Nuclear Engineering and Design 406 (2023) 112227

Contents lists available at ScienceDirect

Nuclear Engineering and Design

journal homepage: www.elsevier.com/locate/nucengdes

Analysis of the short Term-Station Blackout accident at the Peach Bottom

Unit-2 reactor with ASTEC including the estimation of the radiological
impact with JRODOS
Onur Murat a, *, Victor Sanchez-Espinoza a, Fabrizio Gabrielli a, Robert Stieglitz a, Cesar Queral b
a
Karlsruhe Institute of Technology, Institute for Neutron Physics and Reactor Technology, Hermann-von-Helmholtz-Platz 1, D-76344 Eggenstein-Leopoldshafen,
Germany
b
Universidad Politecnica de Madrid, Ramiro de Maeztu, 7, 28040, Madrid, Spain

A B S T R A C T

Driven by the Fukushima accident, the ASTEC code has been extended with new capabilities to describe the BWR-behavior, especially of the core, during severe
accidents with core meltdown. Hence, models for the BWR-typical core components like absorber cross, canister, water rods related to the chemical reactions,
material relocation, and radiative heat transfer were added to ASTEC. To evaluate the prediction capability of ASTEC for BWR, a short-term Station Black-out (ST-
SBO) severe sequence of the Peach Bottom Unit-2 was selected. The goal is to predict the radiological source term with ASTEC and the subsequent radiological impact
using the JRODOS code. For this purpose, the fuel inventory isotope fractions are determined by the ORIGEN-code. It describes the change of the core isotopic
composition during the operation. A comparison of selected parameters of the initial in-vessel phase predicted by ASTEC with the ones of MELCOR shows similar
results. But the vessel failure times and mass of molten material ejected from the core calculated by ASTEC deviates from the one of MELCOR. ASTEC predicts late
oxidation of core structures leading to an accelerated progression of the accident and to an earlier lower head failure compared to the one calculated by MELCOR for
the case of CORBH package. After the containment failure in the drywell head flange, the fraction of the nuclide inventory released to the environment predicted by
ASTEC are similar to the ones of MELCOR. Finally, JRODOS was used to predict the fission product dispersion and radiological impact around the Peach Bottom Unit-
2 plant. The results showed that in the worst-case scenario, a total effective gamma dose rate of 7.2 mSv/h exist.

1. Introduction with the core materials (capture, absorption, scattering), dedicated

codes like ORIGEN are used, in which the Bateman equation is solved.
The severe accidents in Fukushima Daiichi power plants resulted in Various integral codes for severe accidents such as ASTEC (Chate­
large radiological impact emphasizing the importance of an accurate lard, et al., 2016), MELCOR (Humphries, et al., 2017), MAAP (EPRI,
prediction of the radiological source term after severe accidents for the 2010), AC2 (Wielenberg, et al., 2019), SOCRAT (Leonid, et al., 2019),
emergency teams in order to initiate the proper countermeasures to etc. are being developed, improved, and validated worldwide for severe
minimize the consequences. accident simulations in LWRs. They include chemo-physical and math­
The accurate prediction of radiological impact around the site of a ematical models for the main phenomena occurring during the in-
NPP after a severe accident with core degradation requires appropriate vessel, ex-vessel phases as well as in the containment, describing the
computational tools for the following areas: a) prediction of the nuclide release of fission products from the fuel due to the failure of the safety
inventory in the core for a realistic core loading with fuel assemblies barrier, their behavior in the primary and secondary loops, and in the
burnt at different degree, b) prediction of the integral NPP behavior containment. New capabilities added to ASTEC to describe BWR-specific
during the progression of a severe accident and the fission products components in the core (Chatelard, et al., 2017) include rectangular-
release in the core, their transport in the primary/secondary circuits and shaped fuel channel boxes. Simulation of the subchannel geometry
in the containment, c) estimation of the dispersion of the released fission enclosing active core regions in the BWR design with physical models
products and their radiological impact on the citizens and environment. including chemical interactions, material degradation and movement
In order to fulfill the first requirement, namely to describe the change and radiative heat transfer, enabling the extension of the ASTEC’s
of the nuclide inventory due to the fission process in the core, the sub­ capabilities.
sequent radioactive decay and the different interactions of the neutrons Depending on progression of the accident, the integrity of the

* Corresponding author.
E-mail address: onur.murat@partner.kit.edu (O. Murat).

https://doi.org/10.1016/j.nucengdes.2023.112227
Received 30 November 2022; Received in revised form 14 February 2023; Accepted 16 February 2023
Available online 8 March 2023
0029-5493/© 2023 Published by Elsevier B.V.
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

containment may be lost and the loss of the final barrier will result in promising results.
radiological dispersion to the environment. In order to take timely ac­ The Chapter 2 describes the ASTEC model of the Peach Bottom Unit-
tions and evaluate countermeasures for such scenarios, tools such as 2 including core, vessel, containment volumes and cavity and connec­
JRODOS (Ievdin, et al., 2010) have been developed to help decision tions between them. Considered physical phenomena in the model
makers. Applying the results of the severe accident codes to the described and boundary conditions of the model explained. In the next,
dispersion calculations shows the consequences of the simulations for Chapter 3 includes fuel depletion calculation with SCALE-Origen code
the safety of the population and the environment. which is adopted in order to employ corresponding fission product in­
BWR power plant design includes specific components such as fuel ventory with reference study. Results of the selected scenario presented
channel boxes and cross type control blades which increases structural in the Chapter 4 and after that JRODOS implementation concluded the
mass in the core region. Compared to the design of PWR, the higher work by taking into account fission product dispersion and levels of dose
amount of zircaloy, stainless steel, and boron carbide in BWR affects the rate received by the public.
accident progression and core coolability. A larger amount of metallic
structures in the core has the potential to accelerate oxidation kinetics 2. Fuel inventory calculation with ORIGEN
and the release of large amounts of energy during severe accidents and
reflooding transients. An average French PWR 900 design contains Definition of the decay heat and fission product fractions in the core
about 20 tons of zircaloy fuel cladding, about 300 kg of stainless-steel describes how the accident progress in-vessel, ex-vessel and potential
absorber cladding, and has the potential to produce 900 kg of environmental and public hazards. In the previous study (Carbajo,
hydrogen generated by oxidation reactions in the reactor vessel. On the 1994), which is publication of the referenced study report (Carbajo,
other hand, the studied ASTEC model of Peach Bottom Unit-2 BWR4 1993), statement for the burnup level and isotope fraction was made as
design hosts 34.17 tons of zircaloy fuel cladding with 22.64 tons of following; high burnup fuel after long time full power operation. Exact
zircaloy fuel channel boxes, 1265 kg of stainless-steel absorber cladding, burnup level and isotope fraction values were unclear in order to
and 1424 kg of boron carbide absorber material. The simulation results reproduce fuel inventory for the ASTEC model. For this reason, deple­
with hydrogen production of more than 1800 kg during the oxidation tion module of the SCALE V6.2b code system (Wieselquist, et al., 2020),
reactions in-vessel shows the severity potential of the larger amount of ORIGEN was adopted in order to reproduce comparable results in the
metallic structures in the core of the BWR. Moreover, the oxidation previous MELCOR study.
process for the boron carbide absorber material is strongly exothermic ORIGEN code solves the rate equations in order to calculate gener­
and the hydrogen production capacity is about 6–7 times higher ated or depleted isotope concentrations in the fuel resulted by fission,
compared to zircaloy (Adroguer, et al., 2003). decay or transmutation. Problem dependent cross section libraries are
The presence of a high proportion of metallic structures is a chal­ used in the ORIGEN by interpolating pre-generated cross section li­
lenge to core safety, but eutectic interactions between the large pro­ braries. Neutron cross section libraries, which is generated by the Scale
portion of boron carbide absorber material and the stainless steel using by transport codes, can fit any type of fuel configuration and their
cladding can also lead to premature core degradation. The evolution of operating conditions.
eutectic melt between boron carbide and stainless steel around 1500 K Initial fuel loading in Peach Bottom Unit-2 NPP was General Electric
leads to the formation of molten material below the melting point. The 7x7 and one same type of assembly example in ORIGEN was used.
movement of the eutectic melt could interact with the Zircaloy fuel Previous study with MELCOR (Carbajo, 1993) also performed with
channel boxes and the binary system of boron carbide and stainless steel 168.48 tons of fuel materials which is corresponds to the 7x7 assembly
may lead local failures and early reaching of molten material around the type core loading. Average specific power for BWR fuel assemblies used
fuel rods (Steinbrück, 2010). in U.S. recorded as 24 MW/MTU (Hu, et al., 2016) and one assembly
The Primary Containment Vessel (PCV) volume of the BWR design is model in ORIGEN input has 0.1902 tons of uranium. Corresponding
smaller compared to the PWR containment design. To handle severe power level 4.5648 MW for one assembly in ORIGEN was defined and
accident conditions of high pressures and high temperatures, a wet burn time period adjusted as 2 years since the exact operational duration
compartment was added to the BWR PCV structure. The inclusion of a is not specified in MELCOR study.
large volume of water, called suppression pool in the wetwell, plays the Achieved discharge burnup was 17.5 GWd/t and it was compared
role of a condensation pool for excess heat. Steam discharged from the with reference databank for BWR fuel assemblies in U.S. (Fig. 2.1). It
core is directed through safety relief valves (SRVs) from the steam line to was found that 7x7 assembly types were in operation between the early
the suppression pool. In the event of severe accidents, when the fuel rods 1970 s and 1985, and discharge burnup records during this time period
fail and fission products released to the vessel domain, carrier gasses climbs up to about 26 GWd/MTHM. However, not only the 7x7 assembly
transport the fission products to the suppression pool. Thanks to the types but also other types of assemblies in operation after 1975, and
absorption capacity of the liquid water, the excess heat is suppressed especially the 8x8 assembly numbers is almost as many as the 7x7 as­
there and the fission products are retained in the water pool. For this sembly at the early 1980 s. Therefore, to separate the burnup of 7x7
reason, ex-vessel progression of severe accidents will be different in assemblies from that of other assemblies, one may conclude that the
BWR design than standard dry containment atmosphere of PWR design. period from the beginning to 1975 may be appropriate time period to
The main objective of this study is to simulate a complete severe consider. The high burnup peak at the beginning is related to fuel as­
accident transient, starting with the description of the fuel and ending semblies used for research purposes and not used in the power plants.
with the selected accident scenario, including the consideration of the When considering the selected period, the 17.5 GWd/t achieved
radiological consequences. In this way, the simulation capabilities of the discharge burnup are within the range of values recorded for the 7x7
current ASTEC version for BWR design and the obtained results can be BWR fuel assembly type.
evaluated in terms of public safety. Thanks to collaboration between KIT The description of the element groups and the selected elements in
and IRSN, available recent source code was adopted for this study. Se­ previous study with MELCOR and the comparison with the ORIGEN
lection of the plant was Peach Bottom Unit-2 BWR4 Mark-1 design as V6.2b calculation were presented in Table 2.1. The underlined elements
Fukushima Daiichi plants. Also, extensive studies on the Peach Bottom are the selected ones in the reference study and the masses of the
and data availability were one of the major factors for choosing the plant selected elements for the reference study and the ORIGEN results were
design. given. The comparison of the selected element masses in Table 2.1 shows
It is worth to mention that the ASTEC-capability to describe the in- that the selected power and the achieved discharged burnup for a period
vessel phase of BWR-cores was validated in the previous study (Murat, of 2 years produce suitable amount and fraction fission products for the
et al., 2020) using the QUENCH-20 BWR experiment. This study showed ASTEC ISODOP module.Table 4.1.

2
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

Fig. 2.1. Evolution of the discharge burnup and BWR assembly types in the U.S. from 1968 to 2013 (Hu, et al., 2016).

Table 2.1 Table 4.1

Comparison of Selected Element Masses between Reference Study and Scale Design parameters and stationary results of Peach Bottom Unit-2 ASTEC model.
ORIGEN Results. Reference Design Parameters (Larsen, 1978) ASTEC
Class names in Member of elements Reference Study Scale Core Power (MW) 3293 3293
MELCOR (kg) (Carbajo, ORIGEN
Feedwater mass flow rate (kg/s) 1679.68 1673.83
1993) V6.2b (kg)
Total mass flow rate (kg/s) 12914.78 12914.89
Noble gasses Xe, Kr, Rn, He, Ne, Ar, H, 463.7 446.7 Core mass flow rate (kg/s) 11336.75 11336.84
N Bypass mass flow rate (kg/s) 1578.03 1578.05
Alkali metal Cs, Rb, Li, Na, K, Fr, Cu 246.3 256.9 Steam mass flow rate (kg/s) 1685.98 1673.83
Alkaline metal Ba, Sr, Be, Mg, Ca, Ra, Es, 207.52 187.19 Steam temperature at dome (K) 559.29 559.38
Fm Core outlet temperature (K) 560.48 559.33
Halogens I, Br, F, Cl, At 20.93 17.44 Core inlet temperature (K) 548.53 547.39
Chalcogens Te, Se, S, O, Po 40.78 40.99 Feedwater temperature (K) 464.32 464.32
Platinoids Ru, Pd, Rh, Ni, Re, Os, Ir, 306.99 296.6 Dome pressure (MPa) 7.033 7.033
Pt, Au Core outlet pressure (MPa) 7.1564 7.0366
Transition Mo, Tc, Nb, Fe, Cr, Mn, B, 350.69 324.98 Pressure drop over the core (MPa) 0.152 0.1017
metals Co, Ta, W Core inlet pressure (MPa) 7.3084 7.1383
Tetravalents Ce, Zr, Th, Np, Ti, Hf, Pa, 593.65 554.22 Core exit void fraction 0.65 0.71
Pu, C Core avg. void fraction 0.304 0.43
Trivalents La, Pm, Sm, Y, Pr, Nd, Al, 571.03 528.0 RPV water level (m) 14.326 14.278
Sc, Ac, Eu, Gd, Tb, Dy, Ho,
Er, Tm, Yb, Lu, Am, Cm,
Bk, Cf core domain. Fuel cladding structure placement starts from the 0.0379
More volatile Cd, Hg, Pb, Zn, As, Sb, Tl, 1.41 1.284 m up to 4.1023 m level which encloses the active core material of UO2.
main group Bi
Between 0.0589 m and 3.7169 m axial elevations fuel material was
metal
Less volatile Sn, Ag, In, Ga, Ge 8.59 8.917 defined. At the same axial level of UO2, absorber material B4C and
main group cladding stainless steel around were placed. There are 10 uniformly
metals distributed axial mesh in active fuel domain and 2 additional meshes
presents at the top and bottom of it which can be seen in Fig. 5.1. Axial
meshes are not equally divided in all levels, in the figure it was only
3. ASTEC model of Peach Bottom Unit-2 nuclear power plant
shown for representative purposes.Fig. 2.2.Fig. 2.3.Fig. 2.4.Fig. 2.5.
Fig. 4.1.Fig. 4.2.Fig. 4.3.Fig. 4.4.Fig. 4.5.Fig. 4.6.Fig. 4.7.Fig. 4.8.
Dimensional data of the core, fuel rods, pressure vessel and volumes
Fig. 4.9.Fig. 4.10.
of the sections were taken from the document which is prepared by the
Reactor vessel internal radius is 3.1875 m and has thickness of
General Electric Company that includes plant design and operating data
0.1635 m. Inside of the vessel, “downcomer” and “jet” flow channels
for the cycle 1 and 2 (Larsen, 1978). Model definitions, physical modules
defined and separated by “jetpump” solid macro structure in order to
described in the following sections.
carry out an external recirculation flow. Reactor core flow inside of the
Reactor core and pressure vessel of BWR4 type Peach Bottom Unit-2
shroud structure divided into 4 channels which are bypass1, bypass2,
nuclear reactor defined under ICARE and CESAR modules of ASTEC.
bypass3 and bypass4 (Fig. 5.1).Fig. 5.2.
Pressure vessel consist of two structures, which are cylindrical part that
Total number of fuel assemblies in the core is 764, their structure
holds the core structures and hemispherical volume that represents the
type selected as BOX4SIDE rectangular model of ICARE and all assem­
lower plenum. Axial meshing starts from the 0.0 m to 4.1191 m eleva­
blies are selected as Type-1 (General Electric 7x7) initial loading. Fuel
tion level. Fuel channel boxes placed alongside the axial meshing of the

3
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

Fig. 2.2. ICARE radial meshes (left) and axial meshes (right) of the core with fuel channel box (BOX4SIDE) and absorber structure equivalents in the each channel.

12.37 mm.
Inside the fuel assemblies there are also additional flow channels
“core1”, “core2”, “core3” and “core4” which are responsible of the
active core cooling. Therefore, flow channels “bypass1”, “bypass2”,
“bypass3” and “bypass4” that defined inside the “shroud” remains
outside of the fuel assembly and they are treated as bypass flow regions.
Bypass flow sections have no direct contact with fuel rods unless there is
a failure in fuel channel box structures.
Cylindrical control rod structures placed in each bypass channel as 1,
15, 60 and 115 respectively. Stainless steel cladding of control rods has
39.1 mm external diameter. Neutron absorber material B4C with
diameter 32.1 mm placed inside the cladding. During representation of
cross-type absorber blades into the cylindrical shape, mass of B4C and
stainless steel preserved, however definition of cylindrical geometry
caused under estimation of surface area (Chatelard, et al., 2016).
The lower plenum lies axially below the cylindrical part of the vessel.
Domain of the lower plenum lies along the point − 6.376 m to 0.0 m. The
cylindrical portion of the lower plenum begins at 0.0 m in height and
Fig. 2.3. ICARE Lower Plenum elevation points, dimensions and extends to the − 3.188 m point. Below this point the hemispherical
inner structures. portion of the lower plenum stays. The inner radius of the cylindrical
part is 3.1877 m and the wall thickness is 163.5 mm. The wall thickness
assemblies made of Zr placed inside the core channels as 4, 60, 240 and of the hemispherical part is 166.7 mm up to the 39th degree and 214.3
460 respectively. Thickness of rectangular macro structure is 2 mm, and mm thereafter (Fig. 2.2). The number of steel pipes is 185, the outer
have width of the 0.138 m. Number of fuel rods in each fuel assembly is diameter of the pipes is 265 mm and the thickness is 5 mm.
49 and external diameter of Zr clad structure is 14.3 mm with 0.815 mm Definition of the support plate structure can be found the PWR type
thickness. Inside the cladding, diameter of fuel pellet made of UO2 is power plant reference input decks in ASTEC. However, definition of

Fig. 2.4. CESAR volumes, junctions, ICARE core domain and their connections of RPV of Peach Bottom Unit-2 Model.

4
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

Fig. 2.5. CPA zones that represents Primary Containment Vessel and reactor building sections of ASTEC model of Peach Bottom Unit-2 Reactor.

Fig. 4.2. Water level inside the vessel (ICARE domain) and top of the down­
Fig. 4.1. Vessel pressure during automatic SRV-1 and manual SRV operation. comer (DCTOP).

canisters in BWR input deck creates non-concentric sub-channels sepa­ 1993). Drywell, wetwell and vent down zone with their connections
rated by solid structure. Description of the support plates not included each other forms the PCV structure of the reactor. The definition of the
based on the recommendations for definition of plate structure for non- vent pipe in ASTEC only possible if it is located between two CPA zones.
concentric sub-channels. For this reason, the definition of the ZON_1 zone provides a necessary
Recirculation lines of BWR type reactors modeled with CESAR vol­ zone for the connection of the vent pipe, which submerges into the
umes (3 volumes for each recirculation line) and connections between wetwell pool. The F021 vacuum breaker connection ensures that the
them. After the “downcomer” channel, the connections take water from pressure level in wetwell zone does not exceed the pressure level in
the bottom of the channel and direct it to the first volumes of the drywell zone by opening and allowing a mechanical equilibrium be­
recirculation lines (JET_L11 and JET_L21). The water is directed through tween the two zones. The F398 connection between the drywell and the
the second volumes (JET_L12 and JET_L22) with the help of recircula­ refueling bay and the F400 connection between the wetwell and the
tion pumps. Then, the flow path of the recirculation pipes ended with torus compartment simulate the containment failure paths. The first
connections that connect the CESAR volumes (JET_L13 and JET_L23) failure mode, rupture of the wetwell compartment defined for the case
and the ICARE channel “jet”. The jet channel directs water to the lower when the pressure in wetwell reaches 1.2 MPa. The second mode, the
plenum and then to the core region. The water is heated to saturation F398 connection, simulates the rupture of the flange of the drywell head
temperature and reaches the SEP volume of CESAR. The liquid water is when the temperature rises to 644 K and the pressure in the drywell
separated from the SEP volume into the DCTOP volume and then the reaches to the 0.565 MPa. The high temperature in the drywell disturbs
water enters the downcomer channel to recirculate again. the flange structure and destroys the elasticity of the material, and the
The PCV sections and the reactor building volume, their connections high pressure creates paths through the refueling zone. The size of the
and failure modes were constructed using the CPA module of ASTEC, crack in the flange depends on the pressure in the drywell and changes
taking into account the MELCOR design of the previous study (Carbajo, linearly between 0.565 MPa (0.0 m2) and 1.378 MPa (0.04 m2).

5
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

Fig. 4.3. Generated hydrogen mass caused by the oxidation of the structures in the core (left) and Zr structures fuel channel boxes and fuel claddings (right).

Fig. 4.4. Corium mass in lower plenum (left) and ASTEC typical debris configuration in lower plenum (right).

Fig. 4.5. Degradation level and temperature levels on the point of lower head failure.

Connections of the SRVs from steam line volumes of the CESAR thickness is 1.524 m, cavity height where molten material collected is
through to the wetwell zone which hosts the suppression pool can be set 2.457 m, cavity radius is 6.48 m and thickness of the lateral walls is
seen in Fig. 2.4. Since the opening and closing pressures of the SRVs are 0.52 m. Concrete material composition fractions in the cavity 0.338
different from each other four steam line model as it is in the design of CaO, 0.358 SiO2, 0.072 H2O, 0.206 CO2, Al2O3 0.009 and Fe 0.017 and
the reactor was introduced to the ASTEC model. Automatic Depressur­ ablation temperature of the concrete is set to 1503 K. Single cavity
ization System (ADS) consist of operation of the SRV number 1, 2, 3, 5 model of the previous study (Carbajo, 1993) was referenced for the
and 6. cavity model of the ASTEC.
Definition of the single cavity in the ASTEC model was made in the ASTEC FP_HEAT rubric, which includes isotope fractions as well, is
drywell zone. Geometrical definition of the cavity as follows; basemat responsible for the decay heat production. Calculation of the fuel

6
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

Fig. 4.9. Fission products in the CPA Zones.

Fig. 4.6. Drywell Zone pressure up to end of the ASTEC transient simulation. 3.1. Considered physical phenomena in ASTEC model

Modeling of the Peach Bottom Unit-2 in ASTEC requires definition of

physical models which are responsible for heat transfer, chemical in­
teractions, failure criteria and material movement. Selected models and
their definitions for the related domain explained in this chapter.

3.1.1. Heat transfer models

In order to define conductive heat transfer in macro components or
between macro components automatic definition of CONDAUTO was
used. Module RADR definition was also made for radiative heat transfer
between structures in the core up to cavity occurs. Described RADAS­
SEM module takes action to calculate radiative heat transfer after cav­
ities appear. New implementation of the radiative heat transfer
capabilities for the fuel channel boxes in BWRs also was taken into ac­
count thanks to RADASSEM module. In addition, heat transfer mecha­
nism between fluid and macro structures was build by CONVAUTO
option. In order to define in mesh heat transfer models, which includes
molten pool heat transfers between its meshes, POOL option was
Fig. 4.7. Fission products in the core domain. enabled. In POOL option MAYINGER model (default) was selected
which assumes homogeneous temperature in corium pool and uses
specific correlation to find characteristic velocity of the pool in order to
get mixing conductivity.

3.1.2. Chemical reactions and interactions

Oxidation of zirconium with steam was provided by structure ZROX.
Cladding of fuel rods, canisters of fuel bundles and spacer grids were
included in the ZROX and BEST-FIT option was selected. In order to deal
with steel oxidation by steam FEOX structure was defined for external
face of absorber rod cladding and internal faces of the shroud and vessel
structures. Steel oxidation kinetic option was selected default value
MATPRO that considers the correlation of J.WHITE (White, 1967)
studied on type 304 stainless steel. Module BCOX structure models the
boron carbide oxidation by steam. Default option for oxidation kinetic
BEST-FIT considered in range of experimental studies done with VERDI
facility and BOX rig experiments in frame of COLOSS project (Adroguer,
et al., 2003). Eutectic interaction between stainless steel and boron
carbide was activated by BSCC structure. Melting point of boron carbide
is around 2350 ◦ C and for stainless steel is around 1450 ◦ C. Due to
Fig. 4.8. Fission products in the CESAR Volumes. eutectic interaction between boron carbide and stainless steel, formation
of the liquefied stainless-steel layer can be observed around 1200 ◦ C
inventory, in order to make better source term estimation, was per­ which is 200 ◦ C lower than melting point of stainless steel (Nagase, et al.,
formed by using Scale ORIGEN code which detailed explanations can be 1997). Modeling of the boron carbide and stainless-steel interaction by
found in the section 3. BCCS structure of ASTEC is based on the PWR type (rod like) cylindrical
boron carbide rods. Implementation of this module is adaptation of
blade structure as a cylindrical structure. Dissolution of UO2 and ZrO2 by
liquid zircaloy is also considered chemical reaction during severe

7
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

Fig. 4.10. Fission products release to the environment cumulatively.

Fig. 5.1. Mesh grid in 400 km range around the Peach Bottom NPP site and population distribution over grid.

accident conditions. Structure UZRL activated to ensure interaction take and EPMX hoop creep and CRAC burst criterion parameters selected as
place between UO2 fuel material and zircaloy cladding. Chemical in­ 0.25 and 0.5 respectively as indicated as recommended values for
teractions associated with magma component is structured by the reactor applications.
UZOXMAG option. Integrity criteria of the fuel, cladding and canister structures was
defined using by INTE structure. For fuel macro structures, condition
3.1.3. Mechanical behavior would set to DISLOCAT if the temperature was greater than 2500 K.
Mechanical behavior of the UO2 fuel material enclosed with Zircaloy DISLOCAT status of the material starts degradation of the macro struc­
cladding structures are first safety barrier for radioactive material. Their ture. It is not mandatory to contain molten material for degradation state
integrity and related parameters defined under mechanical behavior but can contain MIXTU structure, which is not explicitly user defined
options of the code. layer. DISLOCAT state also allows fission product release and double
CREE structure was defined for creep behavior of the Zircaloy clad­ sized oxidation. Cladding INTE structure conditions was set to two
ding and the model is only available for Zr layer. All cladding groups in branches. If the temperature reached to 2500 K status would set DIS­
each channel included in the structure with the following options. Best LOCAT. In addition, if temperature level exceeded 2300 K and thickness
estimate physical model for burst occurrence EDGAR option was of material layer decreased to 250 µm, status of the cladding would set
selected. Gap pressure between fuel rod and cladding was set to 30 bar DISLOCAT as well. Integrity criteria of the canister structures includes

8
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

activated.

4. Analysis of the ST-SBO for Peach Bottom Unit-2

Steady state results and established transient scenario and ASTEC

predictions presented in this section. CESAR, ICARE, CPA, ISODOP,
SOPHAEROS, RCSMESH, MEDICIS, RUPUICUV, CORIUM and DOSE
modules activated for the calculations. Execution of ST-SBO scenario
resulted around 9 h CPU time with ASTEC V2.2 Revision 6790 M
developer version, which is compiled with GCC 12.2.0 package in
Windows.

4.1. Stationary plant conditions at nominal power

Calculation executed for 500 s and results are presented in Table 4.1.
Discrepancy between inlet pressure resulted higher amount of steam
production in the core and larger void fraction in general. Since the basic
channel application was introduced without jet pump, increase in the
pressure was not observed.

Fig. 5.2. Total effective gamma dose rates of one day simulations for one year
4.2. Short term Station Black-Out accident
(2021) among the interested meshes.

Decision of the severe accident scenario was made based on the

the condition of temperature which greater temperature level than 2500
previous safety assessment studies on BWRs. Even though the referenced
K leads to DISLOCAT status.
study with MELCOR (Carbajo, 1993) considers the SBO accident tran­
DECAUTO automatic model definition was made to provide material
sient, it must be emphasized that the choice of severe accident scenario
movement inside mesh. From one set of structure like rod can be
must be based on a solid foundation.
transferred to another set such as grid in the same mesh domain. In order
Based on previous safety assessments (Kolaczkowski, et al., 1989) the
to deal with melt corium movement in core region MOVEMAG option
loss of offsite power (LOSP) scenario with diesel generator failure and
with LEAK setting enabled. Setting LEAK provides material transition
battery depletion was found to have the greatest impact on the core
from core region to lower plenum domain.
damage frequency of the Peach Bottom Unit-2. This work was also part
Integrity lost criteria for the lower plenum meshes are adopted from
of the complementary study of severe accident risk analysis of five nu­
the reference study (Carbajo, 1993) as when the temperature of the
clear power plants that included Peach Bottom Unit-2 (U.S. Nuclear
mesh reaches 1723 K rupture option in the LOWERPLE structure acti­
Regulatory Commission, 1990), and as part of that work, SBO accidents
vates and mesh disappears.
were again found to have the highest core damage frequencies among
the estimated transients. Another study (Leonard, et al., 2007) also
3.1.4. Physical phenomena in lower plenum
showed that Short-Term Station Blackout (ST-SBO) with a stuck open
Corium melt movement started with MOVEMAG structure and LEAK
relief valve is one of the major contributor to the core damage
setting provides a breach from core to lower plenum and other end of the
frequency.
breach was defined with SLUMP structure. User needs to define SLUMP
Another important point: the calculation time of the SBO scenarios
structure in case of lower plenum domain exist and corium melt
without safety systems was the fastest compared to the other estimated
movement transfers to lower plenum. FRAGLOWE option is enabled to
transients, which contributes to fast simulation runs and easier exami­
activate corium fragmentation in lower plenum when the melt is con­
nation of the results. Based on the studies and conclusions, it was
tacted with water in lower plenum and it undergoes the fragmentation.
decided to follow the same transient with the reference MELCOR work.
Material separation in lower plenum, when corium is collected in and
Initial event was loss of offsite power following with all diesel gen­
metallic composition starts to be collected at the top of the heavier oxide
erators failure and battery depletion. Reactor scram was initiated at time
layer, provided by SEPALOWE structure. Movement of the debris bed
t = 0 with containment isolation which is main steam line and feedwater
particles which are located above the liquid corium pool defined with
line closure. Operation of the SRV-1 for 200 s was assumed which cor­
MOVELOWE structure. Debris particles can penetrate and sink into
responds the time period until operator actions takes place to depres­
corium pool beneath.
surize the reactor. After that, manual operation of the one safety valve
continued between 6.49 MPa and 7.18 MPa. Operation was carried out
3.2. Boundary conditions until water level in core reaches one third of the core level and following
ADS operation initiated. Calculation executed until the basemat in the
Water injection to the vessel described with connections named cavity domain ruptures.
FWATER and, as a boundary condition, 191.17 ◦ C water was fed at a rate
of 419.92 kg/s rate to each feed water volume in CESAR. Pressure 4.2.1. Main results of ST-SBO transient
boundary conditions were described for the volumes of steam pipe in Estimated events during calculations are presented in order
CESAR for steam extraction. The pressure parameter was set to 7.0268 Table 4.2. Each saved sequence represent steps of the propagation of the
MPa to keep pace with the design pressure of the vessel dome during severe accident starting from the failure of the fuel cladding structure up
steady state calculations. Thermal heat exchange was introduced with to rupture of the basemat structure in the cavity.Table 4.3.
the boundary condition HEAT from the volumes CESAR to the CPA zone After the plant is shut down, the decay heat increases the tempera­
drywell in the PCV domain. In order to allow the molten material to ture and pressure inside the vessel. The pressure inside the vessel in­
transfer from the vessel to the containment and cavity, the boundary creases to the opening set point of SRV-1, which has the lowest set point
condition VESPOUR was described. The definition of the cavity was between SRVs, and its automatic operation maintains the pressure be­
already done in drywell zone in CPA. However, to impose direct heating tween set points until manual operation begins. After 200 s, the opera­
of the containment, the boundary conditions DCH and MCCI were tion of one SRV between the 6.49 MPa and 7.18 MPa pressure setpoints

9
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

Table 4.2 production is comparable to that of the fuel claddings. Hydrogen pro­
Sequence of events Peach Bottom Unit-2 ASTEC Model ST-SBO Scenario. duction is an indicator of the level of oxidation reaction of the material,
Sequence of Events Time (s) which indicates the potential of energy release in case of an exothermic
oxidation reaction. The results show that the level of hydrogen pro­
SRV-1 starts operation 0
SRV-1 stops operation, Manual operation of a SRV starts 200 duction mass for fuel channel boxes is not far from the fuel claddings.
First cladding creep rupture, start of the fission product release 1037 Their contribution to the total hydrogen production of 1847.6 kg ac­
First material slump in lower plenum 1142 counts for half of the mass produced.
First slump of corium with fission products in lower plenum 1158 Investigation of the ST-SBO accident scenario in the reference study
First appearance of a cavity in core 1167
Manual operation of a SRV stops, ADS actuates 3867
(Carbajo, 1993) consist of two calculation approaches which MELCOR
First total core uncovery 6772 and MELCOR/CORBH. In order to take into account debris quenching
Lower head vessel failure 13,361 with water in lower plenum MELCOR with CORBH package was used.
PCV head flange failure 13,415 Standalone MELCOR calculation did not include debris quenching and
Basemat rupture 88,142
high temperature debris particles directly heated the lower plenum
structure. In addition, standalone MELCOR uses multi-node cavity and
MELCOR/CORBH deals with only single node cavity which is suited for
Table 4.3 ASTEC as well. The estimated hydrogen production for the stand-alone
Element mass fractions released into the environment. MELCOR case is 500 kg and for MELCOR/CORBH is 600 kg, which is far
Element MELCOR (Carbajo, 1993) ASTEC below the ASTEC prediction.
Xe, Kr 0.06 0.198
The accumulation of the corium, which is called magma structure in
I 0.0075 0.0308 ASTEC terminology, in the lower plenum is shown in Fig. 4.4. It can be
Cs 0.0097 0.0017 seen that mass transition between the heavy, light and oxide layers of
Te 0.0066 9.51E-4 the magma and their construction beginning of the corium buildup. At
Sr 5.0E-4 1.797E-5
time 13,361 s, failure of the lower head vessel occurs and most of the
Ba 0.0005 2.2E-4
Ru 3.0E-8 2.05E-5 molten material is ejected from the lower head into the cavity. The total
La 4.0E-7 1.65E-6 mass of corium ejected from the vessel is predicted to be 152.34 tons. As
Ce 1.0E-7 1.66E-5 can be seen on the left in Fig. 4.4, the MAGMA1 heavy metal layer is not
completely ejected after the vessel failure, there is still about 20 tons of
material in the lower plenum. The amount of failure in the meshes of the
continued to depressurize the system, corresponding to the manual ac­
lower plenum and the degree of degradation of the core domain (ICARE)
tion of the operators. Manual operation of the SRV continued until
at the time of failure of the lower head can also be seen in Fig. 4.5.
initialization of ADS. At 3867 s, the water level in the core reached one-
The time for lower head failure estimated in the reference study
third of the core level and this triggered the ADS operation. Thus, the
(Carbajo, 1993) for the MELCOR case is 9760 s, considering only
triggering of the 5 SRVs resulted in a sudden pressure drop in the vessel
penetration failures. In the MELCOR/CORBH case, penetration failure
(Fig. 4.1). The estimate of ADS triggering after manual actuation of an
was observed in 17,128 s and complete lower head failure in approxi­
SRV in the reference study was 3983 s, which is close to the MELCOR
mately 28,000 s. The stand-alone MELCOR calculation did not estimate
and ASTEC results related to SRV operations.
the total lower head failure and showed an early penetration failure due
Depressurizing the core by removing steam from the vessel results in
to the excess heat of the debris being directly transferred to the lower
a loss of water mass and a drop in the water level in the vessel. As can be
head wall. In contrast, the CORBH package accounts for the quenching
seen in Fig. 4.2, the water level is shown in separate domains. Core
of the debris and the amount of excess heat removed by the water intake
domain where fuel assemblies were inserted is shown with the name
in the lower plenum, resulting in later failure. The mass ejected from the
“Vessel” and the water supply to the core, at the top of the downcomer
lower head is approximately 180 tons for the MELCOR standalone case
volume “DCTOP”. The loss of water mass and the drop in the water level
and 279.52 tons for the MELCOR/CORBH package. A larger amount of
starts at the beginning with the steam extraction by the automatic and
material is expected to be ejected in the case of the CORBH package
manual operation of the SRVs. After the water in the DCTOP volume is
because a complete failure of the lower head is observed. The compar­
depleted, the level in the active core region begins to drop. Activation of
ison between the previous study and the ASTEC results shows major
the ADS at time 3867 s results in a sudden drop in pressure and a flash of
differences. Definition of the lower head structures in the ASTEC model
steam causing a sudden jump in the water level in the core. The complete
consists of cylindrical structures that do not fully correspond to the
discovery of the core is recorded at 6772 s, which can be seen in Fig. 4.2,
complex framework of the lower head of the BWR. For this reason, the
After the water level reaches to below lowest axial mesh in the core
MELCOR model considers a larger material definition and the consid­
domain, the water level record drops to zero because the lower plenum
eration of penetrating tubes in the lower plenum with a larger mass.
representation in ASTEC consists of one mesh and the construction of the
After the total failure of the lower head for MELCOR/CORBH and only
average water level was not considered. The reference study (Carbajo,
the failure of the penetration tubes for the standalone case, a larger mass
1993) does not specifically mention the total water depletion including
was ejected from the lower plenum. In addition, a larger thermal mass
the water level in the lower plenum and the time of core uncovery.
extends the period of lower plenum failure by increasing the total heat
Hydrogen production caused by oxidation of materials in the core is
capacity of the structural system in the lower plenum.
one of the driving mechanisms of core degradation, as this exothermic
After the failure of the vessel, the accident transient continues in the
reaction causes additional energy release in the core. The removal of
PCV domain which deals with phenomena outside the vessel. Extracted
steam by opening the SRVs and the pressure drop create more steam and
steam and fission products from the core inserted in the wetwell zone
a high temperature steam environment that further promotes the
through vent pipe connection that allow the suppression pool scrubbing
oxidation reaction. In Fig. 4.3, the hydrogen mass produced by the
during the SRV and ADS operations. Condensation of the steam reten­
oxidation of Zr, stainless steel, boron carbide, and magma structures are
tion of the fission products with carrier gasses takes place in the sup­
shown separately on the left and the Zr structures of the fuel channel
pression pool and excess mass transfers to the air atmosphere of the
boxes and fuel claddings are shown separately on the right up to the time
wetwell zone and drywell zone. During the ST-SBO scenario high tem­
of the bottom head failure. Since the rectangular fuel channel boxes,
perature and pressure build up in the drywell activated the failure mode
which are BWR-specific structures, are also made of Zircaloy material, it
for the head flange failures in the drywell. In Fig. 4.6, it can be seen that
is important to emphasize whether or not their contribution to hydrogen
pressure rises resulting action of the SRV and ADS operation and vessel

10
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

failure causes sudden increase in the pressure in the drywell since the In order to understand the possible consequences of ST-SBO accident
cavity is defined in the drywell zone. Activated head flange failure mode in Peach Bottom Unit-2, masses of the fission products that estimated in
opens the Valve-398 and allows the fission products to reach environ­ environmental release by ASTEC, converted into JRODOS source term
ment at 13,415 s. Pressure level circulates around 600 kPa in Fig. 4.6 information. Point of release was selected as actual geographical loca­
since the opening on the flange is dependent on the pressure in drywell tion of the Peach Bottom NPP in the databank of the JRODOS and
zone and opening and closure of the connection causes this behavior. meshed grid was constructed around it within 400 km radius (Fig. 5.1).
MELCOR/CORBH calculation was terminated before the contain­ ASTEC simulation showed that release duration of the fission products to
ment failure, and only the stand-alone MELCOR calculation was the environment was about 20.75 h and this duration divided into 30
continued in the previous study. The head flange failure in the drywell min time steps. Amount of released fission products described as source
zone was also the driving mode for the containment failure in the terms for each time interval.
MELCOR study. Head flange failure was observed at 25,515 s, which is The radiological dispersion analysis does not include a detailed study
larger than ASTEC prediction. Standalone MELCOR simulation predic­ of the source term release consequences in terms of public safety. The
tion for the amount of mass ejected from the lower plenum is compa­ main objective of the following part of the study is to examine the sta­
rable to the ASTEC prediction, but the MELCOR simulation only predicts tistical analysis on the considered region over a time period in order to
penetration failure without failure of the entire lower head, and the understand potential consequences based on worst-case approach.
pressure accumulation is slower than the ASTEC model. For this reason, The selection of analysis for each day for JRODOS simulations were
the accumulation of temperature and pressure in the MELCOR model performed for 24 h. In order to be able to perform a statistical analysis
takes longer than in the ASTEC model and the MELCOR estimation of for one year on the selected area with a personal computer in a
PCV failure is in later stages than ASTEC. reasonable time the simulations were performed for one day.
Fission product inventory in the core until vessel failure is shown in As can be seen in Fig. 5.1, the population is widely dispersed along
Fig. 4.7. After the loss of cladding integrity, the fission products begin to the east coast of the United States. Examining each residential area
move from outside the fuel domain. The production of steam and through the grid is difficult to evaluate and draw some conclusions. For
removal by the SRVs transports the fission products along the path. The this reason, only certain meshes that have higher population densities
sudden drop in fission product masses in Fig. 4.7 at time 13,361 s cor­ than others in the region were selected as points of interest. Selected
responds to the failure of the lower head. After that point, the calcula­ meshes on the map as follows; mesh number 5554, 5606, 5658, 5607 for
tion of the ICARE and CESAR modules is stopped and the last stored the New York City, mesh 3209, 3240 and 3241 for the Philadelphia,
parameters remain in the database. mesh 2064 and 2065 for the Washington DC, mesh 2594 for the Balti­
Operation of the SRV steam extraction and fission product trans­ more. Average population density in the New York City is around 17,000
portation also can be tracked in the Fig. 4.8. Large jumps at the begin­ people/km2, Philadelphia has around 8000 people/km2, Washington DC
ning caused by the manual operation of the SRVs and after ADS and Baltimore City have around 5000 people/km2 for their corre­
actuation at the 3867 s there is a slight increase in the fission product sponding meshes.
masses. The simulation of the 24 h requires the weather data from initial time
After the CESAR volumes, which SRV connections were made, CPA to the end of the simulation and any recorded or user defined weather
zones simulates containment for the ex-vessel phenomena and final definition could be described in the JRODOS. An investigation of the
barrier between fission products and environment. weather conditions that could potentially produce the highest dose rate
The accumulation of fission products in the environment is shown in in the selected meshes was performed to pursue the worst-case
Fig. 4.10. The logarithmic scale was used to show the fission product approach. A weather database for the year 2021 was selected and a
masses in one place because the quantities on the y-axis are widely simulation was run for each day from January 1, 2021 to January 1,
separated. The low retention rate of the noble gasses results in a sig­ 2022, without considering early countermeasures. The start time for the
nificant amount of them being released into the atmosphere. At the end simulation was randomly selected for each day.
of the calculation, the total activity released to the atmosphere was Maximum recorded total effective gamma dose rates on each day in
recorded as 1.836E18 Bq. The comparison between the MELCOR 2021 between the interested mesh points are shown in Fig. 5.2.
reference study and the ASTEC model in terms of elemental fractions Maximum dose rate on each day was recorded 178 times in New York
released to the environment is shown in Table 4.3. The magnitude of the City, 98 times in Philadelphia, 6 times in Washington DC, and 48 times
two studies shows that the results are comparable and the simulation in Baltimore. The total number is not equal to 365 because on some days
transient followed in the same direction. the weather conditions favored the selected mesh points as they did not
In order to make a conclusion about released fission products to the receive fission products and for this reason no dose rate was recorded. Of
environment in terms of public safety and to understand possible con­ all the days recorded, the highest dose rate was recorded in mesh
sequences caused by them further analysis needs to be performed. In number 2594, corresponding to Baltimore City, at 7.2 mSv/h when the
order to achieve that JRODOS implemented to continue complementary release of fission products begins at 10:33 am on February 12, 2021.
work. A single simulation was performed to examine the example repre­
senting the highest total effective gamma dose rate in the statistical
5. Statistical analysis of the fission products dispersion and analysis. The calculation was started at 10:33 a.m. on February 12,
worst case scenario approach with JRODOS 2021, and ended 24 h later, excluding early countermeasures. As can be
seen in Fig. 5.3, the affected area of Baltimore City receives no dose
Assessment of the fission product release to the environment is the during the first 12 h because of wind direction carries the fission prod­
next step of the calculation and it was carried out with Java based Real- ucts to the west. However, the wind direction changes and the results
time Online Decision Support code (JRODOS) (Ievdin, et al., 2010). The after 18 and 24 h show that the propagation of the dose rate to the east
RODOS code was developed to meet the needs of decision makers in decreases and the cloud covers Baltimore City, resulting in a total
nuclear accidents. The main objective was to estimate the doses and the effective gamma dose rate of 7.2 mSv/h in mesh 2594, as shown in the
effects of radiological dispersion on the environment and the popula­ statistical results (Fig. 5.2). The clouds cover not only the Baltimore area
tion, taking into account the geographical shape of the area, the distri­ but also other counties to the west of the Peach Bottom NPP. After the
bution of the population, and the weather conditions. The development change of wind direction, the Washington DC area also receives a total
of models and better estimation of radiological dispersion assists deci­ effective gamma dose rate of about 2 mSv/h in this case.
sion makers in conducting analyzes in possible accident scenarios and in The limit on total effective dose equivalent for individuals of the
improving emergency management guidelines. public and for radiation workers in a year has been set by the United

11
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

6. Conclusion and outlook

An overall study of a severe accident was performed, starting with

fuel inventory estimation using Scale-ORIGEN and simulating the
selected ST-SBO accident using ASTEC, and concluding with JRODOS
dispersion analysis around the Peach Bottom unit site. The joint appli­
cation of the codes provided the opportunity to examine the entire ac­
cident in detail and evaluate the results in terms of their impact on
public safety.
Regarding the estimation of accident progression ST-SBO, the
reference study and ASTEC results showed comparable results for vessel
pressure during manual and automatic operation of SRVs. After actua­
tion of ADS and structural degradation, the estimates for vessel failure
and the amount of corium ejected are different. The modeling of the
lower plenum and the amount of structural mass defined there resulted
in inconsistent times for MELCOR and ASTEC. However, the amount of
fission products released into the environment is of the same order of
magnitude. The transport of fission products from the core to the envi­
ronment through the CESAR volumes and CPA zones shows consistent
results compared to the reference work. A lower plenum definition and
mass adjustment could result in simulation results similar to the MEL­
COR study. To extend the severe accident analysis of the BWR design
with ASTEC, new types of fuel assemblies with water rods and a different
design and amount of active fuel material can be considered.
To perform a comprehensive analysis of the effects of the received
dose, the number of affected meshes must be higher and the simulation
period should be larger. Only 24 h of simulations are not sufficient to
draw an accurate conclusion. In addition, the calculations in this study
did not consider the countermeasures that are important for public
safety when fission products are released into the atmosphere. The
possible dispersal pathways of fission products through soil and water
supply, as well as contamination of food and feed, are also issues that
need to be subjected to detailed analysis. Statistical analysis has also
shown that some residences, although more distant than other affected
areas, have a higher risk of fission product transmission and could have a
higher dose rate. The annual regime of winds and rain plays an impor­
tant role in this case, and statistical analysis can provide clues as to
which area has a higher probability of fission products. A decision based
on the distance between the power plant and the residential area may
not provide enough information to interpret the probability of a dose
record in this region. Simulating the statistical analysis for 3–5 years of
meteorological data can provide information on which meshes to
consider. Detailed analysis considering longer simulation periods and
long-term effects of received doses will produce better estimates of
public health and safety.
Conducting a severe accident analysis for a BWR design is a step
toward expanding the scope of ASTEC. Although the ASTEC code has
already implemented boiling water reactor design, the inclusion of
BWR-specific fuel channel structures and subchannel physical phe­
nomena, and the evaluation of the results, demonstrates the capabilities
of the code in addition to the studies mostly performed at PWR design. A
Fig. 5.3. Total effective gamma dose rate map for 6, 12, 18 and 24 h after
first complementary analysis of a BWR design with ASTEC considering
fission products release resulted by ST-SBO accident in Peach Bottom Unit-2
an appropriate fission product inventory, latest physical models and a
NPP within the constructed domain in JRODOS starting from 12th of
February 2021 at 10:33.
dispersion study showed promising results for further BWR studies with
ASTEC.
States Nuclear Regulatory Commission (USNRC) at 1 mSv and 5 mSv,
CRediT authorship contribution statement
respectively (U.S. Nuclear Regulatory Commission, 1991). The com­
parison between the highest recorded dose rate and the annual limits
Onur Murat: Methodology, Software, Writing – original draft,
shows that the dose rate is high enough to exceed the limits, regardless
Visualization. Victor Sanchez-Espinoza: Supervision, Writing – review
of whether the persons are members of the public or radiation workers.
& editing. Fabrizio Gabrielli: Supervision, Writing – review & editing.
This result is not unique to Mesh 2954, which corresponds to Baltimore
Robert Stieglitz: Supervision. Cesar Queral: Supervision, Writing –
City. There are also regions that were not considered in the statistical
review & editing.
analysis where dose rates are as high as in the affected meshes. However,
the selection of points took into account that cities with larger pop­
ulations can potentially have more severe consequences.

12
O. Murat et al. Nuclear Engineering and Design 406 (2023) 112227

Declaration of Competing Interest Epri, 2010. MAAP4 Applications Guidance, Desktop Reference for Using MAAP4
Software, Revision 2. Electric Power Research Institute, Palo Alto, CA.
Hu, J., Gauld, I. C., Peterson, J. L. & Bowman, S. M., 2016. US Commercial Spent Nuclear
The authors declare that they have no known competing financial Fuel Assembly Characteristics: 1968-2013, NUREG/CR-7227, Oak Ridge, TN 37831-
interests or personal relationships that could have appeared to influence 6170: U.S. Nuclear Regulatory Commission.
the work reported in this paper. Humphries, L. et al., 2017. MELCOR Computer Code Manuals, Vol. 1: Primer and Users’
Guide, Verson 2.2.9541, Albuquerque: Sandia National Laboratories.
Ievdin, I., Trybushnyi, D., Zheleznyak, M., Raskob, W., 2010. RODOS re-engineering:
Data availability aims and implementation details. Radioprotection 45, 181–189.
Kolaczkowski, A. M. et al., 1989. Analysis of Core Damage Frequency: Peach Bottom, Unit 2
Internal Events, NUREG/CR-4550 Vol. 4, Albuquerque: Sandia National Laboratories.
The authors do not have permission to share data. Larsen, N., 1978. Core Design and Operating Data for Cycles 1 and 2 of Peach Bottom 2,
NP-563. General Electric Company, California.
Acknowledgement Leonard, M. T., Gauntt, R. O. & Powers, D. A., 2007. Accident Source Terms for Boiling
Water Reactors with High Burnup Cores Calculated Using MELCOR 1.8.5, Albuquerque:
Sandia National Laboratories.
Authors would like to thank Alexandre Bleyer, Lionel Chailan, Pat­ Leonid, A.B., Kirill, S.D., Arkady, E.K., Valery, F.S., 2019. Results of SOCRAT code
rick Chatelard, Patrick Drai, Laurent Laborde and IRSN ASTEC Team; development, validation and applications for NPP safety assessment under severe
accidents. Nuclear Engineering and Design 341, 326–345.
Wolfgang Hering and Christian Staudt for their helps and continues Murat, O., Espinoza, V.S., Wang, S., Stuckert, J., 2020. Preliminary validation of ASTEC
supports. The corresponding author is funded by Republic of Turkey V2.2.b with the QUENCH-20 BWR bundle experiment. Nuclear Engineering and
Ministry of National Education during his Ph.D. studies. Design.
Nagase, F., Uetsuka, H., Otomo, T., 1997. Chemical interactions between B4C and
stainless steel at high temperatures. Journal of Nuclear Materials 245 (1), 52–59.
References Steinbrück, M., 2010. Degradation and oxidation of B4C control rod segments at high
temperatures. Journal of Nuclear Materials 400 (2), 138–150.
Adroguer, B., et al., 2003. Core Loss During a Severe Accident (COLOSS Project), U.S. Nuclear Regulatory Commission, 1991. Radiation Dose Limits for Individual Members
COLOSS Final Extended Report, Part 1. IRSN, Experimental and analytical activities of the Public (Subpart D), 10 C.F.R. § 20.1301. [Online] Available at: [Accessed 16 11
Main outcomes from the project, s.l. 2022].
Carbajo, J.J., 1994. MELCOR sensitivity studies for a low-pressure, short-term station U.S., Nuclear, Regulatory, Commission, 1990. Severe Accident Risks: An Assessment for
blackout at the Peach Bottom plant. Nuclear Engineering and Design 152, 287–317. Five U.S. Nuclear Power Plants. United States Nuclear Regulatory Commission,
Carbajo, J., 1993. Severe Accident Source Term Characteristics for Selected Peach Bottom NUREG-1150, Washington.
Sequences Predicted by the MELCOR Code, NUREG/CR-5942, Oak Ridge, TN 37831- White, J., 1967. Physico-chemical studies of clad UO2 in potential meltdown
6285: U.S. Nuclear Regulatory Commission. enviroments. Nuclear Technology Department Nuclear Energy Division General
Chatelard, P., et al., 2016. Main modelling features of the ASTEC V2.1 major version. Electric, Virginia.
Annals of Nuclear Energy 93, 83–93. Wielenberg, A., et al., 2019. Recent improvements in the system code package AC2 2019
Chatelard, P. et al., 2017. Contribution of ASTEC numerical simulations to the understanding for the safety analysis of nuclear reactors. Nuclear Engineering and Design 354,
of the Fukushima accidents, Vienna: IAEA Workshop on Advances in Understanding 110211.
the Progression of Severe Accidents in Boiling Water Reactors. Wieselquist, W., Lefebvre, R., Jessee, M., 2020. SCALE Code System, ORNL/TM-2005/
39. Oak Ridge National Laboratory, Oak Ridge, TN.

13