Page 1
1, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Contribution from Subrata Chakraborty, Ph.D. (Affiliation: University of California, San
Diego)
Contributed to Rep. Mike Levin’s Task Force (Technical Committee) on San Onofre
Decommissioning (Process, Material and Natural Factors)
Disclaimer: I address these questions based on my analysis of relevant peer-reviewed and
published literature and government documents as referred in the document and listed at the end.
Since we are independently evaluating the entire system, these analyses are NOT based on Holtec
provided analysis, I have only taken information from Holtec’s documents when physical
parameters were required. This is my independent research and UCSD is not directly or indirectly
involved. Figures and Tables are sequentially numbered following the question area and question
number (i.e., Figure 1 of Material Science question 1 is numbered as MS-1-1).
Page 2
2, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
MS-1: What are fuel and canister system (including the concrete) failure mechanisms,
processes, and consequences: Embrittlement, cracking/gouging, corrosion (water, galvanic)?
There are several parts to this question and they are addressed separately as follows.
Fuel failure: There are two fundamental reactions that take place in a reactor that determine the
composition of the fuel: (i) fission of fissile nuclides, such as U-235 and Pu-239, and (ii) neutron
capture and sequential -decay reactions that create transuranium isotopes, mainly Pu-239 from
U-238. Hence, the Pu concentration in the fuel increases with time, and fissile Pu-239 provides up
to one-third of the energy generated in a typical light-water reactor (LWR, SONGS had pressurized
water reactor (PWR), that is a variety of LWR). The final composition of the fuel depends on the
fuel type, chemical composition, level of initial enrichment in U-235, neutron energy spectrum
and the extent of fission of the fuel (the burn-up level).
Figure MS-1-1. Understanding the microstructure of the spent nuclear fuel (SNF) after irradiation
inside the reactor. (a) schematic showing the grain boundary, gas bubbles, oxide layers. (b) Cross-
section of a fuel pellet (diameter = 1 cm), the fractures are created by the steep thermal gradient from
the center of the pellet to its edge. Volatile fission product elements segregate in these fractures and
become part of the instantaneous release fraction. (c) Higher-magnification image of individual grains
in a reactor fuel sample, demonstrating the segregation of fission product gases into bubbles that
decorate the surfaces of grains of UO2 in the fuel. Figure taken from Ewing, Nature Materials, 2015.
Page 3
3, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Figure MS-1-1 shows a structure of reactor irradiated fuel [Ewing, 2015] and the inventory is
shown in Figure MS-1-2 [Bayssie et al., 2009]. The initial level of radioactivity of the irradiated
fuel is very high, caused mainly by the presence of the 3 to 4 atom-% of fission products (e.g., I-
131, Cs-137 and Sr-90) and activation products (for example, Co-60, Ni-63), with a longer-lasting
contribution from long-lived transuranium elements (e.g., Pu-239, Np-237 and Am-241). At the
end of the fuel’s useful life in the reactor, about 96% of the spent nuclear fuel (SNF) is UO2. The
remaining balance consists of fission products, transuranium elements and activation products, but
these elements occur in many different phases, and over differing structural length scales: (a)
fission product gases, such as Xe, I and Kr, occur as finely dispersed bubbles in the fuel grains;
(b) metallic fission products, such as Mo, Tc, Ru, Rh and Pd, form immiscible metallic precipitates
(ε-particles) that are nanometres to micrometres in size; (c) fission products form oxide precipitates
of Rb, Cs, Ba and Zr; (d) some fission product elements, such as Sr, Zr, Nb and lanthanides, can
form solid solutions with the UO2 fuel; and (e) transuranium elements can substitute for U in the
UO2.
The distribution of elements is not homogeneous within a single pellet (Figure MS-1-1) because
of the steep thermal gradients that exist (as high as 1,700 °C at the center of the pellet and
decreasing to 400 °C at its rim). Thermal excursions during reactor operation can additionally
cause a coarsening of the grain size, as well as extensive micro-fracturing (Figure MS-1-1).
Volatile elements, such as Cs and I, may also migrate to grain boundaries, fractures and the gap
between the edge of the fuel pellet and the surrounding metal cladding. The extent of burn-up is
also not uniform across the fuel pellet; higher burn-up at the edge of the pellet leads to higher
Figure MS-1-2. Pie chart showing the inventory in spent nuclear fuel-pellets in moderate burn-up
(taken from Bayssie et al., 2009).
Page 4
4, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
concentrations of Pu-239, an increase in porosity and a reduction in grain size of the UO2 grains
(~0.15 to 0.3 μm). Thus, the SNF has complex chemistry and phase distribution, resulting from its
thermal history, burn-up and initial composition. All these factors have to be considered when
understanding the long-term evolution of fuel [Ewing, 2015].
Summary: The SNF is a complex and actively evolving material, with changing radiotoxicity
and chemical toxicity over time and due care required in handling. Based on the available
literature, it does not seem possible to start a self-sustaining chain reaction within the
canister while stored in normal conditions (i.e., would not reach criticality) and while only
the fuel degradation is considered. However, cladding failure could release radioactivity
inside the canister. If canister systems are deemed to be 100% fail-proof, there will be no
issue with the radioactivity release inside the canister as that would not expose the
surroundings. However, as we will discuss, the canister failure issues in the following section,
the release of radioactivity to the environment remains to be a matter of major concern.
Canister failure:
United States Nuclear Waste Review Board nicely summarized the potentials of dry storage
canister failure modes [Rigby and Members, 2010]. This analysis mostly adopts Review Board’s
analysis (with references therein) with the addition of a few recent pieces of literature and focused
only on the outer surface of the canister (Figure MS-1-3). It is mentioned in NRC’s letter [NRC,
2019] that the canister surfaces were laser peened. This is a new technique and reduces the
possibility of stress corrosion cracking (SCC) [Hackel et al., 2018; Sathyajith and kalainathan,
2015]. In this technique, the outer layer (about a millimeter (mm)) goes through plastic
deformation and reduces the stress. However, according to a white paper [MPR_White_Paper,
2018] prepared by MPR, Inc for Edison, the welds of the canisters are all laser peened, and not the
entire canister. This is contrary to the NRC’s letter [NRC, 2019], which states that the canister
surfaces are laser peened and thus creates a protective coating and resist againt scratching and
gouging during the downloading process. This descriptional difference is significant, while laser
peened welds are robust against SCC as potential corrosion site, as the total surface is not peened,
hence any scratch would be a new potential site for SCC.
Page 5
5, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Atmospheric Chemistry Corrosion Environment. A dry-storage nuclear waste cask or canister
exposed to the atmosphere for decades to a few centuries will be exposed to a
corrosiveenvironment determined by virtually all chemical species in the atmosphere and the
effects of gamma radiation. This exposure will be enhanced by the natural convection cooling of
the canister where air, along with dust, will flow along the metal surface. The flow of air will
continually bring dust into contact and deposition with the metal surface, and the dust will contain
chlorides, nitrates, salts, insoluble inorganics, and organics. Gamma radiation will be present and
will result in chemical reactions occurring in the local air and dust; the result will be the production
of a spectrum of chemical species. Hydrogen chloride, sulfuric acid, nitric acid and virtually all
oxides of nitrogen occur in the atmosphere as a result of chemical reactions. These three acids will
react with calcite and dolomite (magnesium, present in dust) and produce the corresponding
chloride, nitrate and sulfate salts. Along with inorganic compounds, organics are ubiquitous in
atmospheric dust. It is documented that biofilms are capable of influencing electrochemical
processes at the metal surface, often leading to the deterioration of metals referred to as
biocorrosion or microbiologically influenced corrosion. Biofilms typically consist of microbial
cells and their metabolic products including extracellular polymers, and inorganic precipitates.
Page 6
6, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Interaction of biofilms and exopolymers with metal ions has long been proposed as one of the
mechanisms of metal biodeterioration [Beech, 2004].
Coupled with the inorganic and organic chemistry, there is a gamma radiation field at the metal
surface of a dry-storage container. This gamma field should be of interest for susceptibility to
stress corrosion cracking
because of gamma-field-induced chemistry changes and there is
observational evidence regarding that. On top of the radiation field effect, it should be noted that
the chemical reaction rates increase with increasing temperature. All these atmospheric corrosion
factors outlined above are potentially viable for SONGS ISFSI canisters and supply the ingredient
for different types of corrosion outlined below with their effects.
Pitting and Crevice Corrosion. Pitting corrosion is a localized form of corrosion by which
cavities or "holes" are produced in the material. Pitting is considered to be more dangerous than
uniform corrosion damage because it is more difficult to detect, predict and design against. A
small, narrow pit with minimal overall metal loss can lead to the failure of an entire engineering
system. There are several pathways for the initiation of this kind of corrosion- (i) Localized
chemical or mechanical damage to the protective oxide film, water chemistry factors which can
Figure MS-1-3. Schematics of canister cooling systems: (left) canister wall natural diffusive cooling
(inlet and outlet air vents are at the top; (right) flow-dynamics of inner canister cooling (figure taken
from Holtec International).
Page 7
7, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
cause breakdown of a passive film are acidity, low dissolved oxygen concentrations (which tend
to render a protective oxide film less stable) and high concentrations of chloride (as in seawater).
(ii) The presence of non-uniformities in the metal structure. Non-uniformities might come from
external treatment of the surface (scratches and gouges).
Crevice Corrosion refers to the localized attack on a metal surface at, or immediately adjacent to,
the gap or crevice between two joining surfaces. The gap or crevice can be formed between two
metals or a metal and non-metallic material. Outside the gap or without the gap, both metals are
resistant to corrosion. Crevice corrosion is initiated by a difference in concentration of some
chemical constituents, usually oxygen, which set up an electrochemical concentration cell. Outside
of the crevice (the cathode), the oxygen content and the pH are higher - but chlorides are lower.
Chlorides concentrate inside the crevice (the anode), worsening the situation. Ferrous ions form
ferric chloride and attack the stainless steel rapidly. The pH and the oxygen content are lower in
the crevice than in the bulk water solution, just as they are inside a pit.
Stress corrosion cracking (SCC). SCC in metals can occur from the combined effect of tensile
stress and the presence of a water-soluble chloride salt. Bulk liquid water need not be present
because an aqueous environment will always exist on metal surfaces exposed to the atmosphere
due to sorption of moisture from the air. This moisture sorption phenomenon is sometimes called
"physisorbed" water. This type of corrosion also is referred to as chloride-induced stress corrosion
cracking. SCC is considered corrosion with local slip at the crack tip and is often found to initiate
where pitting or crevice corrosion has occurred. Air cooling of stainless steel canister walls means
that the wall is in direct contact with moist air that will contain salt, including chlorides. Austenitic
stainless steels are susceptible to chloride stress corrosion cracking starting on the outside surfaces
in certain humid marine air environments under tensile stress. Residual tensile stresses in the
storage canister are mostly derived from welding between the wall and lids and cold working of
the metal. As the temperature of the canister decreases, condensation can occur resulting in the
Page 8
8, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
accumulation of corrosive media such as chlorides, particularly in marine or industrial
environments.
A metal canister’s corrosion lifetime for CSCC corrosion can be estimated as the time needed for
the canister temperature to decline to the point where condensation is possible, which can be above
100ºC, plus the time for SCC corrosion to initiate, plus the time needed for SCC corrosion to
propagate through the canister wall thickness. Figure MS-1-4 shows the predicted rate of
propagation of cracks once formed in stainless steel (SS316) based on measured values [Ilgen et
al., 2015].
A series of natural exposure and accelerated corrosion tests of conventional stainless steel used
spent fuel canisters were conducted recently by the Central Research Institute of Electric Power
Industry in Japan [Kosaki, 2008]. Natural exposure tests were conducted at Miyakojima, an island,
one of the most corrosive areas in Japan, and accelerated corrosion tests were conducted in an
environment filled with NaCl steam mist at 60ºC and humidity of 95 percent. One set of
experimental tests on types 304 and 316 stainless steel yielded CSCC corrosion initiation times
ranging from about 1.6 to 3 years under natural exposure conditions. Pitting or crevice corrosion
Figure MS-1-4. SCC propagation rates for atmospheric corrosion of 304SS and 316SS. Time to
failure corresponds to the time required to penetrate a 0.625' thick canister wall (llgen et al., 2015).
Page 9
9, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
was found to be a trigger to CSCC because CSCC initiation was found to start from the bottom of
the corrosion area by pitting or crevice corrosion. Under natural exposure conditions, the CSCC
corrosion rate varied from about 0.04 to 0.6 mm/year over a range of residual tensile stresses,
which would take from about 25-375 years to penetrate through a thickness of thin-wall canister
(~0.6 inches).
Cooling system failure via vent blockage. One of the selling points of dry cask storage is its
passive cooling system. The outer surface is being cooled via the diffusive flow of marine airmass
of ambient temperature through inlet vents and relatively hot air exit from the outlet vent (location
of both of these vents are located at the top as shown in Figure MS-1-3). There is a significant
chance of clogging the vents (partially or fully). There is genuine potential for such situations.
There are screens installed in the vents to prevent clogging of the vents and the latest FSAR [FSAR,
2018] proposed mitigation of this situation by cleaning the screen as quickly as possible. This
mitigation strategy only works if the screens accumulate debris over time. Presence of a big
tsunami wave or even mud-slide (as this is a flood basic as mentioned in the geology section) a
huge amount of mud and silt/sand could get pass these screens and accumulate at the bottom of
the UMAX hole and could potentially block the airflow system (partially or fully). This kind of
scenario is not out of question as the region saw extreme landslides in the past decades (especially
during 1997-’98 El-Nino event as described in the Geology section). Such extreme but likely
scenarios are not addressed in FSAR [FSAR, 2018] and if such an event does occur that would
block the air vents and simple cleaning of screen would not be effective mitigation. There are
modeling studies to evaluate the dry cask cooling/ventilation system, one of which shows that
clogging the outlet vents is more crucial than clogging the inlet vents and without the proper
cooling, the fuel temperature might rise above the threshold point (~673 K) degrading the cladding,
Thermal simulation of a container with SNF accident conditions shows that the maximum
temperature can be higher than safety criteria limits [Alyokhina, 2018].
Helium leakage from the steel canister. According to the standard operational procedure, after
loading SNF inside the canister under thespent fuel pool water, canisters are thoroughly dried and
filled with 2-atmospheric pressure of helium and welded shut and leak tested. There always
remains a practical possibility for an undetected leak (tiny and under the detection limit). Also,
there is always a chance of development of micro-level leakage in the canister. Any leak in the
Page 10
10, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
canister would result in helium loss, which is the active material inside the canister to maintain the
effective heat transfer from the spent fuel rod to the canister body [Penalva et al., 2017]. Therefore,
any leaks from the container would result in a cooling degradation (might be over a few years of
time scale) of fuel that might jeopardize fuel integrity if the temperature exceeded a threshold
value.
Summary: The canister failure could occur in several ways- corrosion cracking and air inlet
blockage are the two major ones. The research shows that even SS316L is not corrosion-
resistant. The welds of the canisters used in SONGS are treated with laser peening and that
process is supposed to modify the weld-surfaces and improve the performance against stress
corrosion cracking. However, the entire body of the canisters are not treated via this process,
the scratches and gouges created during the downloading process might create the potential
sites for pitting crevice corrosion. Althrough these canisters are exposed to the corrosive
marine environment and thus prone to atmospheric corrosion. One of the major ways of
blocking air cooling vents is through the flash flood (if not unexpected Tsunami waves).
These events carry silt and sands and the present first line of defense to protect the air vents
are the installed screens, which are inadequate for such debris. In such an event the debris
could fill the bottom of the UMAX cavity and block the airflow. The research shows that the
air vent blockage could potentially raise the temperature of the canister exceeding the safety
limit. There is no mitigation strategy existing in Edison’s FSAR, as the FSAR does not
include such scenario. The SONGS’ ISFSI is located in a flood plain and in the past decades
such an event took place (during 1997-’98 El-Nino year) and must be considered. No study
was found which calculates the internal pressure of the canister due to fuel failure and fission
gas leakage and hydrogen desorption inside the canister, thus it was not possible to perhaps
state the status of the canister if the design criticality is reached due to an off-normal
condition. Moreover, these canisters are welded shut with at least 2-atmosphere of helium.
Though initially leak checked, any undetected leak (tiny and under the detection limit, initial
or post-developed) of the canister would yield a helium loss, which is the active material
inside the canister to maintain the effective heat transfer from the spent fuel rod to the
Page 11
11, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
canister body. Thus, potential leaks from the container would result in a cooling degradation
of fuel that might jeopardize fuel integrity if temperature exceeded a threshold value.
Concrete failure:
United States Nuclear Waste Review Board nicely summarized the potentials of dry storage
canister failure modes [Rigby and Members, 2010]. I mostly adopted their analysis (with references
therein) with the addition of a few recent pieces of literature.
Concrete is used in the overpack of the canisters and in the foundation pads for the stored canisters.
SONGS ISFSI is an underground system and basically a concrete block with cavities for canister
storage. The degradation of concrete is caused by many physical and chemical mechanisms, some
of which act in concert. The age-related mechanisms may degrade concrete and reinforcing steel
during dry storage period, in a couple of decades. Other degradation mechanisms may be
experienced by concrete in contact with the ground, such as the foundation pads, because of
groundwater and chemicals in the soil as outlined below.
Elevated temperature. Concrete that is exposed to high temperatures will lose moisture and
eventually can induce thermomechanical destruction of both the internal cement structure and the
bonds between cement and aggregate. Undesirable chemical reactions between the cement and
aggregates also can occur. All of these types of degradation significantly decrease concrete
strength, elastic stiffness, and toughness. For example, at 150ºC (300ºF), the mean concrete
strength is about 80 percent of its normal value. Generally, the threshold of degradation in the
concrete is at a temperature range of 66ºC to 95ºC (150ºF to 180ºF). Because the concrete in
SONGS ISFSI will reach the temperature in this range [FSAR, 2018], the concrete would need to
handle such temperatures, as well as conditions of natural cooling mechanisms failure. Exposure
to elevated temperature could be a significant age-related degradation mechanism for the concrete
enclosures. Creep of concrete at an elevated temperature, under a sustained load could have an
adverse effect on the concrete structure over time. Furthermore, if air vents are plugged because of
some natural or human-caused off-normal scenario, concrete temperatures could rise significantly
above the normal operating range. Degradation of concrete would affect its strength and might not
be able to withstand earthquake-related shock.
Page 12
12, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Freeze-Thaw Cycles. In warm climates, such as at SONGS, this degradation mechanism will not
be effective.
Leaching of Calcium Hydroxide.
Climates with significant rainfall provide a source of flowing
water on concrete surfaces and in cracks or joints that can leach out calcium hydroxide (white
hydrated lime) from the concrete. Removal of calcium hydroxide can significantly weaken
concrete mechanically and leaching over long periods increases the porosity and permeability of
concrete, making it more susceptible to leaching and freeze-thaw damage. This kind of damage is
also not of importance for SONGS.
Chemical Attack. Chemical attacks on concrete derive from hydrated carbon dioxide in the air
and hydrated sulfates in the soil. Both can lead to local acid solutions in concrete that react with
the high alkalinity calcium hydroxide (lime) of concrete (pH > 12.5) to convert the calcium
hydroxide to calcium carbonate (a process called “carbonation”). Fully “carbonated” concrete
results in concrete pH levels slightly above 8. The carbonation progression inward in concrete
proceeds at a slow but fairly steady rate, on the order of 1-2 mm per year.
Whereas carbonation
tends to decrease porosity and increase concrete density and strength, sulfate attacks usually
increase the porosity and permeability of concrete and can lead to decreased density and
compressive strength. Sulfate reactions are accompanied by expansive stress within the concrete,
which can lead to spalling, cracking, and strength loss.
The presence of chlorides also lowers the
pH of concrete and so can lead to corrosion of reinforcing steel. For concrete that is obscured from
view, soil sulfate attacks may not be noticeable until damage is significant, which is one of the
concerns for the underground storage that is being employed at SONGS.
Corrosion of Reinforcing Steel. For the reinforcing steel bars in concrete to corrode, diffusion of
both chloride ions and carbon dioxide into the concrete is required at particular concentration
thresholds. The high alkalinity of concrete (pH >12.5) protects the embedded reinforcing steel
from corrosion. However, when the local pH near the steel is decreased below the threshold level
of about 11.5, then chemical corrosion initiates more easily. Initiation of chemical rebar corrosion
also requires a chloride threshold concentration, which is on the order of 0.06 percent by weight
of concrete for black steel. The pH reduction occurs by the intrusion of aggressive ions, primarily
chlorides, in the presence of oxygen. The transport of chloride ions in intact concrete occurs by
ionic diffusion according to Fick’s Second Law of Diffusion. This means that the higher the
Page 13
13, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
chloride concentration in the seawater next to concrete, the faster the chloride ion diffusion will
occur in the concrete. Accordingly, concrete surfaces that are continually exposed to salty marine
or industrial chemical environments are most at risk, particularly if cracks are present. Calcium
chloride accelerates the corrosion more than sodium chloride. In addition to the corrosive agents,
the severity of corrosion is influenced by the quality of concrete (cement type, properties of
aggregates, and moisture content), depth of concrete cover over steel, and the permeability of
concrete.
The key outcome of this degradation mechanism is the creation of steel corrosion products that
cause high swelling pressures within the concrete, which often leads to concrete spalling and
cracking. Once rebar corrosion initiates, significant concrete near-surface damage can proceed
quickly (on the order of months or a few years). This minor spalling and cracking usually will not
affect the structural performance of the reinforced concrete significantly until a critical level of
rebar tensile strength is lost. The service life of concrete structures where this is the dominant
degradation mechanism, and it often is, can be predicted by models based on the chloride ion
diffusion. Because rebar is hidden from view by the concrete cover, any initiating degradation of
concrete structures by this mechanism can be detected by making electrochemical measurements
of the reinforcing steel that indicates corrosion, if this capability was designed in advance (none
of the current dry-storage systems relying on concrete have this capability).
The reaction of Aggregates with Alkalines. The alkaline aggregate reaction is a common
concrete-degradation mechanism identified by an irregular cracking pattern on the surface of
concrete. The mechanism can be greatly minimized by a careful selection of aggregate that does
not contain reactive materials. Two types of reaction, alkali-silica and alkali-carbonate have been
identified. Moisture must be available for the chemical reactions to occur. Thus, concrete that is
either consistently wet, or that experiences wet and dry periods, is susceptible.
Creep and Shrinkage. A normal part of concrete behavior, as the concrete dries out, is that it
shrinks in volume, inducing tensile strains and, possibly, shrinkage cracks. Most of the shrinkage
(98 percent) typically occurs during the first few (e.g., five) years of service. The significance of
a shrinkage crack as a potential contributor to degradation depends primarily on its size and
environmental exposure conditions. A crack can allow aggressive agents access to the reinforcing
steel, promoting the possibility of corrosion.
Page 14
14, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Irradiation. The degradation of concrete exposed to neutron and/or gamma radiation is manifested
in many ways. Fast and slow neutrons usually cause aggregate growth, decomposition of water,
and warming of concrete. Gamma radiation affects the cement paste portion of the concrete,
producing heat and causing water migration. The degradation due to nuclear heating and water
loss is more serious than degradation associated with direct radiation damage. This is because
nuclear heating causes the free water within the concrete to evaporate, and both the neutron
shielding, and structural characteristics of the concrete become impaired. As a consequence, the
concrete could experience a decrease in its strength (compressive, tensile, and bonding strengths)
and stiffness (modulus of elasticity) from shrinkage and cracking if the thermal gradient is
excessive. According to the American National Standard ANSI/ANS-6.4-1985, nuclear heating
can be neglected if the incident energy fluxes are less than 1,010 MeV/cm2sec. Under typical dry-
storage conditions, the energy flux is negligible and decreases with time, making this aging
mechanism insignificant.
Summary: As is evidenced by deteriorating concrete structures, atmospheric corrosion and
degradation of concrete structures do occur. The additional influence of heat and radiation
damage can compound environmental damage. Usually, monitoring the condition of
concrete overpacks to identify damage before it becomes significant will be important. Once
the damage is found, a follow-up maintenance program will correct the damage and help
minimize future degradation problems. However, the design of ISFSI at SONGS is such that
visual inspection of the concrete is not possible, as it is an underground system and it is a
block structure, which deters from any kind of inspection. There are already reported
observations of visible metal and concrete degradation of dry-storage systems in areas near
the sea in different storage locations. The dust and chemical species that comprise an
atmospheric-corrosion environment can affect several different mechanisms. The heating
degradation of concrete is more severe for SONGS because of its block structure. The
independently standing concrete overpacks would have better heat dissipation efficiency
than the underground systems as that of SONGS.
Page 15
15, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
MS-2: Current knowledge on Hydride formation, hydride reorientation, H2 formation,
radiation damage of the fuel rods and cladding
The US Nuclear Waste Technical Review Board’s 2010 assessment of dry storage [Rigby and
Members, 2010] is comprehensive documentation on this subject. The present description and
analysis are heavily based on that review and updated accordingly with the most current literature
data. This seems to be a complex subject and many research papers available [Adrych-Brunning
et al., 2018; Cha et al., 2015; Cha et al., 2018; Chu et al., 2008; Colas et al., 2013; Jang and Kim,
2017; Y-J Kim et al., 2015; Lee et al., 2018; Min et al., 2014; Ronald Adamson, 2017; Won et al.,
2014]. There are several mechanistic pathways for fuel and cladding degradation of the spent fuel
during dry storage as shown in Figure MS-2-2.
The Zirconium alloy cladding (the outer coating of the fuel pellets) goes through two heating-
cooling cycles. It gets hot when in the reactor, at that time H2 (produced in the surrounding) get
dissolved in this cladding as the fuel comes out from the reactor to the pond, it starts cooling from
400 °C to 50 °C. During that time hydrogen precipitates as hydride in the outer edge of the circular
cladding. This is the first cycle. During placing the fuel for dry storage, the temperature goes up
very quickly and after the cask is welded-shut, the temperature starts to decrease very slowly, at
this cooling phase, the formed hydrides reorient itself. How much reorientation will take place
depends on several parameters: dissolve hydrogen amount, cooling rate, terminal temperature,
Figure MS-2-1. Sketch showing the mechanisms effecting spent fuel cladding in dry storage (taken
from US Nuclear Waste Technical Board’s Review document (2010).
Page 16
16, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
H2/O2 ratio and neutron damage during its
lifetime. All these factors are
experimentally proven, and the maximum
damage of the cladding happens after 5
years of dry storage as shown in Figure
MS-2-1 [Adrych-Brunning et al., 2018;
Cha et al., 2015; Jang and Kim, 2017; Y-J
Kim et al., 2015; Min et al., 2014; Won et
al., 2014]. Additionally, with increased
burnup, more corrosion-produced
hydrogen will lead to more hydrogen
being absorbed by the cladding. More
hydrogen pickup will lead to more hydride
precipitation and possibly other effects
such as more embrittlement, delayed
hydride cracking, and acceleration of the
corrosion rate [NRC, 2007a]. Many of
these factors may significantly affect the
physical state of the used fuel and cladding
during dry storage.
What happens when hydrides reorient radially? It becomes brittle. Ductility decreases
significantly and technically the fuel is damaged at that point. Since the reorientation depends on
so many factors and not every canister is the same, the reorientation of hydrides would be different
for different canisters and it is hard to make sure that hydride orientation has not taken place
(individually for each canister).
Figure MS-2-1. Micrograph showing orientation of
hydrides in Zircaloy-4 cladding of 230 ppm H2: (a) as-
hydride and (b) after 8 cycles of thermal treatment
(taken from Chu et al., 2008).
Page 17
17, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Table MS-2-1. Cladding degradation effects. The table summarizes the degradation effects of
cladding materials during prolonged dry storage and each degradation effects are also described
in the table [Spykman, 2018].
A quote from NRC document- CNWRA-2012-001 [C NRC, 2012], “No breach of canister was
considered in this analysis, following a cladding breach, internal gas pressures within SNF rods
represent a driving force for the release of gas and fuel particulates into an SNF canister. Initial
and damaged states of SNF, including fuel pellet fracturing. Because the breach sizes and number
of breach sites are small compared to the total surface area of cladding, pathways for the flow of
gas and particulates out of SNF rods are confined, and aerosol dynamics considerations are
necessary to estimate the extent to which SNF particulates, generated within the fuel rod cladding,
can be released from the fuel rods into the canister. Fission product gas and SNF particulates can
both contribute to the radiological source term”.
The formation of hydrites in the zirconium alloy cladding is very common and and in some
condition these hydrides will reorient in the radial direction and that would alleviate creep
deformation and eventually make the cladding material brittle [Rigby and Members, 2010]. The
above description (from NRC) of cladding breach would lead to radiological leaks inside the
Page 18
18, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
canisters as NRC have not considered any canister breach [C NRC, 2012]. Another issue
intertwined with the degradation of zirconium cladding is the allowed peak temperature, because
the tensile hoop stress of the cladding increases with the extreme thermal history of the fuel. Figure
MS-2-3 explains the relationship between the hoop stress and temperature [Kook et al., 2013].
Recent experimental results show that the hgher heat-up temperature and larger tensile hoop stress
generated larger radial hydride fraction. Authors explained the radial hydride fraction by the
combined effects of a hydrogen solubility
difference between heat-up and cool-down
temperatures [Cha et al., 2018]. Unfortunately,
there is no agreed upon internationally allowed
peak cladding temperature. NRC established a
regulation on a cladding peak temperature during
interim dry storage in which maximum
temperature should not exceed 400 °C for all fuel
burn-ups under normal conditions, however, a
higher short-term temperature limit is allowed for low burn-up fuel. Contrary to that, Japan
Nuclear Regulation Authority (NRA) established a different regulation that specifies the allowable
cladding peak temperatures of 250 and 275 °C, depending on fuel burnup [Cha et al., 2018]. This
discrepancy in regulation standard between two leading countries indicates the lack of
understanding of the time evolution of the cladding morphology and chemistry while in dry
storage. The radiation damage of the fuel cladding (by beta, alpha and neutrons [In general, β-
decay is the primary source of radiation during the first 500 years of storage, as it originates from
the shorter-lived fission products]) further complicates the fuel degradation process and that the
results in cladding swelling and He bubble trapping [Adrych-Brunning et al., 2018; J-S Kim et al.,
2017; Ronald Adamson, 2017]. The stress due to rod internal pressure can also induce cladding
degradation such as, stress corrosion cracking, hydride reorientation, and delayed hydride cracking
[Alam and Hellwig, 2008; J-S Kim et al., 2017]. If the fuel degrades during extended storage, it
could be susceptible to damage from the vibration and shocks encountered during transport
operations. The consequences may include release of fission-product gases into the canister or the
cask interior [Rigby and Members, 2010]. This is an important factor and fission-product gases
must be contained during a transportation accident.
Figure MS-2-3. Relationship between
Temperature and Hoop Stress in Cladding.
Page 19
19, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
What is unknown is the degree of residual water remaining in dried fuel, and its ultimate fate and
effects. Radiolysis of the water could allow fuel pellet, cladding, or metal component degradation
if the water or oxygen/hydrogen gas were to escape cladding through pinholes or cracks in
canisters.
Summary: According to available literature the zirconium alloy cladding of SNF are
susceptible to degradation, more so compared to the fuel rods itself. Cladding is an integral
part of the fuel and it is hard to dissociate one from the other. The cladding failure is a major
issue and will result in the release of radioactivity inside the canister. There is a high
possibility that the embrittled cladding would dissociate while retrieving the canister from
the ISFSI for transportation and that would result in unconfined radioactivity inside the
canister. The maximum cladding degradation happens 5 years after the dry storage begins.
Moreover, there is no international consensus of allowable peak cladding temperature
(higher temperature degrades the cladding faster). The US NRC regulation allows peak
temperature to be below 400 °C, whereas the Japan Nuclear Regulation Authority (NRA)
established the regulatory limit to 275 °C. This regulatory discrepancy highlights the lack of
knowledge on time evolution of cladding morphology and chemistry.
Page 20
20, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Two topics (under the GE section) are combined in the following discussion.
GE-1: What is the timeline, vulnerability of ISFSI to coastal processes under current and
future, including: coastal erosion, groundwater, sea level, flood (tsunami), humidity/fog and
weather-related?
GE-2: What is current and future earthquake frequency and magnitude and what is the
impact on the site?
Geomorphology, Coastal Geology and Hydrology of San Onofre State Beach: The
geomorphic features of the southern California coast are the raised marine terraces that rise in steps
from the beach up to the slopes of adjacent mountains. Between Dana Point on the north and the
Mexican Border on the south, marine terraces and beach-ridges record a series of Quaternary sea-
level high-stands superimposed on tectonically rising segments of the California coast. The lowest
terraces are fronted by sea cliffs exposed to both marine and subaerial processes [Kuhn, 2000;
Survey, 1889a]. The San Onofre Area, a 2.5-mile segment of coastline approximately 12.5 miles
southeast of Dana Point, is characterized by reddish-brown Quatenary marine or nonmarine terrace
and alluvial fan deposits, which range from 15 to 30 meters thick. Thick, horizontally-bedded
Pleistocene strata unconformably overly the Monterey Formation (Miocene) throughout this
region. A cross-sectional beach profile of the San Onofre State beach is shown in Figure GE-1-1.
SONGS is located on the coastal terrace.
Erosion of the coastal terrace at San Onofre State Beach and Camp Pendleton is seen to occur
under both natural conditions and, as a result of man-induced accelerated erosional processes.
Subaerial erosional processes, which occur along this coastal segment are: (a) rainfall-induced
landslides, (b) lateral and headward erosion along canyons accelerated by man’s alteration of
existing drainage patterns; (c) gullying of the coastal terrace and cliff face by rain wash, and (d)
surface erosion of the top of the coastal terrace [Kuhn and Shepard, 1984]. Since 1987, the beach
width most markedly increased following the storms of 1992-93 and 1997-98 during two of the
biggest El-Nino events were recorded in these years in recent history [Kuhn, 2000; Young, 2015;
2018]. More than 80% of the cliffs between the SONGS and Target Canyon (Camp Pendleton)
consists of landslides and the landslides appear to be directly related to periods of intense sediment
saturation and large storm swell.
Page 21
21, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
There are no perennial streams in the general vicinity of the SONGS site. However, ephemeral
streams and watercourses exist. The major streams are San Mateo Creek, located approximately 2
miles to the northwest, and San Onofre Creek located approximately 1 mile to the northwest [NRC,
2007b]. The total Groundwater Storage Capacity for this basin is estimated to be 6,500 acre-feet
(DWR 1975; SDCWA 1997). The historical average groundwater production is about 750 acre-
feet /yr and average recharge of reclaimed water is about 500 acre-feet/yr (SDCWA 1997).
Sea Level Rise, High Tide, and Coastal Aquifers: Thermal expansion of the ocean and glacier
melting have been the dominant contributors to 20th-century global mean sea level rise.
Observations since 1971 indicate that thermal expansion and glaciers (excluding Antarctic
glaciers) explain 75% of the observed rise. The contribution of the Greenland and Antarctic ice
sheets has increased since the early 1990s, partly from increased outflow induced by warming of
Figure GE-1-1. Cross-sectional beach profile of San Onofre State beach. SONGS is located at the
coastal Terrace shown in the sketch (not in scale). Figure taken from Kuhn (2000).
Page 22
22, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
the immediately adjacent ocean. Since 1993, when observations of all sea-level components
become available, the sum of contributions equals the observed global mean sea level rise within
uncertainties. Changes in ocean currents, ocean density, and sea level are all tightly coupled such
that changes at one location impact local sea level and sea level far from the location of the initial
change, including changes in sea level at the coast in response to changes in open-ocean
temperature. Although both temperature and salinity changes can contribute significantly to
regional sea-level change, only temperature change produces a significant contribution to global
average ocean volume change due to thermal expansion or contraction [Church et al., 2013].
The global energy balance is a fundamental aspect of the Earth’s climate system. At the top of the
atmosphere, the boundary of the climate system, the balance involves shortwave radiation received
from the Sun, and shortwave radiation reflected, and longwave radiation emitted by the Earth. The
rate of storage of energy in the Earth system must be equal to the net downward radiative flux at
the top of the atmosphere. This energy imbalance is rising due to increased greenhouse gases in
the atmosphere and hence the global increase in temperature. The Intergovernmental Panel on
Climate Change (IPCC) adopted a set of emission scenarios known as ‘representative
concentration pathways’, or RCPs. RCPs consist of four future pathways, named for the associated
radiative forcing (the globally averaged heat-trapping capacity of the atmosphere measured in
watts/square meter) level in 2100 relative to pre-industrial values: RCP 8.5, 6.0, 4.5 and 2.6,
respectively. RCP 8.5 is consistent with a future in which there are no significant global efforts to
limit or reduce emissions. RCP 2.6 is a stringent emissions reduction scenario and assumes that
global greenhouse gas emissions will be significantly curtailed. Under this scenario, global CO2
emissions decline by about 70% between 2015 and 2050, to zero by 2080, and below zero
thereafter. RCP 4.5 and 6 are the two intermediate scenarios and resulted in a similar way.
Page 23
23, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
There is another extreme sea-level rise scenario in the Fourth National Climate Assessment
(known as an H++ scenario), which predicts much higher sea-level rise in the long run compared
to RCP 8.5 [Griggs et al., 2017]. Figure GE-1-1. shows the predicted sea-level rise in different
scenarios at La Jolla (40 miles south of SONGS). By 2030 and 2050, the predicted sea-level rise
for all the scenarios is similar (around 15 cm by 2030 and 25 cm by 2050), but the predicted rise
differs significantly among different scenarios in 2100 and 2150 (Figure GE-1-2). Although long-
term mean sea-level rise by itself will provoke increasing occurrences of coastal lowlands
flooding, over the next several decades it is highly likely that short-term increases in sea level will
continue to be the driver of most of the strongest impacts along the coast of California. Short-term
processes, including Pacific Basin climate fluctuations (Pacific Decadal Oscillation, El Niño
Southern Oscillation, and North Pacific Gyre Oscillation), King tides (perigean high tides),
seasonal cycles, and winter storms, will produce significantly higher water levels than sea-level
rise alone.
Figure GE-1-2. Predicted sea level rise at La Jolla, CA (40 miles south of SONGS) based on different
climatic scenario (emission estimates) described in the text. In all the scenarios the predicted rise is
about the same for 2030 and 2050 and diverges thereafter. The plot shows the increases in different
scenarios in different shades of green. The H++ scenario is the extreme case (shown in dark green)
where the predicted sea level increase is dramatic, about over 260 inches by 2150 (Data taken from [4].
In a conservative scenario, the expected sea-level rise by 2050 is by about 11 inches.
Page 24
24, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Over the recorded era of the 20th and early 21st centuries, most of the significant storm damage to
California’s coastline has occurred during major El Niño events, when elevated sea levels
coincided with storm waves and high tides. The most prominent of those cases were major El Niño
events, for example, 1940-41, 1982-83, and 1997-98, when sea levels were elevated 8 to 12 inches
for several months at a time [Feely, 1987; J. J. Barsugli and P. D. Sardeshmukh, 1999; Yeh et al.,
2009].
High tides along the California coast occur twice daily, typically of uneven amplitude, and are
caused predominantly by the gravitational attraction of the moon and the sun on the Earth’s oceans.
Extreme tides, called spring tides, occur in multi-day clusters twice-monthly at times of the full
and new moon. Additionally, even higher tides occur several times a year and are designated as
perigean high tides, or more popularly “King tides”. These events are now recognized as producing
significant coastal flooding in some well-known areas. The Earth-moon-sun orbital cycles also
amplify tidal ranges every 4.4 and 18.6 years, producing peaks in the monthly high tide that are
about 15 cm and 8 cm, respectively, higher than in the intervening years [Griggs et al., 2017] (see
Figure GE-1-3).
Recently, there are several studies addressing the intrusion of seawater in coastal aquifers [Chun
et al., 2018; Hoover et al., 2017; Singaraja et al., 2018] in the context of sea-level rise. Sea-level
rise related impacts on coastal systems can occur in several ways. Marine inundation will shift the
Figure GE-1-3. Recorded tide surge from San Clemente tide-gauge for last two years. The plot
shows the monthly variation of the tide level along with occasional surges, which are as are as
high as 7 feet from the mean-sea level.
Page 25
25, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
coastline landward, erode beaches, accelerate cliff failure, degrade some coastal habitats, and
potentially damage coastal infrastructure [Young, 2015; 2018]. Recent studies have demonstrated
that sea-level rise also could contribute to saltwater intrusion in coastal regions by raising the
interface between intruding saltwater and overlying freshwater.
Sea-level rise and tidal forcing will cause the water table level to rise in the coastal areas, and the
mean high sea level could approach and ultimately rise above the ground surface [Chun et al.,
2018] specifically in southern California [Hoover et al., 2017]. Moreover, the tides are also known
to affect the groundwater fluctuation in the coastal aquifers [Singaraja et al., 2018].
Geological Setting of the Site and Earthquakes: The SONGS site is located in a seismically
active zone on the Pacific coast, which is part of active Pacific-North America transform plate
boundary, and where the seafloor is deformed by several large oblique-slip fault systems [Legg et
al., 2015]. The offshore Southern California Borderland has undergone dramatic adjustments as
conditions changed from subduction tectonics to transform tectonics, including major Miocene
oblique extension, followed by transpressional fault reactivation. A moderately landward-dipping
San Mateo–Carlsbad (SMC) fault converges downward with the steeper, right-lateral Newport-
Inglewood/ Rose Canyon (NIRC) fault forming a fault wedge as shown in Figure GE-2-1. The
NIRC fault zone is an active strike-slip fault system within the easternmost of the offshore Inner
Continental Borderlands (ICB) fault system and an active component of the Pacific-North America
plate boundary. The best-estimated geologic slip rate of this fault is 1.5 ± 0.5 mm/yr, which is
Page 26
26, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
approximately 5-15% of the estimated 50 mm/yr plate boundary deformation in southern
California [D M Singleton et al., 2018; D M A Singleton, D. C.; Maloney, J. M.; Rockwell, T. K.,
2017]. NIRC poses a significant hazard to coastal Southern California because of its proximity to
some of the most densely populated regions of North America (e.g., San Diego, Orange, and Los
Angeles counties, as well as Tijuana, Mexico). There have been several moderate-sized
(magnitude > 5.5) earthquakes in this region in the last 100 years and the estimated epicenters are
marked in Figure 1 [Legg et al., 2015; Sahakian et al., 2017; Sorlien et al., 2015]. Recently, the
NIRC fault has been mapped in detail and the rapture magnitudes for three different scenarios are
estimated [Sahakian et al., 2017]. These authors assigned a magnitude value up to 7.4 in the
Figure GE-2-1. Geological location map of SONGS. Data taken from [1-5]. The locations
of the epicenters of the previous earthquakes are shown with year of occurrence and
magnitude in parenthesis. The location of the faults and the locations of the epicenters of
the previous earthquakes are shown for illustration purpose only (and might not be precise).
The relative motions of the two plates are also shown.
Page 27
27, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Richter scale for an earthquake from this fault alone without an estimate of the error in this
analysis. The prediction of an earthquake is a fascinating area of research, and the prediction of
the magnitude is a daunting task. As mentioned, given the wedged-shaped fault system between
NIRC and SMC faults, the prediction is more complicated than that from a single fault. In addition
to the offshore faults, there are multiple faults east of the SO-NPP. These include the San Jacinto
fault and Elsinore fault on the east and the famous San Andreas fault northeast of the SO-NPP.
Considering all these complex geological features together, an earthquake ≥ 8 in the Richter scale
is well within the possibility and could not be ruled out for this reason.
Tsunamis are the consequence of oceanic earthquakes where long ocean waves sustained by
gravity that increase in amplitude as water depth decreases. Therefore, such waves are particularly
hazardous along populated coastlines near offshore faults that produce a vertical displacement of
the seafloor and water column. The hazard from earthquake-generated tsunamis offshore of
Southern California has received relatively little attention, however, there is a historical record of
Tsunamis in this region, two of them within the last 100 years [Lander et al., 1993]. A report
prepared by Intersea Research Corporation (for Southern California Edison, 1976) estimated the
maximum height of Tsunami waves of 15-feet height from the mean low sea level [Intersea
Research Corporation, 1976]. The existence of the borderland shelves (Figure GE-2-1) are
considered to be the main barrier towards high Tsunami waves for the SONGS site. About 133
feet high waves were detected during 2011 Japan earthquake [WikiPedia, 2011], however, the
predicted wave heights are about 33 feet at the open ocean [NOAA, 2011]. This discrepancy shows
that the predictions could be significantly off from reality. A recent study [Ryan et al., 2015]
updated the possibility of tsunamis in Southern California triggered by an earthquake of magnitude
7.4 in the Ventura basin (a few tens of miles north of the SONGS), which has the similar dip-slip
faults as described for the SONGS location. With the new studies of the seafloor fault mapping
and the computer simulation of the tectonic movements and triggering of a tsunami, the
understanding of the natural hazards present at the coastal region of southern California has
significantly improved in recent years.
Summary: Effect of the Natural Causes On SONGS ISFSI (GE-1 and GE-2)
Based on the above information, the following points emerge as a concern for underground
dry storage: (i) SONGS is located at a coastal terrace supported by high-erosion prone sea
Page 28
28, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
cliff and coastal bluff. The foundation of the 8-feet sea wall (the average total height of the
barrier is 28-feet from the mean low water level), built to protect SONGS from the ocean
wave action, is on the fragile bluff and, therefore, vulnerable during high erosion events, e.g.,
following El-Nino fueled storm surge (as that happened during 1997-’98 El-Nino). There is
published data on San Onofre state beach and the surrounding area that these areas are
prone to sporadic land slides and erosion.
(ii) SONGS is located in the groundwater catchment basin, which indicates groundwater
activity. The ISFSI is an underground system, the base of which is only 18 inches above the
ground water-table, but surrounded by groundwater activities (because of the catchment
area). Any breach of the storage system potentially would contaminate the groundwater
system specially with long-lived fission-product radionuclides [e.g., Cs-137 (t1/2 = 30 years),
Sr-90 (t1/2 = 29 years), Pu- 239/240 (t1/2 = 24,000 years, 6500 years), I-129 (t1/2 = 16 million
years), Se-79 (t1/2 = 327,000 years), Zr-93 (t1/2 = 1.53 million years), , Pd-107 (t1/2 = 6.5 million
years)] [Chiba et al., 2017]] because owing to their high solubility in water and mobility in
the geosphere [Salvatores et al., 1998] and potentially contaminating the regional watersheds
and estuaries.
(iii) Off-Shore Earthquakes and Tsunamis: SONGS is located in the seismically active zone.
Though a recent study computed a maximum earthquake intensity of 7.3 in Richter scale
[Sahakian et al., 2017], other studies documented earthquake magnitude over 8 for the
similar kind of slip-fault as that exist near San Onofre [Legg et al., 2015]. Tsunamis are the
consequence of oceanic earthquakes where long ocean waves sustained by gravity that
increase in amplitude as water depth decreases. A tsunami triggered by a large earthquake
at the southern California coast cannot be ignored based on the available study as discussed.
As noted, there were huge (more than 4 times) discrepancies between the predicted wave-
height and the actual one for the 2011 Japan tsunami and proves the non-fail-proof model
outcomes for the natural systems. To provide a quantitative estimate, one could
conservatively consider a recurrence interval of 1 in 300 years for such a great tsunami, then
the probability of occurrence of such an event in the next 50 years is about 15 % (i.e., 1- (1-
1/300)50). Since the seawall height is measured from the mean low water level, in a futuristic
scenario of swollen sea (or if the tsunami happened during those high tide seasons, the
effective seawall height would be close to 20 to 21 feet and might be ineffective for stopping
Page 29
29, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
the tsunami waves. The tsunami waves carry sands and debris onshore in high velocity, and
that debris might impact the cooling mechanism of the underground ISFSI and the potential
consequences are discussed in canister failure section (MS-1).
Two topics (under the GE section) are combined in the following discussion.
RM-1: How is risk being defined, what is the probability cut off for considering scenarios
RM-2: What are assumptions used in the risk analysis and are they still valid today? E.g.
transport readiness of canisters, canister requirements after Yucca shut down, etc.
The risk tree analysis is the preferred method of analyzing risk in the nuclear industry while
assessing the risk of dry SNF transportation and storage [Chen et al., 2010; NRC, 2007a; Yun et
al., 2017]. The NRC uses the risk triplet, in which the elements of risk are the scenarios, the
frequencies of the scenarios, and the consequences of the scenarios, where the measure of risk is
the consequences multiplied by the frequency of the consequences and mathematically defined
(measure risk is the consequences multiplied by the frequency of the consequences) [NRC, 2007a]
as shown in Eq. 1.
𝑅 = 𝑓∑ 𝑃𝑛𝐾𝑛𝑚𝑛=1 …….. (Eq. 1)
where R = total risk, f= frequency (1/time), the conditional probability of nth accident scenario,
and consequence of the nth accident. The risk of all plausible accident scenarios in all stages of the
dry cask storage operation can also be derived in the same way.
Some of the recent literature followed a similar risk-free analysis and found negligible risk in the
entire process [Chen et al., 2010; Yun et al., 2017]. Because of the multiplicative nature, the overall
risk of failure of the dry cask storage comes out to be negligibly small. They use a risk-free
approach, which is multiplicative. As an example, suppose a task has 4 steps, and the risk of each
step is between 0 and 1. If only one step of extreme risk (e.g., 1), and other steps are of little risk
(e.g., 0.1, 0.2, and 0.1). The total risk would be 0.002, which means very low risk even though the
first step is of extremely high risk. This kind of analysis does not consider any risk related to the
degradation of the canister [NRC, 2007a].
Page 30
30, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Another method for performing risk analysis is by failure mode and risk analysis framework [Yang
et al., 2011], which is a semi-quantitative method. In this type of analysis, every stage of the
process could be evaluated separately by determining the Risk Probability Number (RPN). RPN
is obtained by multiplying three different factors for every stage: Severity, Occurrence, and
Detection. The three factors are ranked on a scale of 1 to 10. For severity, 10 is extreme; for
occurrence 10 means the high frequency of occurrence and for detection, 1 means low risk due to
high detection ability, and 10 means high risk due to low detection probability. The method is
semi-quantitative because, for most cases, the number (scale factor) assigned is subjective
(intelligent guess based on some prior experience), however, it identifies processes of different
risk levels. The higher the RPN number, the higher is the risk of catastrophe. As an example, Table
I shows estimated RPN values for spent nuclear fuel loading and storage in a marine environment.
Table I. Failure Mode and Risk Analysis for SNF Loading and Storage in the Marine Environment
(This table should be used just as an outline as it is not yet been published).
The risk analysis of these two methods described here yield different outcomes. As an example,
during an incident like dropping a canister during transfer, the risk tree analysis identifies as the
Page 31
31, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
high-risk process, but the failure mode and risk analysis identifies as one of the low-risk ones,
because the detection risk is very low (well-trained personnel could easily detect it without any
technological aids) [Chen et al., 2010; NRC, 2007a; Yun et al., 2017]. When using this analysis
for the SONGS underground storage (ISFSI), tsunamis, followed by the internal degradation of
the canister by different processes (as described), was identified as one of the greatest risks because
the detection of risk is high for these situations. Though the risk estimation methods are subjective,
however, they are important as they provide a semi-quantitative guideline for mitigating risk
factors in SNF handling and storage.
This is an active area of research and there were several papers presented in the Waste Management
Conference (Phoenix, AZ, March 2019) and the common theme of these papers is that the risk tree
analysis is not the perfect method to capture the real risk as being currently followed.
Summary: According to new research, the present nuclear industry-standard of risk analysis
schemes does not represent the real risk. NRC performs risk studies based on risk-tree
analysis and provides three steps of safety criteria: absolutely safe, reasonably safe, unsafe.
These criteria are arbitrary and based on biased judgments.
Page 32
32, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
SH-1: What are the root causes of the near-drop incident and gouging of canisters? Is there
a mitigation strategy that could, could have been, or could be applied for canisters that would
alleviate concerns?
The process of downloading Holtec (MPC 37) canisters inside the UMAX cavity is one of the
many initial processes in storing dry casks. There was a “near-miss” at the SO-NPP on August 3,
2018 and it was investigated by NRC. As workers were lowering one of the 54-tonne (US ton)
canisters packed with SNF (37 fuel assemblies), the canister got caught on the lip of an inner ring
(Figure SH-1-1), hanging by about 0.6 cm. The drop-restraining system was not in place and the
canister was about to fall by about 5.5 m (18 ft). During downloading operation, the canister system
(with overpack) was not visible to the crane operator (due to line of sight radiation shielding).
During the NRC hearing (November 2018) we learned that there was no guide system for
downloading and the crane operator was verbally instructed by few of the downloading crew.
Seeing the slack in the lifting slings, the crew mistakenly concluded that the canister was inside
the cavity. Later from the high radiation reading, they realized that the canister was caught in the
guide-ring of the cavity.
The root cause of this mishap is the improper and inadequate equipment and technology available
to the crew to perform this delicate task and the inadequate training of the crew. This task was
delicate because of several reasons- the weight (54-tonne) and dimension (6 ft in diameter and 18
ft height) pf the cylindrical-shaped canister; the clearance between the canister outer wall and the
inner diameter of the guide-ring was only about ¼- of an inch on all the sides. Figure SH-1-1
schematically shows the situation and also estimates the force vectors using Newtonian physics.
English et al. [English et al., 2019] examined this “near miss” scenario from a basic physics
standpoint. By examining the dropping of the canister in free fall, we have estimated the upper
value of the velocity, kinetic energy, and momentum when the canister crashes into the concrete
at the bottom of the UMAX cavity. The falling canister could hit the concrete floor at a speed of
40 km/hour with the impact force of over 3 x 108 N. This situation is equivalent to that of a fully-
loaded 18-wheeler truck with a gross weight of 54-tonne crashing into reinforced concrete at 40
km/hour. This impact could ruin the canister’s cooling system and potentially cause a large
radiation release. There was no physical drop test performed on this particular kind of canister.
Holtec’s FSAR [FSAR, 2018] states in section 3.4.4.1.4, “Tipover is not an applicable load case
Page 33
33, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
for HI-STORM UMAX. The VVM (i.e., Vertical Ventilated Module) is situated underground and
cannot be moved; therefore, drop and tip-over events are not credible accidents for this design
configuration.” This assertion in safety analysis report indicates that the vendor-Holtec was not
prepared for this kind of accident and thus mitigation policy was not in place. NRC’s letter to the
Edison Company (EA 18-155, dated July 9, 2019) stated, “As part of the equipment enhancements,
the licensee installed a camera on the side of one of the VCT towers. The camera was positioned
to provide an overhead view of the top of the canister as it passed through the transfer cask into
the ISFSI vault. The camera wirelessly displayed the video feed to a monitor that was located next
to the Holtec cask loading supervisor and the SCE oversight specialist”.
There are few studies available regarding drop test of canister from a significant height to
understand the deformation of the canister [Choi et al., 2011; Lin et al., 2015; Wu et al., 2012].
Most of these studies are computer simulation (with a limited drop height) and there are handful
dealing with a physical drop of a reduced size canister with no load condition [Witte et al., 1998].
The motive behind these studies is to acquire knowledge about the deformation of internal
Figure SM-1-1. Schematically shows the stuck situation of the “near-miss” event. Applying the laws
of Newtonian physics, the force vectors are estimated and applying that estimated force the depth and
gouges are also estimated.
Page 34
34, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
channels and heat flow conditions. The heat removal from the active SNF depends on complex
heat-flow dynamics and any deformation might result in insufficient heat removal and that leads
the canister towards criticality. The model simulations show that for a small drop, the canister’s
internal deformation is not severe, however, this conclusion might not readily be applicable for an
18 foot drop (the potential drop height of August 2018 incident).
Based on our analysis [English et al., 2019], installation of a camera as stated above is NOT the
adequate remedy or a mitigation strategy for future loading. Any prevention of metal to metal
scratching during the loading stage should be the goal to avoid scratching and gouging. These
artifacts turned out to be the potential site for chemical corrosion, especially of the hot and the
highly radiative surface and marine environment [Bayssie et al., 2009; Beech, 2004; Lv et al., 2016;
Nilsson, 2012; Padovani et al., 2017; Robertson, 1991; Ul-Hamid et al., 2017].
Another important point is the development of scratches and gouges during downloading because
of metal-to-metal interaction via indentation. The recent visual inspection by Edison (as stated in
the NRC letter to Edison, July 2019) [NRC, 2019] which was performed by a robotic crawler
equipped with navigational cameras and a borescope. The borescope was a flexible camera with
interchangeable tips (a GE equipment). All surface irregularities were recorded and compared to
post-fabrication photos to determine whether the surface irregularities were a result of
downloading operations. They determined that the maximum wear depth of 0.026 inches (0.66
mm). Though the results seem comforting, however the technology measures the depth based on
the computer model and there is no direct measurement. Therefore, these simulated depth
measurements should associate with a large uncertainty which was not stated in the NRC letter
[NRC, 2019] as well as in the GE’s (manufacturer) website. Even if we consider the measured
depth number at face value, the measured depth of 0.026 inches is 160% greater than the ‘worst-
case predicted scratch depth’ as published by Edison [Edison, 2019] using a proprietary computer
simulation model.
Holtec and Edison remotely measured 8 canister surface out of 29 and through a statistical analysis
“determined that the deepest scratch at one location resulting from insertion followed by
withdrawal with a 95 percent probability and 95 percent confidence to be 0.0584 inches (58 mils,
i.e., 1.47 mm), which was still below the ASME code limit of 10 percent (0.0625 inches)”[NRC,
2019]. The above statement is scientifically incorrect because as stated Wikipedia, “a 95%
Page 35
35, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
confidence level does not mean that for a given realized interval there is a 95% probability that the
population parameter lies within the interval (i.e., a 95% probability that the interval covers the
population parameter). According to the strict frequentist interpretation, once an interval is
calculated, this interval either covers the parameter value or it does not; it is no longer a matter of
probability. The 95% probability relates to the reliability of the estimation procedure, not to a
specific calculated interval.” [NIST, 2008; Wikipedia, 2019]. Moreover, they have determined
these statistics with a very small population (i.e., 8) which indicated this 95% confidence limit is
not realistic.
Summary: Imperfect design, inadequate equipment and untrained crew were the root causes
of the August 2018 near-miss accident. A drop during canister loading was not even
considered as a potential accident issue by the contractor as stated in their FSAR. Insufficient
foresight of the vendor/contractor about this kind of potential accident scenario resulted in
no mitigation policy. At present the vendor installed camera (after the incident) to better
monitor the loading process. This mitigation strategy is still inadequate because it would not
prevent metal-to-metal contact during misalignment. This kind of metal-to-metal contact
should be avoided to decrease the risk of scratching and gouging of the stainless-steel surface
(see MS section for further detail). These later created imperfections are potential sites for
degradation of stainless steel through environmental processes (chemical, biofilm, etc)
especially when these stainless-steel surfaces are exposed to high temperatures in a
radiological environment with marine environment surroundings (these canisters are cooled
passively by thermal diffusion of marine air). The consequence of such a huge drop of SNF
loaded canister is not well known. Most studies considered a small drop (of 3 to 4 inches and
not 18 feet). Such a drop would potentially deform the internal channels for fluid flow for
adequate heat transfer from SNF to the surface of the canister. A contact sensor must be
installed to avoid any metal to metal grinding. One has to remember that when a 54-tonne
canister is being downloaded inside a UMAX hole through a sling system, there is no way
that the canister would not swing (basic principle of pendulum) and since the clearance is
small, a small swing would result in metal to metal contact and scratch. The scratch depth
measurement provided by Holtec/Edison is 160% more than their own predicted value, even
though the technique they used does not provide the associated uncertainty. The provided
statistic is erroneous because of the incorrect assertion as stated and due to a very small
Page 36
36, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
sample size (i.e., population). Any real drop of a canister during the downloading process
would be catastrophic because it would deform the internal channels that would disrupt the
fluid dynamics and the system might run towards criticality.
Page 37
37, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
References
Adrych-Brunning, A., M. R. Gilbert, J. C. Sublet, A. Harte, and C. P. Race (2018), Modelling the
interaction of primary irradiation damage and precipitates: Implications for experimental irradiation of
zirconium alloys, Journal of Nuclear Materials, 498, 282-289,
doi:https://doi.org/10.1016/j.jnucmat.2017.10.022.
Alam, A. M., and C. Hellwig (2008), Cladding Tube Deformation Test for Stress Reorientation of
Hydrides, Journal of ASTM International, 5(2).
Alyokhina, S. (2018), Thermal analysis of certain accident conditions of dry spent nuclear fuel storage,
Nuclear Engineering and Technology, 50(5), 717-723, doi:https://doi.org/10.1016/j.net.2018.03.002.
Bayssie, M., D. Dunn, and A. Csontos (2009), Evaluation of austenitic stainless steel dry storage cask
stress corrosion cracking susceptibility, edited by N. R. Commission, NRC.
Beech, I. B. (2004), Corrosion of technical materials in the presence of biofilms—current understanding
and state-of-the art methods of study, International Biodeterioration & Biodegradation, 53(3), 177-183,
doi:https://doi.org/10.1016/S0964-8305(03)00092-1.
Cha, H.-J., K.-N. Jang, J.-H. An, and K.-T. Kim (2015), The effect of hydrogen and oxygen contents on
hydride reorientations of zirconium alloy cladding tubes, Nuclear Engineering and Technology, 47(6),
746-755, doi:https://doi.org/10.1016/j.net.2015.06.004.
Cha, H.-J., K.-N. Jang, and K.-T. Kim (2018), An allowable cladding peak temperature for spent nuclear
fuels in interim dry storage, Journal of Nuclear Materials, 498, 409-420,
doi:https://doi.org/10.1016/j.jnucmat.2017.11.018.
Chen, K. C., K. Ting, Y. C. Li, Y. Y. Chen, W. K. Cheng, W. C. Chen, and C. T. Liu (2010), A study of
the probabilistic risk assessment to the dry storage system of spent nuclear fuel, International Journal of
Pressure Vessels and Piping, 87(1), 17-25, doi:https://doi.org/10.1016/j.ijpvp.2009.11.009.
Chiba, S., T. Wakabayashi, Y. Tachi, N. Takaki, A. Terashima, S. Okumura, and T. Yoshida (2017),
Method to Reduce Long-lived Fission Products by Nuclear Transmutations with Fast Spectrum Reactors,
Scientific Reports, 7(1), 13961, doi:10.1038/s41598-017-14319-7.
Choi, W.-S., J.-E. Jeon, K.-S. Seo, J.-E. Park, G.-S. You, and W.-G. Park (2011), Demonstration drop test
and design enhancement of the CANDU spent fuel storage basket in dry storage facility, Nuclear
Engineering and Design, 241(3), 723-730, doi:https://doi.org/10.1016/j.nucengdes.2010.12.024.
Chu, H. C., S. K. Wu, and R. C. Kuo (2008), Hydride reorientation in Zircaloy-4 cladding, Journal of
Nuclear Materials, 373(1), 319-327, doi:https://doi.org/10.1016/j.jnucmat.2007.06.012.
Chun, J., C. Lim, D. Kim, and J. Kim (2018), Assessing Impacts of Climate Change and Sea-Level Rise
on Seawater Intrusion in a Coastal Aquifer, Water, 10(4), 357.
Church, J. A., et al. (2013), Sea Level Change in Climate Change 2013: The Physical Science Basis.
Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change, edited by T. F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A.
Nauels, Y. Xia, V. Bex and P. M. Midgley, Cambridge University Press, Cambridge, United Kingdom
and New York, NY, USA.
Page 38
38, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Colas, K. B., A. T. Motta, M. R. Daymond, and J. D. Almer (2013), Effect of thermo-mechanical cycling
on zirconium hydride reorientation studied in situ with synchrotron X-ray diffraction, Journal of Nuclear
Materials, 440(1), 586-595, doi:https://doi.org/10.1016/j.jnucmat.2013.04.047.
Edison (2019), Evaluation of Downloading Effects on Multi-Purpose Canister Integrity, edited.
English, T., S. Chakraborty, and L. Hering (2019), San Onofre Nuclear Waste Problems, edited, Samuel
Lawrence Foundation, San Diego.
Ewing, R. C. (2015), Long-term storage of spent nuclear fuel, Nature Materials, 14, 252,
doi:10.1038/nmat4226.
Feely, R. A. (1987), Distribution of chemical tracers in the eastern equatorial Pacific during and after the
1982-1983 El Nino/Southern Oscillation event, J. Geophys. Res., 92, 6545-6558.
FSAR, H. (2018), FINAL SAFETY ANALYSIS REPORT on THE HI-STORM UMAX CANISTER
STORAGE SYSTEM, edited by NRC.
Griggs, G., J. Árvai, D. Cayan, R. DeConto, J. Fox, H. Fricker, R. Kopp, C. Tebaldi, and E. Whiteman
(2017), Rising Seas in California: An Update on Sea-Level Rise Science., edited by C. O. P. C. S. A. T.
W. Group, California Ocean Science Trust.
Hackel, L., J. R. Rankin, A. Rubenchik, W. E. King, and M. Matthews (2018), Laser peening: A tool for
additive manufacturing post-processing, Additive Manufacturing, 24, 67-75,
doi:https://doi.org/10.1016/j.addma.2018.09.013.
Hoover, D. J., K. O. Odigie, P. W. Swarzenski, and P. Barnard (2017), Sea-level rise and coastal
groundwater inundation and shoaling at select sites in California, USA, Journal of Hydrology: Regional
Studies, 11, 234-249, doi:https://doi.org/10.1016/j.ejrh.2015.12.055.
Ilgen, A., C. Bryan, and E. Hardin (2015), Geologic Disposal Requirements Basis for STAD
Specification, edited by U. S. D. o. Energy.
Intersea Research Corporation, p. f. S. C. E. C. (1976), Probable Maximum and Minimum Sea Elevations
at San Onofre Excluding TsunamisRep.
J. J. Barsugli, J. S. W., A. F. Loughe,, and a. Z. T. P. D. Sardeshmukh (1999), The effect of the 1997/98
El Nino on individual large-scale weather events Bulletin of the American Meteorological Society, 80( 7,),
13.
Jang, K.-N., and K.-T. Kim (2017), The effect of neutron irradiation on hydride reorientation and
mechanical property degradation of zirconium alloy cladding, Nuclear Engineering and Technology,
49(7), 1472-1482, doi:https://doi.org/10.1016/j.net.2017.05.006.
Kim, J.-S., J.-D. Hong, Y.-S. Yang, and D.-H. Kook (2017), Rod internal pressure of spent nuclear fuel
and its effects on cladding degradation during dry storage, Journal of Nuclear Materials, 492, 253-259,
doi:https://doi.org/10.1016/j.jnucmat.2017.05.047.
Kim, Y.-J., D.-H. Kook, T.-H. Kim, and J.-S. Kim (2015), Stress and temperature-dependent hydride
reorientation of Zircaloy-4 cladding and its effect on the ductility degradation, Journal of Nuclear Science
and Technology, 52(5), 717-727, doi:10.1080/00223131.2014.978829.
Page 39
39, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Kook, D., J. Choi, J. Kim, and Y. Kim (2013), REVIEW OF SPENT FUEL INTEGRITY
EVALUATION FOR DRY STORAGE, Nuclear Engineering and Technology, 45(1), 115-124,
doi:https://doi.org/10.5516/NET.06.2012.016.
Kosaki, A. (2008), Evaluation method of corrosion lifetime of conventional stainless steel canister under
oceanic air environment, Nuclear Engineering and Design, 238, 1233-1240,
doi:10.1016/j.nucengdes.2007.03.040.
Kuhn, G. G. (2000), Sea cliff, canyon, and coastal terrace erosion between 1887 and 2000: San Onofre
State Beach, Camp Pendleton Marine Corps Base, San Diego County, California, paper presented at 000.
Neotectonics and Coastal Instability: Orange and Northern San Diego Counties, California. Joint field
Conference, Vol. I, AAPG-Pacific Section & SPE-Western Section, Long Beach, California
June 19-22.
Kuhn, G. G., and F. P. Shepard (1984), Sea Cliffs, Beaches, and Coastal Valleys of San Diego County:
Some Amazing Histories and Some Horrifying Implications, UNIVERSITY OF CALIFORNIA PRESS.
Lander, J. F., P. A. Lockridge, and M. J. Kozuch (1993), Tsunamis Affecting the West Coast of the
United States 1806–1992, NGDC Key to Geophysical Records Documentation KGRD, edited by U. S. D.
o. Commer., p. pp. 242, NOAA, Natl. Geophys. Data Cent., Boulder, Colo.
Lee, J.-M., H.-A. Kim, D.-H. Kook, and Y.-S. Kim (2018), A study on the effects of hydrogen content
and peak temperature on threshold stress for hydride reorientation in Zircaloy-4 cladding, Journal of
Nuclear Materials, 509, 285-294, doi:https://doi.org/10.1016/j.jnucmat.2018.07.005.
Legg, M. R., K. M. D., S. Natsumi, and W. D. S. (2015), High‐resolution mapping of two large‐scale
transpressional fault zones in the California Continental Borderland: Santa Cruz‐Catalina Ridge and
Ferrelo faults, Journal of Geophysical Research: Earth Surface, 120(5), 915-942,
doi:doi:10.1002/2014JF003322.
Lin, C.-Y., T.-Y. Wu, and C.-C. Huang (2015), Nonlinear dynamic impact analysis for installing a dry
storage canister into a vertical concrete cask, International Journal of Pressure Vessels and Piping, 131,
22-35, doi:https://doi.org/10.1016/j.ijpvp.2015.04.006.
Lv, W., C. Pan, W. Su, Z. Wang, S. Liu, and C. Wang (2016), Atmospheric corrosion mechanism of 316
stainless steel in simulated marine atmosphere, Corrosion Engineering, Science and Technology, 51(3),
155-162, doi:10.1179/1743278215Y.0000000042.
Min, S.-J., J.-J. Won, and K.-T. Kim (2014), Terminal cool-down temperature-dependent hydride
reorientations in Zr–Nb Alloy claddings under dry storage conditions, Journal of Nuclear Materials,
448(1), 172-183, doi:https://doi.org/10.1016/j.jnucmat.2014.02.007.
MPR_White_Paper (2018), SONGS Dry Used Fuel Storage and Chloride-Induced Stress Corrosion
Cracking
(CISCC)Rep., MPR.
Nilsson, O. (2012), Radiation-induced corrosion of steel, Stockholm, Stockholm.
NIST (2008), 1.3.5.2. Confidence Limits for the Mean, edited by NIST.
Page 40
40, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
NOAA (2011), https://sos.noaa.gov/datasets/japan-tsunami-wave-heights-march-11-2011/, edited.
NRC (2007a), A Pilot Probabilistic Risk Assessment Of a Dry Cask Storage System At a Nuclear Power
Plant, edited by U. S. N. R. Commission.
NRC (2007b), San Onofre 2&3 FSAR, edited by Hydrology.
NRC (2019), NRC SUPPLEMENTAL INSPECTION REPORT 050-00206/2018-006, 050-00361/2018-
006, 050-00362/2018-006, 072-00041/2018-002, edited.
NRC, C. (2012), Potential Releases inside a Spent Nuclear Fuel Dry Storage Cask Due to Impacts:
Relevant Information and Data Needs, edited, NRC, .
Padovani, C., et al. (2017), The corrosion behaviour of candidate container materials for the disposal of
high-level waste and spent fuel – a summary of the state of the art and opportunities for synergies in
future R&D, Corrosion Engineering, Science and Technology, 52(sup1), 227-231,
doi:10.1080/1478422X.2017.1356973.
Penalva, J., F. Feria, and L. E. Herranz (2017), Thermal performance of a concrete cask: Methodology to
model helium leakage from the steel canister, Annals of Nuclear Energy, 108, 229-238,
doi:https://doi.org/10.1016/j.anucene.2017.04.049.
Rigby, D. B., and U. N. W. T. R. B. S. Members (2010), Evaluation of the Technical Basis for Extended
Dry Storage and Transportation of Used Nuclear Fuel, edited, United States Nuclear Waste Technical
Review Board.
Robertson, J. (1991), The mechanism of high temperature aqueous corrosion of stainless steels, Corrosion
Science, 32(4), 443-465, doi:https://doi.org/10.1016/0010-938X(91)90125-9.
Ronald Adamson, M. G., Charles Patterson (2017), Irradiation Growth of Zirconium Alloys: A Review,
Advanced Nuclear Technology International.
Ryan, K. J., E. L. Geist, M. Barall, and D. D. Oglesby (2015), Dynamic models of an earthquake and
tsunami offshore Ventura, California, Geophysical Research Letters, 42(16), 6599-6606,
doi:doi:10.1002/2015GL064507.
Sahakian, V., B. Jayne, D. Neal, H. Alistair, K. Graham, and W. Steve (2017), Seismic constraints on the
architecture of the Newport‐Inglewood/Rose Canyon fault: Implications for the length and magnitude of
future earthquake ruptures, Journal of Geophysical Research: Solid Earth, 122(3), 2085-2105,
doi:doi:10.1002/2016JB013467.
Salvatores, M., I. Slessarev, M. Uematsu, and A. Tchistiakov (1998), The neutronic potential of nuclear
power for long-term radioactivity risk reduction, Progress in Nuclear Energy, 32(3), 471-475,
doi:https://doi.org/10.1016/S0149-1970(97)00055-3.
Sathyajith, S., and S. kalainathan (2015), Effect of Laser Peening without Coating on 316L austenitic
stainless steel, IOP Conference Series: Materials Science and Engineering, 73, 012152,
doi:10.1088/1757-899x/73/1/012152.
Page 41
41, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Singaraja, C., S. Chidambaram, and N. Jacob (2018), A study on the influence of tides on the water table
conditions of the shallow coastal aquifers, Applied Water Science, 8(1), 11, doi:10.1007/s13201-018-
0654-5.
Singleton, D. M., T. K. Rockwell, J. M. Maloney, and D. C. Agnew (2018), Combining paleoseismic and
geodetic data to study the characteristics of an auxiliary plate boundary fault within the California
Borderlands: Earthquake history and preliminary slip rate for the southern section of the Newport-
Inglewood-Rose Canyon fault system, in American Geophysical Union, Fall Meeting 2018, abstract
#T41C-05, edited, San Francisco.
Singleton, D. M. A., D. C.; Maloney, J. M.; Rockwell, T. K. (2017), GPS Measurements of Crustal
Deformation in San Diego, CA: Results from fixed-height monument network and implications for the
Inner Continental Borderlands, in American Geophysical Union, Fall Meeting 2017, abstract #G43B-
0944, edited, San Francisco.
Sorlien, C. C., J. T. Bennett, M.-H. Cormier, B. A. Campbell, C. Nicholson, and R. L. Bauer (2015), Late
Miocene–Quaternary fault evolution and interaction in the southern California Inner Continental
Borderland, Geosphere, 11(4), 1111-1132, doi:10.1130/GES01118.1.
Spykman, G. (2018), Dry storage of spent nuclear fuel and high active waste in Germany—Current
situation and technical aspects on inventories integrity for a prolonged storage time, Nuclear Engineering
and Technology, 50(2), 313-317, doi:https://doi.org/10.1016/j.net.2018.01.009.
Survey, U. S. C. a. G. (1889a), Topography Pacific Coast from Horno Canyon to Las Flores and Canyon
Aliso, California: Topographic Map #T-2015, edited by D. r. t. a. o. f. s. D.C., U.S. Coast and Geodetic
Survey, U.S. National Archives, Wash.
Ul-Hamid, A., H. Saricimen, A. Quddus, A. I. Mohammed, and L. M. Al-Hems (2017), Corrosion study
of SS304 and SS316 alloys in atmospheric, underground and seawater splash zone in the Arabian Gulf, 1-
7 pp., doi:10.1080/1478422X.2016.1213974.
WikiPedia (2011), https://en.wikipedia.org/wiki/2011_T%C5%8Dhoku_earthquake_and_tsunami, edited.
Wikipedia (2019), Condefence Limit, edited.
Witte, M. C., J. Hovingh, G. C. Mok, S. S. Murty, T. F. Chen, and L. E. Fischer (1998), Summary and
Evaluation of Low-Velocity Impact Tests of Solid Steel Billet Onto Concrete Pads, edited by NRC, NRC,
Livermore, CA.
Won, J.-J., M.-S. Kim, and K.-T. Kim (2014), HEAT-UP AND COOL-DOWN TEMPERATURE-
DEPENDENT HYDRIDE REORIENTATION BEHAVIORS IN ZIRCONIUM ALLOY CLADDING
TUBES, Nuclear Engineering and Technology, 46(5), 681-688,
doi:https://doi.org/10.5516/NET.07.2014.052.
Wu, T.-Y., H.-Y. Lee, and L.-C. Kang (2012), Dynamic response analysis of a spent-fuel dry storage cask
under vertical drop accident, Annals of Nuclear Energy, 42, 18-29,
doi:https://doi.org/10.1016/j.anucene.2011.12.016.
Yang, J., H.-Z. Huang, L.-P. He, S.-P. Zhu, and D. Wen (2011), Risk evaluation in failure mode and
effects analysis of aircraft turbine rotor blades using Dempster–Shafer evidence theory under uncertainty,
Engineering Failure Analysis, 18(8), 2084-2092, doi:https://doi.org/10.1016/j.engfailanal.2011.06.014.
Page 42
42, SONGS Task Force (Technical Committee), Contributed by Subrata Chakraborty
Yeh, S.-W., J.-S. Kug, B. Dewitte, M.-H. Kwon, B. P. Kirtman, and F.-F. Jin (2009), El Nino in a
changing climate, Nature, 461(7263), 511-514,
doi:http://www.nature.com/nature/journal/v461/n7263/suppinfo/nature08316_S1.html.
Young, A. P. (2015), Recent deep-seated coastal landsliding at San Onofre State Beach, California,
Geomorphology, 228, 200-212, doi:https://doi.org/10.1016/j.geomorph.2014.08.005.
Young, A. P. (2018), Decadal-scale coastal cliff retreat in southern and central California,
Geomorphology, 300, 164-175, doi:https://doi.org/10.1016/j.geomorph.2017.10.010.
Yun, M., R. Christian, B. G. Kim, B. Almomani, J. Ham, S. Lee, and H. G. Kang (2017), A software tool
for integrated risk assessment of spent fuel transportation and storage, Nuclear Engineering and
Technology, 49(4), 721-733, doi:https://doi.org/10.1016/j.net.2017.01.017.