Skip to content

Commit

Permalink
Added used fuel.
Browse files Browse the repository at this point in the history
  • Loading branch information
Anthony Scopatz committed Jul 30, 2011
1 parent e25970b commit 9820b55
Showing 1 changed file with 27 additions and 27 deletions.
54 changes: 27 additions & 27 deletions one_group_method/one_group_paper.tex
Expand Up @@ -14,7 +14,7 @@ \section{Introduction}
fidelity by fully decoupling the fuel cycle simulation from physics-based reactor burnup calculations
\cite{Jacobson2009}.

Fuel cycle problems revolve around perturbing input paramter values and measuring cooresponding component or
Fuel cycle problems revolve around perturbing input parameter values and measuring corresponding component or
systemic changes which result. For instance, initial fresh fuel compositions in a reactor may be varied
to attain a specific
burnup. Or repository impact may be studied by varying the cooling time before emplacement. Even
Expand All @@ -41,25 +41,25 @@ \section{Introduction}
the burnup, and thus are calculated after other parameters. This tool may then be embedded within
a full fuel cycle model. To demonstrate how such a coupled framework functions, it is applied to
two fuel cycle concepts. Because the neutron production and destruction rates are integral
quatities over all energy, this reactor model is known as the one-group method.
quantities over all energy, this reactor model is known as the one-group method.

The first of the two fuel cycles incorporates a limited uranium recycle strategy. Uranium from a
standard pressurized water reactor (PWR) is reprocessed and blended with enriched natural uranium
(NU) before going into the second core to be burned again. After this burn, it is noteworthy that
further recycling via blending is still possible. In other words, this fuel cycle could be fully
closed with respect to the uranium stream, avoiding the need to dispose of the reprocessed uranium
that comprises over 90\% of the mass of spent PWR fuel on an initial heavy metal (IHM) basis.
that comprises over 90\% of the mass of used PWR fuel on an initial heavy metal (IHM) basis.
This is known as a recyclable uranium (RU) blend and burn fuel cycle since the uranium is reused
in PWRs independently of other actinides. RU blend and burn is an attractive option since it has
the potential to eventually do a near complete burn of all uranium in LWR spent fuel if used in a
fully closed cycle.
the potential to eventually do a near complete burn of all uranium in light water reactor (LWR)
used fuel (UF) if used in a fully closed cycle.

The second fuel cycle that utilizes the burnup tool developed here demonstrates how the tool may
be used to improve the fidelity of the fuel cycle material balance calculations associated with
one of the Advanced Fuel Cycle Research and Development (AFC R\&D) base-case proposals \cite{PC-000555}.
This case is a closed fuel cycle that uses a fast burner reactor (FR) with a conversion ratio of 0.5.
Actinides are continually recycled in the FR while fission products (FP) are removed and
disposed. The reprocessing facility that is built for fast reactor spent fuel has a separation
disposed. The reprocessing facility that is built for fast reactor used fuel has a separation
efficiency of 0.99 for all actinides. This is a meaningful option to explore since fast reactors
are considered to be a key component of long term nuclear energy strategies that close the fuel cycle.

Expand All @@ -70,7 +70,7 @@ \section{Introduction}

As will be seen, many fuel cycle parameters may be calculated from the resulting data. Moreover
due to the dynamic nature of this method, these parameters will be more accurate
than if a static recipe or simple linearization had been used. A static recipe would not be able
than if a static recipe or simple linearizion had been used. A static recipe would not be able
to mix recyclable and low enriched uranium (LEU) streams in a way that respects the dependence of
neutronic characteristics on feed composition, an issue that is inherent to the RU fuel cycle.
Moreover, models that use static recipes may fail to accurately capture the evolution of material
Expand Down Expand Up @@ -412,7 +412,7 @@ \subsubsection{Multiple Batch Cores}
is the fluence at discharge $F\mbox{d}$. After the fluence for each of the burnups has been calculated,
the multiplication factor of the system needs to be known.

To achieve this for a multibatch core, a flux-weighted batch averaging procedure is followed.
To achieve this for a multi-batch core, a flux-weighted batch averaging procedure is followed.
The multiplication factor at each fluence is determined by the $P(F)/D(F)$ ratio. Denote the batch
number that a parameter is associated with by the subscript ``b''. In Figure \ref{1g_fig06},
the three $k_b$s correspond to the three fluences chosen via the three burnup
Expand Down Expand Up @@ -470,7 +470,7 @@ \subsubsection{Multiple Batch Cores}
M_j(F) = \sum_i m_i^F \cdot T_{ij}(F)
\end{equation}
Plugging in the fluence at discharge
$F\mbox{d}$ into equation \ref{1g_trans_Mj} will yield the mass of isotope $j$ in the spent fuel per
$F\mbox{d}$ into equation \ref{1g_trans_Mj} will yield the mass of isotope $j$ in the used fuel per
kilogram of IHM. Performing this operation for all $j$ isotopes of interest will yield the isotopic
output vector for the core. Since fission product masses were included in the transformation matrices,
the fission products at discharge are also known.
Expand All @@ -495,7 +495,7 @@ \subsubsection{Multiple Batch Cores}
In conclusion, the above is an algorithm for finding the maximum discharge burnup and its discharge
composition given only a fuel form represented by $m_i^F$, the pre-generated burnup parameters ($p_i(F)$,
$d_i(F)$, $\mbox{BU}_i(F)$, and $T_{ij}(F)$), the number of batches in the core $B$, and the non-leakage
probability $P_{NL}$. Methods for generating these burnup parameters are disscussed in
probability $P_{NL}$. Methods for generating these burnup parameters are discussed in
\S \ref{1g_sec:gen_BU_param}.


Expand Down Expand Up @@ -547,7 +547,7 @@ \subsection{Burnup Parameter Generation}
Distinct sets of libraries were prepared for 0.5 (used in this study) and 0.25 target
conversion ratio cores. Additionally, separate libraries were prepared at each conversion ratio for
fuels with higher and lower minor actinide (MA) content. The higher MA case is representative of the
feed to a FR if Pu from spent UOX is first burned in MOX, while the lower MA library, the one used in
feed to a FR if Pu from used UOX is first burned in MOX, while the lower MA library, the one used in
this study, reflects the TRU content with no MOX burn.


Expand All @@ -564,8 +564,8 @@ \subsubsection{Hydrogen Cross Section Rescaling}
has experienced. Moreover, hydrogen contributes about 10\% to the total destruction rate in the core.
Thus a significant change in the hydrogen cross section could yield an appreciable error in the model as
formulated above. What follows is more of an adjustment to the hydrogen one-group cross section than a
true mutli-energy group response. However, this adjustment provides an easy to calculate model to
reduce the error induced into the neutron destruction rates by non-represntative hydrogen cross sections.
true multi-energy group response. However, this adjustment provides an easy to calculate model to
reduce the error induced into the neutron destruction rates by non-representative hydrogen cross sections.

To account for these spectral variations an f-factor is introduced that is a function of the burnup
$\mbox{BU}(F)$. $f(F)$ is a unitless factor by which all hydrogen destruction rates $d_H(F)$ are
Expand Down Expand Up @@ -628,7 +628,7 @@ \subsection{Uranium Recycle Fuel Cycle}

LWR UF contains a higher enrichment of \nuc{U}{235} than NU; therefore RU blending may be economically
advantageous over the once-through fuel cycle given that reprocessing is already taking place to recover
TRU. The weight percent of \nuc{U}{235} in the legacy spent fuel of the United States is between
TRU. The weight percent of \nuc{U}{235} in the legacy used fuel of the United States is between
0.8-0.9\% \cite{Schneider2007}.
However, LWR UF also has non-negligible amounts of \nuc{U}{236} bred into it, between 0.3-0.5\%.
As \nuc{U}{236} is a neutron poison, the advantage of the extra \nuc{U}{235} in RU is not immediately
Expand Down Expand Up @@ -685,8 +685,8 @@ \subsection{Fast Burner Reactor Fuel Cycle}
input mass streams as well as all dependent isotopics. However, burnup values of 51 MWd/kg for the LWR and
170 MWd/kg for the FR are used unless otherwise stated. Both of these values were taken from the
VISION \cite{Jacobson2009}
specifications for nominal LWR and FR cases. Similarly, spent fuel from both the LWRs and FRs is allowed to
cool for a time before being reprocessed. The time that spent fuel of any sort is allowed to cool also
specifications for nominal LWR and FR cases. Similarly, used fuel from both the LWRs and FRs is allowed to
cool for a time before being reprocessed. The time that used fuel of any sort is allowed to cool also
affects the isotopics and mass stream material balances. The cooling time is again a parameter
that may be specified within the fuel cycle model for which nominal values were taken here.

Expand Down Expand Up @@ -756,12 +756,12 @@ \section{Fuel Cycle Model Benchmarking \& Results}
amount of time that UF spends in storage cooling, and the fuel cell specifications that are used. Values
for these parameters that were used in this study are presented in Table \ref{1g_table1}. Knowing this information, the
fuel cycle model wraps around the reactor burnup model and invokes it when input streams need to be mixed
or spent fuel compositions must be found. The fuel cycle model then sends the results of the burnup model
or used fuel compositions must be found. The fuel cycle model then sends the results of the burnup model
to the next fuel cycle component (as seen in Figures \ref{1g_fig08} \& \ref{1g_fig09}). Thus the fuel cycle
model coupled with the burnup model developed here produce unique, physics-based values for the material
balances for one or many recycle passes. This represents an advantage over recipe based
simulations which are not able to alter any of the input parameters at will. In this model all inputs are
allowed to change simeltaneously.
allowed to change simultaneously.

\begin{table}[htbp]
\begin{center}
Expand Down Expand Up @@ -881,12 +881,12 @@ \section{Fuel Cycle Model Benchmarking \& Results}
is smaller because the concentration of the much more abundant parent, \nuc{Pu}{241}, is reasonably accurate.

It is also necessary to benchmark the feed stream blending and discharge burnup calculations. This is best
achieved in the context of a comparitive fuel cycle material balance. Material balances of this type
achieved in the context of a comparative fuel cycle material balance. Material balances of this type
generated using the high-fidelity material balance simulation package COSI \cite{Boucher2006} are
available in a number
of OECD systems studies \cite{OECD2002}. A benchmark of the equilibrium material balance for a PWR/FR fleet
has been carried out. In this benchmark, the methodology described here is used to find the enrichment
of PWR fuel to achieve a specified burnup as well as to derive the equilibrium FR fresh and spent fuel
of PWR fuel to achieve a specified burnup as well as to derive the equilibrium FR fresh and used fuel
compositions through the cycle iteration procedure described above. Specific results benchmarked
included the PWR/FR thermal power split, the isotopic composition of the burned fuel, and the fuel cycle
cost derived from the material balance flowsheets.
Expand Down Expand Up @@ -1087,7 +1087,7 @@ \subsection{The Fast Reactor Fuel Cycle}
\index{The Fast Reactor Fuel Cycle@\emph{The Fast Reactor Fuel Cycle}}
\label{1g_sec:FRFC}
A walkthrough follows of how to apply the burnup model algorithm to the closed fast reactor
system described above. The first time a fast reactor system comes online, there exists no
system described above. The first time a fast reactor system comes on-line, there exists no
FR-U and FR-TRU streams from prior cycles so these mass flows and compositions are set to zero.
This leaves only reprocessed DU and LWR-TRU streams to be mixed in order to obtain the burnup
value specified for the fast reactor. These two parameters reduce to only one
Expand Down Expand Up @@ -1131,13 +1131,13 @@ \subsection{The Fast Reactor Fuel Cycle}
made as to what the relative fractions are. These bound the burnups available and the
bisection method is applied to calculate the DU/LWR-TRU ratio that hits the target burnup.

Now that the target burnup has been reached and the fresh fuel burned, it becomes spent fuel once
Now that the target burnup has been reached and the fresh fuel burned, it becomes used fuel once
more. The FR UF is again stored and cooled. These results are then sent to the reprocessing code
where the separation efficiencies are applied. Fuel cycle parameters that are specific to this
pass are now calculated and stored.

This process may continue indefinitely as there are an infinite number of cycles possible. However,
fuel cycle parameters (such as isotopic concentrations in the fresh and spent fuels, uranium and
fuel cycle parameters (such as isotopic concentrations in the fresh and used fuels, uranium and
transuranic masses) typically come to ``equilibrium'' given enough passes through the system.
Two approaches to determining equilibrium status were considered: the stabilization of certain
tracked isotopes or the convergence of the FR fresh fuel transuranic fraction. This TRU fraction
Expand All @@ -1149,7 +1149,7 @@ \subsection{The Fast Reactor Fuel Cycle}
If the fuel cycle study is concerned with the prevalence of isotopes having, for instance, a strong
effect on repository performance it is wiser to develop a list of important species whose convergence
is to be tracked. Then at the start of each pass after the first, the change of these isotopic
concentrations in the spent fuel is calculated. If all isotopes in the list have converged to
concentrations in the used fuel is calculated. If all isotopes in the list have converged to
within some allotted error (perhaps 1\%), then the fuel cycle as a whole is said to have converged.
In the FR case considered here, \nuc{Pu}{239}, \nuc{Pu}{240}, and \nuc{Pu}{242} dominate the TRU by
mass and converge after only a few cycles. Moreover, \nuc{Pu}{242} is the precursor isotope for
Expand All @@ -1174,11 +1174,11 @@ \subsection{The Fast Reactor Fuel Cycle}

The first set of parameters displayed is the input fractions of the four input mass streams
into the fast reactor: DU, LWR-TRU, FR-U, and FR-TRU. This is shown in Figure \ref{1g_fig18}.
The input mass of fuel into the fast reactor comes in some part from LWR spent fuel (1 kg or less)
The input mass of fuel into the fast reactor comes in some part from LWR used fuel (1 kg or less)
and partially from recycled FR fuel. Note that all streams are normalized to 1 kgIHM for each cycle.
The LWR streams here are the parameters that are explicitly iterated over in the fuel cycle model
in order to achieve the target FR burnup. Note that the first cycle is comprised of only LWR
spent fuel (as discussed above) and then these curves quickly level off. Also recall that
used fuel (as discussed above) and then these curves quickly level off. Also recall that
all FR actinide mass (sans what is decayed away in storage and what is lost in reprocessing)
returns to the FR on the next pass. Since the FR converts roughly 20\% of its initial fuel to
fission products, this explains why the sum of FR-U and FR-TRU levels off at about 0.8.
Expand Down Expand Up @@ -1276,7 +1276,7 @@ \subsection{The Fast Reactor Fuel Cycle}
\end{figure}

Finally, individual graphs may be generated for every isotope that is tracked.
The plots display the total mass of the given isotope in both the fresh (input) and spent (output)
The plots display the total mass of the given isotope in both the fresh (input) and used (output)
fast reactor fuel streams. Note that the output masses are given after the additional cooling time.
Samples for \nuc{Pu}{238}, \nuc{Pu}{239}, \nuc{Pu}{240}, \nuc{Cm}{244}, and \nuc{Cm}{244} are shown
in Figures \ref{1g_fig22}-\ref{1g_fig26} respectively. The plutonium figures serve to show that the
Expand Down

0 comments on commit 9820b55

Please sign in to comment.