Dynamic structural response of reinforced concrete dry storage casks subjected to impact considering material degradation

Highlights

• High-fidelity finite element modeling of dry cask storage systems was performed.

• Material models and structural response under impact was calibrated using experimental data.

• The effect of material degradation on structural performance was investigated using a newly proposed pushover analysis.

Abstract

The performance of vertical concrete casks (VCC) for storage of spent nuclear fuel (SNF) under hypothetical accident conditions is of particular importance given the planned extensions of service-lives of such structures in the absence of a permanent disposal solution. Numerical models are created in this paper to simulate the structural response of a generic VCC in the cases of end-drop (handling drop) and tip-over considering material degradation. A concrete damage plasticity model (CDP) was calibrated using standard material tests to capture the non-linear behavior of concrete. Similarly, the steel properties and the properties of the soil supporting the cask foundation were based on material test data. The finite element models (FEM) were validated using experimental data from end-drop and tip-over tests of a 1/3-scale VCC performed by the same authors and reported elsewhere. The numerical results are compared with the experimental measurements in terms of velocity, acceleration, damage, and failure modes to gain confidence in the numerical results. Thereafter, a push-over analysis method was proposed and performed on the VCC for the first time to investigate the effect of different levels of material degradation on the structural response of VCC. The reduction in the concrete reinforcing steel area and bond degradation due to corrosion, and concrete strength reduction due to effects such as alkali-silica reactivity were considered as the variables for the parametric analysis. The results are interpreted to understand the changes in the performance of VCC in the long-term when subjected to hypothetical accident conditions in comparison to intact as built conditions.

Introduction

Vertical concrete casks (VCC) are free standing cylindrical shaped vaults used to store spent nuclear fuel (SNF) from nuclear power plants (NPP). They function as shielding and thermal barriers and also protect the radioactive waste material inside a stainless steel canister that sits within the VCC from hypothetical accidents. A typical VCC is made of reinforced concrete (RC) enclosing a steel liner and it is approximately 6 m in height and 2 m in diameter (see Fig. 1). The loaded weight of a VCC, depending on the configuration and capacity, is about 160 tons. As shown in Fig. 1, the cask has an RC overpack that encloses the steel liner. The steel base plate is attached to the concrete with multiple shear studs and it has the intakes for air circulation. The composite lid assembly, which is essentially a basket filled with concrete and welded to the main circular plate, is attached to the cask body with structural bolts for shielding purposes. Typically, ASTM C36, (2014a) carbon steel is used for all the steel parts in the VCC including the base plate, liner and the lid. These steel parts are protected against weathering by corrosion resistant coatings. In this paper, the focus is on the performance of the VCC under material degradation and impact; therefore, the internals of the SNF canister have not been modeled in detail.

The standard review plan for dry storage systems issued by United States Nuclear Regulatory Commission (NRC) (NUREG, 1997) requires structural evaluation under natural phenomena (e.g., earthquakes, tornados) and hypothetical accidents (e.g., end-drop, non-mechanistic tip-over). The tip-over event could be caused by an extreme event such as a flood, a tsunami or a strong earthquake. The end drop may happen due to mishandling during the loading or transportation of the cask from the fuel handling facility to the storage pad. The maximum acceleration experienced by the cask during impact is an important measure for the design and damage evaluation of the canister, which holds the SNF assemblies.

Concrete is widely used in construction of nuclear facilities and auxiliary structures for its strength, shielding ability and low cost of production. Since casks are very heavy structures, local availability of concrete is a significant advantage for the VCC over metal only casks. Therefore, VCC is one of the most popular casks currently used in the U.S. As of 2016, there were more than 1100 VCC in service at Independent Spent Fuel Storage Installations (ISFSI) in the United States (Jones, 2015). The initial licensing term for these casks has been 20 years from the date of loading. Since the Yucca Mountain permanent repository program has been suspended, the dry storage has become the only option for storage of commercial SNF in the United States after removal from the spent fuel pools. Licenses for VCC could be renewed for a period not to exceed 40 years (U.S. NRC, 2016). In the certificate of compliance renewal application, NRC requires that the aging effects are adequately managed for the period of extended operation. It is mandatory for the cask manufacturers to validate that structures, systems and components important to safety will remain functional for the extended operation period according to NRC regulations, i.e., 10 CFR 72.240 (U.S. NRC, 2017). The degradation mechanisms nuclear energy related RC structures may face are summarized in Table 1.

Table 1 shows all the degradation mechanism possible for the nuclear related concrete structures. A comprehensive literature review and extensive material tests were performed by the authors prior to the tip-over tests for better understanding of the aging processes (Zhou et al., 2014, Attar et al., 2016). Based on these studies, the chloride attack and alkali–silica reactivity (ASR) were found to be the most relevant aging mechanisms, which are assessed in the current research program. Steel corrosion causes considerable problems in reinforced concrete structures. Formation of corrosion products may cause cracking and spalling, reduction of reinforcing steel bar (rebar) area and loss in interfacial bond strength. These effects may increase deflections and reduce ultimate strength (Rodriguez et al., 1997, Cabrera, 1996). The reaction of calcium hydroxide (Ca(OH)₂) in cement with carbon dioxide (CO₂) in the air produces calcium carbonate (CaCO₃). This process is known as the carbonation and it reduces the pH of pore solution of concrete. In the presence of chloride ions or following carbonation of concrete, an expansive corrosion product forms around the rebar leading to cracking, reduction of bond strength and reduction of steel cross sectional area. Since steel plates used for the liner, base plate and the lid are covered with corrosion-resistant coatings, they are assumed to be immune to weathering in this study and the corrosion is limited to the rebar. The material properties of concrete could also be degraded by ASR. The reaction occurs in concrete when alkali from cement react with reactive aggregates to form a siliceous gel. The gel expands by absorption of water and develops a network of microcracks in the concrete. The entire process causes degradation in the form of reduction in strength and modulus of elasticity (Ahmed et al., 2003, Smaoui et al., 2005). Both steel and concrete degradation is covered in this paper. The creep and shrinkage is also investigated by the authors in separate studies (Champiri et al., 2016, Champiri et al., 2015).

Comprehensive experimental and numerical studies were performed by McConnell (1993), and by Witte (1997) and Witte et al. (1998) at Lawrence Livermore and Sandia National Laboratories. McConnell (1993) performed series of tests including end-drops of a 1/3-scale steel billet and a full-scale steel cask from different drop heights. These tests were continued by Witte et al. (1998) on the scaled steel billet with more end-drop, side-drop and tip-over loading scenarios at Lawrence Livermore National Laboratory. They calibrated finite element models (FEM) based on the experimental data. The maximum g load was estimated assuming a rigid behavior for the pad and the estimated g load was applied quasi-statically to demonstrate structural integrity. Methods were also proposed to apply the results to a full-size cask. In addition, finite element analyses of a full-scale generic cask was performed for side-drop and tip-over events and maximum accelerations in the cask were determined for each case. Teng et al. (2003) performed numerical simulation of inclined-drop of a steel cask and estimated the stresses at various locations. The results showed that edge impact from a height of 1.2 m, falling at a 30 degree incline onto a rigid surface is safe as the outer shell remained elastic during the impact. Shah et al. (2003) prepared a detailed model of a tip-over event for HI-STORM 100 cask (Holtec International, 2017) considering a multi-layer soil under the cask foundation. It was concluded that the structural integrity of the canister has significant margins of safety during a tip-over event. However, the variation of material properties of the cask and energy absorption of the pad and foundation were not investigated. Huang and Wu (2009) performed tip-over analysis on a VCC using LS-DYNA (Hallquist, 2006). The acceleration at canister support disk and plastic strain of concrete overpack, considering the effect of bond between the liner and concrete overpack, was estimated. It was found that the modeling of the interface between the VCC and steel liner has a substantial effect on the acceleration. Furthermore, the inclusion of concrete fragmentation significantly decreased the maximum acceleration.

In this study, first an explicit FEM is developed using ABAQUS (Dassault Systemes, 2016) and the results are compared with experimental data from end-drop and tip-over test previously published by the authors (Hanifehzadeh et al., 2017). Two major sources of energy dissipation during the impact were the concrete damage and the deformation of the soil. A well-established concrete damage plasticity (CDP) and a Mohr-Coulomb model were calibrated based on actual material test data and used for the concrete and soil, respectively. After gaining confidence in the FEM through comparisons with the experimentally measured data, a push over analysis is proposed and conducted to evaluate the effect of steel corrosion and concrete degradation on the performance of VCC.