Thermal Analysis of Spent Nuclear Fuel Shipping Cask

Açelya Deniz Gökselkınav, Research Assist. Hüseyin Ayhan, Assist. Prof. Dr. Şule Ergün,

Hacettepe University Nuclear Engineering Department, Beytepe, 06800 Ankara Turkey.

sulergun@hacettepe.edu.tr, 90 312 297 7300

ABSTRACT

In this study, a computational fluid dynamics (CFD) thermal analysis was performed for the TN-24P cask. For the

analysis, ANSYS Fluent as a CFD tool was selected since it has the proper finite volume methods to realistically

simulate the thermal behavior of shipping casks.

For the analysis, spent fuels discharged from pressurized water reactors (PWRs) were modeled. In the model, there

are 24 PWR spent fuel assemblies loaded in the TN-24P cask. The fuel design is assumed to be similar to standard

Westinghouse 15x15 rod design. Total heat (decay) generated in the cask was estimated to be 20.6 kW. To input the

axial power profile required to calculate the heat flux, a User Defined Function was generated. Fuel storage space

(canister) is filled with Helium gas to cool spent nuclear fuel. In the cask, heat transfer occurs through the heat

conduction by helium and basket, natural circulation driven by gravity, and thermal radiation in the complex

geometry. In the canister region, laminar flow model with Boussinesq approximation is used to simulate the natural

circulation. The helium domain was assumed symmetric in the model. For thermal radiation, the Discrete Ordinates

(DO) model was chosen in the presented study due to its accuracy and capability of parallel processing. In typical

vertical TN-24P dry storage cask system consist of two nested cask. Between inner and outer cask is in the air. Air

inlet section is at the bottom side of cask and outlet ventilation is at top of cask. At this region, turbulence regime

occurs and turbulence is modeled by using k-epsilon model.

The analysis include small scaled and full scaled model. In small scale model, geometry is defined rectangular to

make mesh generation easy and to validate the analysis tools using the experimental data. In the full-scale

simulation, the results of analysis and experimental data for peak clad temperature (PCT) were compared.

Key Words: TN-24P dry storage cask, CFD, thermal analysis, PCT, air blockage.

1. Introduction

A dry-storage system (DSS) is an effective long-term storage solution for spent nuclear fuel

(SNF) management. DSSs are preferred for a regional solution since they are: easy to operate,

small or passive and low maintenance systems, open for all kinds of fuel, causing less secondary

waste, and simple for accident prevention [8]. In the past years, for these systems safety studies

performed covered extreme conditions: aircraft crush, fire, earthquake, cask burial, cask tip-over,

fuel cladding breach [5].

The most important thermal design objective for a DSS is to remove decay heat from fuel

assemblies and maintain a peak clad temperature below specific limits. The NUREG- 1536

guidelines for DSSs indicate that the maximum cladding temperature should be <400 0C for

normal conditions, and <570 0C for off-normal and accident conditions [2]. The temperature in a

DSS impacts the integrity of the fuel cladding and structure material [3]. And also PCT is a key

parameter governing the licensing and regulation of dry storage systems (NRC: 10 CFR- Part

72).

The commercial CFD code Fluent had been used to analyze the model flow characteristics and

the results were compared with the published experimental data of Yoo, No, et al[1]. A

numerical simulation of flow and heat transfer in a dual purpose dry storage cask system TN24

family is performed and the results were compared with the experimental data to assess the

validity of the computational approach. The measurements of steady state temperature at ambient

pressure is available. The computational results corresponded to a laminar model with

Boussinesq approximation.

The TN24P cask operates on the basic principles of buoyancy driven natural convection where

fuel storage space (canister) is filled with Helium gas, and heat transfer occurs through the heat

conduction by helium and basket, and thermal radiation in the complex geometry, and natural

circulation driven by gravity where cooler air enters the air inlet at bottom of the cask and hot air

exists the cask at the top.

2. Description of the TN24P Cask

The Transnucleaire TN-24 family is one of the dual-purpose storage and transport cask system

and metal cask. The TN24 cask family has been developed for the transport and storage of spent

fuel assemblies with a cooling time of 5 to 10 years. Each cask version complies with the IAEA

transport regulations [7]. TN 24 family metallic dual-purpose casks are operation in USA, Japan,

Belgium, Switzerland, and upcoming casks in Germany, Italy [5]. Metal Dual Purpose (DP)

casks have both storage and transport licenses. The basic storage only technologies (vaults,

casks, and silos), some cask and silo technologies are also being used or developed for dual-

purpose (storage and transport) or multi-purpose (storage, transport and disposal) applications.

Dry storage facilities generally remove decay heat by passive cooling and have low operating

costs. Metal casks are massive containers used in transport, storage and the eventual disposal of

spent fuel. The structural materials for metal casks may be forged steel, nodular cast iron or a

steel/lead sandwich structure. They are fitted with an integral internal basket or sealed metal

canister, which provides structural strength as well as the assurance of sub-criticality. Metal

casks usually have a double lid closure system that may be bolted or seal welded and may be

monitored for leak tightness [4].

The thermal analysis performed for the dry storage casks assumes the use of helium as a cover

gas. In addition, the use of an inert gas (helium) is to ensure long-term maintenance of fuel clad

integrity [9]. Because helium is not corrosive and has a high thermal conductivity, it is used as an

inert fill gas in both fuel rods (within the cladding) and fuel storage canisters or containers.

A helium atmosphere during dry storage is used to prevent the possibility of fuel oxidation and

gross ruptures due to unzipping, and corrosion of internal canister components. The typical

helium fill gas pressure of LWR fuel rods is between 1.5 to 3.5 MPa at 20ºC for PWRs. PWR

internal rod pressures at the start of dry storage to lie between about 3 to 4.5 MPa at 25ºC [10].

The TN-24P spent fuel storage cask consists of a forged steel body for structural integrity and

gamma shielding, surrounded by a resin layer for neutron shielding. The resin layer is enclosed

in a smooth steel outer shell. The cask geometric cross section and overall features are given in

Figure 1[6].

Figure 1: Views of TN-24P PWR Spent Fuel Storage Cask (Dimensions in centimeters).

The simplified figure of shipping cask is shown in Figure 2 [10].

Figure 2: Typical Vertical Dry Cask-Storage System.

Graphic Courtesy of Holtec International, Inc.

3. Small-scale Simulation

In small-scale simulation geometry is defined rectangular to make mesh generation easy and to

validate the analysis tools using the experimental data. Firstly we used GAMBIT which is a

single integrated Pre-processor for CFD analysis, generates mesh for all fluent solvers. Domain

is drawn 2D triangular face mesh then sweep faces to render 3D Tet/Hybrid volume mesh. The

boundary and operation conditions, defined fluid properties, refined the grid viewing are listed in

Tables 1-6 [1].

Table 1: The geometry of the small-scale model.

Small-scale model

Body Height 1.3 m

Outer diameter 0.3 m

Fuel Fuel rod array 1x1

Heated length 1 m

Decay heat 220 W

Basket Width 0.18 m

Thickness 0.089 m

Table 2: TN24P cask test matrix and experimental results.

Orientation Vertical

Operating pressure 1.5 bar

Ambient Temperature 18 0C

Cask heat 20.6 kW

Meas. Guide tube temp. 214 0C

Est. PCT 221 0C

Backfill gas Helium

Table 3: Thermal conductivities of the materials.

Thermal conductivities (W/mK)

Steel cask body 41.5

Aluminum basket 206

Steel shell 41.5

Table 4: Material properties.

Materials He Air Al Steel

Properties Density, kg/m3 0.1625 1.225 2719 8030

Absorption coefficient, 1/m 1.00E-06 0 1 0.00514

Refractive index 1.000035 1.000277 1.31575 2.5

Thermal expansion coefficient, 1/K 0.002 - - -

Table 5: Numerical models for the TN24P cask simulation.

Numerical models for the TN24P cask simulation

Physical Models

Density model Boussinesq approximation

Viscous model Laminar (He region)/ k-ε Turbulent

Thermal radiation model Discrete Ordinates (DO)

Angular discretization 4x4

Pixelation 3x3

Numerical method

Pressure-velocity coupling SIMPLE

Discretization scheme

Gradient Green-Grass Cell Based

Pressure Body Force Weighted

Momentum First Order Upwind

Energy First Order Upwind

Discrete ordinates First Order Upwind

Round-off

Double precision

Table 6: Boundary conditions.

In small-scale simulation 1 fuel rod was modeled to execute the solution quickly. Energy

equation is opened in Fluent model. The canister is sealed and filled with helium gas in the

ambient temperature. Canister is in vertical position. Helium region in basket is modeled with

laminar flow model because of Rayleigh number is 8.67*108 smaller than 109. Between inner

and outer casks there is air. Air inlet section is at the bottom side of cask and outlet ventilation is

at the top of cask. Velocity inlet of air is 0.5 m/s. At this region, turbulence regime occurs and

turbulence is modeled by using k-epsilon model. In order to consider the buoyancy-driven flow

in the cask, the pressure discretization scheme was fixed as Body Force Weighted. Heat flux,

giving sinusoidal axial heat flux profile, of heated rods calculated by local volumetric heat

generation rate at position by generating a UDF. At Figure 3 shows configuration of small-scale

model with selecting a plane which cut down z-direction of domain. The PCT along the plane is

approximately 500 K (227 0C) as presented in Figure 4.

Figure 3: View of grid and cross-section of the plane in Fluent.

B.C

Air inlet Velocity inlet

Air Outlet Pressure outlet

Heated rods Wall

Cask body Wall

Can Wall

Figure 4: Temperature distribution in the plane.

4. Full-scale Simulation (TN24P Cask Simulation)

In this section of the study, full scale simulation of the TN24P cask was performed. By assuming

that the helium flow would be symmetric, the numerical domain was simulated by modeling its

1/8 section. The grid of the TN24P cask was generated by GAMBIT software. The two-

dimensional grid on the cross-section of the TN24P cask was generated and extruded along the

axial direction to consist of the three-dimensional grid. The number of grids for the TN24P cask

was 10 times less than published experimental of Yoo, No, et al[1] which is shown in Figure 5.

Spent fuel rods were lumped to 3x3 assemblies. Equivalent fuel rod diameter is 0.0535 m. All of

the boundary and initial conditions of small-scale simulation are valid at TN24P cask simulation.

(a) (b)

Figure 5:Grid generation for the TN24P cask.(a) rod assembly array, (b)full view of cask.

Specifications of TN24P cask and assembly are shown in Table 7 and 8 [1].

Table 7: The specifications of TN24P cask.

TN24P Cask

Height 5063 mm

Outer diameter 2281 mm

Loaded weight 100 t

Body Material Forged steel

Gamma shielding Steel

Neutron shielding Solid

Fuel type PWR W.H. 15x15

Source of assembly Surry

Burnup 29-32 GWD/MTU

Cooling time 4.2 years

Fuel Enrichment 2.9-3.2%

Decay heat Assembly 832-919 W

Average 860 W

Cask total 20.6 kW

Material Aluminum

Basket Critical control Borated aluminum/Poison rods

Table 8: Specifications of the W.H. 15x15 assembly.

Item Dimension

Fuel rod Outer diameter 10.7 mm

Length 3857 mm

Pitch 14.3 mm

Fuel assembly Array 15x15

Width 214 mm

Length 4058 mm

No.of fuel rods 221

Figure 6: View of mesh in Fluent.

Figure 7: Temperature distribution in the plane.

The Figure 6 shows configuration of full-scale model with selecting a plane which cut down

slide of 1/8 domain where is close to hottest point of heated rods. The PCT along the plane is

approximately 511 K (238 0C) as presented in Figure 7.

5. Conclusions

In this study, thermal analysis was performed for the TN-24P cask in ANSYS Fluent. Small and

full-scale simulation was performed to observe one of the critical parameters of dry storage

system, that is PCT. The results of analysis and experimental data for peak clad temperature

(PCT) were compared as shown in Figures 8 and 9. As shown in these figures the calculated data

for the PCT is 238 0C whereas the measured PCT is 221 0C [1].

Figure 8: Axial temperature distribution at the line of hottest region in the TN24P cask.

Figure 9: Axial temperature distribution at the cavity regions and PCT in the assembly

[1].

As shown in Figure 8, axial temperature distribution begin and finish at approximately 210 0C

because of selecting the line of hottest region in the TN24P cask is not as same D1 line in Figure

9. Another factor of the difference is number of grids for the TN24P cask, mesh quality and

iteration numbers.

With this study, by using the limited computational effort, a conservative simulation of PCT for

the TN24P cask was obtained. The conservative value is less than the limit value defined for the

design license of a DSS.

References

[1] Yoo, S. H., No, H. C., Kim, H. M., Lee, E. H., 2010. Full-scope simulation of a dry storage

cask using computational fluid dynamics. Nuclear Engineering and Design 240, 4111-4122.

[2] U.S. NRC, 2009. Standard Review Plan for Spent Fuel Dry Storage Systems at General

License Facility. NUREG-1536 Revisions 1A.

[3] Tseng, Y. S., Wang, J. R., Tsai, F. P., Cheng, Y. H., Shih, C., 2011. Thermal design

investigation of a new tube-type dry-storage system through CFD simulations. Annals of Nuclear

Energy 38, 1088-1097.

[4] IAEA-TECDOC-1532, January 2007. Operation and Maintenance of Spent Fuel Storage and

Transportation Casks/Containers.

[5] Issard, H., 2-6 September 2012. Reactor Fuel Performance- Spent Fuel and Transportation,

Dry storage reliable solutions for the management of spent nuclear fuel in the long term.

European Nuclear Society, TopFuel 2012, UK.

[6] Broadhead, B.L., Tang, J.S., Childs, R.L., Parks, C.V., H. Taniuchi, May 1995. Evaluation of

shielding analysis methods in spent fuel cask environments. EPRI TR-104329 Research Project

3290-02 Final Report.

[7] IAEA-TECDOC-1100, July 1999. Survey of wet and dry spent fuel storage.

[8] Dyck, H.P., July 1999. IAEA-SM-352/42-Regional Spent Fuel Storage Facility (RSFSF),

IAEA-TECDOC-1089 Storage of Spent Fuel from Power Reactors

[9] U.S. NRC, 10 CFR Part 72, Certificate Number: 1005.

[10] United States Nuclear Waste Technical Review Board, December 2010. Evaluation of the

Technical Basis for Extended Dry Storage and Transportation of Used Nuclear Fuel.

[11] Rolanda, V., Chiguerb, M., Guénonc, Y. Dry storage technologies: Keys to choosing among

metal casks, concrete shielded steel canister modules and vaults. IAEA-CN-102/14.

[12] FLUENT Inc, 2009. FLUENT 12 User’s Guide.