Progress in Nuclear Energy 134 (2021) 103674

Contents lists available at ScienceDirect

Progress in Nuclear Energy

journal homepage: http://www.elsevier.com/locate/pnucene

Modeling of atmospheric dispersion and radiation dose for a hypothetical

accident in radioisotope production facility
Hesham Elkhatib a, *, Mohammed A. Awad a, Mahamed A. El-Samanoudy b
a
Nuclear Reactors Departement, Nuclear Research Center, Egyptian Atomic Energy Authority, Cairo, Egypt
b
Faculty of Engineering, Ain Shams University, Cairo, Egypt

A R T I C L E I N F O A B S T R A C T

Keywords: Atmospheric dispersion modeling and radiological safety analysis is performed for public outside radioisotope
Modeling production facility (RPF) in case of hypothetical radioactive Iodine spilling and leakage from hot cell. Potential
Hypothetical human error is expected and column that occupies Iodine may be broken causing it to spill on the radioactive cell
Hotspot
floor. Ventilation exhaust system is dedicated to extract dispersed material through dedicated filters before gases
Radioactive release
Filtration
expelled outside the facility. Two scenarios are performed in this paper; the first one is predicting the dispersion
Effective dose equivalent with good filtration from extract ventilation system, while the other is with loss efficiency of filtration compo­
nents. The spilled radioiodine is the source term, and the HotSpot 3.1, Health Physics code designed by LLNL
(Carestiaet al, 2016) is used to provide required calculation tool to assess and evaluate emergency situation
including radioactive nuclides release to illustrate the transport modeling which is then applied to estimate the
total effective dose equivalent (TEDE) in different Pasquill stability classes, and how it would be transferred to
human body depending on downwind distance and radionuclide activity. The adopted methodology uses
dominant site-general meteorological data and theories of dispersion models to study the impact of hypothetical
dispersion and release to the environment from the selected radionuclide and assess how such dispersion may
have radiological worse impact on public.

1. Introduction doses were calculated for different environmental stability classes, with
average site-specific weather conditions.
Total Effective Dose Equivalent (TEDE) and Committed Effective D. Di Giovanni et al. (Di Giovanniet al, 2014) illustrates the impacts
Dose Equivalent (CEDE) are main variables to identify risks in humans. of two various cases of intentional release by modeling of different
Total effective dose equivalent is a Nuclear Regulatory Commission scenarios. The first case studies the release of Cs-137 from chimney of
(NRC) term that combines the effects of both the internal and external cement production industry, while the other studies the release of
exposures. It is the sum of the deep dose equivalent (DDE), dose to the Strontium 90 from the explosion of a dirty bomb. A simulation of a
skin to a depth of 1 cm, from external gamma radiation and the CEDE Radiological Dispersal Device (RDD) involving Cesium-137 is consid­
from internal exposure. ered and analyzed by V. M. Bulhosa1 et al. (Bulhosa1et al., 2017),
Accidents in spent nuclear fuel reprocessing plants are a major coupling the outputs of the HotSpot Health Physics Code to evaluate
problem for the safety of individuals, operators and the environment, in Accidents Involving Radioactive Materials and the epidemiological
addition to the plants security. Mariachiara et al. (Carestiaet al, 2016) equations of the Radiation Effects Research Foundation (RERF) to sup­
demonstrates the potential to use a free license code to model radio­ port the decision making process. The results of the modeling is used to
logical diffusion after an accident in these types of plants, which permits help quantify the number of general individuals allocated to areas of
defining routes of escape for the workers possibly involved in the fallout. higher radiological risk and of interest to medical care by providing the
Atmospheric dispersion modeling and radiation dose prediction were scientific data to develop a more appropriate approach to risk and its
conducted by S. S Raza et al. (Raza and Iqbal, 2005). Committed communication to the affected population.
Effective Doses (CEDs) to the people at different downwind locations Atmospheric dispersion simulation and radiation dose calculations
were predicted using a health physics computer code HotSpot and the were performed by Bo Cao et al. (Caoet al, 2016) for a hypothetical

* Corresponding author.
E-mail address: [email protected] (H. Elkhatib).

https://doi.org/10.1016/j.pnucene.2021.103674
Received 27 August 2020; Received in revised form 22 January 2021; Accepted 1 February 2021
Available online 9 February 2021
0149-1970/© 2021 Elsevier Ltd. All rights reserved.
H. Elkhatib et al. Progress in Nuclear Energy 134 (2021) 103674

AP1000 Steam Generator Tube Rupture (SGTR) accident using HotSpot significant difference compared to other areas in Cairo. The average
code 3.03. TEDE, the respiratory time-integrated air concentration, and wind velocity is 3–5 m s¡1 during non-monsoon. The maximum wind
the ground concentration are predicted for different atmospheric sta­ speed during considered period is in March with the reading of 6.8 m
bility classes, and the released radionuclides might be transported to s¡1. March is an inter-monsoon month, where the wind direction turned
long distances but have not any bad impact on the public. from north direction to the opposite direction. As illustrated in Fig. 1
The release investigation of nuclear power plant candidate site CS1 is below, the most redundant wind orientation in site blows towards south
performed by S. D. Shamsuddin et al. (Shamsuddinet al, 2017), using from north direction, throughout the considered year. Table (1) shows
HotSpot software based on Fukushima Accident in 2011 and it illustrates the Meteorological Pasquill stability classes’ description in RPF site A, B,
that the radionuclide dispersion can reach as far as 80 km from the site. C, and F while other classes are not present. Stability class F is the most
However, estimated yearly effective dose is not more than 1 mSv limit, prevailing class and occupies 41.8% of the distribution. Stability con­
which is considered acceptable in the point of view of radiological dition classes A, B, and C take the percentage values lower than class F
health risk for human and the environment. Fagner C. et al. (Rotheret al, and have values 19.82%, 24.99%, and 13.49% respectively.
2016) demonstrated a convergence of three different capabilities to
evaluate risk and help rapid safe resource efficient response. The com­ 2.1. Radiation dose and release calculations methodology
bined methodology supported procedures of risk reduction and resource
allocation through work checklist, the designation and entity of At the time of the postulated accident, radionuclide inventory was
response teams, and site location for operation. In the illustrative RDD calculated by Monte Carlo MCNP code and the integrated source term
scenario, the contaminated area for protection, evacuation, and for each radionuclide was then estimated. Next, the Atmospheric release
long-term public concern was strongest for calm and steady weather is accounted by using the general Gaussian plume model in HotSpot
conditions, whilst close-quarter responders faced highest dose rates for code, a Health Physics code from Lawrence Livermore National Labo­
neutral atmospheric conditions. Generally, the risks to women re­ ratory used as a computational tool to perform the atmospheric trans­
sponders were available to be extremely more than for men, and the port modeling. As fission products are released to the atmosphere, they
risks to 20-year-old responders were three times that of their 60-year-old escape from the filters at an air flow rate of 140 m3 h− 1, in case of loss of
counterparts for similar exposure because the risk depends on the in­ filtration and picked by the wind where they are immediately trans­
terval period between age of free risk exposure (amin) and maximum age ported from the release stack point to downwind receptor locations,
(amax), so the higher the ages at initial exposure the shorter the lifetime towards the predominant wind direction, and following the specified
segment up to old age (World Health Organization, 2013). About 160 meteorological data of the site. Radiation dose calculations are then
PBq of I-131 and aerosol were released into the atmosphere during the conducted and final results assessed.
accident at the Fukushima Dai-ichi Nuclear Power Plant (Lebel et al,
2016). These releases were discovered and measured by monitoring 3. Source term in normal operation
stations throughout Japan, and from these measurements, key insights
can be made about the various chemical forms that were released, beside Approximate activities emitted from general process in facility per
persistence of each in the environment. Radionuclides were discovered year are predicted yearly. Table 2 (Raza and Iqbal, 2005) below shows
from the Fukushima nuclear accident at Fukuoka, Japan, 1000 km west the maximum allowable radioactivity released into the atmosphere by
of the Fukushima Dai-ichi nuclear power complex (Momoshima, 2011). each radionuclide released annually during the normal operation of the
Iodine-131 was first discovered 3 days after the accident depicting that it Facility. The table is built taking in consideration the representative
was possibly transported depressively because of local weather condi­ nuclides and assuming the continuous release of one nuclide at a time.
tions, and not general air circulation. The maximum concentrations,
5.07 MBq m− 3 for I-131, 4.04 MBq m− 3 for Cs-134 and 4.12 MBq m− 3 for 4. Iodine hot cell
Cs-137, were recorded in particles collected on April 6, 2011. Gaseous
I-131 accounted for 30%–67% of the total I-131 content. An atmospheric The Iodine 131 Hot Cell is designed for the weekly production of 10
accidental dispersion calculation is predominantly based on Gaussian Ci of carrier-free I-131 solution and its further fractionation and cap­
plume Hotspot code (Abdelhady, 2015). The impact of the dedicated sulation. The tightness of the cell prevents leaks to the outside. The inner
filters in reducting the radiological consequences following an acci­ volume of the radioactive cell is divided into two sections, one taken up
dental release of radionuclides was performed for all probable stability by the production sector proper, and the other by the fractionation and
classes and the results illustrate that the total effective dose reduced by capsulation equipment. Sidewalls are built in aluminum, the floor in
90% in case of good working of filtration. The radiation dose estimation stainless steel, the partition wall in acrylic, and the roof in aluminum,
and radiological consequences of a hypothetical AP1000 SGTR accident with acrylic windows for lighting purposes. The ventilation system of
by HotSpot code 3.03. After the hypothetical AP1000 SGTR accident, the I-131 cell is equipped with special nuclear grade filters. It’s venti­
TEDE the respiratory time-integrated air concentration, and the ground lation system has been connected to the central ventilation system for
deposition are calculated. A procedure calculation combined two real­ other cells and glove boxes.
istic potential scenarios in the case of a terrorist attack (Pereiraet al., The floor of every airtight box has doors joining the box and the hot
2016), involving the explosion of a dirty bomb in the Olympic Village corridor. The Active Solid Element Transfer Equipment is placed there to
during the Rio 2016 Olympic Games. Two representative radionuclides transfer materials. The system is designed to guarantee the presence of
were considered, Cs-137 and Am-241, using the US NARAC HotSpot shielding and tightness during the operations. The design and con­
code and data obtained from ICRP 60, showing the maximum effective struction of this system follows criteria equivalent to those applied to the
dose distance of 10 m from the explosion spot, and the maximum Mo-99 cells. The activated carbon filter performance curve (Farr, 2012)
Effective dose of 8900 mSv for Am-241. is illustrated in Fig. 2 which depicts increase of pressure loss across the
filter. This filter is dedicated to adsorb released radioactive Xenon and
2. Metrological conditions of the RPF site Iodine in case of emergency situations. All components exposed to these
types of gases are equipped with these types of filters to clean expelled
Generally, facility site has warm and sunny in most daylights, while air before exit to environment. Pressure deferential manometers are
its nights are moderate the whole year, with low potential and proba­ attached on filters to measure and monitor the pressure drop develop­
bility to rain through the year. Temperature in site ranges from 23 ◦ C to ment with time.
41 ◦ C, with humidity more than 40 percent. The investigation of mete­ The pressure difference across the filter increase means the blocking
orological information for the site (Barakatet al., 1992), shows no of the filter media and it should be replaced in case of realizing

2
H. Elkhatib et al. Progress in Nuclear Energy 134 (2021) 103674

Fig. 1. Wind rose in Google earth.

Table 1
Meteorological Pasquill stability classes description in RPF site.
Stability Class Wind speed m/sec. Percentage

A 3 19.82%
B 4 24.99%
C 7 13.39%
F 3 41.80%

Table 2
Release limit for the different radionuclides.
Isotope Activity (Ci) Isotope Activity (Ci)

Kr-83m 3.93E-02 Xe-133 2.66E+01

Kr-85m 2.51E-03 Xe-135m 5.40E-01
Kr-85 2.77E-01 I-135 2.19E-02 Fig. 2. Filter performance curve.
I-131 5.66E-02 Xe-135 6.75E+00
I-132 1.23E-01 Xe-131m 2.70E-02
I-133 1.11E-01
Table 3
Activity released per batch from gas storage tanks to the charcoal beds.
maximum pressure drop recommended by the manufacturer. The zero Isotope Activity (Ci) Isotope Activity (Ci)
pressure difference means cutting media of the filter and the projectile Kr-85 9.99E-02 Xe-133m 3.7
air flows freely through the filter without any filtration from the media Xe-131m 5.35E-01 I-131 6.37
and the filter should be replaced. The dedicated ventilation keeps the I-132 6.52E-02
environment within the I-131 hot cell at lower pressure with respect to
the outside environment, and provides the necessary air renewal. The
4.1. Source term in abnormal operation
charcoal filter is designed to act as first containment of the I-131 that
may be eventually released during the process. The system is connected
As the Iodine-131 can be produced by fission every two weeks and
to the central cell ventilation system, which has a second containment as
about 30 Ci is transferred to cell. Human or operational error may occur,
charcoal filter bank. The I-131 Hot Cell release or exposure event rep­
causing most of these activities to be spilled and then constitutes the
resents an unlikely incident with low probability that could be registered
source term. The dedicated ventilation system extracts this activity
when certain uncommon sequences of events take place. Due to that the
through filters which act also as a radioactive source. The HotSpot
design of the I-131 hot cell meets all safety requirements (see Table 3).
models the release results from this accidental situation and assessment

3
H. Elkhatib et al. Progress in Nuclear Energy 134 (2021) 103674

is presented while filtration is available as well as its unavailability. u(z): Wind speed at reference height z
u(H): Wind speed at height H
5. Mathematical model p: Factor from depends on stability class.
Buoyant plume rise is calculated according to relation of buoyancy
The source part is the scalar amount of radioactive dispersed mate­ flux, F below:
rial, in grams or curies, released to the air. The initial source part is the ( )
Ta
amount of radioactive dispersed materials driven by air at the accident F = R2 * g * u* 1 − (4)
Ts
source. The initial respirable source part is the amount of radioactive
nuclides driven by air at the accident source that is effectively inhaled by
where:
respiration organ. Lesser source terms are determined by applying
G: Gravitational acceleration.
filtration or deposition factors to the initial source term.
R: Stack radius.
The airborne pathway is of major interest for non-reactor nuclear
Ta: Ambient air temperature.
facilities. The dose from the inhalation process pathway represents the
Ts: Stack effluent temperature.
overall dose. The airborne source term is typically calculated by the
The effective release height, H, due to buoyancy plume rise is
following five-component linear equation (Pasquill, 1961):
determined as follows for different atmospheric stability classifications
Source ​ Term = (M)AR × (D)R × (A)RF × (R)F × (L)PF (1) (Briggs, 1975)

1.6*F 3 *X 3
1 2

where. H=h+ (5)

(M)AR = Material-at-Risk (curies or grams). u(h)
(D)R = Damage Ratio.
where.
(A)RF = Amount of releases airborne (Rate of released airborne).
X: Distance associated with final plume height.
(R)F = Respirable Fraction, and.
The effective release height due to momentum plume rise is deter­
(L)PF = Leak path Factor.
mined as follows, for atmospheric stability classifications A, B, C and D,
Radioactive dispersion and deposition in the ambient is usually
(Unstable to Neutral) (Briggs, 1975)
modeled and analyzed using Gaussian plume equation. HotSpot uses this
equation to project the dispersion and deposition of radionuclide in the 6*u*R
H=h+ (6)
ambient. The Gaussian dispersion model (Pasquill, 1961) is shown in u(h)
equation below
( ( )2 )[ ( ( )2 ) And for stability class E and F,
Q 1 y 1 (z − H)
C(x, y, z) = exp − exp − ( )1
2.π .σy .σz .u 2 σy 2 σz C 3 −
*S (7)
1
( ( )2 )] [ H = h + 1.5 6

1 (z + H) λx ] u(H)
+exp − exp − DF(x) (2)
2 σz u
where,
where. C = 0.25(2*R*u)2
C: Concentration at a given position.
Q: Source emission rate. S = 0.000875 and 0.00175 for stability classes E and F.
H: Effective release height, The effective release height is estimated based on above considered
y: Crosswind direction, equation and equal 31.85 m.
x: Downwind direction,
z: Vertical direction 6. Results and discussion
λ: Radioactive decay constant (s− 1).
DF(x): Plume Depletion factor. After the hypothetical accident in Iodine hot cell, TEDE, particles
While σy and σz are the standard deviations of the concentration concentration, and deposition of particles on ground are predicted by
distribution and frequently referred as dispersion or diffusion coefficient HotSpot 3.1 as it depends on downwind distance and performed in
which indicates the propagation of the plume in the y and z directions Figs. 3 and 4. It can be seen from results that ground surface deposition,
respectively; Normally, the values of σy and σz are calculated based on and TEDE are very minimal, and the highest value of TEDE is 4E-05Sv
the stability of the atmosphere and they increase as the distance x in­ and maximum ground deposition value is 109 kBq m¡2 occurred at
creases from the source.

5.1. Effective release height

Plume rise can occur because of the wind velocity of on the stack top
position and resulted buoyancy from air expelled from the stack emis­
sion, but the temperature differential between the stack effluent and the
surrounding air is neglected. The rise of the plume results in an increase
in the release height.
H = h + Δh

where.
H: Physical tack height.
Δh: An increase of the plume height due to velocity and buoyancy
( )p
H
u(H) = u(z) (3)
z
Fig. 3. TEDE variation vs. downwind distance from release point.

4
H. Elkhatib et al. Progress in Nuclear Energy 134 (2021) 103674

300 m in metrological condition class F. According to figures and

deeming the IAEA dose limits defined by the IAEA-TECDOC-93, it can be
noticed that there are almost no hazard from radiation impacts within
surrounding area in the case of stability class F. The TEDE rises sharply
from 0.04 km to 0.3 km and peaked at 2.2E-05Sv. After wards, the TEDE
decreases with the increase of the distance and the rate of slop slows
down gradually until the value has fallen to a very low values. The
maximum TEDE value in case of good filtration and good duty from
ventilation system is still far below the annual permissible limits of 1
mSv for the public as set in IAEA Safety Report Series number 115. That
is to say the released radionuclides might be transported to a long dis­
tance but it hasn’t any bad impacts or consequences on the public.
The prediction in Figs. 5 and 6 shows total effective dose equivalent
and ground deposition in case of filtration failure in ventilation system
which allow the radionuclides to be released outside causing the
maximum values of TEDE move slightly up to 4.5 E− 04Sv and ground Fig. 5. TEDE variation vs. downwind distance from release point.
deposition to 330 kBq m¡2.
Inner: 1.0E-04 Sv (0.017 km2) Middle: 5.0E-05 Sv (0.053 km2) Outer:
5.0E-06 Sv (0.72 km2).
Results above from the simulations in two cases are slightly different
and illustrates the positive function from ventilation system to avoid the
release and protect the public, as with filtration the TEDE predicted in
Figs. 3 and 5 as 4.0E-05 Sv at 0.3 km from the stack while the value is
moved up to 4.5E-04Sv at 0.3 km with filtration failure.
Figs. 4 and 6 show the ground deposition values of Iodine and the
prediction shows that the maximum values in case with filtration and
without filtration are respectively 110 kBq m¡2 and 330 kBq m¡2 at the
same location 0.3 km from stack, and as illustrated from contour plots in
Fig. 7 in case of filtration failure, the area around the center of emission
in vicinity of facility have TEDE value 1.0E-4Sv while it is about 5.0E-
06Sv far from the facility. Fig. 8 depicts clearly the TEDE in various
statues including also the normal operation statues.
Fig. 9 shows the distribution of committed equivalent dose to thyroid
organ in two cases, with filtration and without filtration as it is depen­
dent on downwind distance. Committed effective dose equivalentCEDE Fig. 6. Plume centerline ground deposition vs. wind rose distance.
is a value received by any individual due to settling down at the dedi­
cated area and exposed to the radioactive material release. This amount
is the total fraction of the ground deposition of radionuclides. It is seen
from result that with good filtration the dose received by thyroid is
reduced by amount more than 90% and according to the these results,
the calculated TEDE in the different distance from the facility are
considerably lower than the permissible dose limits of public, which
agrees with national regulatory limits.

7. Conclusion

This study introduced different statues of hypothetical radiological

release to atmosphere due to expected human error considering

Fig. 7. TEDE contour plot without filtration for class B.

atmosphere function of the Pasquill class and wind speed. The evalua­
tion of filtration is studied and compared with no filtration case to show
the annual dose for public. The radioactive material for evaluation was
selected based on data from the process sector. Gaussian plume Hotspot
code is employed to predict the release in all stability classes and
available metrological data. The predicted results show that:

1. In case of normal operation with good filtration, the maximum public

total effective dose offsite is far from dose limit for all metrological
stability classes which a proof of good performance of ventilation.
Fig. 4. Plume centerline ground deposition vs. wind rose distance.

5
H. Elkhatib et al. Progress in Nuclear Energy 134 (2021) 103674

show that they are under permissible committed dose limits but it
found relatively high without filtration in tense of more 90% over
filtration case.

Declaration of competing interest

The authors declare that they have no known competing financial

interests or personal relationships that could have appeared to influence
the work reported in this paper.

References

Abdelhady, A., 2015. Public dose exposure resulting from plate-melt hypothetical
accident”. Arab Journal of Nuclear Science and Applications 48–1, 53–59.
Barakat, M., et al., 1992. Report of Basic Characteristics of MTR Site. National Center for
Nuclear Safety and Radiation Control.
Briggs, G.A., 1975. Plume rise predictions. In: Lectures on Air Pollution and
Environmental Impact Analyses, Workshop Proceedings. American Meteorology
Fig. 8. TEDE of different statues for class B. Society, Boston, MA, pp. 59–111.
Bulhosa1, V.M., et al., 2017. Assessment actions and communication of solid cancer
development risk in scenarios RDD based on computational simulation. In:
International Joint Conference.
Cao, Bo, et al., 2016. Radiation Dose Calculations for a Hypothetical Accident in
Xianning Nuclear Power Plant Science and Technology of Nuclear Installations, vol.
2016. Hindawi Publishing Corporation Science and Technology of Nuclear
Installations, p. 6. Article ID 3105878.
Carestia, Mariachiara, et al., 2016. Use of the HotSpot Code for safety and security
analysis in nuclear power plants. A case study” Environmental Engineering and
Management Journal 15 (4), 905–912.
Di Giovanni, D., et al., 2014. Two Realistic Scenarios of Intentional Release of
Radionuclides Cs- 137, Sr-90 the Use of the HotSpot Code to Forecast Contamination
Extent, vol. 10, p. 106. E-ISSN: 2224-3496.
Farr, Camfil, 2012. Air Filtration Solutions for Nuclear Industry”.
Lebel, L.S., et al., 2016. Radioiodine in the atmosphere after the Fukushima Dai-ichi
nuclear accident”. Journal of Environmental Radioactive 151 (1), 82–93.
Momoshima, N., 2011. Atmospheric radionuclides transported to Fukuoka, Japan remote
from the Fukushima Dai-ichi nuclear power complex following the nuclear
accident”. J. Environ. Radioact. 111, 28–32.
Pasquill, F., 1961. The estimation of the dispersion of windborne material. Meteorol.
Mag. 90 (No), 1063.
Pereira, J.F., et al., 2016. Dirty bomb radiological simulations: two explosion scenarios
Fig. 9. Target Thyroid committed equivalent dose as a function of down­ using the Rio 2016 Olympic Games Athletes” Village as a model. Brazilian journal of
wind location. radiation sciences 6–2, 1–18.
Raza, S.S., Iqbal, M., 2005. Atmospheric dispersion modeling for an accidental release
from the Pakistan Research Reactor-1 (PARR-1)”. Ann. Nucl. Energy 32 (11),
2. In case of loss of dedicated filters the maximum public total effective 1157–1166.
Rother, Fagner C., et al., 2016. Radiological risk assessment by convergence
dose is higher but still less than the limit.
methodology model in RDD scenarios. Risk Anal. 36 (11).
3. Intension should be considered if the facility decides to increase and Shamsuddin, S.D., et al., 2017. Radioactive dispersion analysis for hypothetical nuclear
expand the production from Iodine as the TEDE exceeds the limit to power plant (NPP) candidate site in Perak state” Malaysia. In: EPJ Web of
be 1.2 msv. Conferences, vol. 156.
World Health Organization, 2013. Health Risk Assessment from the Nuclear Accident
4. The committed equivalent doses of thyroid organs are predicted for after the 2011 Great East Japan Earthquake and Tsunami Based on a Preliminary
the two cases, with available filtration and without it and the results Dose Estimation, ISBN 9789241505130.