CANDU reactor as minor actinide/thorium burner with uniform

power density in the fuel bundle

Su mer Sahin
a,
*
, Senay Yalcn
b
, Kadir Yldz
c
, Hac Mehmet Sahin
a
, Adem Acr
a
,
Necmettin Sahin
c
a
Gazi U

niversitesi, Teknik Eg itim Faku ltesi, Teknikokullar, Ankara, Turkey

b
Bahc es ehir U

niversitesi, Mu hendislik Faku ltesi, Bes iktas ,

_
Istanbul, Turkey
c
Aksaray U

niversitesi, Mu hendislik Faku ltesi, Aksaray 68000, Turkey

Received 22 March 2007; received in revised form 2 August 2007; accepted 3 August 2007
Available online 18 September 2007
Abstract
A CANDU reactor fueled with a mixed fuel made of thoria (ThO
2
) and nuclear waste actinides has been investigated. The mixed fuel
composition has been varied in radial direction to achieve a uniform power distribution and fuel burn-up in the fuel bundle.
The best fuel compositions with respect to power attening as well as long term reactivity have been found by mixing thoria with 14%
minor actinides in form of MAO
2
in the central fuel bundle and with a radially decreasing MAO
2
content at discrete levels down to 2% at
the periphery. Furthermore, as an alternative fuel composition, 5% UO
2
has been added to the mixed fuel for the sake of a higher degree
of nuclear safeguarding through denaturing the
233
U component with
238
U. The temporal variation of the criticality k
1
and the burn-up
values of the reactor have been calculated for a period of 10 years, operated at full power. The criticality starts at time zero near to
k
1
= 1.24 for both fuel compositions. A sharp decrease of the criticality has been observed during the rst year as a consequence
of rapid plutonium burnout in the actinide fuel. The criticality becomes quasi-constant after the 2nd year after sucient
233
U is accu-
mulated and remains close to k
1,end
= 1.06 over 10 years. Quasi-uniform power generation density has been realized in the fuel bun-
dle throughout the reactor operation.
Very high burn-up could be achieved with the same fuel material (up to 200,000 MW D/MT), provided that the fuel rod claddings
would be replaced periodically. This would reduce fuel fabrication costs and nuclear waste mass for nal disposal per unit energy
drastically.
2007 Elsevier Ltd. All rights reserved.
1. Introduction
1.1. Nuclear waste
Nuclear power plants are producing nuclear waste mate-
rials in substantial quantities, which are classied as ssion
products, minor actinides (MA) and structural materials,
where can be dened the most hazardous radioactive waste
products because of their long term, high level radioactiv-
ity. Table 1 shows the composition of MA in the spent fuel
of a light water reactor. A signicant fraction of reactor
waste MA consists of diverse plutonium isotopes, which
represents serious public and political concern with respect
to misuse of this plutonium and also accidental release of
highly radiotoxic material in to the environment. Hence,
precautions to keep plutonium under strong security have
paramount importance.
Some nuclear waste actinides (
239
Pu,
241
Pu,
242m
Am and
245
Cm) are ssile fuel under thermal neutron spectrum, and
all of them become ssile fuel under fast neutron spectrum.
Some other MA are moderately ssile even under thermal
neutron spectrum (
238
Pu,
241
Am and
244
Cm). Hence, it is
worth to investigate the incineration of MA in conven-
tional commercial reactors.
0306-4549/$ - see front matter 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.anucene.2007.08.003
*
Corresponding author. Tel./fax: +90 312 212 43 04.
E-mail address: sumer.sahin@bahcesehir.edu.tr (S. Sahin).
www.elsevier.com/locate/anucene
Available online at www.sciencedirect.com
Annals of Nuclear Energy 35 (2008) 690703
annals of
NUCLEAR ENERGY
1.2. Thorium as alternative nuclear fuel
World thorium reserves are estimated to be about three
times more abundant than the natural uranium reserves.
Utilization of thorium would make possible to exploit
new nuclear fuel reserves. Further advantages of thorium
would be enhanced nuclear safeguarding because thorium
will be converted to uranium, a signicantly benign mate-
rial compared to plutonium with respect to safeguarding
considerations as well as radio toxicity.
As thorium by itself is not a nuclear fuel, the ssionable
components of MA must serve as a booster ssile fuel
material in a thorium fueled thermal reactor at start up.
Thorium cycle produces
233
U which, from a non-prolif-
eration point of view, is more acceptable compared to plu-
tonium for two reasons. Firstly, it is contaminated with
232
U, which decays to give highly radioactive daughter
products with very hard c-ray radiation, which has
already a very high level of deterrence and would make
handling and diversion dicult. Secondly, the
233
U could
easily be denaturized by adding some
238
U to the thorium,
which would then eliminate the possibility to make a crit-
ical assembly with the bred fuel. In future, thorium may
play a larger role in the nuclear fuel cycle. The superior
nuclear properties of
233
U over
235
U and
239
Pu in thermal
reactors imply potential advantages to
233
U in advanced
terrestrial reactors as well as in compact space craft
reactors.
1.3. Utilization potential of MA and thorium in CANDU
reactors
There has been quite a number of works with respect to
the commercial utilization of thorium in CANDU nuclear
reactors. Higher neutron economy of (CANDU) reactors
compared to light water reactors (LWR)s gives a funda-
mental advantage for thorium fuel. With a uranium enrich-
ment grade of >1%, utilization of thorium in CANDU
reactors can become viable (Boczar et al., 2002a,b; Cri-
toph, 1976; Jagannathan et al., 2001; Loewen et al.,
2001). Other studies have investigated the utilization of
LWR spent fuel in CANDU reactors mixed with thoria
(ThO
2
) (Sahin et al., 2004a,b).
A series of studies have been conducted on the Direct
Use of spent Pressurized water reactor (PWR) fuel In
CANDU reactors (DUPIC) by Korean researchers in a
wide spectrum, which covered a wide spectrum of physical
and technological aspects, including fuel management sce-
nario and fuel performance (Choi et al., 1997).
In the present work, the neutronic analysis of a
CANDU reactor with a mixed fuel made of thorium and
MA is presented. The utilization of MA in combination
with thorium has two main purposes. Firstly, it will enable
the incineration of MA and so the reduction of the long liv-
ing nuclear waste material. Furthermore, it will allow the
exploitation of rich world thorium reserves. For enhanced
inherent safeguarding purposes, 5% natural uranium will
be added to the mixed Th/MA fuel. Furthermore, one of
the main goals of this work will also be to seek uniform
power generation in the fuel bundle.
2. Description of the problem
The investigations in the present study are based on a
GENTILLY-II design (Woodhead and Ingolfsrud, 1975)
modied to a CANDU-thorium reactor, as described in
Altnok (1993), Alkan (2003), IAEA (1979). The reactor
core has 380 fuel bundles. Fig. 1 shows a typical CANDU
fuel channel where the fuel bundle is contained in the cal-
andria tube. The calandria tubes, made of zircaloy-2 (Petrie
et al., 2004a), are arranged in square arrays in D
2
O moder-
ator at low temperature (<71 C) and at 100 kPa pres-
sure. The nuclear heat generated in the fuel rods is
transferred via the D
2
O coolant contained in the pressure
tubes to the power conversion system. Pressure tubes are
made of ZrNb (Ma, 1983). D
2
O coolant enters the tubes
at 266 C and leaves at 310 C at a pressure of 10 MPa
(Lamarsh and Baratta, 2001). The pressure and calandria
tubes are separated through a gap in order to assure an
eective heat barrier between D
2
O moderator at low tem-
perature and D
2
O coolant at high temperature. The gap
is lled with CO
2
gas.
In the CANDU reactor, each fuel channel contains 37
fuel rods in the fuel bundle zone, as shown in Fig. 2. The
cladding of the rod is made of zircaloy-4 (Petrie et al.,
2004a). Table 2 shows the composition of the reactor struc-
tural materials. During the investigations, a series of dier-
ent fuel compositions have been applied consisting of the
following main components:
1. The rst group of the mixed fuel is composed of thoria
(ThO
2
) as the main fuel and few % of MA as booster
material for reactor criticality, where the uranium iso-
topes and ssion products have been removed from
the LWR spent fuel. The booster fuel contains mainly
Table 1
The composition of MA in the spent fuel of a light water reactor
Isotopes Mass (kg/year)
per unit PWR
a
237
Np 15.1
238
Pu 16.1
239
Pu 205
240
Pu 120
241
Pu 72.7
242
Pu 41.6
241
Am 6
242m
Am 0.00793
243
Am 21.8
244
Cm 15.6
245
Cm 1.74
33 MWd/kg, 32.5% thermal eciency, 150 days after discharge.
Nuclear Chemical Engineering (Manson et al., 1981, p. 370, Table 8.5).
a
Pressurised-water reactor, fuel with plutonium recycle, 1000-Mwe
reactor, 80% capacity factor.
S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703 691
reactor grade plutonium isotopes with some other
actinides, such as neptunium, americium and curium
isotopes.
2. The second group of the mixed fuel is composed of tho-
ria (ThO
2
) as the main fuel and few % of the same MA,
as in item 1, as booster material for reactor criticality.
Furthermore, 5% natural UO
2
has been added to dena-
turize the new
233
U fuel with
238
U. The latter will assure
enhanced nuclear safeguarding of the uranium compo-
nent in addition to the very hard c-ray radiation, origi-
nating from
232
U.
The isotopic composition of the minor actinides has
been adopted from Manson et al. (1981) and the atomic
densities of the minor actinides in the ThO
2
/MAO
2
mixed
fuel in dierent rows of a CANDU fuel bundle are shown
in Table 3. In the mixed fuel, minor actinides will serve in
earlier operational years as booster fuel until sucient
quantities of
233
U is built up to carry the reactor criticality.
This would allow to burn signicant quantities of thorium
fuel, and consequently would lead to a wide exploitation of
world thorium reserves, based on the well-established con-
ventional CANDU reactor technology. In that case, large
development costs and eorts to develop a new technology
line for thorium exploitation could be avoided.
3. Numerical calculations
The neutron transport calculations have been performed
with the help of SCALE5 system (Petrie, 2004b), in S
8
-P
3
approximation by solving the Boltzmann transport equa-
tion in one-dimensional geometry with the transport code
XSDRNPM using the 238-neutron groups data library,
derived from ENDF/B-V (Jordan et al., 2001). The library
has 148 fast and 90 thermal (above 3 eV) groups, and
10.3378
11.2064
12.90
13.1794
Calandria tube
CO
2
Moderator
32.243
Fuel
bundle
28.575
I
II
Pressure tube
Fig. 1. Cross-sectional view of the fuel channel: I original CANDU
square lattice cell. II equivalent diameter, used in calculations (dimen-
sions are in centimeters, not in scale).
13.081
29.769
57.506
86.614
Fuel Rod
Coolant
Pressure tube
D
o
D
i
Fuel
Clad
Coolant/Moderator
a b
I
II
III
IV
Basic fuel cell
D
i
=12.243; D
o
= 13.081
Fig. 2. Placement of 37-fuel rods in the bundle (dimensions are in millimeters, not in scale).
Table 2
Main structure materials in CANDU reactors
Material Elements Weight fraction in alloy (%)
Zircaloy-4 (Petrie et al., 2004a) Zr 98.23
Sn 1.45
Fe 0.21
Cr 0.11
Hf 0.01
Zr2.5Nb (Ma, 1983) Zr 97.5
Nb 2.5
Zircaloy-2 (Petrie et al., 2004a) Cr 0.1
Fe 0.135
Zr 98.25
Sn 1.45
Ni 0.055
Hf 0.01
692 S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703
enables very ne energy resolution for multi-group neutron
transport calculations. The cross-section sets are available
for the discrete temperatures of 300 K, 600 K and 900 K,
which have applied for the moderator, coolant and fuel,
respectively. The resonance self-shielding weighted cross-
sections have been processed with the help of the CSAS
control module using BONAMI for unresolved resonances
and NITAWL-III for resolved resonances.
3.1. Considerations for the lattice criticality and burn-up
calculations
The initial fuel charge starts with a high k
e
value for a
CANDU reactor. As the CANDU reactor uses an on-
power refueling scheme to provide sucient reactivity dur-
ing reactor operation without reactor shutdown, a modest
excess reactivity will be sucient for long-time operation.
The on-power refueling results in a continuous distribution
of fuel burn-up in the core. A fuel bundle can be reshued
in the core until discharge reactivity k
1,end
is attained. The
k
1,end
for a CANDU fuel channel is typically 1.051.06
considering the leakage from the core and parasitic absorp-
tion in the core (see reference Sahin et al., 2004b, chapter
2.2). In the present study, multiplication factors (k
1
) for
the fuel channel are calculated under consideration the
axial leakage via a pseudo absorption with the help of a
axial buckling factor for an approximate core height of
594 cm.
In a series of successive calculations, the temporal vari-
ation of the atomic densities N of the ssionable fuel iso-
topes in a CANDU fuel bundle during reactor operation
are evaluated for discrete time intervals Dt = 10 days,
under consideration of the variations of the actinide
nuclide densities through nuclear reactions via neutron
capture and radio-active transformation processes of the
actinides, which are shown in Fig. 3, in detail.
Atomic density variations of the actinides in the course
of breeding reactions are calculated as
DN
d
Dt N
m1
Z
r
d;m1
E UE dE Dt k
m2
N
m2
1
Indices denote daughter (d) and mother (m) isotopes
Atomic density variations in the course of depletion
reactions are calculated as
DN Dt N
Z
r
dep
E UE dE Dt k N 2
Eqs. (1) and (2) consider the variations of the atomic num-
ber densities of the ssionable nuclides by nuclear transmu-
tations and radioactive decay. The atomic densities have
been calculated, considering spatial variations in the neu-
tron spectrum and the atomic densities of all ssionable
isotopes over the radial coordinate in the fuel bundle.
3.2. Lattice calculations for the fuel rod and fuel channel
Resonance self shielding calculations for all nuclides,
both actinide fuels and structural materials and the follow-
ing cross-section processing are performed with the help
BONAMI and NITAWL-III at rst step for the CANDU
fuel rod as a micro lattice, shown in Fig. 2b. This code sys-
tem allows cell calculations in cylindrical geometry, with a
maximum of three zones dened as fuel, cladding and mod-
erator regions, and apply Bondarenko method for reso-
nance treatment.
One-dimensional neutron transport calculations are
conducted with the help of the S
N
transport code
XSDRNPM for the CANDU fuel channel in Fig. 1, as a
representative macro lattice for the reactor core. The qua-
dratic fuel channel is transformed to a volume- and mate-
rial equivalent cylindrical macro lattice for each fuel rod
row in radial direction separately. Due to the irregularities
of the CANDU bundle in Fig. 2a, moderator volume per
fuel rod reveals some minor variations in radial direction
Table 3
Composition and atomic densities of the ThO
2
/MAO
2
mixed fuel in dierent rows of a CANDU fuel bundle
Elements Atomic densities (10
24
cm
3
)
1st row 2nd row 3rd row 4th row
232
Th 8.03054E03 8.40401E03 8.87095E03 9.15106E03
16
O 1.86757E02 1.86757E02 1.86757E02 1.86757E02
237
Np 4.27493E05 3.05352E05 1.52675E05 6.10700E06
238
Pu 4.55211E05 3.25151E05 1.62575E05 6.50300E06
239
Pu 5.79229E04 4.13735E04 2.06867E04 8.27474E05
240
Pu 3.39077E04 2.42198E04 1.21099E04 4.84386E05
241
Pu 2.05720E04 1.46943E04 7.34716E05 2.93887E05
242
Pu 1.18180E04 8.44142E05 4.22071E05 1.68829E05
241
Am 1.75081E05 1.25058E05 6.25291E06 2.50117E06
242m
Am 2.24380E07 1.60271E07 8.01357E08 3.20542E08
243
Am 6.17145E05 4.40818E05 2.20409E05 8.81635E06
244
Cm 4.42078E05 3.15770E05 1.57885E05 6.31540E06
245
Cm 5.10653E06 3.64752E06 1.82376E06 7.29505E07
*
*
: Read as 7.29505 10
07
.
S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703 693
in the fuel zone, and the arrangement of the 37 fuel rods
does not comply exactly with a geometrically hexagonal
structure. A general purpose code system, such as
SCALE5, may at rst appear not directly applicable for
CANDU congurations due to irregularity of the CANDU
bundles. However, SCALE5 allows a multitude of fast
computer runs for a dierentiated analysis of nuclear
energy production in one-dimensional geometry with a
very ne energy resolution in neutron 238-groups.
In order to consider the irregularity of the basic cell
geometry in the CANDU fuel bundle, separate micro-cell
calculations are conducted for each fuel rod row, namely
from 1st one to 4th one in the bundle, under consideration
of variable moderator volume for each individual rod.
During time calculations, the variations of the densities
of all ssionable nuclides are calculated for each fuel rod
location separately under consideration of resonance self
shielding eects and neutron feed-back at every discrete
time step of Dt = 10 days. In this way, four dierent
cross-section sets have been processed in the fuel zone in
order to take into account minor regional spectral uctua-
tions in the fuel bundle.
Following the lattice calculations for the basic fuel rod
cell, the micro-cell weighted and resonance treated cross-
sections are then used for the neutron transport calcula-
tions with the S
N
code XSDRNPM for the fuel channel.
The cluster geometry is divided into separate zones with
similar neutronic characteristics, such as, fuel zone (4 radial
regions with minor variations of moderator/fuel ratio),
structures, CO
2
gap and moderator zones, as shown in
Fig. 1. During the lattice calculations, a reecting bound-
ary condition in the centre and a white boundary condition
at the outer periphery of the cylindrical moderator region
are applied.
The entire calculation procedure described in this chap-
ter has been repeated for each discrete time step of Dt = 10
days over 10 full power reactor operation period, namely
starting from the resonance self shielding calculations for
new temporally and spatially variable actinide densities in
the fuel zone.
For reactor dimensions exceeding several neutron mean
free paths, one fuel channel can be considered as a repre-
sentative lattice element for the entire reactor core to fol-
low temporal evaluation of burn-up characteristics.
3.3. Criticality and burn-up of the basic reactor lattice
In large LWRs and CANDU reactors, k
1
value of a
macro lattice is very close to k
e
of the entire reactor. In
previous work, the multiplication factor k
1
of the
CANDU macro lattice and the multiplication factor k
e
of the CANDU reactor including core and reector with
nite dimensions and under consideration of the neutron
leakage have been calculated for dierent fuel composi-
tions. Comparisons of corresponding k
1
and k
e
have
shown that the dierence between k
1
and k
e
remains
<1% for all fuel compositions, both at the beginning of life
(BOL) and at the end of life (EOL) of the reactor (Sahin
et al., 2004a). This is also a good indication that the neu-
tron leakage is very low due to the great dimensions of
the CANDU reactor coupled with an ecient neutron
reector. Hence, criticality and burn-up values of the fuel
channel with a white boundary condition at the periphery
can represent the reactor core. Time dependent calculations
are conducted for a constant total neutron ux in n/cm
2
at
the initial total reactor power level of 640 MW
el
and
2180 MW
th
(IAEA, 1979) using two dierent mixed fuel
compositions, described in Chapter 2.
234
U
238
Np
2.1 days
239
Pu
2.4410
4
yr
237
Np
2.1410
6
yr
233
U
1.6210
5
yr
233
Pa
27 days
232
Th
1.4510
10
yr
240
Pu
6850 yr
238
Pu
86.4 yr
242
Am
152 yr
244
Cm
17.6 yr
242
Cm
163 days
245
Cm
9.310
3
yr
241
Am
433 yr
242
Am
16 h
243
Am
7400 yr
(n,)
(n,2n)

-
(n,)
(n,)
(n,)
(n,)
(n,)
(n,)

-
(n,2n)
(n,2n)
(n,)
(n,)
(n,)

%35
%65
%84
235
U
(n, )
(n, ) 237
U
6.75 days
(n, )

-
238
U
(n,2n)
239
U
23.5 min
(n, )
239
Np
2.35 days

-
236
U
241
Pu
14.4 yr
(n,)
(n,)
242
Pu
3.7610
5
(n,2n)
(n,)
+
%16
Fig. 3. Major nuclear reactions and radioactive transformation processes in the course of reactor operation.
694 S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703
3.3.1. Time evolution of the lattice criticality
Previous work has shown that utilization of uniform
mixed fuel composition with plutonium components in
all fuel rods of a reactor lattice would lead to a rapid
decrease of ssion power generation in the lattice towards
the central fuel rows (4-fold) due to the radial depression
of thermal ux (Sahin et al., 2006). Non-uniform power
generation in the lattice would lead to thermal stresses
and non-uniform fuel utilization. Furthermore, tempera-
ture limits at the peripheral rows would force the reactor
to run at low load factors and increase cost of electricity.
Hence, a uniform power generation is to be preferred. In
this work, the MAO
2
fraction in mixed fuel is continuously
reduced at radial direction in the lattice.
A variety of fuel composition combinations have been
applied in dierent runs seeking a quasi-uniform ssion
power generation in the bundle, with decreasing MAO
2
fraction from the central rod (row # I in Fig. 2) outwards,
keeping in mind that the acceptable innite lattice reactiv-
ity for the fuel bundle should be around k
1
= 1.051.06.
Some of the investigated variable ThO
2
/MAO
2
composi-
tions in successive radial rows of the fuel rods are cited
in the legend of Fig. 4, where the temporal variations of
reactor lattice criticality k
1
over an operation period of
10 years are shown for the selected and most promising fuel
compositions.
At start up, the reactor criticality will be supported by
the ssile components of the actinide isotopes, mainly
through
239
Pu and to some degree through
241
Pu. Hence,
a higher MAO
2
content in the mixed fuel leads to higher
lattice criticality in the early years of reactor operation.
However, most of the ssile actinides will be burnt out
almost totally in the early years and there share on ssile
neutron generation decreases gradually in the course of
reactor operation, along with a rapid criticality drop,
depending on the content of the actinide fraction in the ini-
tial mixed fuel.
In long term, the contribution of the newly generated
and accumulated
233
U isotope to the criticality will domi-
nate more and more, which is directly produced through
the neutron capture in
232
Th. Therefore higher ThO
2
and
consequently lower MAO
2
content in the mixed fuel will
lead to higher criticality values in later years of the reactor
operation due to the higher
233
U production rate in the lat-
tice. For that reason, a crossover of the criticality values
can be distinguished in Fig. 4, where a mixed fuel with
lower MAO
2
content catches that one with a higher
MAO
2
content in later years.
Recommended composition with respect to long term
criticality and quasi-constant power generation in the bun-
dle turns out to be variable ThO
2
/MAO
2
fractions in radial
direction in the fuel rods decreasing from 14% MAO
2
in
the central row down to 2% MAO
2
at the periphery in a
fuel channel, indicated in the legend of Fig. 4 as mode l.
Table 3 shows the selected fuel compositions and the corre-
sponding atomic densities during the rest of the study,
where a k
1
value 1.051.06 can be sustained over 10
years.
After the determination of the most suitable ThO
2
/
MAO
2
composition prole, 5% natural UO
2
has been
added to the fuel to denaturize the
233
U production with
238
U, in situ. Fig. 5 shows the innite lattice criticality k
1
and the fuel burn-up values as a function of the plant oper-
ation period by full power for the radially variable mixed
fuel compositions, selected under considerations mentioned
in Section 2. The reactor starts by k
1
= 1.2411 and 1.2373
for the ThO
2
/MAO
2
and ThO
2
/MAO
2
/UO
2
mixed fuel
compositions, respectively, which are indicated in the leg-
0 2 4 6 8 10
1
1.05
1.1
1.15
1.2
1.25
1.3
1.35
1.4
1.45
Operation time (years)
k

0 2 4 6 8 10
1. 03
1. 04
1. 05
1. 06
1. 07
1. 08
1. 09
Fig. 4. Temporal variation of the lattice criticality k
1
.
r
: 1st row: 92% ThO
2
+ 8% MA; 2nd row: 93% ThO
2
+ 7% MA; 3rd row: 94% ThO
2
+ 6% MA;
4th row: 95% ThO
2
+ 5% MA.
s
: 1st row: 91% ThO
2
+ 9% MA; 2nd row: 92% ThO
2
+ 8% MA; 3rd row: 93% ThO
2
+ 7% MA; 4th row: 94%
ThO
2
+ 6% MA.
t
: 1st row: 90% ThO
2
+ 10% MA; 2nd row: 91% ThO
2
+ 9% MA; 3rd row: 92% ThO
2
+ 8% MA; 4th row: 93% ThO
2
+ 7% MA.
u
: 1st
row: 86% ThO
2
+ 14% MA; 2nd row: 90% ThO
2
+ 10% MA; 3rd row: 95% ThO
2
+ 5% MA; 4th row: 98% ThO
2
+ 2% MA.
S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703 695
end of Fig. 5. One can observe a steep criticality drop in the
1st year along with the burn-up of plutonium in the acti-
nide component. Once the amount of
233
U will reach a cer-
tain level, the criticality is then stabilized, namely after the
3rd year of plant operation period. It remains k
1
> 1.06
over 10 years for the ThO
2
/MAO
2
mixed fuel, and is still
very close to k
1
= 1.06 for the ThO
2
/MAO
2
/UO
2
mixed
fuel. Addition of 5% UO
2
decreases k
1
to some degree
because of the reduction of the precious
233
U fuel density.
However, the ThO
2
/MAO
2
/UO
2
fuel can still be consid-
ered sucient for a continuous CANDU reactor operation.
At start-up, the reactor will have relatively large reactiv-
ity margins (20%). Several possibilities can be considered
to compensate excess reactivity:
Utilization of a multitude of control rods. This would
consume a great fraction of the neutrons and may lead
to severe neutron ux distortions in the core.
Utilization of burnable poison in form of boric acid.
This would also consume a great fraction of the precious
excess neutrons.
A series of transmutation fuel (TF) rods containing
ThO
2
can be inserted. The excess neutrons can convert
the additional
232
Th in
233
U which then can be fed into
the fuel cycle in later stages during the renewal of the
fuel rod cladding in the course of multiple fuel utiliza-
tion. The TF rods can be inserted in such a way to sup-
port the ssion power attening in the reactor core.
However, this is out of the scope of the present work.
3.3.2. Burn-up considerations
One can recognize in Fig. 5 that the reactor can attain
very high burn-up levels over 10 years with same fuel mate-
rial by sucient criticality, viz., >200,000 MW D/MT. The
excellent neutron economy of a CANDU reactor allows a
continuous reactor operation until a signicant fraction
of thorium will be burnt up. This process can continue until
material damage criteria would limit the use of the same
fuel rod.
Conservative burn-up values in a heavy water reactor
(HWR), LWR, and FBR are known to be around
<10,000 (7000), 30,00040,000, and 100,000 MW D/
MT, respectively. These correspond to fuel consumption
rates of 142, 3325 and 10 g/MW D for the respective reac-
tor types. Previous studies have reported that the peak dis-
charge exposure must apparently be in excess of
200,000 MW D/MT in order to reach burn-ups that are
high enough to fully exploit the economic potential of a
mixed-oxide (MOX) fuel system (Leggertt and Omberg,
1987). Although such burn-up levels have not yet been
achieved in conventional reactors, there is no reason to
expect that such performance cannot be attained (Waltar
and Deitrich, 1988). These observations may be true for
fast reactors or pebble bed modular reactors, however,
and probably not for the fuel elements of current CANDU
reactors and LWRs. But the trends in MOX fuel develop-
ment indicate that UO
2
fuel elements could be designed to
withstand the necessary burn-ups for potential advantages
(Leggertt and Omberg, 1987; Waltar and Deitrich, 1988).
Standard CANDU reactors fueled with natural uranium
can reach burn-up only below 10,000 (7000) MW D/MT
due to the rapid fuel depletion and related criticality loss
(IAEA, 1979). On the other hand, a CANDUreactor, fueled
with the investigated mixed ThO
2
/UO
2
/MAO
2
fuel, could
attain a burn-up level of 40,000 and 100,000 MW D/
MT in 2 and 5 years, comparable to that in a LWR and fast
breeder reactor (FBR), respectively, as shown in Fig. 5.
Then, for these burn-up values, fuel consumption rates
would shrink down to 25 and 10 g/MW D from
142 g/MW D by utilization of nuclear waste actinides
instead of natural uranium in a CANDU reactor.
Although such high burn-up values are not applicable for
the present CANDU reactors, previous work by Leggertt
and Omberg (1987), Waltar and Deitrich (1988) report
about the prospects of fuel element development to with-
stand burn-up levels up to >200,000 MW D/MT. Theoret-
ically, sucient reactor criticality could still be maintained
with minor actinide waste/thorium mixed fuel for burn-up
levels as high as 200,000 MW D/MT with the same fuel
material.
These high burn-up grades for a long fuel life-time might
be realized by replacing the fuel rod cladding at denite
0 2 4 6 8 10
1
1.05
1.1
1.15
1.2
1.25
0
50000
100000
150000
200000
250000
k

Operation time (years)

B
u
r
n
-
U
p

(
M
W
d
a
y
/
t
o
n
e
)

k=1.06
Fig. 5. Temporal variation of the lattice criticality k
1
and the fuel burn-
up grade. Solid lines: 1st row: 86% ThO
2
+ 14% MAO
2
; 2nd row: 90%
ThO
2
+ 10% MAO
2
; 3rd row: 95% ThO
2
+ 5% MAO
2
; 4th row: 98%
ThO
2
+ 2% MAO
2
. Dashed lines: 1st row: 81% ThO
2
+ 14% MAO
2
+ 5%
UO
2
; 2nd row: 85% ThO
2
+ 10% MAO
2
+ 5% UO
2
; 3rd row: 90%
ThO
2
+ 5% MAO
2
+ 5% UO
2
; 4th row: 93% ThO
2
+ 2% MAO
2
+ 5%
UO
2
.
696 S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703
intervals (for example after every 50,000100,000 MW D/
MT) accompanied with the removal of gaseous ssion
products through a simplied reprocessing process. Then,
the same fuel material could be used until a great fraction
of the thorium fuel is burnt up. This would reduce the
nuclear waste per unit energy output drastically with a fuel
consumption rate as low as 5 g/MW D.
3.4. Fissile fuel density
A series of ssionable isotopes transformations via neu-
tron interaction and radioactive decay will be observed
during reactor operation, as outlined in Fig. 3. In a
CANDU reactor, most of the thermal neutrons will be
absorbed in the peripheral fuel rods and only a small frac-
tion of them can reach the central fuel rod. Hence, fuel
burn-up and transformation proceed at the periphery faster
than in the center.
Fig. 6 shows the variation of the weight densities of the
dominant actinide isotopes in the 18 peripheral fuel rods
in the bundle for the mixed fuel compositions, as indicated
in the legend. These rods start with the lowest plutonium
content, which falls rapidly, except
242
Pu in the rst and sec-
ond year of plant operation time. After the 3rd year,
242
Pu
remains as the sole plutonium isotope in the mixed fuel.
Most of the plutonium isotopes are converted to higher
actinides, such as
243
Am and
244
Cm. One can further
observe a rapid built up of
233
U, which will be gradually
degraded with
234
U, a positive aspect with respect to nuclear
safeguarding. Fig. 7 shows the temporal variation of the
weight densities of other actinide isotopes in the peripheral
fuel rods. A modest
235
U accumulation is observed.
Through nuclear transformations, some precious nuclear
fuels with very high thermal ssion cross-sections are con-
tinuously produced, where
242m
Am is burnt up in situ
totally, but a tiny amount of
245
Cm remains always present.
Utilization potential of
242m
Am and
245
Cm for space craft
reactors is discussed in earlier work (Rubbia, 1998; Ronen
et al., 1998; Sahin and Calinon, 1985; Sahin, 1989).
The central rod starts with higher plutonium content
and the respective isotopes reveal a signicantly dierent
temporal scenario, as depicted in Fig. 8. Although the even
ssile isotopes
239
Pu and
241
Pu are consumed relatively fast,
the odd isotope
240
Pu holds longer, whereas
242
Pu becomes
the dominating isotope over long term. This would support
the nuclear safeguarding aspects of plutonium, but its
radio-toxicity hazard potential remains present. There will
be gradual build up of
234
U and a signicant amount of
higher actinides in form of
243
Am and
244
Cm. The deple-
tion of other actinide isotopes in the central row is shown
in Fig. 9, where some precious
245
Cm will be present
throughout the entire operation period.
The alternative fuel with 5% natural-UO
2
reveals a sim-
ilar temporal behavior, as shown in Figs. 1013. One can
recognize in Fig. 10 that
233
U becomes rapidly the domi-
nant ssile isotope in the peripheral rods. On the other
hand, in the central rod,
242
Pu becomes the dominating iso-
tope, Fig. 12. The weight densities of
233
U,
243
Am and
244
Cm become more or less comparable. The modest
235
U
fraction in the initial charged will be burnt up, but its
supply becomes greater than its consumption through
0 2 4 6 8 10
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
Operation time (years)
F
i
s
s
i
l
e

i
s
o
t
o
p
e
s

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
233
U
239
Pu
234
U
244
Cm
241
Pu
242
Pu
240
Pu
243
Am
Fig. 6. Density variations in the peripheral row of the main ssionable
isotopes for the bundle with 1st row: 86% ThO
2
+ 14% MAO
2
; 2nd row:
90% ThO
2
+ 10% MAO
2
; 3rd row: 95% ThO
2
+ 5% MAO
2
; 4th row: 98%
ThO
2
+ 2% MAO
2
.
0 2 4 6 8 10
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
Operation time (years)
F
i
s
s
i
l
e

i
s
o
t
o
p
e
s

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
242m
Am
238
Pu
235
U
241
Am
245
Cm
237
Np
Fig. 7. Density variations of the peripheral row of the main ssionable
isotopes in the bundle (legend as Fig. 6).
S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703 697
successive neutron capture in
233
U, so that one can observe,
in Figs. 11 and 13, a swing of
235
U density with increasing
tendency after certain time. However, its presence as ssile
isotope remains always very minor; so that addition of nat-
ural-UO
2
will have a negative eect on the reactor
criticality.
Common aspect for the actinides is that most of them
will burn-up. At the end phase, there will be an accumula-
tion of some heavy isotopes, such as,
242
Pu,
243
Am and
244
Cm. Previous work has indicated that they could be used
as excellent ssionable fuels under the hard neutron spec-
trum of a fusionssion (hybrid) blanket (Kumar and
0 2 4 6 8 10
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
Operation time (years)
F
i
s
s
i
l
e

i
s
o
t
o
p
e
s

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
233
U
239
Pu
234
U
244
Cm
241
Pu
242
Pu
240
Pu
243
Am
Fig. 10. Density variations in the peripheral row of the main ssionable
isotopes for the bundle with 1st row: 81% ThO
2
+ 14% MA + 5% UO
2
;
2nd row: 85% ThO
2
+ 10% MA + 5% UO
2
; 3rd row: 90% ThO
2
+ 5%
MA + 5% UO
2
; 4th row: 93% ThO
2
+ 2% MA + 5% UO
2
.
0 2 4 6 8 10
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
Operation time (years)
F
i
s
s
i
l
e

i
s
o
t
o
p
e
s

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
242m
Am
238
Pu
235
U
241
Am
245
Cm
237
Np
Fig. 11. Density variations in the peripheral row of the main ssionable
isotopes for the bundle (legend as Fig. 10).
0 2 4 6 8 10
0
0.1
0.2
0.3
0.4
0.5
Operation time (years)
F
i
s
s
i
l
e

i
s
o
t
o
p
e
s

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
233
U
239
Pu
234
U
244
Cm
241
Pu
242
Pu
240
Pu
243
Am
Fig. 8. Density variations in the central row of the main ssionable
isotopes for the bundle (legend as Fig. 6).
0 2 4 6 8 10
0
0.01
0.02
0.03
0.04
Operation time (years)
F
i
s
s
i
l
e

i
s
o
t
o
p
e
s

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
242m
Am
238
Pu
235
U
241
Am
245
Cm
237
Np
Fig. 9. Density variations of the central row of the main ssionable
isotopes in the bundle (legend as Fig. 6).
698 S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703
Sahin, 1983, 1984; Sahin and Kumar, 1984a,b; Sahin and
Al-Kusayer, 1986). A rigorous incineration of all actinides
of the CANDU spent fuel in a fusion-ssion (hybrid) reac-
tor is reported in earlier work (Sahin and Yapc, 1989;
Sahin et al., 1991). The same application would also be
possible with accelerator driven systems.
The initial fertile isotopes
232
Th and
238
U are consumed
continuously over the plant operation. Gradual depletion
of their metallic density is depicted in Fig. 14 for the respec-
tive mixed fuel compositions. The depletion of non-ssile
isotopes reduces the macroscopic absorption cross-section
in the reactor core and has a positive eect on k
1
. As most
of the initial actinides are burnt up in early years, the plant
energy will mainly be supplied by thorium consumption,
keeping in mind that two neutrons will be needed to burn
thorium fuel.
Fig. 15 shows the variation of the accumulated densities
of main ssile isotopes (
233
U +
235
U +
239
Pu +
241
Pu) in
the central and in the peripheral fuel rods over a power
plant operation period of 10 years. In earlier years, one
can observe a sharp decrease of the cumulative ssile den-
sities, where the initial
239
Pu and
241
Pu are burnt up rap-
idly. However, the decrease of the ssile material will
strongly slow down after few years so that fuel generation
begins to recover almost all fuel depletion. Such a CANDU
reactor becomes then practically a thorium burner after the
2nd year. Continuous depletion of the ssile isotopes is
compensated by reduced absorption in
232
Th and
238
U so
that k
1
1.06 can be maintained over >10 years. Theoret-
ically, the reactor criticality could be sustained until the
entire thorium fuel is burnt up, provided that the fuel rods
could be fabricated to withstand such high burn-up levels
(>200,000 MW D/MT), as reported in Leggertt and
Omberg (1987).
0 2 4 6 8 10
0
0.1
0.2
0.3
0.4
0.5
Operation time (years)
F
i
s
s
i
l
e

i
s
o
t
o
p
e
s

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
233
U
239
Pu
234
U
244
Cm
241
Pu
242
Pu
240
Pu
243
Am
Fig. 12. Density variations in the central row of the main ssionable
isotopes for the bundle (legend as Fig. 10).
0 2 4 6 8 10
0
0.01
0.02
0.03
0.04
Operation time (years)
F
i
s
s
i
l
e

i
s
o
t
o
p
e
s

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
242m
Am
238
Pu
235
U
241
Am
245
Cm
237
Np
Fig. 13. Density variations in the central row of the main ssionable
isotopes for the bundle (legend as Fig. 10).
0 2 4 6 8 10
0
1
2
3
4
5
6
7
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
2
3
2
T
h

m
e
t
a
l
l
i
c

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
Operation time (years)
2
3
8
U

m
e
t
a
l
l
i
c

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
238
U
Fig. 14. Gradual depletion of the metallic density of the dominant
232
Th
and
238
U isotopes.
r
: 1st row: 86% ThO
2
+ 14% MA; 2nd row: 90%
ThO
2
+ 10% MA; 3rd row: 95% ThO
2
+ 5% MA; 4th row: 98%
ThO
2
+ 2% MA.
s
: 1st row: 81% ThO
2
+ 14% MA + 5% UO
2
; 2nd
row: 85% ThO
2
+ 10% MA + 5% UO
2
; 3rd row: 90% ThO
2
+ 5%
MA + 5% UO
2
; 4th row: 93% ThO
2
+ 2% MA + 5% UO
2
. (Solid lines,
in the central fuel row; broken lines, in the peripheral fuel row.)
S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703 699
In early stages of plant operation,
239
Pu will be the dom-
inating ssile fuel in this mixed fuel, which will be replaced
by
233
U gradually. At that point, a remark is warranted
concerning reactor control. The delayed neutron fraction
in
233
U,
235
U and
239
Pu for thermal is = 0.00264,
0.0065 and 0.00221, respectively (ANL-5800, 1963). Hence,
special attention and more care are needed for reactor con-
trol with
233
U and
239
Pu fuel.
3.5. Fission power prole in the bundle
The spatial ssion rate prole and consequently the s-
sion power generation function in a fuel channel have pri-
mary importance for total reactor power output, because
temperature limits in the fuel rod claddings are imposed
by material considerations. By uniform fuel loading, higher
heat generation would occur at the periphery, and must be
kept below the limits for the maximum permissible temper-
ature so that the reactor operation has to be at lower power
load (Sahin et al., 2004b, 2006). For economical and tech-
nological reasons, a uniform power ssion generation will
be more benecial, which is sought in this work.
Fig. 16 shows the ssion rate in radial direction in a fuel
channel, which is directly proportional to the ssion power
density (FPD). One can recognize sharp peaking factors in
early stages of plant operation (BOL and 1 year), which is
caused by the strong depression of the soft neutron spec-
trum of a heavy water moderated reactor due to the pres-
ence of the plutonium fuel with high thermal neutron
absorption. As the plutonium density decreases in radial
direction in the corresponding rows, the homogenized cross
sets vary in dierent fuel zones, and a discontinuity occurs
by passing from the internal rows outwards. Along with the
continuous depletion of plutonium in later stages of plant
operation, peaking factors will be smoothed towards
EOL, where ssion is dominated by
233
UO
2
fuel.
The behavior of ssion power distribution of a standard
235
UO
2
of a CANDU was evaluated in Fig. 8 of Sahin et al.
(2004a) and in Fig. 5 of Sahin et al. (2004b). With uranium
fuel only, the picking factors remain at modest level, simi-
lar to those towards EOL in the present study.
The specic shape of the power distribution with pluto-
nium fuel leads to non-uniform power generation in the
same fuel rod at BOL up to a factor of 2 in the 3rd
row. The mechanical fuel behavior will then need special
care. Thermo-hydraulically coupled stress analysis must
be performed in case of a practical application of this type
of fuel, which is out of the scope of the present study.
Concerning power attening between dierent rows,
even at BOL with the chosen composition of the mixed
fuel, the FPD is almost quasi-uniform. The quasi-unifor-
mity of the FPD between rows can be observed over the
entire operation period, i.e., the fuel in the core is burnt
also uniformly. At the EOL (after 10 years), one can
observe a slight increase of the power production towards
periphery. After the 2nd year,
233
U becomes the dominant
ssile isotope in the reactor. As the neutron absorption in
uranium is less than in plutonium, the non-uniformity of
FPD remains low. The reactor can operate close the nom-
inal power over the entire life-time.
Fig. 16. The ssion power production density in the fuel zone of the
bundle with 1st row: 81% ThO
2
+ 14% MA + 5% UO
2
; 2nd row: 85%
ThO
2
+ 10% MA + 5% UO
2
; 3rd row: 90% ThO
2
+ 5% MA + 5% UO
2
;
4th row: 93% ThO
2
+ 2% MA + 5% UO
2
; Solid lines (BOL);
dashed lines (1 year); dash-
dot lines (EOL).
0 2 4 6 8 10
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.55
0.6
0.65
Operation time (years)
T
o
t
a
l

f
i
s
s
i
l
e

i
s
o
t
o
p
e
s

d
e
n
s
i
t
y

(
g
r
/
c
m
3
)
Fig. 15. Temporal variation of the accumulated densities of ssile isotopes
(
233
U +
235
U +
239
Pu +
241
Pu) in the fuel bundle.
r
: 1st row: 86%
ThO
2
+ 14% MA; 2nd row: 90% ThO
2
+ 10% MA; 3rd row: 95%
ThO
2
+ 5% MA; 4th row: 98% ThO
2
+ 2% MA.
s
: 1st row: 81%
ThO
2
+ 14% MA + 5% UO
2
; 2nd row: 85% ThO
2
+ 10% MA + 5% UO
2
;
3rd row: 90% ThO
2
+ 5% MA + 5% UO
2
; 4th row: 93% ThO
2
+ 2%
MA + 5% UO
2
. (Solid lines, in the central fuel row; broken lines, in the
peripheral fuel row.)
700 S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703
The fuel assembly average power is normalized to 1 at
BOL by constant ux (n/cm
2
s) and is depicted as horizon-
tal lines. The external rows determine the average power,
because of the higher fuel content. One can recognize that
the average power reveals minor variations over plant oper-
ation, and remains around 1 by constant neutron ux
level. This is a positive factor for long term reactor control.
4. Safeguard aspects of the mixed fuel
In the course of reactor operation, the fuel charge will
produce uranium and plutonium components in situ, both
of which are naturally managed during the entire operation
period of the reactor in a way that avoids all proliferation
concerns. The non-proliferation concern of the actinides
may arise mainly
during the multiple utilization of the same fuel in the
reactor
and in the course of the fuel discharge from the reactor
to the storage.
Very reliable and inherently assured safeguarding of the
fuel for both elements is described below in detail.
4.1. Plutonium component
The most ecient way to denaturize plutonium is to
increase the amount of even
240
Pu and
242
Pu isotopes. Then,
their high spontaneous ssion rate continuously generates
abundant neutron background which would then lead to
a pre-detonation of the critical assembly and prevent an e-
cient nuclear explosion (Meyer et al., 1977). The spontane-
ous ssion half-life T
1/2
of the plutoniumisotopes is given in
Manson et al. (1981) as T
1/2
= 5.5
*
10
15
, 1.4
*
10
11
and
7.1
*
10
10
years for
239
Pu,
240
Pu and
242
Pu, respectively.
Hence, the intensity of spontaneous ssion neutron genera-
tion in
240
Pu and
242
Pu is about 40,000 and 77,000 times
higher than in
239
Pu so that the presence of a moderate level
(>5%) of
240
Pu and/or
242
Pu would already be sucient to
denaturize the plutonium to a non-prolic level.
The initial charge of nuclear actinides is already strongly
denaturized, as one can see in Figs. 6, 8, 10 and 12. In all
stages of reactor operation, there are inherently suciently
high
240
Pu and
242
Pu isotopes generated in the mixed fuel.
After few years, the
239
Pu isotope will almost diminish
totally and only a small fraction of ssile
241
Pu will be pres-
ent in the mixed fuel so that the
240
Pu and
242
Pu will always
be dominating isotopes in plutonium. The naturally occur-
ring denaturing of the generated plutonium in the reactor is
always kept below any proliferation level over entire oper-
ation period.
4.2. Uranium component
From the viewpoint of criticality, a small quantity of
natural UO
2
(5%) in the mixed fuel would guarantee the
presence of suciently high amount of the non-ssile
238
U isotope in the generated uranium fuel in order to
degrade the uranium component to a low to moderate
enriched fuel. This will then absolutely prevent to make a
nuclear bomb from uranium. One can assume that a ura-
nium enrichment grade of <20% will be suciently dena-
turized, although it is doubtful whether even higher
uranium enrichment grades up to <50 could ever be used
to build a useful nuclear warhead. Fig. 17 shows total ura-
nium enrichment grade in the fuel over reactor operation,
which can be dened as the (
233
U+
235
U)/(total uranium
isotopes) ratio in the fuel containing 5% UO
2
in the initial
charge. It begins to increase along with accumulation of the
233
U quantity. One can see clearly that the uranium enrich-
ment grade remains just below 20% over the entire reactor
operation and will not be suitable for military applications
so that the insertion of natural 5% UO
2
to the mixed fuel
will be sucient to denaturize the uranium component of
the mixed fuel.
With natural uranium in the mixture, some minor quan-
tities of new
239
Pu will be generated. However, Figs. 10 and
12 depict that there are suciently high
240
Pu and
242
Pu
isotopes present in plutonium also for this fuel to prevent
any military application.
On the other hand, even in the case if no natural UO
2
is
contained in the fuel, there will be always some
232
U pro-
duction in the uranium fuel as an inevitable byproduct dur-
ing the production cycle of
233
U. Then
232
U, even at minor
quantities, becomes already a suciently high deterrent
factor in handling and diversion due to the hard c-ray
emission, as mentioned in the introduction. In fact, alone
0 2 4 6 8 10
0
2
4
6
8
10
12
14
16
18
20
22
Operation time (years)
(
2
3
3
U
+
2
3
5
U
/
T
o
t
a
l

U
r
a
n
i
u
m

I
s
o
t
o
p
e
s
)

R
a
t
i
o

%

Fig. 17. (
233
U+
235
U/total uranium isotopes) ratio in the fuel bundle with
1st row: 81% ThO
2
+ 14% MA + 5% UO
2
; 2nd row: 85% ThO
2
+ 10%
MA + 5% UO
2
; 3rd row: 90% ThO
2
+ 5% MA + 5% UO
2
; 4th row: 93%
ThO
2
+ 2% MA + 5% UO
2
. (Solid line: in the central fuel row; dashed
line: in the peripheral fuel row.)
S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703 701
the presence of
232
U in the fuel can be considered qualied
for safeguarding. It can be noted that denaturizing of
233
U
through natural uranium will cause a series of concessions,
such as a decrease of the fuel cycle eectiveness, an increase
of parasitic neutron capture and ssile material consump-
tion, and nally a certain degree of decrease in the multipli-
cation factor k
1
, as one can see clearly in Fig. 5 where the
broken lines denote lower k
1
and burn-up values for the
mixed fuel with denaturized
233
U.
In summary, the fuel is denaturized at multiple levels,
viz., through the hard c-ray emission of
232
U, through
non-ssile character of the
238
U,
240
Pu and
242
Pu isotopes,
and through high spontaneous ssion rate in the
240
Pu and
242
Pu isotopes.
5. Conclusions
Utilization of a mixed fuel in a CANDU reactor, con-
sisting of ThO
2
/MAO
2
components with variable frac-
tional composition will allow operating the reactor over
unprecedented long periods without fuel renewal. An
appropriate variation of the fractions of ThO
2
/MAO
2
in
the fuel bundle would lead power attening and almost
uniform fuel utilization. Very high burn-up levels
(200,000 MW D/MT) could be attained for a given fuel
mass, which would reduce drastically the fuel fabrication
costs as well as the nuclear waste mass per unit energy out-
put, if fuel rods could be fabricated to endure such high
burn-up levels. The most obvious advantages can be sum-
marized, as follows:
1. Incineration of the most harmful nuclear waste actini-
des, to a great degree.
2. Higher nuclear fuel exploitation. Utilization of a class of
nuclear waste as useful fuel will lead to negative fuel
cost, which will be a unique feature in the energy mar-
ket. Conventional reactor operators will pay the cus-
tomer for acquiring nuclear waste actinides.
3. Commercial utilization of thorium as nuclear fuel. After
2 years, the reactor will begin practically to operate as
a thorium burner.
4. Signicant reduction of the fuel rod fabrication costs.
5. Drastic reduction of nuclear fuel reprocessing per unit of
total energy production.
6. Plutonium component in the fuel remains always non-
prolic.
7. Addition of few % (5%) of natural UO
2
to the mixed
fuel will provide full inherent safeguarding also for the
generated uranium over the entire plant operation time.
8. Finally, drastic reduction of the nuclear waste material
per unit energy output for nal waste disposal through
extended burn-up.
References
Altnok, T., 1993. Neutronic analysis of CANDU reactors operating with
rejuvenated spent fuel in catalysed fusion reactors, PhD Thesis in
Turkish, Gazi University, Institute of Science and Technology,
Ankara, Turkey.
Alkan, M., 2003. Reutilization of nuclear fuel wastes, PhD Thesis in
Turkish, Gazi University, Institute of Science and Technology,
Ankara, Turkey.
ANL-5800, 1963. Reactor Physics Constants, second ed. United States
Atomic Energy Commission.
Boczar, P.G., Chan, P.S.W., Dyck, G.R., Ellis, R.J., Jones, R.T., Sullivan,
J.D., Taylor, P., 2002a. Thorium fuel-cycle studies for CANDU
reactors, thorium fuel utilization: options and trends. In: Proceedings
of Three IEAE Meetings Held in Vienna in 1997, 1998 and 1999,
IEAE-TECDOC-1319, International Atomic Energy Agency, Vienna,
Austria, pp. 2541.
Boczar, P.G., Dyck, G.R., Chan, P.S.W., Buss, D.B., 2002b. Recent
advances in thorium fuel cycles for CANDU reactors, thorium
fuel utilization: options and trends. In: Proceedings of Three
IEAE Meetings Held in Vienna in 1997, 1998 and 1999, IEAE-
TECDOC-1319, International Atomic Energy Agency, Vienna,
Austria, pp. 104120.
Critoph, E., 1976. Prospects for self-sucient equilibrium thorium cycles
in CANDU reactors, Report AECL-5501, Atomic Energy of Canada
Ltd., Canada.
Choi, H.B., Rhee, B.W., Park, H.S., 1997. Physics study on direct use of
spent pressurized water reactor fuel in CANDU (DUPIC). Nuclear
Science and Engineering 126, 80.
IAEA., 1979. Status and prospects of thermal breeders and their eect on
fuel utilization, Technical Report Series No. 195, International Atomic
Energy Agency, Vienna, Austria.
Jordan W.C., Bowman, S.M., Hollenbach, D.F., 2001. Scale Cross-Section
Libraries, NUREG/CR-0200, Revision7, 3, sectionM4, ORNL/NUREG/
CSD-2/V3/R7, Oak Ridge National Laboratory, Oak Ridge, TN, USA.
Jagannathan, V., Usha, P., Karthikeyan, R., Ganesan, S., Jain, R.P.,
Kamat, S.U., 2001. ATBR-A thorium breeder reactor concept for
early induction of thorium in an enriched uranium reactor. Nuclear
Technology 133, 131.
Kumar, A., Sahin, S., 1983. Investigation of neutronics of some (D,T)
driven experimental thorium hybrid blankets with actinide multipliers.
Journal of Fusion Energy 3, 185197.
Kumar, A., Sahin, S., 1984. (Deuteriumdeuterium) Driven experimental
hybrid blankets and their neutronic analysis. Fusion Technology 6,
225239.
Lamarsh, J.R., Baratta, A.J., 2001. Introduction to Nuclear Engineering,
third ed. Prentice Hall, Upper Saddle River, NJ 07458.
Loewen, E.P., Wilson, R.D., Hohorst, J.K., Kumar, A.S., 2001.
Preliminary frapcon-3th steady-state fuel analysis of ThO
2
and UO
2
fuel mixtures. Nuclear Technology 136, 261277.
Leggertt, R.D., Omberg, R.P., 1987. Mixed oxide fuel development. In:
Proceedings of the International Conference on Fast Breeder Systems:
Experience Gained and Path to Economical Power Generation,
Richland, Washington, September 1317, American Nuclear Society.
Manson, B., Pigford, T.H., Levi, H.W., 1981. Nuclear Chemical
Engineering. McGraw-Hill, New York.
Ma, B.M., 1983. Nuclear Reactor Materials and Applications. Van
Nostrand Reinhold Company Limited, Molly Millars Lane, Woking-
ham, Berkshire, England.
Meyer, W., Loyalka, S.K., Nelson, W.E., Williams, R.W., 1977. The
homemade nuclear bomb syndrome. Nuclear Safety 18, 427.
Petrie, L.M., Fox, P.B., Lucius, K., 2004a. Standard Composition
Library, NUREG/CR-0200, Revision 6, 3, Section M8, ORNL/
NUREG/CSD-2/Volume III/Revision 7, Oak Ridge National Labo-
ratory, Oak Ridge, TN, USA.
Petrie, L.M., 2004b. SCALE5-Scale System Driver, NUREG/CR-0200,
Revision 7, Volume III, Section M1, ORNL/NUREG/CSD-2/V3/R7,
Oak Ridge National Laboratory.
Rubbia, C., 1998. Neutrons in highly diusive transparent medium: an
eective neutron storage device. In: Proceedings of The Ninth
International Conference on Emerging Nuclear Energy Systems,
Keynote, Tel-Aviv, Israel (June 28July 2, 1998), p. 4.
702 S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703
Ronen, Y., Aboudy, M., Regev, D., 1998. A novel method for energy
production using
242m
Am as a nuclear fuel. In: Proceedings of The
Ninth International Conference on Emerging Nuclear Energy Systems,
Tel-Aviv, Israel (June 28July 2, 1998), pp. 5966.
Sahin, S., Kumar, A., 1984a. Fast hybrid thermionic blankets with
actinide waste fuel. Nuclear Technology/Fusion 5, 374381.
Sahin, S., Kumar, A., 1984b. Neutronic analysis of (deuteriumtritium)
driven experimental hybrid blankets. Fusion Technology 6, 97108.
Sahin, S., Calinon, R., 1985. Criticality of curium assemblies. Atomkern-
energie/Kerntechnik 46, 4549.
Sahin, S., Al-Kusayer, T., 1986.
244
Cm as multiplier and breeder in a ThO
2
hybrid blanket driven by a (deuteriumtritium) source. Fusion
Technology 10, 12971302.
Sahin, S., Yapc, H., 1989. Investigation of the neutronic potential of
moderated and fast (D,T) hybrid blankets for rejuvenation of
CANDU spent fuel. Fusion Technology 16, 331345.
Sahin, S., 1989. Power attening in a hybrid blanket using nuclear waste
actinides. Kerntechnik 53, 285290.
Sahin, S., Baltacoglu, E., Yapc, H., 1991. Potential of a catalyzed fusion
driven hybrid reactor for the regeneration of CANDU spent fuel.
Fusion Technology 20, 2639.
Sahin, S., Sahin, H.M., Alkan, M., Yldz, K., 2004a. An assessment of
thorium and spent LWR-fuel utilization potential in CANDU
reactors. Energy Conversion and Management 45, 10671085.
Sahin, S., Yldz, K., Acr, A., 2004b. Power attening in the fuel bundle of
a CANDU reactor. Nuclear Engineering and Design 232, 718.
Sahin, S., Yldz, K., Sahin, H.M., Sahin, N., Acr, A., 2006. Increased fuel
burn up in a CANDU thorium reactor using weapon grade plutonium.
Nuclear Engineering and Design 236 (17), 17801790.
Waltar, A.E., Deitrich, L.W., 1988. Status of research on key LMR safety
issues. Nuclear Safety 29, 125.
Woodhead, L.W., Ingolfsrud, L.J., 1975. Performance of Canadian
Commercial Nuclear Units and Heavy Water Plants. In: Paper
Presented at European Nuclear Conference, 21 April 1975, Paris,
France. Summary in Transaction American Nuclear Society, vol. 20, p.
160.
S. Sahin et al. / Annals of Nuclear Energy 35 (2008) 690703 703