This repository has been archived by the owner on Aug 21, 2019. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 4
/
Copy pathconclusion.tex
105 lines (95 loc) · 6.67 KB
/
conclusion.tex
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
\section{Discussion and conclusions}
This work introduces the open source \gls{MSR} simulation package SaltProc.
SaltProc expands the capability of SERPENT2, the continuous-energy Monte Carlo
code to include online reprocessing modeling capabilities
\cite{rykhlevskii_arfc/saltproc:_2018}. Benefits of SaltProc include
generic geometry modeling, multi-flow capabilities, time-dependent feed and
removal rates, and the ability to specify removal efficiency. The main goal of
this work has
been to demonstrate SaltProc's capability to find the equilibrium fuel salt
composition (where equilibrium is defined as when the number densities of major
isotopes vary by less than 1\% over several years). A secondary goal has been to
compare predicted operational and safety parameters (e.g., neutron energy
spectrum, power and breeding distribution, temperature coefficients of
reactivity) of the \gls{MSBR} at startup and equilibrium state. A tertiary goal
has been to demonstrate benefits of continuous fission products removal for
thermal \gls{MSR} design.
To achieve these goals, a full-core high-fidelity benchmark model of the \gls{MSBR}
was implemented in SERPENT2. The full-core model was used instead of the
the simplified single-cell model \cite{betzler_molten_2017,
rykhlevskii_online_2017, betzler_fuel_2018} to precisely describe the
two-region \gls{MSBR} concept design sufficiently to accurately represent
breeding in the outer core zone. When running depletion calculations, the most
important fission products and $^{233}$Pa are removed while fertile and fissile
materials are added to the fuel salt every 3 days. Meanwhile, the removal
interval for the rare earths, volatile fluorides, and seminoble metals was greater
than month a (50 days), which caused effective multiplication factor fluctuation.
\subsection{Equilibrium state search}
The results of this study indicate that the effective multiplication factor
slowly decreases from 1.075 and reaches 1.02 at equilibrium after approximately
6 years of operation. At the same time, the concentrations of $^{233}$U, $^{232}$Th,
$^{233}$Pa, $^{232}$Pa stabilized after approximately 2500 days of operation.
Particularly, $^{233}$U number density equilibrates\footnote{fluctuates less
than 0.8\%} after 16 years of operation. Consequently, the core reaches the quasi-equilibrium state after 16 years of operation. However, a wide variety of nuclides,
including fissile isotopes (e.g. $^{233}$U, $^{239}$Pu) and non-fissile strong
absorbers (e.g. $^{234}$U), continue accumulating in the core.
\subsection{Spectral shift}
We also found that the neutron energy spectrum grew harder as the core
approaches equilibrium because significant heavy fission products accumulated in
the \gls{MSBR} core. Moreover, the neutron energy spectrum in the central core
region is much softer than in the outer core region due to lower
moderator-to-fuel ratio in the outer zone, and this distribution remains stable
during reactor operation. Finally, the epithermal or thermal spectrum is needed
to effectively breed $^{233}$U from $^{232}$Th because radiative capture cross
section of thorium-232 monotonically decreases from $10^{-10}$ MeV to $10^{-5}$
MeV. A harder spectrum in the outer core region tends to significantly increase
resonance absorption in thorium and decrease the absorptions in fissile and
structural materials.
The spatial power distribution in the \gls{MSBR} shows that 98\% of the fission
power is generated in central zone I, and neutron energy spectral shift did not
cause any notable changes in a power distribution. The spatial distribution of
neutron capture reaction rate for fertile $^{232}$Th, corresponding to breeding in
the core, confirms that most of the breeding occurs in an outer,
undermoderated, region of the \gls{MSBR} core. Finally, the average $^{232}$Th
refill rate throughout 60 years of operation is approximately 2.40 kg/day or
100 g/GWh$_e$.
We compared the safety parameters for the initial fuel loading and
equilibrium compositions using the SERPENT2 Monte Carlo code.
The total temperature coefficient
is large and negative at startup and equilibrium but the magnitude decreases
throughout reactor operation from $-3.10$ to $-0.94$ pcm/K as the spectrum
hardens. The moderator
temperature coefficient is positive and also decreases during fuel depletion.
The reactivity control system efficiency analysis showed that the safety rod integral
worth decreases by approximately 16.2\% over 16 years of operation, while
graphite rod integral worth remains constant. Therefore, neutron energy
spectrum hardening during fuel salt depletion has an undesirable impact on
\gls{MSBR} stability and controllability, and should be taken into
consideration in further analysis of transient accident scenarios.
\subsection{Benefits of fission product removal}
The \gls{MSBR} core performance benefits from the removal of volatile gases,
noble metals, and rare earths from the fuel salt.
Moreover, immediate removal of volatile gases (e.g., xenon) and noble metals
increased reactivity by approximately 7500 pcm over a 10-year
timeframe. In contrast, the effect of relatively slower removal of rare earth
elements (every 50 days cycle instead of 3 days) has less impact (5500 pcm) on
the core reactivity after 10 years of operation. An additional study
is needed to establish neutronic and economic tradeoffs of removing each element.
\subsection{Future work}
SaltProc-SERPENT coupled simulation efforts could progress in a
number of different directions. First optimization of reprocessing parameters (e.g. time step, feeding rate,
protactinium removal rate) could establish the best fuel utilization, breeding
ratio, or safety characteristics for various designs. This might be performed with a parameter sweeping
outer loop which would change an input parameter by a small increment, run the
simulation and analyze output to determine optimal configuration. Alternatively,
the existing RAVEN optimization framework \cite{alfonsi_raven_2013} might be
employed for such optimization studies.
Only the batch-wise online reprocessing approach has been treated in this
work. However, the SERPENT2 Monte Carlo code was extended for
continuous online fuel reprocessing simulation \cite{aufiero_extended_2013}.
This extension must be verified against existing SaltProc/SERPENT or
ChemTriton/SCALE packages, and could be employed for immediate removal of
fission product gases (e.g., Xe, Kr) which have a strong negative impact on
core lifetime and breeding efficiency. Finally, using the built-in SERPENT2
Monte Carlo code online reprocessing \& refueling material burnup routine would
significantly speed up computer-intensive full-core depletion simulations.