Population Dose
Expectation Value
(man-rem)
6.8 x 10"! to 6.8 x 10"?
1.3 x 10"J to 1.3 x 10
6.8 x lO'.to 6.8 ,
1.3 x 10"' to 1.3 x 10
1.6 x ID'S
2.3 x 10";
1.6 x 10":
2.3 x 10"5
1.6 x 101,
7.5 x lO"'
2.4 x 10"4 to 9.0 x 10"7
2.0 x ID'S to 7.5 x 10"?
2.8 x 10"' to 1.1 x 10"3
1.0 x 102 to 2.1 x 101
16
Population Dose
per 1000 MWe-yr.
(man-rem)
1.6 x 10"? to 1.6 x 10",
3.0 x 10"s to 3.0 x 10""
1.6 to 1.6 x 10-1
3.0 x ID'3 to 3.0 x 10
3.7 x 10"?
5.3 x 10"?
3.7 x 10 "5
5.3 x 10"'
3.7 x 10" I
1.7 x 10" J
5.6 x 10"6 to 2.1 x 10"8
4.7 x 10"? to 1.7 x 10"]?
6.5 x 10"S to 2.6 x 10" "
2.4 to 5.0 x 10"1
3.7 x 10 I1 ,
Health Risk
per 1000 MHe-yr.
(I of excess cancers)
1.0 x 10"6 to 1.0 x 10*7
1.0 x 10"4 to 1.0 x 10*5
3.4 x 10"12
3.4 x 10"10
1.5 x 10"5
2.2 x 10"9 to 8.4 x 10~12
4.2 x 10"'2 to 1.6 x 10"14
1.7 x 10"4 to 4.7 x 10"5
-------
all postulated explosions In reprocessing constitute roughly 1/8 of the
total risk at the high end of the range of likelihoods and as much as 1/2
at the low end. However, the likelihoods estimated for explosions 1n
reprocessing are at least one order of magnitude, and, possibly as high
as two orders of magnitude smaller than corresponding estimates for ex-
plosion likelihoods 1n other Industries. Further work in this area would
contribute more confidence to these results.
210
-------
5.6 Mixed Oxide Fuel Fabrication
The results for accidents considered in mixed oxide fuel fabrication are
given in Table 5-6. The consequences and risks associated with postulated
multiple failures of the exhaust HEPA filters dominate the total risk
from this component of the fuel cycle. This is because, with the exception
of criticality, the sources are particulate and the failure of the two
building exhaust HEPA filters is estimated to increase the release to the
environment by a factor of 105 (whereas the probability of simultaneous
failure of the dual exhaust HEPA filters is estimated to be 10 /demand).
Although the risks associated with a few of the accidents are relatively
insignificant (glove failure and criticality), several of the remaining
accidents contribute in roughly equal amounts to the total risk from
accidents in mixed oxide fuel fabrication.
211
-------
ro
ro
Accident
A.I Explosion In Oxidation-
Reduction Scrap Furnace
a. Normal HEPA Filtration
b. HEPA Filter Failure
B.I Major Facility Fire
a. Nomal HEPA Filtration
b. HEPA Filter Failure
6.." Fire In Wast* Compaction
Glove Box
.1. Normal HEPA Filtration
b. HEPA Filter Failure"
B.J Ion-Exchange Resin Fire
a. Normal HEPA Filtration
b. HEPA Filter Failure
TABLE 5-6
ENVIRONMENTAL RISKS FROM ACCIDENTS IN MIXED OXIDE FUEL FABRICATION
Population Dose
for Generic Plant
(m.in-rem)
1.5 (bone)
3.1 x 10;2 (T.B.)
1.5 x 10, (bone
3.1 x 10J (T.B.)
7.6 x 101 (bone)
1.6 (T.B.)
7.6 x 10* (bone)
1.4 x 105 (T.B.)
(plus 25 short-term
deaths)
1.5 x 10"' (bone)
3.1 x ID!3 (T.B.)
1.5 x 10, (bone)
3.1 x 10* (T.B.)
4.5 x 10"! (bone)
9.2 x ID/1 (T.B.)
4.5 x 10, (bone)
9.2 x 10' (T.B.)
Accident
Likelihood
(plant-yr.)-'
5 x 10"2 to ,
,2 x 10"J
5 x 10 to ,
2 x 10"6
*»2 x 10"4
~>2 x 10"7
A* 10"2
A»10"5
10"' to 10"4
10"4 to 10"7
Population
Dose
Expectation Value
(man-rem)
7
1
7
1
1
3
1
2
1
3
1
3
4
9
4
9
.5
.6
.5
.6
.5
.2
.5
.8
.5
.1
.5
.1
.5
.2
.5
.2
xlO-f
x 10
to 3.0
x 10"T
x 10"?
x 10'4
x 10"2
x 10"?
x 10";
x 10",
x 10"3
x 10"?
x 10"*
to 4.5
x 10-2
to 3.0 x 10"?
to 6.2 x 10"b
x 10-' ,
to 6.2 x 10"J
to 4.5 x 10"5
to 9.2 x 10"'
x 10-3
to 9.2 x 10"*
Population
per
Dose
1000 MUe-yr.
(man-rem)
3.
7.
3.
7.
6.
1.
6.
1.
6.
1.
6.
1.
2.
4.
2.
4.
3
0
3
0
5
4
5
2
5
3
5
3
0
0
0
0
x 10"?
x 10 ,
x 10",
x 10"3
x 10"?
x 10,
x 10",
x 10
x 10"5
x ID'S
x 10":
x 10"4
x 10"?
x 10 i
x 10",
x 10"J
to
to
to
to
to
to
to
to
1.3 x
2.7 x
1.3 x
2.7 x
2.0 x
4.0 x
2.0 x
4.0 x
<
w:|
10.4
10 *
%
10 J
10"6
Health Risk
per 1000~HBe-yr.
(I of excess cancers)
,-9
6.7 x 10"° to 2.6 x 10'
6.7 x 10"6 to 2.6 x 10"7
1.3 x 10"
1.2 x 10
,-6
n-7
(plus 2.2 x 10 short-
term deaths)
1.3 x 10'
1.3 x 10
4.0 x 10"8 to 4.0 x 10'11
4.0 x 10"6 to 4.0 x 10'9
-------
ro
Occident
B.4 Olssolver Hre 1n Scrap
Recovery .
a. Normal HEPA Filtration
b. HEPA Filter Failure
C.I Glove Failure
a. Normal HEPA Filtration
b. HEPA Filter Failure
C.2 Severe Glove Box Damage
a. Norffl.il HEPA Filtration
b. HEPA Filter Failure
0.1 Crltlcallty
a. Normal HEPA Filtration
b. HEPA Filter Failure
Totals (bone)
(thyroid)
(T.B.)
Population Dose
for Generic Plant
(iwn-rem)
7.6 (bone)
1.6 x 10;' (T.B.)
7.6 x
1.6 x
6.1 x
1.3 x
(bone)
(T.B.)
(bone)
(T.B.)
6.1 x 10' (bone)
1.3 (T.B.)
3.0 (bone)
6.1 x 10;* (T.B.)
3.0 x 10, (bone)
6.1 x 103 (T.B.
13 (thyroid)
3.8 x 10:1 (T.B.)
2.0 x 10, (bone
4.2 x 10Z (T.B.)
Accident
Likelihood
(plant-yr.)-'
8.0 x 10"3 to ,
63.0 x 10"5
8.0 x 10 ° to „
3.0 x 10"B
TABLE 5-6
(continued)
Population Dose
Expectation Value
(man-rem)
7.6 x 10"?
1.6 x 10"J
7.6 .
1.6 x 10 '
6.1 x 10"?
1.3 x 10",
6.1 x 10"'
1.3 x 10"3
3.0 x 10"?
6.1 x 10"4
3.0
6.1 x 10 "-
1.0 x 10"} to 3.9 x 10"?
3.0 x 10 , to 1.1 x 10";
1.6 x 10"' to 6.0 x 10"c
3.4 x 10" to 1.3 x 10
2.5 x lol.to 1.3 x 101 ,
Population Dose
per 1000 HWe-yr.
(man-rem)
3.3 x 10"?
7.0 x 10 ,
3.3 x 10",
7.0 x 10
2.7 x 10"5
5.7 x 10",
2.7 x 10"£
5.7 x 10"3
1.3 x 10"?
2.7 x 10"?
1.3 x 10"!
2.7 x 10"J
4.3 x 10"3 to 1.7 x 10",
1.3 x 10", to 4.8 x 10"t
7.0 x 10"i to 2.6 x 10 7
1.5 x 10"* to 5.7 x 10"'
1.1 to 5.5 x 10"1 _5
Health Risk
per 1000 HUe-yr.
(# of excess cancers)
6.7 x 10
6.7 x 10
5.4 x 10
5.4 x 10"
,-10
2.6 x 10
2.6 x 10
5.2 x 10"' to 3.2 x 10"
1.9 x 10 to 1.2 x 10
-2
3.1 x 10"7 to 1.2 x 10"'
1.4 x 10"7 to 5.3 x 10"10
2.2 x 10"5 to 1.1 x 10"5
(plus 2.2 xlO"7 short-
term deaths)
-------
5.7 Plutonium Storage
The results for the critlcality accident considered 1n Plutonium storage
are given in Table 5-7. Bone is the critical organ for this release,
because a relatively small number of fissions is postulated (1017) and a
small fraction of the plutonium in a container is assumed to be released
by the energy evolved in the excursion. Accordingly, the risk associated
with the postulated multiple failure in the exhaust HEPA filters dominates
the normal HEPA operation case. The risk associated with this accident
is in the range of the other postulated criticalities in the fuel cycle,
but is, nonetheless, relatively insignificant.
214
-------
TABLE 5-7
ENVIRONMENTAL RISKS FROM ACCIDENTS IN PLUTONIUM STORAGE
Accident
Population Dose Accident
for Generic Plant Likelihood
(man-rein) (plant-yr.)"'
Population Dose
Expectation Value
(man-rem)
Population Dose
per 1000 HWe-yr.
(man-rem)
Health R1»t
per 1000 HBe-yr.
(I of excess cancers)
D.I CriticalHy
ro a. Normal HEPA Filtration
b. HEPA Filter Failure
Totals
bone)
T.B.)
2.8 (bone)
8.8 x 10-1 (T.B.)
2.8 x 1<£ (bone)
5.7 x 103 (T.B.)
,v8 x ID"5
x 10
,-8
2.2 x 10
7.0 x 10
-5
2.2 x 10"?
4.6 x 10"*
2.2 x 10"?
5.3 x 10
3.9 x 10
1.2 x 10
3.9 x ID'S
8.1 x 10"'
3.9 x 10"
9.3 x 10"
5.2 x 10
,-u
7.8 x 10
,-10
8.3 x 10
-10
-------
5.8 Transportation
The results for the accidents considered in transportation are given in
Table 5-8. The highest consequence accident is the postulated criticality
in Pu02 transport, for which 1% of the plutonium or approximately 60 grams,
is assumed to be expelled from one of the containers from the energy re-
leased in the excursion. This estimate is based upon little corroborative
data; however, the risk associated with this accident is relatively insig-
nificant. The predicted highest risk accident is the collision involving
irradiated fuel. However, the uncertainty in this estimate is large, as
seen by the range in the predicted health risk. The predicted risk associ-
ated with shipments of UFg cylinders is also relatively significant. It
should be noted, however, that as there is essentially no data base from
which to evaluate potential releases from severe transportation accidents
involving radioactive materials, the estimates contained here are highly
uncertain.
216
-------
TABLE 5-8
ENVIRONMENTAL RISKS FROM ACCIDENTS IN TRANSPORTATION
Accident
1- Leakage of coolant from
Irradiated fuel cask
*• Improperly closed pluto-
nluir oxide container
3. Release from a collision
Involving natural UF,
o
4. Release from a collision
Involving enriched UF,
o
5- Release from a collision
PJ Involving Irradiated fuel
~-J 6. Release from a collision
Involving Irradiated fuel
followed by release of
fuel from the cask
7. Release from a collision
Involving plutonlum oxtde
8. Critical Hy of
untrradlated fuel
9. CrltlcalHy of enriched
uo2
10. CritlcaHty of Pu02
Total
Population Dose Accident
for Generic Shipment Likelihood ,
(man-rem) (shipment)
7.2 x 10"* (bone) #3 x 10"4
5.8 x 10"* (T.B.)
56 (bone) IxlO"3 to 4xlO"4
1.1 (T.B.)
38,003 (lung) /*>4 x 10"6
200 (T.B.)
1.4 x 105 (lung) fjk x 10"6
660 (T.B.)
190,000 (G.I.) 9xlO"6 to 2xlO"8
19,000 (T.B.)
200,000 (G.I.) 9xlO~8 to 2xlO"10
27,000 (T.B.)
(plus 14 short-term
deaths)
7.0 x 104 (bone) 3xlO"6 to 2xlO"9
1.4 x 10J (T.B.)
3.8 (T.B.) SxlO"8 to 2xlO"10
11.6 (thyroid) 8xlO"8 to SxlO"11
4.0 (T.B.)
1.2 x 106 (bone) 3x1 O"8 to 2x1 O"11
20,000 (T.B.)
Population Dose
Expectation Value
(man-rem)
2.2 x 10"'
1.8 x 10"'
5.6xlO"i; to 2.2x10'?
l.lxlO"J to 4.4x10"*
1.5 x 10"'
8.0 x 10"
8.4 x 10"'
4.0 x 10" J
1.7 to 3.8xlO"3 .
1.7x10-' to 3.8x10"*
1.8x10"^ to 4.0x10"!
2.4xlO"J to 5.4x10
2.1x10"! to 1.4x10"?
4.2xlO~3 to 2.8xlO"6
1.9xlO"7 to 7.6x10"'°
9.3x10"! to 5.8xlO"!2
3.2x10"' to 2.0x10"
3.6x10"* to 2.4xlO"5
7.5x10"' to 5.0x10"'
Population Dose
per 1000 MWe-yr.
(man-rem)
2.0 x 10"*
1.7 x 10"6
1.6x10"! to 6.2x10"?
3.1xlO"J to 1.2xlO"J
1.6
8.4 x 10~J
2.6 ,
1.2 x 10"*
16 to 3.5xlO"2,
1.6 to 3.5xlO"J
1.7x10"! to 3.7xlO"5
2.2x10"' to S.OxlO"5
5.8x10"! to 3.9x10"?.
1.2x10"' to 7.8x10
1.4xlO"6 to 5.7x10"*
5.3xlO"5 to 3.3x10"?
1.8x10 to 1.1x10
1.0x10"' to 6.6xlO"5
2.1x10 to 1.4x10
4.2 (lung)
16 to 035 (G.I.L
1.7 to ?.5 x lO'^T.B.
Health Risk
per 1000 Mwe-yr.
(I of excess cancer*)
6.8 x 10"'°
3.1x10"' to 1.2xlO"6
6.7 x 10"5
l,.l x 10"4
l.SxlO"3 to 3.4xlO"6
I.SxlO"5 to 4,.0x10"8 .
(plus 1.2x10-' to 2.6x10"°
short-term deaths)
1.2xlO"5 to 7.7x10"'
5.6x10"'° to 2.3xlO"12
9.4x10"'° to 5.8x10"13
2.0xlO"6 to 1.3x10"'
1.7 x 10"3to 1.8 x 10"4
(plus 1.2x10-5 to
) 2. 6xlO-8 short-tern
deaths)
-------
5.9 Overall Fuel Cycle Risks and Comparisons
The total risks from the accidents considered in each component of the
supporting LWR fuel cycle are summarized and aggregated in Table 5-9.
Because of the uncertainties associated with the linear, non-threshold
dose-response relationship, population dose is given in addition to
somatic health effects. Genetic effects are even more uncertain than
somatic effects; however, based upon the correlation given in Section
3.3 from Reference 12, the number of predicted congenital defects result-
ing from accidents in the LWR supporting fuel cycle is in the range of
5.4 x 10 to 1.3 x 10 per 1000 MWe-year, nearly an order of magni-
tude lower than the somatic risk.
Transportation accidents dominate the total accident risk, whereas the
risk from accidents in mining, milling, and plutonium storage are rela-
tively insignificant. The risks from the remaining components of the
fuel cycle, with the exception of spent fuel reprocessing, are roughly
of equal orders of magnitude, although the risk from uranium fuel fabri-
cation is negligible at the low end of the predicted range, characterized
by the existence of a HEPA filter in the building exhaust system.
It is of interest to note that despite the relatively large range of
values given for a number of individual fuel cycle components, the
ranges in the totals, with the exception of uranium fuel fabrication,
are relatively small.
218
-------
TABLE 5-9
TOTAL ENVIRONMENTAL HEALTH RISKS FROM ACCIDENTS
IN THE LWR SUPPORTING FUEL CYCLE
Fuel Cycle Component
Uranium Mining
Uranium Milling
UFC Conversion
o
Enrichment
Uranium Fuel
Fabrication
Reprocessing
Mixed Oxide
Fabrication
Plutonium Storage
Transportation
Totals
Population Dose per
10OP MWe-yrT
(man-rem)
0
.015 (bone)
.001 (T.B.)
.97 to .11 (lung)
.0056 to .00076 (T.B.)
.75 to .53 (lung)
.0037 to .0025 (T.B.)
2.1 to .0021 (lung)
.010 to 4.8 x 10-5(T.B.
.37 (lung)
2.4 to .50 (G.I.)
.0063 to .0028 (T.B.) ,
1.1 to .55 (bone)
.019 to .012 (T.B.)
3.9 x 10~7 (bone)
9.3 x 10~' (T.B.)
4.2 (lung)
16 to .035 (G.I.)
1.7 to .025 (T.B.)
8.4 to 5.2 (lung)
18 to .54 (G.I.)
1.1 to .57 (bone)
1.8 to .044 (T.B.)
Somatic Health Risk per
1000 MWe-yrT
(# of excess cancers)
0
5.9 x 10"7 to 5.6 x 10"7
4.1 x 10"5 to 4.8 x 10"6
3.1 x 10"5 to 2.2 x 10"5
8.9 x 10"5 to 1.6 x 10"7
1.7 x 10"4 to 4.7 x 10"5
2.2 x 10"5 to 1.1 x 10"5
-10
8.3 x 10
1.7 x 10"3 to 1.8 x 10"4
2.1 x 10"3 to 2.7 x 10"4
219
-------
The uncertainty in the results for total estimated health risk is composed
of three main components. These are 1) the aggregated uncertainty associ-
ated with the estimated source terms and accident likelihoods broken down
in Section 4; 2) the uncertainty associated with the dose conversion model
discussed in Section 3.2; and 3) the uncertainty associated with the dose-
response model discussed in Section 3.3. Although an estimate for a source
term and likelihood associated with any particular accident may be in error
by several orders of magnitude, it is estimated that, for the class of
accidents considered in this study, the range of uncertainty in the aggre-
gated expectation value of consequences in the total supporting fuel cycle
is roughly an order of magnitude, with a reasonable degree of confidence.
The range of uncertainty associated with the dose conversion may also be
as high as an order of magnitude, but considering the degree of conser-
vatism factored into these models, the results of the dose conversion are
most likely biased toward the high side. The combined uncertainty from
the source term/likelihood estimates and from dose conversion, may be
as high as a factor of 10 to TOO, and the combined results are probably
also biased toward the high side.
The conversion to health effects using the linear, non-threshold dose-
response hypothesis is also highly uncertain. The degree of realism
associated with this hypothesis is the subject of a continuing debate
within the scientific community. It is generally held that the model
is conservative; however, no attempt will be made here to speculate
220
-------
on the extent of uncertainty associated with its application.
Our estimates do. not include the risk associated with certain "class 9"
accidents, in particular site-related events such as tornados, hurricanes,
floods, or missile impacts. The risk associated with these events has
been dismissed in comparison with process initiated events in earlier
9
studies. Moreover, this study does not address the risks associated with
the management of nuclear wastes, including the storage of high level
wastes as liquids or solids at the reprocessing facility.
The results given in this section are normalized to the LWR fuel cycle
incorporating the recycle of plutonium. Should plutonium not be recycled,
the LWR fuel cycle would be altered as shown in Figure 2-1 and quanti-
fied in Tables 2-5 and 2-6 (assuming the maintenance of reprocessing and
recycle of recovered uranium). Renormalizing to the non-recycle case,
and taking into account the slightly altered, source term from reprocessing,
the total risk given in Table 5-9 would be relatively unchanged at the
high end of the range and increased by roughly 30% at the low end in the
absence of plutonium recycle. This result stems from the increase in the
required amount of UFg conversion, enrichment, and transportation in the
absence of plutonium recycle, which compensates for the reduction in risk
from mixed oxide fuel fabrication and shipments of spent fuel.
The risks from accidents in the LWR supporting fuel cycle are compared
with those associated with normal operations in Table 5-10. The normal
operations source terms used in this evaluation are compiled in Appendix
B. The source terms were converted to population dose and health risk
using the same methodology adopted for accidents, and as discussed in
Section 3.
22,1
-------
ro
ro
ro
TABLE 5-10
COMPARISON BETWEEN ENVIRONMENTAL HEALTH RISKS FROM ACCIDENTS AND
FROM NORMAL OPERATIONS OF THE LWR FUEL CYCLE
Risks from Normal Operations
Fuel Cycle Component
Uranium Mining
Uranium Milling
UFC Conversion
0
Enrichment
Uranium Fuel Fabrication
Dan *»/>/* a e c -i r\n
Population Dose per
1000 MWe-year
(man-ran)
1.4 x wl (lung)
2.2 x 10, (bone)
7.2 x 1(T (T.B.)
4.0 x 10^ (lung)
4.4 x 10, (bone)
1.7 x 10J (T.B.)
0.81 (lung)
2.0 x 10"2 (T.B.)
1.1 (lung)
1.9 x 10"2 (T.B.)
1.3 (lung)
6.2 x ID'3 (T.B.)
icnn fi-Ki/,-n-i/-M
Health Risk per
1000 MWe-year
(# of excess cai
3.3 x 10"1
8.0 x 10"1*
3.9 x 10"5
5.1 x 10"5
5.4x 10"5
•3 c « in~l
Mixed Oxide Fabrication
Plutonium Storage
Transportation
Totals
Reactor
790 (T.B.)
2.7 (bone)
.057 (T.B.)
0
0.35 (T.B.)
1500 (thyroid)
5400 (lung)
3200 (T.B.)
36 (thyroid)
0.94 (T.B.)
5.5 x 10
-5
1.4 x 10
-4
1.5
2.5 x 10
-3
Risks from Accidents
Health Risk per
1000 MWe-year
(# of excess cancers)
5.9 x 10"7 to 5.6 x 10"7
4.1 x 10"5 to 4.8 x 10"6
3.1 x 10"5 to 2.2 x 10"5
8.9 x 10"5 to 1.6 x 10"7
1.7 x 10"4 to 4.7 x 10"5
2.2 x 10"5 to 1.1 x 10"5
8.3 x ID'10
1.7 x IP'3 to 1.8 x IP"4
2.1 X 10"3 to 2.7 x 10"4
* Control of the tailings pile (covering the pile after the mill has been shut down) would reduce this value to
___ O
** Control of C-14 emissions to }% of normal release coupled with proposed EPA radiation protection control on Kr-85,
1-129 and plutonium would reduce Ihii value to 1.3 x 10"2.
-------
It is seen that the total health risk from the accidents considered in
this study is orders of magnitude lower than the health risk associated
with normal operations of the supporting fuel cycle, and comparable or
lower than the health risk associated with normal operation of the reactor.
The principal contributors to the risk associated with normal opera-
tions of the supporting fuel cycle are the mining, milling, and repro-
cessing components. Additional controls on the emissions from repro-
cessing and milling could reduce the, risk from normal operations of the
supporting fuel cycle by roughly a factor of four. Still the risk from
accidents would be negligible in comparison.*
It is, however, interesting to note that, with the exception of mining,
milling, reprocessing and transportation, the risk from accidents in each
component of'the supporting fuel cycle is -roughly comparable or possibly
larger than from normal operations. For these fuel cycle components,
then, the results of this study indicate that accidents have the effect
of increasing the risks associated with normal operations by roughly a
factor of two.
The estimates of health risks from both normal operations and from acci-
dents compared in Table 5-10 are made on the basis of the same dose con-
version models and dose-response relationship. On a relative basis,
* This conclusion does not take into consideration the theoretical in-
crease in the risk of accidents resulting from potential releases
associated with these new control measures.
223
-------
then, the uncertainty in this comparison derives primarily from uncertain-
ties in the normal operations source terms and the expectation Values of
the acciderit consequences. Assuming that the former contributes insignif-
icantly to the overall uncertainty, the range of uncertainty in the
comparison is estimated to be roughly an order of magnitude.
The risks associated with reactor accidents have been evaluated by the
6
Rasmussen study. The preliminary results of this study indicate that
the approximate societal risks associated with the annual operation of
a generic 1000 MWe-LWR consist of 4 x. 10"4 acute fatalities, 8 x 10"4
3 *
acute illnesses, and 3 x 10 latent cancers. This places the somatic
/
health risk from reactor accidents slightly in excess of both the. risk
from normal reactor operation and the risk associated with accidents in
the supporting fuel cycle.
It is also of interest to compare these results with estimates of occupa-
tional health risks associated with the generation of nuclear electric
power. Reference 52 compares occupational health risks across the
various fuel cycles. A total occupational health risk of 8 x 10
malignancies is attributed to the annual operation of a generic LWR.
_o
This is further subdivided into 1 x 10 occupational malignancies from
* This latent cancer estimate is based upon a linear dose-response
conversion factor of 100 cancers per 10$ man-rem, lower by at least
a factor of four than the conversion factor used in this study.
224
-------
,-2
uranium mining and 7 x 10 malignancies from all other fuel cycle steps.
These risks are also well in excess of the estimated risk to the general
population from accidents in the supporting fuel cycle.
* These estimates are based upon linear dose-response conversion factors
of 10-4 lung cancers per miner - WLM and 200 x 10-6 malignancies per
man-rad. This latter number is at least a factor of two lower than
the conversion factor used in this study.
-------
6. CONCLUSIONS AND RECOMMENDATIONS
FOR FUTURE WORK
The somatic health risk associated with accidents in the fuel cycle
supporting the annual operation of a 1000 MWe LWR is estimated to be of
_3
the order of roughly 10 excess cancers. This result is synthesized from
nominal radiological source terms and accident likelihood data compiled
from a number of diverse sources and subjected to interpretation, renor-
malization, and revision.
The estimate is subject to uncertainties associated with the dose con-
version model and the dose-response relationship, in addition to the
vagaries of the consequence expectation value estimate. Nevertheless,
assuming the validity of the linear, non-threshold dose-response hypothesis,
the uncertainty in the aggregate risk estimate is considered to be roughly
one to two orders of magnitude, with a reasonable degree of confidence.
Accidents considered to fall within the "class 9" category, in particular,
site-related events such as tornados, hurricanes, flood, or missile
impacts, were not included in this assessment. Nor, for that matter,
were accidental releases associated with radioactive waste management,
including the storage of high level wastes at the reprocessing facility.
Comparisons with the risk from normal operations of the supporting fuel
cycle and with occupational risks indicate that, on'the basis of the annual
operation of a 1000 MWe LWR, the risk from accidents in the supporting fuel
cycle is orders of magnitude lov/er. On the same basis, the risk from
accidents in the supporting fuel cycle is also slightly lower than
226
-------
that from reactor accidents, based upon the preliminary results of the
Rasmussen study, and comparable to that from normal reactor operation.
Transportation accidents dominate the total accident risk, whereas the
risk from accidents in mining, milling, and plutonium storage are rela-
tively insignificant. The risks from the remaining components of the
fuel cycle, with the exception of spent fuel reprocessing, are roughly
of equal orders of magnitude. Moreover, the risks from accidents in
uranium hexafluoride plants, enrichment facilities, uranium fuel fabri-
cation, and mixed oxide fabrication plants, albeit small, are roughly
comparable to those from normal operations.
A more comprehensive scoping analysis would include the risks associated
with site-induced and other high consequence, low probability ("class 9")
accidents. For example, although reprocessing plants and current designs
for mixed oxide fabrication plants are hardened to withstand "design
basis" natural disasters, the likelihood of exceeding the magnitude of
these events and the attendant consequences should be assessed. Similar-
ly, since the HEPA filters in these plants constitute an essential final
barrier in the protection of the environment, a more detailed assessment
of HEPA filter failure probabilities would appear warranted.
Accidents at the front end of the fuel cycle have the potential of re-
leasing large quantities of uranium to the environment. The likelihood
of a tornado dispersing nearly the entire inventory of uranium dioxide
at a uranium fuel fabrication plant, for example, should be assessed.
Risks associated with proposed and postulated waste management alterna-
tives have not been considered here,although this area is the subject of an
227
-------
extensive effort currently sponsored by the Energy Research and Develop-
ment Administration. Nor have we assessed the likelihoods or consequences
of accidents at the reprocessing facility involving the interim storage
of high level wastes. In particular, the risk associated w.ith a postula-
ted loss-of-coolant to the liquid waste storage tanks should be assessed.
Accidental releases of chemicals at fuel cycle facilities also have the
potential of producing environmental health effects. These include nitric
acid and hydrogen fluoride, and in particular the HF chemically produced
by the reaction of uranium hexafluoride with humid air. A more extensive
scoping study should address the risks associated with these releases.
The accident probabilities adopted for this study are derived from two
basic sources. For the front end of the fuel cycle, where considerable
experience exists, incidents on-record have been utilized, whenever
possible, to derive accident likelihoods. Existing data compilations,
however, are not comprehensive, and a more complete study would incor-
porate additional data extracted from ERDA and NRC compliance files, and
from facility operating records, if made available.
For the back end of the fuel cycle, where little operating experience
exists, we have relied heavily on past theoretical studies. Most of
these, however, were also scoping investigations, and meagre resources
have been devoted to investigating individual accident likelihoods on
a realistic basis for actual design situations. In particular, more
confidence could be placed in these results if accident scenarios assoc-
iated with spent fuel reprocessing and transportation were examined in
considerably more detail.
228
-------
7. REFERENCES
1. Public Law 91-190, National Environmental Policy Act of 1969, 83
Stat. 852.
2. Theoretical Possibilities and Consequences of Major Accidents in
Large Nuclear Power Plants, Division of Civilian Application, U.S.
A.E.C., WASH-740, 1957.
3. C.E. Guthrie and J.P. Nichols, "Theoretical Possibilities and Con-
sequences of Major Accidents in U-233 and Pu-239 Fuel Fabrication
and Radioisotope Processing Plants," Oak Ridge National Laboratory
Report ORNL-3441, April, 1964.
4. Final Environmental Statement Concerning Proposed Rule Making Action:
Numerical Guidt for the Design Objectives and Limiting Conditions
for Operation * Meet the Criterion "As Low as Practicable" for
Radioactive Material in Light Water-Cooled Nuclear Power Reactor
Effluents. Directorate of Regulatory Standards, U.S. Atomic Energy
Commission, WASH-1258, July, 1973.
5. Generic Environmental Statement Mixed Oxide Fuel (Recycle Uranium,
in Light Water-Cooled Reactors), U.S. Atomic Energy Commission, "
Fuels and Materials, Directorate of Licensing, WASH-1327 (Draft),
August, 1974.
6. Reactor Safety Study, An Assessment of Accident Risks in U.S.
Commercial Nuclear Power Plants, U.S. Atomic Energy Commission,
WASH-1400 (Draft), August, 1974.
7. Environmental Survey of the Uranium Fuel Cycle, U.S. Atomic Energy
Commission, Fuels and Materials, Directorate of Licensing, WASH-1248,
April, 1974.
8. Considerations in the Assessment of the Consequences of Effluents
from Mixed Oxide Fuel Fabrication Plants, Battelle Northwest Labora-
tories Report BNWL-1697, June, 1973.
9. Hazards Analysis of a Generic Fuel Reprocessing Facility, Prepared
by Science Applications, Inc. for the Environmental Protection
Agency, Task Order No. 68-01-1121, May, 1974.
10. Environmental Survey of Transportation of Radioactive Materials to
and from Nuclear Power Plants; U.S. Atomic Energy Commission,
Directorate of Regulatory Standards, WASH-1238, December 1972.
11. "Environmental Analysis of the Uranium Fuel Cycle," Part I - Fuel
Supply, U.S. Environmental Protection Agency, EPA-520/9-73-003-B,
October, 1973.
229
-------
12. Approaches to Population Protection in the Case of Nuclear Accidents,
Office of Radiation Programs, U.S. Environmental Protection Agency,
Washington, D.C., November, 1974.
13. The Effects on Populations of Exposure to Low Levels of Ionizing
Radiation. Report of the Advisory Committee on the Biological
Effects of Ionizing Radiation (BEIR Report). Division of Medical
Sciences, National Academy of Sciences, National Research Council,
Washington, D.C., November, 1972.
14. W.D. Rowe, "An 'Anatomy' of Risk," Environmental Protection Agency,
Washington, D.C., March, 1975.
15. Review of the Current State of Radiation Protection Philosophy,
National Council on Radiation Protection and Measurements, NCRP
Report No. 43, January, 1975.
16. Environmental Statement Light Water Breeder Reactor Program,
Volume IV, Fuel Cycle, U.S. Energy Research and Development Admini-
stration, WASH-1541 (Draft), July, 1975.
17. J.A. Simmons, "Risk Assessment of Storage and Transport of LNG and
LP-Gas," Final Report under Contract 68-01-2695 for the Environ-
mental Protection Agency, November 25, 1974.
18. Environmental Radiation Dose Commitment: An Application to the
Nuclear Power Industry, U.S. Environmental Protection Agency,
EPA-520/4-73-002, February, 1974.
19. Recommendations of the International Commission on Radiological ProtectionJ
Brit. J. Radio1.,Suppl.6 (1955). (Meeting of the International Congress of
Radiology held in Copenhagen, Denmark, July 1953.)
20. Recommendation of the International Commission on Radiological
Protection, ICRP Publication 2, Pergamon Press, Oxford (1959).
21. Recommendations of the International Commission on Radiological
Protection, ICRP Publication 6, Pergamon Press, Oxford (1962).
22. Recommendations of the International Commission on Radiological
Protection, ICRP Publication 10, Report of Committee IV, Pergamon
Press (1968).
23. W.D. Turner, S.V. Kaye and P.S. Rohwer, "EXREM and INREM Computer
Codes for Estimating Radiation Doses to Populations from Construc-
tion of a Sea-Level Canal with Nuclear Explosives," Union Carbide
Corporation Report K-1752 (1968).
24. U.S. Department of Health, Education, and Welfare, "Radiological
Health Handbook," Revised Edition, (1970).
230
-------
25. Deposition and Retention Models for Internal Dosimetry of the Human
Respiratory Tract, Health Physics 12: 173-207 (1966).
26. W.S. Snyder, M.R. Ford, C.G. Warner, and H.L. Fisher, Journal of
Nuclear Medicine, Pamphlet No. 5, 1969.
27. Y.C. Ng., et. al., "Prediction of the Maximum Dosage to Man from the
Fallout of Nuclear Devices, Handbook for Estimating the Maximum In-
ternal Dose from Radionuclides Released to the Bioshpere," UCRL 50163,
part IV, 1968.
28. TERMOD output listing, "Daily Intake by Man Via Terrestrial Pathways."
29. R.S. Booth, S.V. Kaye and P.S. Rohwer, "A System Analysis Methodology
for Predicting Dose to Man from a Radioactivity Contaminated Terres-
trial Environment," Proceedings Third National Symposium on Radio-
ecology, (1971).
30. Report to the American Physical Society by the Study Group on Light-
Water Reactor Safety,Reviews of Modern Physics, Vol. 47, Supplement
No. 1, Summer, 1975.
31. Operational Accidents and Radiation Exposure Experience within the
U.S. Atomic Energy Commission; 1943-1970; U.S. Atomic Energy Commis-
sion, Division of Operational Safety, WASH-1192, 1971.
32. Nuclear Safety Information Center, computer file search on failures
at fuel cycle facilities, private communication from R.L. Scott,
Oak Ridge National Laboratory, May 13, 1975.
33. W.R. Stratton, "A Review of Criticality Incidents," Los Alamos
Scientific Laboratory Report LA-3611, 1967.
34. H.C. Paxton, "Criticality Control in Operations with Fissile
Material," Los Alamos Scientific Laboratory Report LA-3366 (Rev.),
1972.
35. J. Rothfleisch, Nuclear Regulatory Commission, personal communica-
tion, May, 1975.
36. R.C. Merritt, "The Extractive Metallurgy of Uranium," Colorado
School of Mines Research Institute, 1971.
37. Summary Report: Phase I Study of Inactive Uranium Mill Sites and
Tailings Piles; Report on visits to 21 inactive uranium mill sites
by representatives of AEC and EPA; unpublished; October, 1974.
38. Letter from R.A. Winkel, Superintendent Oak Ridge Gaseous Diffusion
Plant, to Mr. C.A. Keller, U.S. Atomic Energy Commission, "UFfi Re-
lease Occurrences 1969-1973," April 8, 1974. °
231
-------
39. Letter from C.D. Tabor, General Manager Goodyear Atomic Corporation,
to Mr. C.A. Keller, U.S. Atomic Energy Commission, "UFC Release
Losses - 1954-1973," April 11, 1974. b
40. Allied Chemical Corporation, Metropolis, Illinois, NRC Docket No.
40-3392.
41. Kerr-McGee Corporation, Sequoyah, Oklahoma, NRC Docket No. 40-8027.
42. Proposed Final Environmental Statement Liquid Metal Fast Breeder
Reactor Program, U.S. Atomic Energy Commission, WASH-1535, Decem-
ber, 1974.
43. AEC Gaseous Diffusion Plant Operations, ORD-684, January 1972.
44. M.J. Bell, "ORIGEN, The Oak Ridge Radioisotope Generation and
Depletion Code," ORNL-4628, May, 1973.
45. Applicant's Supplemental Environmental Report, Uranium Oxide Plant;
U.S. Atomic Energy Commission, Special Nuclear Material License No.
SNM-1227, Docket No. 70-1257, JN-14 Add 1, October, 1971.
46. Allied Gulf Nuclear Services, Inc., "Safety Analysis Report -
Barnwell Nuclear Fuel Plant," U.S. AEC Docket No. 50-332, October
1973.
47. K.J. Schneider and A.M. Platt, editors, "High-Level Radioactive
Waste Management Alternatives," Vol. 1, BNWL-1900, May, 1974.
48. Transportation Accident Risks in the Nuclear Power Industry: 1975-
2020, Prepared for the Office of Radiation Programs, U.S. Environ-
mental Protection Agency by the Nuclear and Systems Sciences Group
of Holmes and Narver, Inc., NSS 8191.1, November, 1974.
49. J. Mishima, "Fractional Airborne Release of Plutonium Under Ship-
ping Accident Conditions," Proceedings of the International Sym-
posium on Packaging and Transportation of Radioactive Materials,
CONF-740901-P3, September, 1974.
50. C.L. Brown and S.W. Heaberlin, "Importance of Quality Control in
Plutonium Packaging Loading," Proceedings of the International
Symposium on Packaging and Transportation of Radioactive Materials,
CONF-740901-P3, September, 1974.
51. An Assessment of the Risk of Transporting Plutonium Oxide and Liquid
Plutonium Nitrate by Truck; Battelle Northwest Laboratories Report
BNWL-1846, BNWL-1846, August, 1975.
52. Comparative Risk-Cost-Benefit Study of Alternative Sources of Elec-
trical Energy; U.S. Atonic Energy Commission, Division of Reactor
Research and Development, WASH-1224, December, 1974.
232
-------
APPENDIX A
SO VEM DOSE OMUTMENT FACTORS FROM A ONE CURIE RELEASE
ASSMUW A WIJUTIW DENSITY OF 1W 7
** U'ISt
BY NUCLlOi-ALl PAIMfcAYS *•
*• PUPUIATIU* l>08E
NUCL10E T.HOliY
H 3
C t«
NA 22
NA 21
AR 39
AR III
CM 51
HN 50
f-E 55
tf. 59
CU 56
CO 60
in 65
IN 69M
ZN 69
BR 80
BR 80H
BR 82
1 9.04-04
1 0,0 X
1 «,15E 01
1 l,5i6X
1 l.nst 02
1 3.873X
1 7.6/E-02
1 0,0031
1 6,93E«02
1 0.003X
1 0,0 X
1 0.002X
1 7,9«E 00
1 0.294X
1 I.77E 00
1 0.065X
1 2,«6E 00
1 0.09U
1 2,27k 00
1 l.«6fc 02
1 5.393X
1 l.76t 01
t 0.651X
1 7.MHE-03
1 0,0 X
1 «,9ut-06
1 u.O X
i 2.7-06
1
1 1.3-04
i
] 8.5-02
BI-rHACT
1 9.04-04
1 0,0 X
i «.m 01
t 2.212X
1 8.21E 01
1 U.377X
1 1.621-01
1 0.009X
1 6.93E-02
1 O.OOIX
1 l.HOE-O"
1 0,0 X
1 1.12E-01
i o.ooex
1 1.07E 01
1 0.57SX
1 n,02IX
1 9.B5E 00
1 0.52SX
1 b.63t 00
1 0.300X
1 l.bht 02
1 6.SOSX
1 t>,"ft 01
1 l.l'7E-OI
1 O.I'IIAX
1 9.79E-OU
1 0.0 X
1 2.7-06
i
l 1.3-04
t
j 8.5-02
(MANHEMJ **
1 9.04-04
1 0.0 X
1 a.lbt 0|
1 0.18UX
1 6,70E 01
1 0.770X
1 6.VHE-02
1 0,0 X
1 6,931-02
1 0,0 X
1 0,0 X
1 0,0 X
1 7, ME 00
1 0.090X
1 S.6UF.-01
1 9.0U7X
1 2.62E 00
1 0.030X
1 1.96E 00
1 0.023X
1 l.iut 02
1 1.661X
1 l,23t 01
1 0.1««>X
1 0.0 X
1 u.O X
1 2.7-06
l
i 1.3-04
i
J 8.5-02
1 9.04-04
1 0.0 X
t «,ISt 01
1 1.69IIX
1 6.70E 01
1 3.056X
1 6.9UE-02
1 0,003X
1 0.003X
1 0,0 X
1 i,98E>02
1 0.002X
1 (J.SBE 00
1 0.392X
i 1,0 at oo
i o,o«9x
1 «,13E 00
i c.ieex
1 2,09f 00
1 0.095X
1 l.««t 02
1 6,5«9X
1 U.42E 01
1 l.OIE-0?
1 V.U X
1 0.0 X
i 2.7-06
1
i 1.3-04
i
1 8.5-02
t
1
-------
BR 83
BR 84
KH 83M
KH 85*
KH 85
KH 87
KH HB
KH 89
KH 90
Htt 8b
Hit 87
RB 89
SH K9
SH 90
3H 91
Y 90
Y 9}M
y 9i
r 92
111 ""
ZH 97
"" "5
i 3.6-05 1
1 l
i 1.1-04 ,
1 '
1 l,«2k»U9 1
1 O.U X 1
t 0.0 X I
1 2.t2t-U« •
t O.U X 1
( 0.0 X 1
1 U,lltl~0* 1
1 0,0 X 1
1 9, OUfOb 1
1 0,0 X 1
1 I,a7t-06/l
1 0,U X 1
1 U.31E UO 1
i O.I'm i
1 l,18t Ul 1
1 0.437X 1
1 5.3-05 '
1 l
1 2,6bt-Ul 1
1 0.010X 1
1 5,21t 01 1
1 1.927X »
1 O.U X 1
1 I.»it-o3 1
i o.o x l
1 5.9U-04 1
1 0,0 X 1
1 2,|4t-02 1
1 U.U X 1
1 u.o X 1
1 i>,95h yO l
1 0,0 t 1
i 7,<*4t--ol i
1 U.U29X 1
3.6-05
1.1-04
OtO X
o.u x
2,«2fc-0«
O.'i X
7,*«fc-05
o.u x.
0.0 X
9,04t-06
o.u x
0.0 X
l,92t 00
0.102X
2.38t 00
0.127X
5.3-05
7. tot ou
0.378X
8, lit OH
0.448X
3.50t-0ft-
U.002X
3.921-01
0.021X
1.07t-03
U.O X
7.29t 0»
O.itm
0 . (' 0 1 X
l.ti't Ul
l.m-llt
l.Vht 01
! 3.6-05 i
i i
1 1.1-04 i
1 l
1 i,4 X
w.oil -»I
u,"37X
| 3.6-05
1
i 1.1-04
l
1 l.«Zt-o9
1 0.0 X
1 %2*t-0*
1 0.0 X
1 2,«2t-0*l
1 0.0 X
1 7,bU£«Ob
1 0.0 X
1 4,2) X
, 8.3-05
1
1 l.Z-04
l
1 1, bit-US
1 0.0 X
1 0.0 X
1 0.0 X
1 7.77t-05
1 0,0 'I
1 «,2
-------
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
I
1
1
1
X O'O
Kft-lfi'f
X O'O
x o'o
•0-146*»
X O'O
»0-1SB*i
X O'O
lfl-199'l
XtfO'O
toto'o
20-169*9
X600'0
X 0*0
• 0-1I2M
X2IO'0
20-1BD'9
S 0*0
SO-1iO'&
»il2*Z
10 16SM
XOOfl'f
to lOt'2
X O'O
O'O
I O'O
90-309*t
X980'0
I0«36l'9
X200'0
20-124M
*0if 0
00 16S'2
« O'O
•0-196'6
t»00'0
20-164*2
X940'0
I0>1ff6'f
X O'O
f 0-1170't
1
1
1
1
1
1
1
1
1
1
1
t
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
X O'O 1
l»J-4?i'l 1
X O'O 1
90-lflf W 1
X O'O 1
no-j?6'f i
X O'O 1
»0-3i»'l 1
X O'O 1
•fl-346'2 1
Xif.0'0 1
tO-llt'2 1
XOIO'O 1
20-3I ' 0
1 90-1/^'n
i x o'n
1 no-io«'t
1 X ft' ii
1 nO-4hf 1
1 X O'O
1 f>u-1f9'<>
i xfoo'o
1 Ki-lflH'2
1 X O'fl
1 20-426§9
1 XI40'0
1 00 10n*B
1 X O'O
1 no-lnh'I
1 X4i0'0
1 00 ?f>n'9
1 X O'O
1 X66l'0
1 10 42Z'l
1 X962'fl
1 10 194*2
1 X9tlt*0
1 10 192*1
1 X fl'O
1 «0-19fi'S
i xoln'n
1 tO-496*H
1 X O'O
1 20-124't
1 XO^fl'O
1 00 12/M
1 X O'O
1 no*l«6*6
1 XtOO'O
1 IO-J67'!
| XfOO'O
1 lfl-49f'2
1 X O'O
1 t''-lB?'l
I X n ' u
i nn-nf I
1 X O'O
1 X 0 ' 1)
1 nn-lox'f
1 X o'o
t no»?6f* !
i X n • (i
1 nO-l{9*?
1 X640»0
1 XwflO'O
i io-ns't
1 X94IM
1 10 lit'?
1 X O'O
1 SO-190'f
1 XiSH'O
1 10 419*1
1 X O'O
1 fO-l46*6
1 till'?
1 10 1i6*(
1 X96t*6
1 20 ItiM
1 X9fS*ff
1 20 162*9
1 t O'O
i nO-9K't
1 XB4»M
I to luu't
I X2ln'0
1 10-lff 2
I X464'n
1 10 129*9
1 X 0*0
1 f 0-364* 1
1 X699*0
1 10 l6?'l
1 Xi9i'U
1 10 1HO*I
1 X O'u
i ro-its'2
1 X 0*0
i no-4if i
1 X o'fl
1 X 'I'D
1 t>0-'40H't
i x o'o
i no-i6f i
i x o'o
i nn-lf9'7
I XOIO'O
1 I0«1ti*2
1 Xtno'fl
1 ?0"124'i
1 X2t»fl'0
1 00 IRI'I
1 X O'U
1 X6?0'0
I«t0-H6*i
1 x o'o
I no-looM
1 to 39c;'|
1 X9Mi'0
1 to 1fl'2
1 oo isn't
1 X O'O
| «0>lfi4'n
1 Xi?fl'0
1 tO«1(7fi
1 XIOO'O
I 20-3io'f
1 X2tO*0
1 |0-129*«
1 X 0*0
1 fO-410'1
1 X4flO'o
1 tO»1»n'l
I xioo'o
1 I0-1i6*1
1 X 0*0
1 fO-49'l'l
1
1 Sfl
1
1 w4ft
1
1 fSt
1
1 H(f 1
1
1 xlft
1
1 2(1
1
i «in
i
1 H62I
1
1 621
1
1
1 i2t
1 621
1
1 HOtl
1 901
1 wCOt
1
1 (0!
1
1 66
1
1 66
1
1 H66
1
1 i6
1
1 Mitt
1
1 i6
IX
IX
IX
IX
3X
11
11
,1
,1
?i
,i
PS
9»
n«
MM
n«
nn
31
31
3i
31
HN
-------
xt 1)7
It 1)H
I 128
1 l?9
I 130
I 1)1
I ,52
1 D)
I 134
I U5
CS 1)«M
C9 l)«
CS 1)6
CS 1)7
CS 138
0* tau
t. 1.0
Ct HI
Ct 1«)
Ct l««
PR 1«3
PM 1U7
1
1
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
9,«St-l)7 1
0.0 X 1
0,0 X 1
9.2-05 1
1
),)Ht 01 1
1.25UX 1
1.6-02 |
1
0,020t-u2 I
o.o X 1
8.87C-UJ 1
U.U X 1
8,78t Ul 1
J.2«9X 1
1,071 02 1
J.954X 1
I, UUt 02 1
5tJ19X 1
1.3-04 |
1
l,65t-02 1
0,0 X 1
l.W-01 1
0.0002
0.0 X
7.B6E-03
0.0 X
7.07E 01
1.079»
i.ret oo
0.027X
l,«9t u2
2,272*
1.3-04
0,0 X
0.002X
V,«*f -02
O.uOIX
o.o x
1.1 Ut 00
0.017X
o.r x
0, 00 IX
1 9,45t»07
• o.o x
1 2,S2t-OS
1 0,0 X
| 3.8-04
i i,m 04
t 9),)1»X
1 1.3-01
1 2,69t 02
1 «,02t-02
1 0.0 X
1 1.571 00
1 O.OUX
1 1.8-03
1
1 1.671-01
1 0.001X
1 5,701-0)
1 0.0 X
1 ).54E 01
1 0.2&2X
1 8.131-01
1 O.OOtX
1 1,171 02
1 1.3-04
1 '
1 7,09t-06
1 O.o X
1 I.IBt-01
1 0,0 X
1 ti.2St-U?
1 0.0 X
1 l.Ult-02
1 0,0 X
1 M.^t-UI
1 0.003X
1 U.U
1 " . ii X
1 0,0
I 0,0 x
1
1
1
1
1
t
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0,0 X
2.S2E-0*
0.0 X
9.6-05
I.4»E 01
1.8S2X
1.8-02
t,2«t-oi
0.016X
e,i7t«o)
O.OOIX
2, 571-02
0.00)X
2.8-04
1,768-02
0.002X
6,911-OJ
0.0 X
5.06E 01
6,455*
1.01E 00
0.129X
1,264 01
1.4-04
3,)l(-0t
0.042*
O.UI9X
0.020X
t, 961-02
fl.0n«x
2.b9(. OU
o.J«2X
0.007X
l,56t-01
u,n2ox
1
1
1
1
1
1
1
1
1
1
1
1
I
I
1
1
1
1
1
1
1
1
1
1
1
I
1
i
I
i
I
I
i
I
I
I
i
I
9.281-U6 1
0,0 X 1
4.04E-05 1
0,0 X 1
9.2-05 '
1
l,42t U| 1
2.030X 1
1.6-02 j
9.62E-02 1
0.014X 1
6,701-0* >
0.0 X 1
l,fl)t.02 1
O.OOIX 1
2.6-04 1
1
1.471-02 1
0.002* 1
9.70E-0) 1
0,0 X 1
).54t 01 1
S.OS3X 1
8.131-01 1
0,116* 1
1,1 71 02 1
16.786X 1
1.3-04 ]
7.09t-06 1
0.0 X 1
i.iet-ot i
O.UI7X 1
8.2Jf-02 1
0.012* 1
i.mt-02 i
0.002* 1
4.)9t»ul 1
0.06)X 1
0.0 1
0,0 X 1
0.0 1
o.o X 1
236
-------
PM J«7 I 1. 621-02 I 1.7»fc OV I «.H«!t«OJ I «,BOE»02 I 6.B7E-02 I 0,0 I l.%6t-01 I 0,0 I
I 0,0 I I 0.09SX I 0.006% I 0.002k I 0.001X I 0,0 X I 0.020X I 0.0 X I
P« l«9 I 5,0«t-0« I l.22t-QV I 6,0it»0l I l.OU-OJ t J.ME-OJ t J,S5t-06 t l.OOE-02 t l.lbfc'Ob t
I 0.0 X I 0.007X I O.C X I 0,0 X I 0,0 X I 0,0 X I 0,001* I 0,0 X I
NO 1«7 I 5,bU»J2 I l.l»t 00 I 7,7UE>02 I tt.6Jfc-02 I 6.99E-08 I 5,)2E«02 I 9,«6E«02 I 5.J21-02 I
I 0.002X I O.C6JX I 0.0 X I 0,00«X I 0.001X I 0,0 X I 0.018X I 0.008X I
SM ISI I 2,«<)t-02 I 9.75t-ol I S.0«t-01 I 6.2*1-02 I 9,«S(.02 I i.2«l-OI I 1.0»t«01 I t,2«t-02 I
I 0,tUl» I O.OirfX I U.006X I O.OOUX I O.OOIX I 0,0 X I 0.014X I fl.OOJX I
8« Ibi I 2, HE- 0« I 5,9bt-02 I 3,02t-0j I S,02E-Oi I 7,05t«oa I l.JOE-05 I 6.02E-01 I I, 101-05 I
i o.o * i o.oasx i o.o x i o.u x i o.o x i o.o x i o.o x i 0,0 x i
EU li« I I.H9E U2 I 2,USt 02 I !,9Vt 02 I l,90t 02 I l,9«f 02 I t ,89f 02 I l,9tl 02 I 1.89E 02 I
I 7.002X I ll.i'SKX i 2.2<»bX I B.6«7X I 2.9ibX I l,i«5* I 2«,169X I 26.978X I
EU l« I J.MBf 00 I 1.20E 00 t 1.77E 00 I J.5U 00 I 1.6JI 00 I J,«^E 00 I 1.61E 00 t J.«E 00~l
I O.OS5X I O.I7IX I 0.02UX I 0,06-»X I 0,02bX I O.OIOX I 0.205X I 0.208X I
PB 210 I l,70t 01 I 2.m 01 I 5.25E 01 I J.20E 01 I 5.69E OJ I 3.»c"oe"i"" "II 'm't'l'tlk"""
I 62.M2X I 1.270X I 60.677X I S«,791X I 66.817X I O.OllX I «,9'72X I 0,«5«I I
237
-------
** Uu8t 3UMMAMK bY M'CLII>fc»Al.t. PATHWAYS ••
• * PI'PI'LATUIN 003E
(MANMtM) • *
U1«TH»CT
Llvtw
KIDNEY
1HYHU10
U'NG
SKIN
at
61
PH
PO
RN
Ru
HA
RA
RA
RA
AC
AC
fH
tH
TH
TH
TM
210
212
212
2IP
220
2?2
22)
22«
226
22*
2<>7
228
227
22B
2)0
2)1
2)2
1 <>.i'Sf-01
1 0.0 X
1 I.JHfc-'D
1 0,0 X
1 2,")t-02
1 U.D X
1 l.7«E 01
1 u. Oil IX
1 -i.bBE-Ub
1 II, U X
i 2.15-00
1 0,0 X
1 1.72t 02
1 0,Ol«X
1 7,i2E 00
1 0,0 X
1 2,79t 0«
1 2.279X-
1 l,J3t 0«
1 1.1'IUX
1 (j.but 02
1 u.OSUX
1 ),9Bt-02
1 0.0 X
1 t,£"3l 01
1 u.ui>2X
1 9.0)E 01
1 w.on/X
1 ), X
5.09-02
o.o x
«.)»t 01
n.i>0«x
9.57E 00
0.002X
5.0)E 01
(1.009X
1.22E 02
o,n2)X
2 . 1 it 01
u.ooax
2.H4E-02
U.o x
9. 9bE 00
O.IJ02X
).9ilt 01
O.Od7X
1,1'E 01
0.002X
*.»Ut»02
n,i' X
2tb«»t ili>
0.0481
1 9,b)E-01
1 0.0 X
i 2,m-o)
1 0,0 x
1 «
1 0.050Z
1 1.70E 00
1 0.0 X
1 7.SOE-0)
i o.o x
1 5.)2t.»n2
'l 0,0 x
1 1,«U 02
1 0.010X
1 5,98t.ofc
i n.o x
1 1.62-00
1 0.0 X
1 0.0
1 o.o x
1 ). 501-02
1,0.0 X
1 ).3it 02
t 0,02)X
1 H.20E 01
1 n.oofeX
1 l,7fct o)
1 0.122X
1 l.DE-01
1 n.O X
1 0.6ME-01
I u.o x
1 S . « U f 01
I o.uo«X
1 7.06E 02
1 O.OU^X
1 S.2"E-02
• 0 . il X
1 7.9<»t 02
1 0.55Z
1 7.92E 00
1 0.0 X
1 9.81E-02
1 0.0 X
1 «,9«f-0t
1 0.0 X
I b.2)( 02
1 0.007X
t 1.02E-03
t 0,0 X
1 7.58-00
1 0,0 x
1 0,0
I o.o x
1 %,7«t-02
1 0.0 X
1 1,->H 0)
1 0.022X
1 8.7)t 01
1 0.001X
1 b.OOE 02
1 0.008X
1 5.SOE-02
1 U.O X
1 5.BOE 00
1 0.0 X
1 2.10E P?
1 0.00)X
t ?,52E 0)
1 0,0 *«X
1 ^.'Jlt-Ol
t 0,0 X
1 2,7)E 0)
| 0.039Z
1 0.0 1
1 0,0 X 1
1 I.56E-OU 1
1 0,0 X 1
1 ), 801-05 |
1 0,0 X 1
1 0.0 1
1 0.0 X 1
1 b.S<4E>06 1
1 0,0 X 1
1 7.25-03 1
1 0,0 X 1
1 0,0 1
1 0,0 X t
1 ).26C-02 1
1 0.002X 1
1 U.73E 00 1
1 0.28UX 1
1 8.06k 01 1
i tt.Biex i
1 0.0 1
1 0.0 X 1
1 8.0JE-OJ 1
1 0,0 X 1
10,0 t
i o.o x i
1 6.52E 00 1
1 0,.}9<>X 1
1 I.^St 00 1
1 U.09JX 1
1 ).)5E-OS 1
1 (>.(' X 1
1 ^,"91 o? I
1 14.965X |
7.01 01
0.01-jX
1.35 00
0.0 X
1.36 01
o.oojx
1.96 03
0,«)2X
2.9«t«02
0,0 X
4.12-00
0.002X
1.31 03
0.299X
*.93 02
0.108X
2.94 01
O.OJ9X
1.72 04
«.)7«X
1.82 M
a.oout
2.58 01
O.OObX
1.24 02
0.«)AX
4.68 03
16.S07X
3.20 03
1l,)l(iX
6.33 -03
0.0 X
2.69 03
10.3831
1 0,0
1 0,0 X
1 1.S6E-04
1 0,0 X
| J.BOt-0)
1 0,0 X
1 0.0
1 0,0 X
1 5.5UE-06
1 0.0 X
1 7.25-03
1 0.0 1
1 0,0
1 0.0 X
1 ).2«E-02
1 0.002X
t «.7)E 00
1 0.28UX
1 B.06E 01
1 U.M6X
1 0.0
1 0.0 X
1 6.0)E-OJ
1 0.0 X
1 0,0
i n.o x
1 6.52E 00
1 fl,)92X
1 1.5SE 00
1 0.09JX
1 J.ibl-Oi
1 n.o x
1 2.avi 02
1 14.9651
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
238
-------
TM
PA
PA
PA
PA
PA
U
U
u
u
u
u
u
u
U 2
NP
HP
PU
Pu
PU
PO
230
230
231
233
230*
210
230
232
233
230
»i
236
237
238
39
237
239
«>3B
239
2
2 . 60f*X
b.SOE-Ui?
0.0 X
l.OOE Ul
n,(>02X
1.0 IE 01
o.no2X
l.uoE ul
".0 X
1 2.556 00
1 0.0 X
1 5.266 03
1 O.OI7X
1 3.13f 07
1 t. 02
1 0.0 X
1 6,03t 02
i o.uo2x
1 1.B96 m
1 O.U X
1 6.2S6-02
I U.O X
1 1.89 02
I o.uoox
1 7.5-06
I
1 9.V3L 03
1 (I . ii 5 1 X
1 l.SVE-02
1 O.U X
t I.iBt 0«
1 l.61t 00
I 0.051*
1 1,61 t 00
i n.usit
i J.»vt u?
1 0 . J f 1 X
1 2.216 00
1 0.0 X
1 0.0
1 0.0 X
i I.OIE OA
1 97.000X
1 2.10E 00
I o.o x
1 5.87E-00
1 0.0 X
1 2.5*t-02
1 0.0 X
1 O.U
1 0,0 X
1 7,7fl6 01
1 O.U05X
1 1.90E UO
1 0,0 X
1 1,1 9fc 00
t u.o x
i i.m 02
1 O.OOBX
1 7.966-01
i o.o x
1 6,08t-02
1 0,0 X
1 7.96-01
1 0,060X
1 6.9-06
i
i 7. 301 02
1 O.i'StX
1 U.O X
1 1.6U 03
1 U. 1 12X
1 2.0/f 03
P.iVIX
1.6<>t 03
O.D21X
2, Snt ul
I'.O X
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
t
1
f
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2,m oo i
0.131X 1
0.0 1
0.0 X 1
0.0 t
0,0 X 1
2.101-01 1
0,ol IX 1
8.76E-09 1
O.U X 1
2.556-02 1
0.002X t
0,0 1
0.0 X 1
7.50E 01 1
1,94(6 00 1
0.1I6X 1
i.m oo i
0.071X 1
1.03E 02 1
6.16UX 1
7.96E-01 1
o.oonx i
6.081-02 1
o.ooox i
7.96-01 1
6.9-06 I
1
1.13E 0? 1
6.791X 1
1.4136-02 1
(1 , V X 1
6.9<>E-01 1
O.OU2X 1
't . 0 ? 0 X 1
7.USI-01 1
0 . 0 1 « X 1
2.90 00
o.oiex
3.65 02
o.onox
3.97 03
0.875X
2.11 00
0.002X
2.94-01
0,0 X'
9.22-02
0.0 X
2.78 03
0.613X
1.96 04
4I.832X
3.84 03
0.862X
3.84 03
0.8S6X
3.94 03
1.572X
3.84 03
0.651X
1.47-01
0.0 X
3.08 03
7.U2X
1.2-05
3.95 03
1.50-01
O.U X
3.B5 03
3.3X
4.34 00
Q.uOfcX
1 2.196 00 1
1 O.I31X 1
1 0.0 1
1 0,0 X 1
1 0.0 1
1 0,0 X t
1 2.106-01 1
1 0.013X 1
1 8.761-09 I
1 0.0 X 1
1 Z.S56-02 1
1 0.002X 1
1 0.0 1
1 0,0 X 1
1 7.506 01 1
1 -0.525X 1
1 1.906 00 1
1 0.1I6X 1
1 1.196 00 1
1 0.071X 1
1 1.03t 02 1
1 6.160X 1
1 7.966-01 1
1 O.U08X 1
1 b. 066-02 1
1 O.OOOX 1
1 7.96-01 |
1 55.20RX 1
I 6.9-06 1
1 t
1 1.13E 02 1
1 6.791X 1
1 1.036-02 1
1 0.0 X 1
1 6,926-01 1
1 n.002X I
1 1.41(16-01 I
i n , n 2 0 X I
1 7.05E-01 1
1 O.OU2X I
1 3.0SI-M i
1 0.f>|t*X 1
239
-------
PU 242
Pu 213
Pn fun
AM 241
AM 24£M
AN 212
AM fUT,
AM Sun
CM 242
CM 243
CM 244
CM 24b
CM 2«7
CM 24R
CM lUt
1 3,
1 0
1 0
1 4,
1 0
1 1.
1 0
1 1.
1 0
1 1.
1 U
1 4.
1 0
1 0
' 7,
1 0
• 2.
1 0
1 0
I J.
1 0
1 0
1 0
1 0
24f 02
,o2bX
.0 X
04P 02
.033X
4»h 02
.udBX
30E 02
,U27X
bbt-02
.0 X
OPE 02
.033X
.0 X
Sbt 00
.0 X
37^ 02
,om
H4f. 02
.OlbX
lit 02
.027X
Mt 02
.027X
7 ft 03
,22b»
.0 X
1 1
1
1 3
1
1 2
1
1 2
1
1 3
1
1 1
1
1 H
1
1 3
1
1 1
1
1 1
1
1 1
1
1 1
1
1 1
1
1 3
1
1 7
1
,OU£ 01
0.002X
,b9f -03
0,0 X
,91t 01
o.oosx
,79E 01
0,no5X
,27E DO
o.o X
.49E-02
o.o x
,8ot 01
o.o x
,20E 01
O.OUlX
.Bit 01
n , U o a X
,21E 01
U , u 0 2 X
,olf 01
0.002X
,53t 01
».003X
.Hit 02
0,05bX
0,0 X
1 l.blt 04 1
1 O.oblX 1
1 1.32k.n3 1
1 0.0 X 1
1 l.b2E 0« 1
1 O.OblX 1
1 4.9«E 03 1
1 O.OlbX 1
1 4,9bt OJ 1
1 O.OlbX 1
1 l,V7t-01 1
i o.o x i
1 b.Oit OJ 1
1 O.OlbX 1
1 5.90t"03 t
1 0,0 X 1
1 1 , 1BE 02 1
1 0,0 X 1
1 4.
-------
APPENDIX B
SOURCE TERMS FROM NORMAL OPERATIONS
B .1 Normal Operations Source Term from Mining
To Atmosphere
Rn 222 1.3 x 104 Ci/yr
B .2 Normal Operations Source Term from Milling
To Atmosphere
Nuclide Activity Released (C1/yr)
U 238 6.2 x 10~2
U 234 6.2 x 10"2
U 235 2.8 x 10"3
Th 234 3.3 x TO"3
Th 230 8.8 x 10"3
Ra 226 5.9 x 10"3
Rn 222 2.2 x 103
241
-------
B.3 Normal Operations Source Terra from UFC Conversion
o
To Atmosphere
Nuclide Activity Released (Ci/yr)
U 234 1.6 x 10'2
U 235 6.6 x 10"4
U 238 1.6 x 10"2
Th 230 4.0 x 10"4
Th 234 1.0 x 10"2
Ra 226 4.6 x 10"5
Rn 222 1.6 x 10"4
To Watercourse
U 234 5.5 x 10"1
U 235 2.5 x 10"2
U 238 5.5 x 10"1
Th 230 1.8 x 10"1
Th 234 5.1 x 10"1
Ra 226 5.3 x 10"3
242
-------
B.4 Normal Operations Source Term from Enrichment
To Atmosphere
Nuclide
U 232
U 233
U 234
U 235
U 236
U 238
Activity Released (Ci/yr)
3.4 x 10"3
7.6 x 10
1.2 x 10
5.8 x 10
9.4 x 10
1.7 x 10
-6
-1
-3
-3
-2
To Watercourse
U 232
U 233
U 234
U 235
U 236
U 238
.19
.00042
6.36
.19
.52
.95
243
-------
B. 5 Normal Operations Source Term from Uranium Fuel Fabrication
To Atmosphere
Nuclide Activity Released (Ci/yr)
U 234 4.7 x 10"3
U 235 1.3 x 10"4
U 236 2.0 x 10~4
U 238 5.8 x 10"4
Th 231 1.3 x 10"4
Th 234 5.8 x 10"4
To Watercourse
U 234 1.1
U 235 3.2 x 10"2
U 236 4.7 x 10~2
U 238 1.4 x 10"1
Th 231 3.2 x 10"2
Th 234 1.4 x 10"1
244
-------
B.6 Normal Operations Source Term from Reprocessing
To Atmosphere
Nuclide Activity Released (Ci/yr)
H 3 1.10 x 106
C 14 7.00 x 102
Kr 85 1.50 x 107
I 129 2.90
I 131 3.50 x 101
Ru 103 1.10
Ru 106 6.60
Sr 89 2.50 x 10"1
Sr 90 2.20 x 10"1
Y 90 2.20 x 10"1
Y 91 4.20 x 10"1
Zr 95 7.50 x 10"1
Nb 95 1.40
Ag 110m 8.40 x 10"3
Sb 125 2.60 x 10"2
Te 127m 1.80 x 10"2
Te 129m 6.60 x 10"3
Cs 134 6.30 x 10"1
Cs 137 3.20 x 10"1
Ce 141 1.40 x 10"1
Ce 144 2.20
245
-------
To Atmosphere
Nucllde
Pm 147
Eu 154
Eu 155
U 232
U 234
U 235
U 236
U 237
U 238
Pu 238
Pu 239
Pu 240
Pu 241
Pu 242
Am 241
Am 243
Cm 242
Cm 244
Ccontlnued)
Activity Released (Ci/yr)
2.90 x 10"1
2.10 x 10"2
2.00 x 10"2
3.20 x 10"4
2.60 x 10"2
4.80 x 10"4
1.10 x 10"2
1.30 x 10"1
9.20 x 10"3
5.20 x 10"2
2.80 x 10"3
5.10 x 10~3
1.30
2.80 x 10"5
1.10 x 10"3
2.60 x 10"4
1.40 x 10"1
5.70 x 10"2
2.46
-------
B .7 Normal Operations Source Term from Mixed Oxide Fuel Fabrication
To Atmosphere
Nucfide Activity Released (Ci/yr)
Pu 238 4.2 x TO"3
Pu 239 2.4 x 10~4
Pu 240 4.2 x 10"4
Pu 241 8.8 x TO"2
Pu 242 2.2 x 10"6
To Watercourse
Pu 238 2.7 x 10"2
Pu 239 1.5 x 10"3
Pu 240 2.7 x 10"3
Pu 241 5.6 x 10"1
Pu 242 1.4 x 10~4
247
-------
B.8 Normal Operations Source Term from Transportation
No radiological effluents to the envlornment. Direct shine doses to the
general public scaled from results given 1n Reference 16.
*
B.9 Normal Operations Source Term from Reactor
To Atmosphere
Nuclide Activity Released (C1/yr)
Kr 83m 3.3
Kr 85m 35
Kr 85 590
Kr 87 13
Kr 88 22
Kr 89 40
Xe 131m 57
Xe 133m 15
Xe 133 1800
Xe 135m 17
Xe 135 44
Xe 137 84
Xe 138 73
I 131 .13
I 133 .52
* Assumed to be comprised of 2/3 PWR's plus 1/3 BWR's.
248
-------
To Watercourse
Nuclide
Activity Released (Ci/yr)
Corrosion and Activation
Products
24Na
32p
33p
51Cr
54Mn
56Mn
55Fe
59Fe
58Co
60Co
63,.
92Nb
117mSn
121Sn
185,
187W
237u
239NP
241Pu
.0001
.00002
.0001
.0003
.0001
.00004
.001
.0004
.005
.0004
.00002
.00008
.00002
—
. 00001
.0008
—
.0003
. 00001
Fission Products
82
86
Br
Rb
.00002
249
-------
To Watercourse
(continued)
Nuclide
Fission Products
(continued)
Sr
90
90y
91Sr
91my
91 „
92
Sr
93,
97mNb
97Nb
99MO
""To
103Ru
103mRh
105Rh
Ru
106
106Rh
127nVe
Activity Released (C1/yr)
.0025
.00014
.004
.0009
.0006
.11
.005
.066
.00004
.00003
.00001
.00001
.00001
.04
.04
.00003
.00003
.00002
.00001
.00001
.00001
250
-------
To Watercourse
(continued)
Nuclide Activity Released (C1/yr)
Fission Products
(continued)
127Te .00002
.00006
129Te .00003
130I .0001
131 "Ve .00006
131Te .00002
[ .08
132Te .0007
132I .0014
133I .026
134Cs .0026
135I .0041
136Cs .0012
137Cs .0018
137mBa .0017
140Ba .004
140La .0025
141 La .00003
141Ce .00009
143Ce .00005
.00003
25 li
-------
To Watercourse
(continued)
Nucllde
Fission Products
(continued)
144,
Activity Released (Cl/yr)
144
147.
TCe
Pr
153,
FNd
3Pm
JSm
Tritium
All Others
.00002
.00002
.00001
.00001
240
.00005
252
ft US GOVERNMENT PRINTING OFFICE 1978— 260-880/91
------- |