1. Introduction

A pressurized water reactor spent fuel transfer cask is a piece of operational equipment used in the process of withdrawing spent nuclear fuel from the nuclear power plant wet storage tanks, loading it into canisters, which is then dried, sealed, and relocated to further storage facilities. The transfer cask with canister is tied to the tie-down device and transported horizontally to the storage module. After removing the upper and lower impact limiters, the transfer cask is placed vertically on top of the storage cylinder in the storage module, and the lower lid of the transfer cask is opened to store the canister in the storage cylinder. The main purpose of this study to design a transfer cask for the transport of spent fuels to vertical Dry Storage Modules (DSM) specifically developed for the dry storage of spent fuels at nuclear power facilities, and to ensure the reliability of the equipment by evaluating its structural safety under different code requirements. In order to assess the structural integrity of the transfer cask, three types of load criteria were applied for the structural analysis, based on designs defined in relation to the confinement boundary between the transfer cask and the canister installed within the transfer cask, as ‘normal’, ‘off-normal’, and ‘accident’ conditions.

2. Transfer Cask Specification

The transfer cask is a metallic container with a height of 5,300 mm, external diameter of 2,170 mm, and internal diameter of 1,760 mm, protected by impact limiters at the top and bottom of the cask. The transfer cask weighs 97.0 tons including the impact limiters, and 90.2 tons without them. The total weight of the canister with its inner components is 39.2 tons. The main body of the transfer cask is a cylindrical shell structure, with lead installed in between the carbon steel shell to function as an internal and external shield. Neutron shields surround the outside of the carbon steel shell. A total of four trunnions are attached at the top and bottom of the sides for handling and binding of the transfer cask and a bottom lid is placed that can be opened and closed for loading the canisters into the storage modules. A number of components are installed within the canister to ensure the critical, thermal, and structural stability of the spent nuclear fuel. Fig. 1 provides a schematic diagram of the transfer cask.

Fig. 1

Schematic diagram of the transfer cask.

3. Design Criteria

A transfer cask needs to maintain its intended functions without compromising its key safety features in all accident or natural conditions, including the internal criticality of the canister and the radiation shielding functionality of the confinement and transfer casks. The design criteria and design values of the transfer cask are categorized into ‘normal’, ‘off-normal’, and ‘accident’ conditions and summarized in Table 1.

A load combination was carried out for the transfer cask in accordance with NUREG-2215 and ANSI/ANS 57.9 [7], based on the results of the structural integrity test results, to conduct stress testing. The combined load was selected a actual load on the transfer cask and canister. The canister pressure condition is added to the transfer cask load combination.

The load combinations are categorized into ‘normal’, ‘off-normal’, and ‘accident’ conditions and summarized in Table 2.

4. Allowable Stress

The confinement structure of the canister is maintained by the base plate, shell, lid, port cover, and welds. Once spent nuclear fuel is loaded into the canister, the canister lid and closure ring are welded to seal the containment structure, which is then placed within the module structure. The weld between the canister shell and lid and between the canister lid and port cover becomes the primary confinement boundary membrane while the closure ring ensures the secondary confinement membrane. The confinement boundary of the canister is shown in Fig. 1(a).

The allowable stress intensities for testing the confinement structure in normal and off-normal conditions are defined in ASME B&PV Code, Sec.Ⅲ, Div.1, Subsec.NB, while requirements outlined in ASME B&PV Code, Sec.Ⅲ, Div.1, App F were applied for the accident condition. The allowable stress intensities are summarized in Table 3 [8,9]. The allowable stress of the transfer cask is calculated by considering the actual temperature under normal, off-normal and accident conditions. While the canister shell and base plate are joined by full fusion welding, the canister lid is fused to the shell by partial fusion welding. The stress intensity of the partial fusion-welded confinement boundary was evaluated by applying a reduction coefficient of 0.8 in accordance with NUREG-2215.

The non-confinement structure is composed of structural components not included in the primary boundary structure, which include the inner and outer shells of the transfer cask body. The non-confinement structure was designed to allow permanent alterations of its shape. The stress intensity limits of the non-confinement structure in the normal and off-normal conditions are specified in ASME B&PV Code, Sec.Ⅲ, Div.1, Subsec.NF, which are summarized in Tables 4 and 5 [10].

The internal structure components were designed with the requirements outlined in ASME B&PV Code, Sec.Ⅲ, Div.1, Subsec.NG, and the allowable stress intensities applied for evaluating the internal structure components are shown in Table 6 [11]. The Level A and B limits specified in ASME B&PV Code, Sec.Ⅲ, Div.1, Subsec.NG were used as the allowable stress intensities for the normal and off-normal conditions. Under the hypothetical accident condition, the non-confinement structure and the internal structure components were assessed by implementing the plastic analysis method defined in the ASME B&PV Code, Sec.Ⅲ, Div.1, App F, while applying the same criteria to the allowable stress intensity.

Impact limiters are installed at the top and bottom of the cask to absorb the impact energy created under drop conditions during horizontal transportation. The shock absorbers consist of a casing produced from stainless steel and the shock absorption wood materials that fill up the inside. No evaluation was performed of the impact limiters, as they are installed only for the purpose of protecting the transfer cask under drop conditions while in transport.

5. Analysis Model

A general-purpose computer program ABAQUS 2017 was used to perform the structural analysis of the transfer cask. An identical structural analysis model was employed with differing initial and boundary conditions to conduct an analysis of each of the load conditions [12]. The analytical model used for the transfer cask is outlined in Fig. 2. The analytical model was configured as a 1/2 model, considering the symmetric nature of the transfer cask and canister. The analytical model for the transfer cask was configured with 253,182 node points and 174,200 solid elements, while the canister was modeled with 688,851 node points and 366,498 solid elements. The solid elements used in the analysis consisted of rectangular 8-node elements (C3D8R, 8-node linear brick, reduced integration with hourglass control). The dynamic analysis model was implemented with the general contact criteria provided by ABAQUS, while surface to surface conditions were used for the static analysis model. The canister, which also functions as the confinement boundary, was given elastic properties, while the non-confinement boundary of the transfer cask and internal structural components were modeled with carbon properties. The cask lid was located so that the axis of the lid was aligned to that of the container. The lid bolts were located at the center positions of the bolt holes. The casing and the fillers of the impact limiters were modeled separately. By not separating the filler layers, the cushioning buffer was modeled as a continuous mass with its parts distinguished by the plates. Both the casing and filler materials were modeled as solid elements. Components that do not affect structural performance, such as port covers, are not modeled.

Fig. 2

Analysis model of the transfer cask (ABAQUS).

The mechanical properties of the key materials used in the structural analysis of the transfer cask referred to ASME B&PV Code Sec.Ⅱ Part A and Part D. The cask body was produced with carbon steel while components besides the cask body were produced with stainless steel [13,14]. Balsa wood is applied as a absorber material, and the compressive strength of balsa wood is 12.01 MPa and the maximum strain is about 0.76. The absorber material breaks due to a significantly increase in stress after the maximum strain.

The analysis of the boundary conditions was performed in line with the analytical criteria of the normal, offnormal, and accident conditions, by firmly fixing part of the transfer cask. Axisymmetric constraints were applied to the symmetrical boundary in the center of the transfer cask, while the rigid base plate was fully constrained to not allow any displacement or changes to the degree of freedom. Full constraint conditions were also applied to the points where the lid bolts were applied.

6. Load and Analysis Results

6.1 Normal Conditions

Under normal conditions, static, active, handling, pressure, and thermal loads were evaluated according to NUREG-2215, and the structural safety was assessed through the combination of these loads.

The static load was 51.0 tons, excluding the canister, based on the vertical load of the transfer cask weight. Active load refers to the load that may occur while in transport, which considers the canisters installed in the cask. Thus, the total weight of the transfer cask applied for active load was 90.2 tons. Moreover, during the handling process, a transfer cask needs to be able to withstand the load that is generated during handling. Therefore, the handling load of the cask was evaluated while the cask was fixed to binding devices. In accordance with ASME NOG-1, the handling load was calculated with a factor of 115% to the self-weight of the transfer cask. There is no build up of internal pressure in the cask body as it has an open structure with upper and lower lids. Thus, only the pressure within the loaded canister inside the transfer cask was considered, where an internal pressure load of 0.5951 MPa was applied, which was calculated from the results of the thermal analysis. The thermal load was calculated from the results of the thermal analysis at 33.3℃, the highest monthly average temperature in Korea, while the temperature distributions of the transfer cask and canister were used as the initial conditions for the structural safety analysis.

The stress test with the load combinations under normal conditions evaluated the membrane, bending, and secondary stress as for the Level A evaluation standards of ASME B&PV Code, Sec.Ⅲ, Div. 1, Subsec.NB, NF. Table 7 provides the stress evaluation results for LC.1 and LC.2, the load combinations under normal conditions. The evaluation results show safety factors above 1.1 for all components, with the biggest stress of 525.3 MPa observed at the bottom lid bolt of the transfer cask. Of the pre-stress applied to the bottom lid bolt, the axial stress is about 218.0 MPa, with a maximum stress of about 362.5 MPa. The internal structural components, which are intended to maintain the spacing and protect spent nuclear fuel, showed relatively high safety factors for the load combinations under normal conditions.

6.2 Off-normal Conditions

Under the off-normal conditions, pressure, wind, and thermal loads of the transfer cask were evaluated according to NUREG-2215 and the structural safety was assessed using the different load combinations.

For the internal canister pressure of the transfer cask, 0.6197 MPa was calculated from the results of the thermal analysis. The wind load on the transfer cask was based on a conservative base wind speed of 60 m∙sec⁻¹ (10 m average wind speed) on a 100-year recurrence interval according to the building structure standards [4]. For the design wind loads, a horizontal wind load of 3,278.9 N/m² and a vertical wind load of 4,380.3 N/m² were applied to the transfer casks. The thermal load was calculated from the results of the thermal analysis, based on the maximum (36.6℃) and minimum (−15.6℃) daily average temperatures in Korea. The temperature distribution of the transfer cask obtained from the thermal analysis was used as the initial conditions for the structural safety analysis.

While Level B evaluation criteria of ASME B&PV Code, Sec.Ⅲ, Div. 1, Subsec.NB, NF has to be applied, the stress test with the load combinations under off-normal conditions evaluated the membrane, bending, and secondary stress, as was performed for Level A evaluation standards, as a conservative measure to be aligned with the normal conditions. Table 8 provides the stress evaluation results for LC.3 and LC.4, the load combinations under the off-normal conditions. The evaluation results show safety factors above 1.1 for all components, with the biggest stress of 528.4 MPa observed at the bottom lid bolt of the transfer cask. Similar to the normal conditions, the internal structural components showed relatively high safety factors.

6.3 Hypothetical Accident Conditions

Under accident conditions, the pressure, drop, seismic loads of the transfer casks were evaluated according to NUREG-2215 and the structural safety was assessed using the different load combinations.

The pressure load under accident conditions assumes an internal pressure created due to a fire, where 100% of the spent nuclear fuel is damaged and 100% of the gas contained in the damaged spent fuel rods are leaked out to the insides of the canister, and where 30% of the damaged nuclear fuel releases fissile gas. There is no build up of internal pressure in the cask body due to its open structure. Therefore, under accident conditions, only the pressure within the loaded canister of the transfer cask was considered, where an internal pressure of 0.8466 MPa, calculated from the results of the thermal analysis, was applied. For the thermal load under accident conditions, a conservative temperature distribution between the maximum daily temperature of 41℃ in Korea or that of the fire conditions was used as the initial conditions for the structural safety and stress evaluations. Secondary stress was excluded from the assessment as no guidelines are provided for allowable stress intensity for secondary stress.

The transfer cask needs to be able to withstand accidental drops during handling. For accidental drops of the casks in horizontal transport conditions, the impact limiters are attached to the lid and base of the cask, and the drop height was conservatively assumed when the cask was detached from the binders. The drop height was assumed to be 2.0 m, the height of the trailers (1.2 m) used in nuclear power facilities plus an additional buffer. The drop load from a drop height of 2.0 m applied 6,264 mm∙sec⁻¹, the speed of the base plate just before the moment of collision, as the initial condition. The drop surface was conservatively assumed to be a rigid plate. The seismic load was calculated based on the designed response spectrum of maximum ground acceleration of 0.3 g in the horizontal and vertical directions, in accordance with US NRC RG 1.60 and the steel damping ratio of 4%, in accordance with US NRC RG 1.61 [5,6]. Since the natural frequency of the transfer cask was above 33.0 Hz, a maximum acceleration of 0.3 g has to be applied to the seismic load in both the horizontal and vertical directions. However, more conservative accelerations of 1 g in the horizontal direction and 2 g in the vertical direction, including self-weight, were applied.

Stress evaluations for load combinations under accident conditions were conducted in line with the Level D evaluation criteria of ASME B&PV Code, Sec.Ⅲ, Div. 1, Subsec. NB, NF, to assess the membrane and bending stress while excluding the secondary stress. The results are shown in Tables 9 and 10. All structural components showed safety factors of 1.2 or above from the evaluation results, with the lowest value obtained from the basket of LC.7. The largest stress was found from the upper lid bolt of the cask in LC.7, which included the drop conditions. Relatively significant stress was applied to the internal components due to seismic load and drop impact. Fig. 3 shows the stress distribution on the transfer cask and canister under seismic and drop conditions. Under drop conditions, the biggest stress was applied to the lid weld due to the impact on the heavy canister lid, while the biggest stress was created in the lid weld as well as the top and bottom of the canister shell under seismic conditions.

Fig. 3

Stress distribution of seismic and drop.

7. Conclusions

This study performed a structural safety evaluation of a transfer cask, based on design standards and design loads. Structural analysis was performed on the transfer cask under normal, off-normal, and accident conditions with differing load conditions. Stress was evaluated for the results of different load combinations for each load condition against the allowable stress intensity outlined in the ASME B&PV Codes.

In summary, all structural components exhibited evaluation results that were below the allowable stress intensities in all normal, off-normal, and accident conditions. Consequently, the transfer cask satisfies the design requirements under all conditions. Based on the evaluation results of this study, a detailed design of the transfer cask for use in light water reactor spent fuel dry storage facilities will be drafted.