Content uploaded by Denia Djokic
Author content
All content in this area was uploaded by Denia Djokic on May 30, 2016
Content may be subject to copyright.
ENVIRONMENTAL IMPACT OF YUCCA MOUNTAIN REPOSITORY AFTER UREX+1A SEPARATION
Denia DJOKIC and Joonhong AHN
Department of Nuclear Engineering, University of California, Berkeley, CA 94720-1730, ahn@nuc.berkeley.edu
ABSTRACT
The environmental impact of Yucca Mountain Repository (YMR), expressed as the radiotoxicity of
radionuclides released from failed waste packages, has been evaluated for the case of partitioning and
vitrification the 63,000 MT of commercial spent nuclear fuel (CSNF) currently designated for disposal. A
parametric study on the effect of fuel cycle parameters on environmental impact has also been conducted.
CSNF inventory has been evaluated by using ORIGEN2. UREX+1a separation is considered as the base
case, and the removal of individual nuclide groups has also been investigated. Of particular interest is the
effect of Cs/Sr removal on the waste loading of a canister. An existing waste-conditioning model for high-
level liquid waste (HLLW) solidification with borosilicate glass is used to determine the composition and
waste loading of a vitrified waste canister by a set of waste loading constraints. A previously developed
release model has been applied to evaluate the environmental impact of the vitrified waste packages in
YMR. Numerical results show that while the removal of Tc and Cs/Sr does not have an effect on the
environmental impact profile, it is linearly sensitive to separation efficiency of actinides. The uncertainty
associated with the environmental impact resulting from the uncertainties of radionuclide solubility values
has been computed. It was determined that in the case of U, TRU, Tc, and Cs/Sr inventory reduction by a
factor of 100 and subsequent vitrification, the repository footprint would decrease by a factor of 3.4,
implying that 3.4 times more electricity generation could be accommodated while the environmental
impact is up to a factor of 100 smaller than the direct disposal case.
I. INTRODUCTION
The storage capacity of Yucca Mountain Repository is
currently limited at 63,000 metric tons of initial heavy
metal (MTHM) of commercial spent nuclear fuel. Since the
current CSNF inventory in the United States is already up
to about 50,000 MT, the limit of waste mass emplacement
will soon be reached. This would leave no room for future
nuclear industry expansion. To be able to have
advancements in nuclear energy development, more
repository space needs to be effectively utilized. The
question arises whether partitioning of spent fuel could
allow for a more effective use of repository space.
I.A. Background
Previously, a study1,2was conducted to evaluate the
environmental impact of Yucca Mountain Repository by
direct disposal of 63,000 metric tons (MT) of CSNF in
7,886 waste packages. The study also evaluated the
environmental impact of 7,000 MT equivalent of Defence
Wastes. The environmental impact is defined as the total
radiotoxicity of radionuclides released from failed waste
packages, based on a release model that was developed for
that study.
A different study3 describes a waste-conditioning
model, based on linear programming principles and high-
level waste (HLW) vitrification constraints, to determine
the composition of a vitrified HLW package after a
separation process. Using both these tools, the next logical
step is to determine the environmental impact of partitioned
and vitrified HLW.
I.B. Study Objectives
The purpose of this study is to determine the
environmental impact of partitioning and vitrification of
63,000 MT of CSNF by the UREX+1a (Ref. 4) separation
scheme.
In this study, the two aforementioned models
developed previously have been combined, and the
environmental impact has been evaluated in the case of
partitioning of all 63,000 MT of CSNF after UREX+1a
separation, with the assumption that the resulting HLW is
vitrified in a borosilicate glass matrix. A parametric study
has been performed to investigate the effects of cooling
times at different stages after discharge and after separation
on the environmental impact and repository capacity. The
effects of removal of Tc and Cs/Sr in the UREX+1a scheme
have also been studied. With the results of this parametric
survey, we have explored for fuel cycle scheme(s) to
minimize environmental impact. We have also determined
140Global 2007, Boise, Idaho, September 9-13, 2007
how repository capacity could be expanded with minimum
environmental impact increase. Lastly, an uncertainty
analysis is necessary to determine whether different
deterministic cases are justified to be treated as separate
cases given repository parameter uncertainties.
For purposes of comparison between direct disposal
and a partitioning and vitrification scenario from an
environmental impact point of view, we assume that the
present conditions of the repository at Yucca Mountain
remain unchanged.
II. MODELS AND ASSUMPTIONS
The analysis starts with the determination of CSNF
compositions. While the CSNF compositions are partially
reported in the Final Environmental Impact Statement
Report5, the complete inventory of spent fuel has been
obtained by using ORIGEN 2.2 (Ref. 6). Keeping track of
all isotopes in the spent fuel avoids underestimation of mass
and heat generation rate of the waste package. The same
conditions as given in Ref. 5 are used, i.e. 50 GWd/MTHM
burnup and enrichment of 4.3% PWR-UO2 fuel.
II.A. Separation Process
After discharge from the reactor, the spent fuel is left to
cool for a certain amount of time Tcool_before, after which it
undergoes separation by the UREX+1a scheme. This
scheme extracts U, Tc, Cs/Sr, and TRU with a given
separation efficiency into different streams. All fission
products (FP) are assumed to be left in the high-level liquid
waste stream. The separations process requires the addition
of process chemicals. Corrosion products are also included
from the stainless steel structural materials in CSNF
assemblies. To account for these, values of these secondary
wastes included in the high-level liquid waste stream per
MTHM are taken from the PUREX process and assumed to
be 22.3 kg-Na, 0.393 kg-P (process chemicals) and 4.2 kg-
Fe, 1.1 kg-Ni, 1.09 kg-Cr (corrosion products).7 After
partitioning, the HLLW is left to cool some additional
amount of time Tcool_after before it is vitrified.
During separation, the following elements are not
retained in the waste stream because they are volatile: H,
He, C, Ne, Cl, Ar, Kr, I, Xe, Rn.
For parametric investigation purposes, each of the
scenarios considered incorporates a different combination
of the following parameter values. The effects of separation
efficiencies of 95%, 99, and 99.9% have been explored, as
well as that of variation of both cooling times between 0
and 25 years. The effects on waste package number and
environmental impact due to removal of Tc and Cs/Sr as
compared to the U and TRU-only removal case. Thus, the
three removal scheme cases are: (Case 1) U, TRU, Cs/Sr,
and Tc removal, (Case 2) U, TRU, and Cs/Sr removal, and
(Case 3) U and TRU removal only.
II.B. Waste-conditioning Model
The waste-conditioning model developed in Ref. 3
determines the composition of vitrified HLW after
partitioning of spent fuel, as well as the number of waste
canisters per MTHM. A linear programming approach is
used to maximize the waste mass loading per canister,
given a set of constraints. Derivation of constraints are
described in Ref. 3 and Ref. 8. The constraints are
summarized as follows:
(1) (Mass constraint) Total mass of a waste canister
must be smaller than 2,500 kg.
(2) (Volume constraint) The volume of the vitrified
HLW must be between 80% and 100% of the canister
volume c
V. Considering that in reality 100% is not possible,
we set 98% as an arbitrary upper bound. No gap is assumed
between the lateral surface of the vitrified HLW and the
inner lateral surface of the canister. If it is smaller than
100%, the vacant space is assumed at the top of the canister.
(3) (Waste-loading constraint) The mass fraction of
the glass frit must be between 70% and 85%. This means
that the mass fraction of radionuclide oxides in vitrified
HLW ranges between 30 and 15%.
(4) (Temperature constraint) The highest temperature
in vitrified HLW must be lower than 400°C.
(5) (Pu constraint) The concentration of Pu in vitrified
HLW must be smaller than 2.5 kg/m3.
(6) (MoO3 constraint) The mass fraction of Mo oxide
in vitrified HLW must be smaller than 2 wt%.
(7) (Na2O constraint) The upper bound for the mass
fraction of Na oxide in vitrified HLW is 10%.
For the case of UREX+1a separation at 99% and 15
year cooling time after discharge, a graphical representation
of the linear programming model is shown in Fig. 1.
Fig. 1. Graphical representation of linear programming
model for determination of waste loading of a vitrified
canister. The case shown is for UREX+1a separation (Case
1) at 99% for all extracted components at 15 year cooling
time before partitioning. The shaded area represents the
feasible solution space.
141Global 2007, Boise, Idaho, September 9-13, 2007
Thus, by maximizing the waste loading of a canister
while still meeting all these constraints, we can determine
the mass of waste MW and the mass of glass matrix MG in
one waste canister. The total number of waste canisters and
the radionuclide composition of each package can so be
determined. The dimensions of the canister containing the
vitrified HLW are assumed to be the same as those for
Defence HLW (Refs. 3, 8).
II.C. Environmental Impact Model
For final disposal, it is assumed that five of these
vitrified HLW canisters are contained in a single waste
package for emplacement, which has dimensions of the Co-
disposal waste package (Ref. 5).a Thus, we can determine
the number of waste packages needed to place 63,000 MT
of CSNF into YMR. For each case, the number of waste
packages depends on the parametric case, as described in
Section III.
The environmental impact of YMR is computed by
using a code developed in a previous study1,2. The model
for environmental impact determination computes the total
amount of transuranic elements and fission products
released (by congruent or solubility-limited release) into the
environment in the case of simultaneous waste package
failure. The repository input parameter values are taken
from Ref. 2. The pore velocity of groundwater assumes a
value of 0.77 m/yr. The porosity of the host rock is assumed
to be 10%. The diffusion coefficient was assumed to be the
same for each element at a value of 0.03 m2/yr. For
evaluation of deterministic cases for both CSNF and
vitrified HLW cases, the high solubility limit shown in
Table 1 was conservatively assumed2.
The values of the repository parameters reported in Ref.
2 have been used to compare the results of disposal of HLW
and CSNF on a common basis. To determine to what extent
the environmental impact of deterministically evaluated
cases differ, an uncertainty analysis has been performed
based on repository parameter uncertainties. Because
solubility is considered to be the most uncertain parameter
in this model and has the widest range of values, the effects
of solubility uncertainty on uncertainty associated with the
environmental impact was considered.
A lognormal distribution of solubility was assumed,
based on the high and low solubility limits summarized in
Ref. 2. These limits are assumed to be, respectively, the 95th
and 5th percentiles of the assumed distribution, thus
aIt is assumed here that the waste package containing five
canisters of vitrified HLW will be stored in an interim storage
before it is placed in the Yucca Mountain Repository, to meet the
heat emission requirement. The initial heat power from a waste
package in YMR is limited at 11.8 kW maximum per package.9 So,
the waste package after UREX+ considered here is assumed to
cool until the heat emission rate becomes smaller than this value.
In other words, the package in this study is determined to be
compatible with YMR requirements.
enabling us to reconstruct the solubility probability
distribution for each element. We have evaluated the
environmental impact for the chosen cases using 300
realizations for each case, based on randomly sampled
solubility values for each element from the corresponding
distribution.
Table 1: Assumed Values for Solubility
Elements Min (mol/m3) Max (mol/m3) Notes
C, Cl, H,
Sr, Tc, Rh,
Ru, Cd, Sb,
I, Cs, Pm,
Eu
1.00E+03
These are
assumed to be
released
congruently with
the waste matrix,
so the solubility
has been set to a
sufficiently large
value.
Se 1.00E+01 1.00E+02
Pd 9.40E-03 9.40E-01
These values have
been taken from
TSPA-VA 10.
Zr 6.80E-07
Nb 1.00E-04
Sn 5.01E-05
Th 1.00E-02
Ra 2.30E-03
For these
elements, in
TSPA-SR 11,
fixed values are
given.
Sm 2.00E-05 1.90E+02
Ni 1.40E-03 3.10E+03
Pu 1.00E-07 2.00E-01
Pa 1.00E-07 1.00E-02
Pb 1.00E-07 1.00E-02
Cm 2.00E-05 1.90E+02
U 5.01E-04 3.98E-01
Am 2.00E-05 1.90E+02
Np 2.00E-03 1.58E+01
Ac 2.00E-05 1.90E+02
For these
elements, in
TSPA-SR 11,
ranges are given
as shown.
Si 1.30E-01 kg/m3 For silica 12
III. NUMERICAL RESULTS AND DISCUSSIONS
III.A. Base-case Comparison
We first have evaluated a base-case environmental
impact, where CSNF is cooled for 15 years, then partitioned
by UREX+1a (Case 1), and the resultant high-level liquid
waste is immediately vitrified. In this case, the maximum
waste loading has been determined to be 28.2% of the mass
of vitrified HLW in a canister. The number of canisters of
vitrified HLW has been obtained as 0.184 per MT of CSNF.
For 63,000 MTHM spent fuel, the total number of canisters
is calculated to be 11,590. If we assume that the Co-
Disposal package is used, which contains five canisters,
142Global 2007, Boise, Idaho, September 9-13, 2007
then, the total number of waste packages is 2320, compared
with the direct disposal case of 7890 CSNF packages.
The environmental impact evaluation for the base case
in Fig. 2 shows a reduction factor of up to 100 as compared
to the direct disposal case. This corresponds to the
separation efficiency of 99%. The environmental impact
from the HLW packages after 99% separation is smaller
than that of Defence Wastes shown in Ref. 2. Thus, the
environmental impact of the whole repository in this case is
dominated by that of Defence wastes, and not by HLW
from 99% separation of CSNF.
Fig. 2. Environmental impact profile comparison between
direct disposal and partitioning & vitrification (Case 1, 99%
separation efficiency, 15 year cooling time after discharge)
cases. The radiotoxicity in the environment of select
radionuclides is shown for the latter case. We can see that
the environmental impact is dominated by actinides such as
Pu-239, Np-237, U-236, and Th-232.
Fig. 2 also shows that actinides, and not fission
products such as Tc, govern the environmental impact curve
for the 99% removal case. Only at higher separation
efficiencies could the environmental impact profile become
sensitive to Tc removal.In this figure, impact from I-129 is
not included because it is removed from the HLLW stream
at an early stage of UREX+ process. This radionuclide,
however, has the half-life of 17 million years, and thus will
be released to the environment eventually with any type of
solidification materials applied.
It is interesting to note the difference in the shape of
the environmental impact profiles for the CSNF and the
HLW cases. Each peak of the CSNF curve corresponds to
each of the peaks governed by the actinides shown for the
HLW case, but the relative height of the peaks is different
for each case. The difference arises mainly due to the
behavior of the Pu-239 peak. Because there is a larger
initial inventory of Pu-239 per package (see Table 2) in the
CSNF case than in the HLW case described, a greater
amount of Pu precipitate forms on the surface of the CSNF
waste matrix due to the low solubility of Pu, and the
relative amount per package released is smaller than the
HLW case. In this latter case, the smaller inventory of Pu
means that a smaller precipitate forms and that Pu is
released almost congruently with the matrix. This relative
release rate difference accounts for the difference in curve
shape.This can be confirmed by Fig. 5, where the peak of
Pu-239 is proportionally decreased by increasing the
separation nefficiency.
III.B. Parametric Investigation
The parametric investigations demonstrate the effect of
different cooling times and UREX+ schemes on
environmental impact. It has been observed (see Figure 3)
that in almost all cases, waiting for a total of 15 years to let
the spent fuel or HLW cool, the removal of the heat-
emitting fission products Cs and Sr does not allow for a
larger waste mass loading. This is because the temperature
constraint3 is not an active constraint in the linear
programming model after 15 years (or even less in most
cases).
Fig. 3: Example of cooling time effects on waste package
number. This figure shows the number of waste packages
vs. cooling time before separation, with cooling time after
separation as a varied parameter, for Case 1 removal with
99% separation efficiency. It can be seen that even after
waiting for a short time before or after separation, the
number of waste packages quickly approaches a limit.
143Global 2007, Boise, Idaho, September 9-13, 2007
It is the Mo mass fraction constraint that becomes the
governing constraint after just a small amount of cooling
time (depending on the case), or by immediate removal of
Cs/Sr. This indicates that removal of heat-emitting
radionuclides may not necessarily have a significant effect
on waste loading as expected.
At best, separation and vitrification yields 2320 waste
packages, which is a lower limit, in this model. This fixes
the repository footprint at ~3.4 times smaller than the
original footprint by direct disposal. For environmental
impact case comparisons, we thus choose the base-case
cooling time value of 15 years before and no cooling after
partitioning, because this is consistent with Ref. 5.
Table 2 shows a comparison of inventories of each
waste package at emplacement of CSNF and vitrified HLW
for each separation efficiency case. For direct disposal, we
have 7886 waste packages, for the 95% separation
efficiency case we have 2994 waste packages, and both the
99% and 99.9% separation efficiency cases have reached
the lower limit of 2324 waste packages.
It has also been shown that after 15 years of total
cooling time for the cases of 99% and 99.9% separation
efficiency, the number of waste packages stays the same at
2320, i.e. the aforementioned lower bound. For the case of
95% separation efficiency, the waste package number is
increased by about 600, due to the Pu constraint becoming
active. Thus, it is important to achieve a target of at least
99% separation efficiency.
It has been shown in Fig. 4 that for a fixed separation
efficiency of all removed elements, separation of the
individual element groups Tc and Cs/Sr do not have any
significant effect on the environmental impact time profile.
However, the removal of Cs/Sr could potentially reduce the
HLW interim storage time or potentially even allow for
closer packing in the repository, if the waste package meets
the thermal repository requirements without having to
spend time in interim storage. Within the scope of this study,
however, YMR conditions and configuration are fixed for
purposes of comparison.
Fig. 5 shows that the variation in separation efficiency
of all removed elements affects the environmental impact
profile significantly. From Fig. 5 we can see that the
environmental impact profile is directly proportional to the
separation efficiency. The three separation efficiency cases
are distinctly different in the deterministic evaluation.Note,
however, that the repository parameters, such as solubilities,
sorption distribution coefficients, groundwater velocity,
porosity of the host rock, the package failure time, are
subject to significant uncertainties. Thus, it remains to be
shown whether an uncertainty associated with the
environmental impact upholds the clear difference between
these three cases(see Section III.C).
Fig. 4. Environmental impact profile comparison between
separation schemes (Cases 1, 2, and 3) for 99% separation
efficiency and 15 years cooling time after discharge.
Virtually no difference is seen between the cases.
Fig. 5. Environmental impact profile comparison between
separation efficiencies of 95%, 99%, and 99.9% for Case 1
removal scheme. This figure demonstrates the sensitivity of
environmental impact on separation efficiency.
144Global 2007, Boise, Idaho, September 9-13, 2007
Table 2: Radionuclide Inventory of Co-Disposal Waste
Packages Partitioned and Vitrified HLW at the Time of
Emplacement in the Repository
(4.3% enriched UO2, 50 GWd/MTHM, 15 year cooling before
partitioning, removal of U, TRU, Tc, Cs/Sr (UREX+1a
partitioning scheme), Process Chemicals and Corrosion Products
are included. High solubility case.)
Inventory Per Package (mol/package) at the
time of emplacement in the repository
Nuclide Half-life
(yr)
MPC
(Ci/m3)CSNF Vitrified HLW
Separation
Efficiency 95% 99% 99.9%
N
o. of
p
ackages 7886 2994 2324 2324
Fission Products
H-3 1.23E+01 1.00E-03 9.71E-02 0.00E+00 0.00E+00 0.00E+00
C-14 5.70E+03 3.00E-05 2.26E-05 0.00E+00 0.00E+00 0.00E+00
Cl-36 3.00E+05 2.00E-05 6.75E-12 0.00E+00 0.00E+00 0.00E+00
Fe-55 2.70E+00 1.00E-04 1.68E-04 0.00E+00 0.00E+00 0.00E+00
Co-60 5.30E+00 3.00E-06 3.91E-05 0.00E+00 0.00E+00 0.00E+00
Ni-59 7.60E+04 3.00E-04 9.68E-02 0.00E+00 0.00E+00 0.00E+00
Ni-63 1.00E+02 1.00E-04 1.62E-02 0.00E+00 0.00E+00 0.00E+00
Se-79 6.50E+04 8.00E-06 8.85E-01 2.33E+00 3.00E+00 3.00E+00
Kr-85 1.07E+01 1.00E+00 1.24E+00 0.00E+00 0.00E+00 0.00E+00
Sr-90 2.90E+01 5.00E-07 4.86E+01 6.40E+00 1.65E+00 1.65E-01
Zr-93 1.50E+06 4.00E-05 9.15E+01 2.41E+02 3.10E+02 3.10E+02
Nb-93m 1.60E+01 2.00E-04 4.45E-04 6.73E-04 8.66E-04 8.66E-04
Nb-94 2.40E+04 1.00E-05 9.57E-05 2.53E-04 3.25E-04 3.25E-04
Tc-99 2.10E+05 6.00E-05 8.86E+01 1.17E+01 3.01E+00 3.01E-01
Rh-102 2.90E+00 8.00E-06 4.30E-06 1.14E-05 1.47E-05 1.47E-05
Ru-106 1.00E+00 3.00E-06 5.86E-04 1.55E-03 1.99E-03 1.99E-03
Pd-107 6.50E+06 5.00E-04 2.62E+01 6.92E+01 8.91E+01 8.91E+01
Cd-113m 1.40E+01 5.00E-07 1.51E-02 3.53E-02 4.55E-02 4.55E-02
Sb-125 2.80E+00 3.00E-05 2.92E-02 7.70E-02 9.91E-02 9.91E-02
Sn-126 1.00E+06 4.00E-06 2.60E+00 6.84E+00 8.82E+00 8.82E+00
I-129 1.70E+07 2.00E-07 1.65E+01 0.00E+00 0.00E+00 0.00E+00
Cs-134 2.10E+00 9.00E-07 8.93E-02 1.18E-02 3.03E-03 3.03E-04
Cs-135 2.30E+06 1.00E-05 3.18E+01 4.19E+00 1.08E+00 1.08E-01
Cs-137 3.00E+01 1.00E-06 7.40E+01 9.75E+00 2.51E+00 2.51E-01
Pm-147 2.60E+00 7.00E-05 1.58E-01 4.15E-01 5.35E-01 5.35E-01
Sm-151 9.00E+01 2.00E-04 9.07E-01 2.39E+00 3.08E+00 3.08E+00
Eu-154 8.60E+00 7.00E-06 1.11E+00 2.93E+00 3.78E+00 3.78E+00
Eu-155 4.80E+00 5.00E-05 1.70E-01 4.47E-01 5.76E-01 5.76E-01
Actinides and decay daughters
Cm-247 1.56E+07 2.00E-08 2.47E-04 3.25E-05 8.38E-06 8.38E-07
Cm-246 4.80E+03 2.00E-08 1.99E-02 2.62E-03 6.74E-04 6.74E-05
Cm-245 8.50E+03 2.00E-08 1.35E-01 1.77E-02 4.57E-03 4.57E-04
Cm-244 1.80E+01 3.00E-08 1.65E+00 2.18E-01 5.62E-02 5.62E-03
Cm-243 2.90E+01 3.00E-08 1.98E-02 2.61E-03 6.72E-04 6.72E-05
Cm-242 4.50E-01 7.00E-07 1.78E-04 2.34E-05 6.03E-06 6.03E-07
Am-243 7.40E+03 2.00E-08 7.56E+00 9.96E-01 2.57E-01 2.57E-02
Am-242m 1.40E+02 2.00E-08 7.34E-02 9.67E-03 2.49E-03 2.49E-04
Am-241 4.30E+02 2.00E-08 3.45E+01 4.55E+00 1.17E+00 1.17E-01
Pu-243 5.69E-04 2.00E-04 8.94E-15 1.18E-15 3.04E-16 3.04E-17
Pu-242 3.80E+05 2.00E-08 2.97E+01 3.91E+00 1.01E+00 1.01E-01
Pu-241 1.40E+01 1.00E-06 3.13E+01 4.12E+00 1.06E+00 1.06E-01
Pu-240 6.50E+03 2.00E-08 7.22E+01 9.51E+00 2.45E+00 2.45E-01
Pu-239 2.40E+04 2.00E-08 1.83E+02 2.41E+01 6.20E+00 6.20E-01
Pu-238 8.80E+01 2.00E-08 9.87E+00 1.30E+00 3.35E-01 3.35E-02
Np-239 6.44E-03 2.00E-05 6.61E-06 8.70E-07 2.24E-07 2.24E-08
Np-237 2.10E+06 2.00E-08 2.76E+01 3.64E+00 9.36E-01 9.36E-02
U-238 4.50E+09 3.00E-07 3.10E+04 0.00E+00 0.00E+00 0.00E+00
U-236 2.30E+07 3.00E-07 1.92E+02 0.00E+00 0.00E+00 0.00E+00
U-235 7.00E+08 3.00E-07 2.52E+02 0.00E+00 0.00E+00 0.00E+00
U-234 2.50E+05 3.00E-07 1.35E+00 0.00E+00 0.00E+00 0.00E+00
U-233 1.60E+05 3.00E-07 1.49E-04 0.00E+00 0.00E+00 0.00E+00
Pa-234 2.23E-06 3.00E-05 6.89E-12 4.07E-11 5.24E-11 5.24E-11
Pa-231 3.30E+04 6.00E-09 4.44E-06 1.17E-05 1.51E-05 1.51E-05
Th-234 6.60E-02 5.00E-06 4.58E-07 1.21E-06 1.55E-06 1.55E-06
Th-232 1.40E+10 3.00E-08 9.73E-05 2.56E-04 3.30E-04 3.30E-04
Th-231 2.91E-03 5.00E-05 1.04E-09 2.75E-09 3.54E-09 3.54E-09
Th-230 7.50E+04 1.00E-07 3.17E-05 8.30E-05 1.07E-04 1.07E-04
Th-229 7.90E+03 2.00E-08 2.99E-08 7.94E-08 1.02E-07 1.02E-07
Th-228 1.91E+00 2.00E-07 2.67E-06 7.04E-06 9.07E-06 9.07E-06
Ac-228 7.00E-04 3.00E-05 2.58E-18 6.79E-18 8.75E-18 8.75E-18
Ac-227 2.17E+01 5.00E-09 7.07E-10 1.87E-09 2.40E-09 2.40E-09
Ra-228 6.70E+00 6.00E-08 2.47E-14 6.50E-14 8.38E-14 8.38E-14
Ra-226 1.60E+03 6.00E-08 1.57E-09 4.10E-09 5.28E-09 5.28E-09
Rn-222 1.41E-02 1.00E+00 1.03E-14 0.00E+00 0.00E+00 0.00E+00
Po-218 5.80E-06 1.00E+00 5.68E-18 1.49E-17 1.91E-17 1.91E-17
Po-214 5.20E-12 1.00E+00 4.99E-17 1.33E-23 1.72E-23 1.72E-23
Pb-214 5.10E-05 1.00E-04 5.10E-24 1.30E-16 1.68E-16 1.68E-16
Pb-210 2.23E+01 1.00E-08 3.70E-12 9.70E-12 1.25E-11 1.25E-11
Bi-214 3.75E-05 3.00E-04 3.70E-17 9.69E-17 1.25E-16 1.25E-16
Total FP & Actinides 3.22E+04 4.10E+02 4.41E+02 4.21E+02
Process Chemicals
Na ---- ---- ---- 2.04E+04 2.63E+04 2.63E+04
P ---- ---- ---- 2.67E+02 3.44E+02 3.44E+02
Corrosion Products
Fe ---- ---- ---- 1.58E+03 2.04E+03 2.04E+03
Ni ---- ---- ---- 3.94E+02 5.08E+02 5.08E+02
Cr ---- ---- ---- 4.41E+02 5.68E+02 5.68E+02
Matrix
Silica ---- ---- ---- 4.73E+03 4.92E+03 5.06E+03
Total 3.22E+04 3.30E+04 4.00E+04 4.03E+04
III.C. Uncertainty Analysis
Fig. 6 shows the effects of solubility variation on
environmental impact profiles of separation efficiency cases.
Three representative cases of 95%, 99%, and 99.9%
separation efficiency have been chosen on which to perform
an uncertainty analysis. Also, the high solubility and low
solubility cases for CSNF disposal are shown.
It can be seen from Fig. 6 that the deterministic cases
for three different separation efficiencies are still
distinguishable. The higher solubility realizations of the
99.9% and 99% cases overlap at early timescales with the
lower solubility realizations of 99% and 95%, respectively,
but at later times they are distinguishable. While more
rigorous statistical analyses need to be performed to finally
conclude that the effects of separation efficiency will not
“bury” in uncertainty resulting from solubility uncertainty,
this preliminary exploration indicates that importance of
separation efficiency effects.
The uncertainty of the environmental impact profile
decreases with time as the contents of the waste canister are
diminished. Also, the uncertainty of the highest separation
efficiency case is less than the cases for 95% and 99% at
later timescales. Both observations are basically due to the
145Global 2007, Boise, Idaho, September 9-13, 2007
fact that the less mass there is in a waste package, the less
radionuclides with which the solubility uncertainties are
associated with are released. Also, in the definition of the
environmental impact adopted in this study, transport of
radionuclides in the geosphere is not included. Thus, the
main source of uncertainty results from the parameters that
determine the rate of release of radionuclides.
Fig. 6 also shows the lower solubility and upper
solubility bound for the CSNF case. At timescales up to 107
years, the uncertainty of the environmental impact profile
has a large range. In fact, the environmental impact
uncertainty distribution due to CSNF envelops most of the
environmental impact profiles for the partitioned and
vitrified cases. At larger times than 107 years, the
environmental impact profiles of the latter cases have a
significantly lower tail value than that of the CSNF cases.
The CSNF uncertainty in the said range of time is
larger than that of HLW due to the uncertainty associated
with the uranium solubility, whereas thesolubility of silica
in the HLW matrix is not assigned an uncertainty. This
observation indicates a potential benefit of partitioning and
vitrification in that the uncertainty with the impact could be
significantly reduced if the uncertainty associated with the
engineered-barrier parameters (in this case the waste matrix
solubility) could be made smaller.
Fig. 6. Effects of variations of solubilities on the
environmental impact of vitrified HLW for three cases of
separation efficiencies as compared to that of CSNF waste
packages. The three HLW cases are relatively
distinguishable, while the environmental impact uncertainty
of the CSNF case envelopes the three former cases.
IV. CONCLUDING REMARKS
A study evaluating the environmental impact of
reprocessing 63,000 MT of CSNF has been performed. A
parametric study showing the effects of separation
efficiency, cooling times, and separation schemes has also
been done. The results show that a reduction in actinide
inventory in the vitrified HLW means that the repository
capacity could potentially be increased by a factor of 3-4,
without exceeding the environmental impact of the current
repository configuration by up to a factor of 100. Assuming
the same footprint of the repository, YMR could
accommodate 3.4 times greater electricity generation while
still having a much smaller environmental impact than that
of the direct disposal case.
It was found that removal of Cs/Sr and Tc do not have
a significant influence on environmental impact reduction.
Also, if we let the spent fuel cool at least 15 years after
discharge in the 99% and 99.9% separation efficiency cases,
the waste loading is not affected by Cs/Sr removal due to
the disappearance of the limiting temperature constraint.
An uncertainty analysis was performed with three
representative distinct deterministic cases. The effects of
solubility uncertainty have been evaluated and the
uncertainty in the environmental impact has been evaluated.
These cases are significantly different from each except for
at short timescalesafter the package failure. When
compared to the uncertainty range of the environmental
impact of CSNF direct disposal case, we have observed that
the effects of partitioning and vitrification of spent fuel are
enveloped by the spread of CSNF environmental impact
values.
V. FUTURE WORK
Further study is required for justified comparison of
environmental impact to the direct disposal case. A more
detailed analysis of waste mass flow is needed. The
destination of removed actinides, Cs/Sr, and Tc must be
considered, as must be the volatile elements lost during
partitioning, as well as secondary HLW. Only when
considering the environmental impact of the whole fuel
cycle can a fair comparison be made.
Also, the effect of removal of Cs/Sr on the linear heat
load of YMR drift tunnels13 was not considered in this
study. Investigating the reduction in repository footprint
and/or configuration due to combining waste conditioning
models and thermal performance models is relevant.
REFERENCES
1. J. AHN and L. ACKERMAN, Environmental Impact
of Yucca Mountain Repository, Transactions, 93, 731-
732, 2005 Winter Meeting, Washington, D.C.,
146Global 2007, Boise, Idaho, September 9-13, 2007
November 13-17, 2005, American Nuclear Society,
2005.
2. J. AHN, Environmental Impact of Yucca Mountain
Repository in the Case of Canister Failure, Nuclear
Technology,157, 87-105, 2007.
3. J. AHN and M. CHEON. Linear Programming
Approach for Optimization of Radionuclide Loading in
Vitrified HLW, Nuclear Technology,156, 303-319,
(2006).
4. G. F. VANDEGRIFT, M. C. REGALBUTO, et. al.,
Lab-Scale Demonstration of the UREX+ Process.
WM’04 Conference, Tucson, AZ.
5. U.S. Department of Energy, Office of Civilian
Radioactive Waste Management, Final Environmental
Impact Statement for a Geologic Repository for the
Disposal of Spent Nuclear Fuel and High-Level
Radioactive Waste at Yucca Mountain, Nye County,
Nevada, DOE/EIS-0250, (Feb. 2002)
6. A.G. GROFF, ORIGEN2: A Versatile Computer Code
for Calculating the Nuclide Compositions and
Characteristics of Nuclear Materials, Nuclear
Technology, Vol. 62, p.335, (1983).
7. M. BENEDICT, T. H. PIGFORD, and H. LEVI,
Nuclear Chemical Engineering, 2nd ed., McGraw-Hill,
1981.
8. U.S. Department of Energy, Office of Environmental
Management, Waste Acceptance Product
Specifications for Vitrified High-Level Waste Forms,
EM-WAPS Rev. 2, DOE/EM-0093 (Dec. 1996).
9. L. H. BARRETT, Letter to J. COHON, Nuclear Waste
Technical Review Board May 30, 2001; available on
the internet at http://www.nwtrb.gov/corr/de053001.pdf
10. US DOE, OCRWM, Viability Assessment of a
Repository at Yucca Mountain, DOE/RW-0508, 1998.
11. Total System Performance Assessment - Analyses for
Disposal of Commercial and DOE Waste Inventories at
Yucca Mountain - Input to Final Environmental Impact
Statement and Site Suitability Evaluation REV 00 ICN
02, 2001.
12. W. J. O'CONNELL, W. L. BOURCIER, J.
GANSEMER, and T. S. UENG, Performance
Assessment Modeling for Savannah River Glass HLW
Disposal in a Potential Repository at Yucca Mountain,
American Chemical Society National
Meeting/Symposium on Science and Technology for
Disposal of Radioactive Tank Wastes, Las Vegas, NV,
September 8-11, 1997.
13. R. WIGELAND, T.H. BAUER, T.H. FANNING, and
E.E. MORRIS, Separations and Transmutation Criteria
to Improve Utilization of a Geologic Repository,
Nuclear Technology,154, (April 2006)
147Global 2007, Boise, Idaho, September 9-13, 2007
