Criticality safety of fuel debris, particularly MCCI(Molten-Core-Concrete-Interaction) products, is one of the major safety issues for decommissioning of Fukushima Daiichi Nuclear Power Station. Criticality or subcriticality condition of the fuel debris is still uncertain; its composition, location, neutron moderation, etc. are not yet confirmed.
Trang 1Study of experimental core con figuration of the modified STACY for
measurement of criticality characteristics of fuel debris
Satoshi Gunjia,*, Kotaro Tonoikea, Kazuhiko Izawab, Hiroki Sonob
a Japan Atomic Energy Agency, Nuclear Safety Research Center, Shirakata 2-4, Tokai-mura, Ibaraki, Japan
b Japan Atomic Energy Agency, Department of Fukushima Technology, Shirakata 2-4, Tokai-mura, Ibaraki, Japan
a r t i c l e i n f o
Article history:
Received 26 August 2016
Received in revised form
10 February 2017
Accepted 1 March 2017
Available online 13 April 2017
Keywords:
Fuel debris
MCCI product
Modified STACY
Fukushima Daiichi nuclear power station
Critical experiment
a b s t r a c t
Criticality safety of fuel debris, particularly MCCI(Molten-Core-Concrete-Interaction) products, is one of the major safety issues for decommissioning of Fukushima Daiichi Nuclear Power Station Criticality or subcriticality condition of the fuel debris is still uncertain; its composition, location, neutron moderation, etc are not yet confirmed The effectiveness of neutron poison in cooling water is also uncertain for use
as a criticality control of fuel debris A database of computational models is being built by Japan Atomic Energy Agency (JAEA), covering a wide range of possible conditions of such composition, neutron moderation, etc., to facilitate assessing criticality characteristics once fuel debris samples are taken and their conditions are known The computational models also include uncertainties which are to be clar-ified by critical experiments These experiments are planned and will be conducted by JAEA with the modified STACY(STAtic experiment Critical facilitY) and samples to simulate fuel debris compositions Each of the samples will be cladded by a zircalloy tube whose outer shape is compatible with the fuel rod
of STACY and loaded into an array of the fuel rods This report introduces a study of experimental core configurations to measure the reactivity worth of samples simulating MCCI products Parameters to be varied in the computation models for the experimental series are:
It is concluded that the measurement is feasible in both under- and over-moderated conditions Additionally, the required amount of samples was estimated
© 2017 The Authors Published by Elsevier Ltd This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Uranium dioxide with235U enrichments of 3, 4, and 5 wt.%,
Concrete volume fraction in the samples of 0, 20, 40, 60, and
80%, and
Porosity of the samples filled from 0 to 80% where the sample
void isfilled with water
1 Introduction
Criticality safety is one of the major safety issues for defueling of
the damaged reactors in Fukushima Daiichi Nuclear Power Station
(1F-NPS) (Tonoike et al., 2013) Criticality control method of fuel
debris must be established in the mid- or long-term process of
defueling and decommissioning A significant difference, from the
view point of criticality control, between situations in the 1F-NPS
reactors and the Three Mile Island Unit 2 reactor (TMI-2) is that
cooling water for fuel debris in the 1F-NPS reactors cannot be poisoned continuously as was done in the TMI-2 Because the cooling water is notflowing in a closed loop, the destination of injected water is not known Therefore, it is difficult to manage concentrations of the poison in the cooling water for the purpose of criticality control
It would be necessary that a mitigation-based criticality control method is adopted for decommissioning of 1F-NPS For this pur-pose, it is necessary to get the criticality characteristics of fuel debris However, the actual fuel debris in the reactors has not yet been observed and it is difficult to obtain accurate information on its composition, location, neutron moderation, etc (Tonoike et al.,
2015) This situation leads to large uncertainty in estimation of criticality characteristics, and criticality or subcriticality condition
of the fuel debris Therefore, a database of computational models for possible criticality characteristics of the fuel debris is being built which will help to predict in which condition critical events may occur (Tonoike et al., 2015)
Most of criticality characteristics of fuel debris have not been
* Corresponding author.
E-mail address: gunji.satoshi74@jaea.go.jp (S Gunji).
Contents lists available atScienceDirect Progress in Nuclear Energy
j o u r n a l h o m e p a g e : w w w e l s e v i e r c o m / l o c a t e / p n u c e n e
http://dx.doi.org/10.1016/j.pnucene.2017.03.002
0149-1970/© 2017 The Authors Published by Elsevier Ltd This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ).
Trang 2evaluated before the accident of 1F-NPS, especially, that of
molten-core-concrete-interaction (MCCI) product Molten core might drop
from the pressure vessels of the reactors on the concretefloors in
the containment vessels, where MCCI products might be produced
MCCI product has a small neutron absorption cross-section, and
may be porous and contain water when submerged In fact, it has
been confirmed that the containment vessel floors are submerged
(Status of Fukushima Daiichi Nuclear Power Station (2015)) In past
studies, criticality characteristics of MCCI products had been
eval-uated only by computations Some of them suggest small critical
mass and high boron concentration in cooling water that guarantee
subcriticality (Izawa et al., 2012) Detail study of criticality
charac-teristics of MCCI products, therefore, should be conducted,
including criticality experiments, for establishment of criticality
control or criticality risk assessments
Critical experiments are being planned to validate such
high-accuracy computations to support criticality safety or criticality
risk evaluation of defueling in 1F-NPS that will change a volume
ratio of fuel debris and water It will be conducted at the modified
Static Experiment Critical Facility (STACY) with samples
simu-lating fuel debris compositions (Tonoike et al., 2015) In this paper,
experimental core configurations with samples of MCCI products
were considered based on recent knowledge of criticality
char-acteristics of fuel debris The amount of samples to be prepared
was determined as function of their reactivity worth and the
dif-ferences of the core critical water heights This is important
because there are limitations on the insertion reactivity and the
critical water height in particular specifications of the critical
facility
2 Analysis and experimental conditions
2.1 Submerged MCCI product
The moderation condition of critical experiments should be
varied, however, more widely because hydrogen in the concrete
would contribute to neutron moderation and because the amount
of hydrogen in actual MCCI products is still unknown A series of
analyses of criticality characteristics of MCCI products was shown
in Ref 5, where infinite multiplication factors of MCCI products
with235U enrichments of 3, 4, and 5 wt.% were computed in
ho-mogeneous and heterogeneous conditions The results indicate that
optimum moderation conditions of those MCCI products would be
at the volume ratios of moderator to fuel (Vm/Vf) of 0.2e4 Vm/Vfis
expressed by following equation(1), this is specifically the
modi-fied STACY design based This value means ratio of water volume to
fuel pellet volume in an active core Therefore, it does not consider
water contents in samples
Vm
Vf¼Volume of moderator waterVolume of UO
2
(1)
2.2 Outline of the modified STACY
The modification of the STACY is now underway at Japan Atomic
Energy Agency (JAEA) in order to accumulate fundamental
exper-imental data relating to the criticality control for fuel debris
handling in 1F-NPS The modified STACY is designed to be a
tank-type light-water-moderated critical assembly, whose first
criti-cality is expected in FY2018 (Sono et al., 2015; Miyoshi et al., 2015)
An overview of the modified STACY is shown inFig 1 Each fuel rod
will consist of UO2pellets with a diameter of 8.2 mm and a235U
enrichment of 5 wt.%, and a zircalloy cladding with an
outer-diameter of 9.5 mm The stack height of the pellets will be
1420 mm
The modified STACY will be operated by means of filling the tank with water, which works as neutron moderator and reflector, from the bottom of the core tank Reactivity will be adjusted by adjusting the water height Vm/Vfof the core will be varied by changing fuel rod interval For example, Vm/Vfis 1.2 when a fuel rod interval in a square lattice is 11.5 mm (Izawa et al., 2015; Sakon et al., 2015), which was selected as the experimental core configuration in this study
2.3 Experimental core configurations Two experimental core configurations were studied by using the MCNP 5.1 code system (X-5 Monte Carlo Team, 2003; Brown et al.,
2009) and the nuclear data library JENDL-4.0 (Shibata et al., 2011)
In order to have statistical error of less than 0.015 %Dk (~2 cent1),
5 107effective neutron histories was used for each calculation The effective critical water heights of the two core configurations were approximately 1000 mm
One configuration was designed to have a hard neutron spec-trum and an under-moderation configuration, which is shown in Fig 2 The configuration consists of 701 fuel rods arrayed in a uniform square lattice with the interval of 11.5 mm Its critical water height is estimated to be 990 mm
The other configuration has a soft neutron spectrum and an over-moderation configuration The configuration consists of 400 fuel rods in total and is divided into two regions as shown inFig 3 The“driver region” is an array of fuel rods with the same interval of 11.5 mm that surrounds the“test region” The “test region” consists
of 85 fuel rods arrayed more sparsely with 84 positions left vacant (filled with water), an effective interval of 16.3 mm, and Vm/Vfof about 3.7 Its critical water height is estimated to be 935 mm The neutron spectra at the center of each experimental core
Fig 1 Concept of modified STACY.
1 The effective delayed neutron fraction was calculated by SRAC/TWODANT;
b ¼ 0.0075.
S Gunji et al / Progress in Nuclear Energy 101 (2017) 321e328 322
Trang 3condition are shown inFig 4 The thermal neutronflux of the
over-moderation configuration was about 3 times as large as that of the
under-moderation configuration
These are considered the base arrays with only fuel rods in the
array and a water height of approximately 1000 mm To determine the reactivity of the samples, each sample was modeled in the base array and the reactivity worth was obtained from the change in keff Effective multiplication factors (keff) were 1.00035± 0.00011 and 1.00357± 0.00011, respectively, for the under-moderation and the over-moderation configurations
The relation between the water height and keffof each con fig-uration are shown inFig 5 The reactivity worth per water level change of each configuration was estimated to be about 0.62 ¢/mm
or 0.63 ¢/mm at a water height of 1000 mm The accuracy of the water height gauge of the modified STACY will be ±0.2 mm Therefore, the reactivity worth derived from a water height dif-ference will have an accuracy of±0.1 ¢, which will be acceptable as the experimental precision Thus, the reactivity worth of pseudo fuel debris samples in each experimental configuration should be greater than 0.3 ¢ to be distinguished from zero
Geometrical buckling of each experimental configuration should be minimally changed The limitation of the changes was determined to be less than 1% This limitation was set to determine the experimental limit, and there is no reason based on quantitative consideration Under these experimental conditions, the change of the water heights should be less than 100 mm
Therefore, favorable change of water height in reactivity worth measurement should be from 0.5 to 100 mm, which corresponds to reactivity worth of pseudo fuel debris samples from 0.3 to 62 ¢
Fig 2 “Under-moderation” experimental core configurations in square lattice of the
modified STACY.
Fig 3 “Over-moderation” experimental core configurations in square lattice of the
modified STACY.
Fig 4 Neutron spectra at the center of each core configuration.
Fig 5 Relations the effective water height and k eff of each core.
Trang 43 Reactivity worth of samples
3.1 Samples for reactivity worth measurements
The samples of pseudo fuel debris simulating MCCI products
were modeled using several compositions which were made of
uranium dioxide with235U enrichments of 3, 4, and 5 wt.% fuel; and
a concrete A list of sample types in this study is shown inTable 1
The composition of the concrete in this study is shown inTable 2
and its density is 2.3 g/cm3 Concrete volume fractions in the
samples were 0, 20, 40, 60, and 80% Additionally, porosities of
these samples were varied from 0 to 80%, which werefilled with
water Concrete volume fraction is expressed by following equation
(2), and porosity is expressed by following equation (3),
respectively;
Concrete volume fractionð%Þ ¼Volume of MCCI productVolume of concrete
100
(2)
100
(3) Ideal sample loading conditions, based on reactivity worths for
each loading, were evaluated forfive arrays with 1, 5, 5, 9, and 13
MCCI products samples loaded in the patterns illustrated inFig 6
The in the test region of the under-moderation configuration fuel
rods were replaced with sample rods For the over-moderation
configuration, the samples were inserted into vacant positions in the test region
Values of keffwere computed for arrays of fuel rods and the samples Reactivity worths were estimated by comparing the keff values and those of the base arrays The estimated relative reac-tivity worth of the pseudo fuel debris whose235U enrichment is
4 wt.% are shown in the following sub-sections
3.2 Relative reactivity worths by changing the concrete volume fraction
Fig 7andFig 8show the computation results of relative reac-tivity worth dependency of the concrete volume fraction in each configuration They are the results of the samples based on using the235U enrichment of 4 wt.% fuels and their porosities are 0% Fig 7shows that the changing of the concrete volume fraction has a big impact on the reactivity worths in the under-moderation configuration The reactivity worth of the samples with no con-crete, shown in thefigure, was negative because the235U enrich-ment of 4 wt.% was lower than the 5 wt.% enrichenrich-ment of fuel rods The absolute value was, however, small and the worth turned to positive if the concrete volume fraction is beyond 40% There was a tendency that reactivity worths increase into the positive according
to increase the concrete volume fraction It is considered that the water in the concrete contributed to moderate of neutron in these configuration
Fig 8 shows that for the over-moderation configuration, the reactivity worths are negative for all of the patterns because the samples excluded the moderator water There was a tendency that reactivity worths increase into the negative according to the con-crete volume fraction increase The positive reactivity worths should have been inserted, because the moderation conditions of
Table 2
The composition of the concrete in this study.
Element Number density [atoms/b cm] Element Number density [atoms/b cm] Element Number density [atoms/b cm]
Table 1
A list of the reactivity worth samples and their specifications.
MCCI product with zircalloy cladding 1420 mm 1, 5a, 5b, 9, and 13
Parameters;
235 U enrichment (3, 4, and 5 wt.%)
Concrete volume fraction (0, 20, 40, 60, and 80%)
Porosity (0, 20, 40, 60, and 80%)
S Gunji et al / Progress in Nuclear Energy 101 (2017) 321e328 324
Trang 5this case were close to the suitable moderation condition by
removing water However, contrary to expectations, the reactivity
worths remained negative Maybe, these results show that dry
condition is nearly optimum moderation condition Furthermore,
in this configuration, it is also concluded that loading of up to 5
samples will be suitable to measure their reactivity worth because
change in the critical water height is too much for a larger number
of samples
Table 3andTable 4summarize the reactivity worth per each sample in each moderator condition In addition, samples of 100% concrete and water with zircalloy cladding are shown as references
In the under-moderation configuration, positive reactivity worths were inserted by increase of the concrete volume fraction, and more reactivity worth was inserted by insertion of the 100% concrete sample.Table 3shows the effect of the replacement of the fuel rod of the water sample is approximately 12 ¢, and that of the 100% concrete sample is approximately 3.3 ¢ in each insertion pattern And 4 wt% fuel rods (see Concrete volume 0%) have negative reactivity worths in each insertion pattern Moreover, the maximum reactivity worth was inserted by swapping the fuel rods for water holes In this configuration, the reactivity worths per rod were almost the same for sample types in each insertion pattern
On the other hand, in the over-moderation configuration, small negative reactivity worths were inserted by increase of the concrete volume fraction.Table 4shows the effect of the replacement of the fuel rod of the water sample is approximately2.5 ¢, and that of the 100% concrete sample is approximately 10¢ in each insertion pattern Sensitivities of both the insertion pattern and the concrete volume fraction for the reactivity worths were small
3.3 Relative reactivity worths by changing the porosity Fig 9 and Fig 10 show the computation results of relative reactivity worth depend on changing the porosities of the sample in
“Pattern 5a” for several concrete volume fraction in each configu-ration They are the results of the samples based on using the235U enrichment of 4 wt.% fuels The relative reactivity worths in each configuration has proportional relations to porosities
Fig 9shows that the increasing of the porosities have moder-ation effects, therefore, the samples has a positive reactivity worth About 40 ¢ positive reactivities occurred by the porosity increased from 0 to 80% in the under-moderation configuration The effect of porosity changing is dominant than that of the concrete volume fraction changing, because the amount of hydrogen differ by one order of magnitude between two parameters
Fig 10shows that about 25 ¢ positive reactivities occurred by the porosity increased from 0 to 80% in the over-moderation configuration This results show that this experimental core configuration is not enough “over-moderation”, because the posi-tive reactivity worths were inserted by increasing of water content 3.4 Additional analysis for over-moderation core configuration
In section 3.3, it has turned out that“over-moderation” core configuration was not have enough moderation ability Therefore, a new over-moderation experimental core configuration which
Fig 8 Relative reactivity of the pseudo fuel debris samples by changing the concrete
volume fraction in the over-moderation configuration.
Table 3
The reactivity worth per sample rod in the under-moderation configuration (Unit: cent/rod).
a All sample has zircalloy cladding.
b
Fig 7 Relative reactivity of the pseudo fuel debris samples by changing the concrete
volume fraction in the under-moderation configuration.
Trang 6shifted array of fuel rods was considered This configuration is
shown inFig 11 In this configuration, local Vm/Vf(¼3.7) at the test
region do not change by insertion of the reactivity worth samples
The relative reactivity worths of the samples by changing the
concrete volume fraction and the porosities are shown inFigs 12
and 13, respectively As considered in section3.3, some features
of the over-moderation were seen in this core configuration The
increase of moderator water by increasing of the concrete volume
Fig 9 Relative reactivity of the samples in the under-moderated “Pattern 5a”
configuration.
Fig 10 Relative reactivity of the samples in the over-moderated “Pattern 5a”
configuration.
Fig 11 A new “Over-moderation” experimental core configurations in square lattice of the modified STACY.
Table 4
The reactivity worth per sample rod in the over-moderation configuration (Unit: cent/rod).
a All sample has zircalloy cladding.
b References.
Fig 12 Relative reactivity of the pseudo fuel debris samples by changing the concrete volume fraction in the new over-moderated configuration.
S Gunji et al / Progress in Nuclear Energy 101 (2017) 321e328 326
Trang 7fraction or the porosities caused insertion of negative reactivity
worths Especially, in these graphs, the relations of the concrete
volume fraction and the reactivity worth or the porosity and the
reactivity worth are characterized by not being linear
Table 5shows the reactivity worth per each sample in each
moderator condition with references In this configuration,
nega-tive reactivity worths were inserted by increasing of the concrete
volume fraction, and more negative reactivity worth was inserted
by insertion of the 100% concrete sample Moreover, minimum
negative reactivity worths were inserted by swapping the fuel rods
for the 100% uranium fuel without water These features in the over
moderation configuration have not seen in the past
“over-moder-ation” configuration described in section3.3
4 Conclusions
As a part of design works of critical experiments, core con
figu-rations to measured reactivity worth of MCCI products were
stud-ied It was found that the measurements using the modified STACY
in under- and over-moderation configurations with pseudo fuel
debris simulating MCCI products are feasible because the worth can
be estimated with enough accuracy from change of the critical
water height The suitable loading numbers of the samples were
estimated From these results, it is possible to determine the
amount of the pseudo fuel debris sample which should be
prepared It was revealed that the experimental“over-moderation” core conditions in this study was not enough over-moderation condition for the sample of pseudo fuel debris Therefore the
“new” over-moderation core configuration was analyzed in this paper This configuration was good to evaluate of the criticality characteristics for high concrete volume fraction samples
5 Further studies The experiment plans drafting in the modified STACY is carried out continuously Further discussion is necessary on criticality characteristics of the 1F-NPS fuel debris For example, water con-tent of fuel debris, MCCI products, and usage of neutron absorber materials should be studied before the experiment using the modified STACY In this paper, a combination of enriched uranium fuel, concrete and water was considered as a first plan, other combinations (burnup, cladding, steel construction, control rod, and so on) should be studied in near future It is scheduled to conduct the actual measurements of reactivity worth for those materials using the modified STACY after FY 2020
Acknowledgments This report includes results of the contract work funded by the Nuclear Regulation Authority (NRA)/the Secretariat of NRA of Japan
References
Brown, F.B., et al., 2009 MCNP5e1.51 Release Notes LA-UR-09-00384, LANL, USA Izawa, K., et al., 2012 Infinite multiplication factor of low-enriched UO 2 -concrete system J Nucl Sci Technol 49 (11), 1043e1047
Izawa, K., et al., 2015 Design of Water-moderated Heterogeneous Cores in New STACY Facility through JAEA/IRSN Collaboration Proceeding of ICNC 2015, Charlotte, North Carolina, USA, September 13e17, pp 965e976
Miyoshi, Y., et al., 2015 Present Status of STACY Modification Program and Fundamental Nuclear Properties of Experimental Cores Related to Fuel Debris Criticality Proceeding of ICNC 2015, Charlotte, North Carolina, USA, September 13e17, pp 1308e1319
Sakon, A., et al., 2015 Representability Evaluation of Fuel Debris Nuclear Charac-teristics by Heterogeneous Core of STACY Proceeding of ICNC 2015, Charlotte, North Carolina, USA, September 13e17, pp 1320e1330
Shibata, K., et al., 2011 JENDL-4.0: a new library for nuclear science and engi-neering J Nucl Sci Technol 48 (1), 1e30
Sono, H., et al., 2015 Modification of the STACY critical facility for experimental study on fuel debris criticality control Chap In: Nuclear Back-end and Trans-mutation Technology for Waste Disposal, vol 22 Springer, pp 261e268 Status of Fukushima Daiichi Nuclear Power Station, 2015 available online URL http://www.tepco.co.jp/en/nu/fukushima-np/index-e
Tonoike, K., et al., 2013 Major Safety and Operational Concerns for Fuel Debris Criticality Control Proceeding of GLOBAL 2013, Salt Lake City, Utah, USA, September 29-October 3, pp 729e735
Fig 13 Relative reactivity of the samples in the new over-moderated “Pattern 5a”
configuration.
Table 5
The reactivity worth per sample rod in the new over-moderated configuration (Unit: cent/rod).
a All sample has zircalloy cladding.
b References.
Trang 8Tonoike, K., et al., 2015a Options of principles of fuel debris criticality control in
Fukushima Daiichi reactors Chap In: Nuclear Back-end and Transmutation
Technology for Waste Disposal, vol 21 Springer, pp 251e259
Tonoike, K., et al., 2015b Study on Criticality Control of Fuel Debris by Japan Atomic
Energy Agency to Support Nuclear Regulation Authority of Japan Proceeding of
ICNC 2015, Charlotte, North Carolina, USA, September 13e17, pp 20e27
Tonoike, K., et al., 2015c Criticality Characteristics of MCCI Products Possibly Pro-duced in Reactors of Fukushima Daiichi Nuclear Power Station Proceeding of ICNC 2015, Charlotte, North Carolina, USA, September 13e17, pp 292e300 X-5 Monte Carlo Team, 2003 MCNP e a General Monte Carlo N-particle Transport Code, Version 5 LA-UR-03-1987, LANL, USA
S Gunji et al / Progress in Nuclear Energy 101 (2017) 321e328 328