A numerical study of thermal behavior of CASTOR RBMK-1500 cask under fire conditions

https://doi.org/10.1016/j.nucengdes.2021.111131Get rights and content

Highlights

  • Temperature distribution in CASTOR RBMK-1500 cask is provided.

  • Thermal behavior of cask under fire accident conditions was analyzed.

  • Local sensitivity analysis was performed.

  • The fuel load max temperature is not exceeding limiting value.

Abstract

Casks are widely used for interim storage of spent nuclear fuel after its pre-storage in water pools. Interim storage of spent nuclear fuel is a very important part in the whole cycle of nuclear energy generation, so the casks must be designed according to relevant safety standards. Like criticality and doses, thermal safety is an important condition for spent nuclear fuel storage in a dry spent nuclear fuel storage facility and should be analyzed thoroughly. During the entire period of interim storage of spent nuclear fuel, normal operation and emergency situations should be investigated. Analysis of various accident scenarios is one of the means to prevent damage of storage casks and spent nuclear fuel. There are many publications on thermal cask behavior during normal operation conditions, yet publications on cask behavior during accidents are rather limited. There are a few publications presenting fire impact analysis of thermal behavior of casks during transportation. Therefore, this paper presents a numerical study that has been performed to evaluate thermal behavior of a cask under fire condition loaded with spent nuclear fuel stored in an open type dry storage facility pre-stored in a water pool for five years. Thermal performance of a 2D vertical CASTOR RBMK-1500 cask under such conditions (steady-state, fire, post-fire) has been numerically studied by using the ANSYS Fluent code. Also, a local sensitivity analysis of parameters that mostly influence the cask’s temperature was performed.

Introduction

During operation, nuclear power plants generate spent nuclear fuel (SNF) and radioactive waste of different activity levels. Proper management of these wastes is a very important and not straightforward issue. It is recognized now that geological disposal of SNF (after recycling or not) and long-lived radioactive waste could be the final step in their management.

Spent nuclear fuel should be properly stored before recycling or disposal in geological repositories. Because of decay of radioactivity and decrease in decay heat release, duration of the storage period determines how easy the management of such waste is. The spent nuclear fuel assemblies that were used in the reactor core and now have been retrieved emit a lot of heat and are highly radioactive and therefore must be stored in special pools (usually located at the reactor site) until their heat and radioactivity reduce. In addition, water in the pools acts as a barrier against radiation and to disperse heat from spent fuel.

When it comes to interim storage, dry storage is a widely applied and ensures spent fuel cooling under different conditions (Wataru et al., 2008, Herranz et al., 2015). Unlike in wet storage systems, which use water, the dry storage technology employs inert gas or air as a coolant with passive cooling, and a neutron and gamma radiation shielding material. The technology is safer compared to other alternatives, does not depend on external power sources, and its management is cost-effective (Jeong and Bang, 2016). The lifetime of such interim storage facilities is usually 50 years (Alyokhina et al., 2015, Vatulin et al., 2003).

Casks can be placed inside specially designed buildings or in open sites, so called open storage facilities. Inside a building with casks, atmospheric air has no influence on SNF. In some storage buildings central cooling systems are installed, so the temperature in such buildings remains stable most of the time. In open storage facilities, such atmospheric conditions as temperature, wind and humidity are an important factor in SNF cooling, and therefore their influence on the temperature of the SNF inside storage casks must be considered (Alyokhina and Kostikov, 2017).

Dry storage casks can be made from metal or reinforced concrete, but the latter is more economically advantageous (Wataru et al., 2008, Hanifehzadeh et al., 2018). Some casks can serve a dual purpose; i.e. can be used for SNF storage and transportation. Such casks must protect the environment and people against radiation exposure (Bang et al., 2016, Frano et al., 2011) and therefore must meet the requirements of the International Atomic Energy Agency (IAEA) (IAEA, 2018).

At the SNF management facility of an NPP, fuel assemblies are loaded in a cask that is then sealed and transferred for interim storage to an SNF storage facility. Normally, the assemblies inside the cask stay in an inert gas medium to protect them against corrosion and to help with the decay heat removal (Koyanbayev et al., 2019). During the dry storage period, the SNF must maintain its integrity. During this period the decay heat is passively dissipated to the atmosphere (Lee et al., 2016).

There are currently two types of SNF storage casks at the market: ventilated and non-ventilated. There is a wide range of investigations performed in ventilated casks under normal storage conditions. For instance, publications (Tseng et al., 2011, Li and Liu, 2016, Benavides et al., 2019) discuss the results of thermal analysis performed for the casks with BWR/PWR SNF. For WWER-1000 SNF placed in the ventilated concrete casks also rather detailed analysis of the thermal processes have been performed. Recent publications (Alyokhina et al., 2015, Alyokhina and Kostikov, 2017) could be indicated for this issue.

Non-ventilated metallic casks have been investigated in recent publications (Yoo et al., 2010, Creeret et al., 1987, Brewster et al., 2012). Also, publications (Poškas et al., 2006, Poškas et al., 2017) provide thermal analysis of RBMK-1500 SNF in metallic (cast iron) and metal-concrete non-ventilated casks. In Koyanbayev et al., 2019, Cho et al., 2015 a detailed modeling for small-scale model was performed with the Fluent code down to the fuel rods. The modeling for this small-scale model aimed to compare the effective thermal conductivity and porous media approximation for modeling of the fuel assemblies.

Publications on thermal performance analysis of SNF casks during fire are rather limited. Publication (Frano et al., 2011) aimed to evaluate integrity of a non-ventilated cask with SNF under normal storage conditions and in a case of an accident during transportation, e. g., an impact of a severe fire. During its transportation, the cask with fuel element baskets is positioned horizontally and has water or air inside of it to serve as a primary coolant. The thermal analyses for steady-state and transient conditions were carried out with ANSYS 3D, a finite element code, and aimed to identify the maximum temperature of the fuel inside the cask. Geometrical differences between the lid and the bottom of the cask cavity were not taken into consideration, and therefore transversal and longitudinal symmetries were applied. Nearly 135 000 elements were used for the modeling. The researchers took into account all heat transfer modes and thermal loads. The solar irradiation on the outside surface of the cask was maximum (as stated by rules of the IAEA) for 12 h. The analysis of the results revealed that the maximum values of the cask packaging temperature had not exceeded the set temperature limits, and therefore the integrity of the tested packaging system had been ensured.

In Bajwa (2002) an analysis of SNF under severe fire conditions during transportation is presented. The fire conditions were assumed from a real fire that happened in the U.S. in a tunnel during transportation of hazardous materials. Temperatures in the tunnel during the accident reached as high as 815 °C. According to the current U.S. Nuclear Regulatory Commission regulations, transportation containers for hazardous waste must be assessed for a fire with an average fire temperature of at least 800 °C for at least 30 min. The researchers developed a 2D model of a cask (7500 elements) in the ANSYS code. To investigate all heat transfer mechanisms (convection, radiation and conduction) taking part in the process, a comprehensive model of a fuel assembly (with rod fill gasses, fuel cladding and pallets) was developed to define it’s effective conductivity. Then, a model was used with two homogenized fuel (fuel assembly and fuel basket) regions for which effective conductivity values were applied. At first, the modeling was performed for normal (before fire) steady-state conditions. Afterwards, for transient conditions a fire and post-fire periods were modeled for 20 h. The analysis presented in the paper shows that the transportation cask’s performance in case of fire with temperature of 816 °C for at least 7 h with no spent fuel cladding failure is guaranteed.

Paper Sanyal et al. (2011) presents a design of a non-ventilated transportation cask that was analyzed numerically in relation to an accidental fire. The study aimed to develop a more advanced design and analysis methodology for transportation casks under thermal conditions. The investigation showed that modeling taking into consideration only conductive melting of lead shielding does not provide realistic data. Instead it should be closely coupled with an accurate turbulent model that could show the influence of high Rayleigh number convection in a cask.

In Bang et al. (2016) the results of experimental investigations carried out in an open pool fire with a one-sixth piece of a dual-purpose cask were presented. The cask was designed for transportation of the PWR SNF assemblies. The experimental approach had to help to assess what effect on the thermal behavior of the dual purpose non-ventilated cask had a neutron shielding and fins under fire conditions. Along with the pool fire test, a numerical analysis was also performed. The thermal analysis was carried out using the ANSYS Fluent code and showed that the modeled tempertatures are higher compared to the data from the experiments.

In Geraldini and Lorenzo (2016) numerical analyses results for different cases of nuclear waste overpack fires are presented. Also, two prototypes were tested so as to compare their results with the obtained values of numerical analyses. The comparison showed matching results.

In this paper, a numerical modeling has been carried out to evaluate heat removal from a metallic (cast iron) non-ventilated CASTOR RBMK-1500 cask, which is stored in an open type dry storage facility for SNF located at the Ignalina NPP site, and which is under accidental fire conditions. The modeling progresses as follows: firstly the distribution of an initial steady-state temperature in the cask under normal conditions (before fire) is determined; then, external conditions representing fire with temperature of 800 °C for 30 min are introduced; finally, a post-fire cool down analysis is performed. The modeling was performed using the ANSYS Fluent code.

Section snippets

Spent nuclear fuel storage cask

The Ignalina Nuclear Power Plant (NPP) had two RBMK-1500 reactors in operation till 2010 that now are under decommissioning. The dry cask storage technology was selected as the best option for interim storage of SNF from these reactors for a period of 50 years. After pre-storage in water pools (usually at least 5 years), SNF is loaded into non-ventilated metal (cast iron) or metal-concrete casks.

This paper discusses heat removal from a metal (cast iron) CASTOR RBMK-1500 non-ventilated cask. The

Results

As it has been indicated above, the thermal analysis of the cask has been performed under normal conditions (before fire), during fire at 800 °C for 30 min. and under post-fire conditions.

Fig. 2 presents temperature variations inside the cask and in the cask body for the mentioned conditions. As Fig. 2a shows, before the fire, the maximum temperature is in the center of the fuel load. The temperature decreases receding from the center in axial as well as in radial directions. The temperatures

Sensitivity analysis

To assess the impact of the parameter uncertainties on the cask temperatures, a local sensitivity analysis was performed. The selected parameters and their justification are presented below:

  • Decay heat of SNF. It depends on fuel enrichment, position in the reactor during the operation, etc.

  • The effective thermal conductivity coefficient of fuel load. It is defined by a special basket with fuel rod bundles and surrounding He. Its effective conductivity coefficient is defined experimentally with

Conclusions

Modeling of a CASTOR RBMK-1500 cask placed in an open type interim storage facility and analysis of its thermal state in a 30 min duration fire (800 °C) and during the post-fire period have led to the following conclusions:

  • 1.

    During the fire, the fuel load maximum temperature is not really changing, but it is increasing in the post-fire period and reaches its maximum at day 7 after the end of the fire.

  • 2.

    During the fire, the cask body accumulates heat, which heats the fuel load and is also dissipated

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

CRediT authorship contribution statement

Robertas Poškas: Writing - original draft, Visualization. Povilas Poškas: Conceptualization, Writing - review & editing. Kęstutis Račkaitis: Investigation, Writing - original draft, Visualization. Renoldas Zujus: . : Methodology, Formal analysis.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Dr. Robertas Poškas is Chief Researcher in Nuclear Engineering Laboratory at Lithuanian Energy Institute and Associate Professor at Kaunas University of Technology, Lithuania. He holds PhD in Technological Sciences in 2003. His main research interests are heat transfer and hydrodynamics, and environmental pollution reduction. He has published nearly 100 articles in journals and proceedings.

References (32)

Cited by (0)

Dr. Robertas Poškas is Chief Researcher in Nuclear Engineering Laboratory at Lithuanian Energy Institute and Associate Professor at Kaunas University of Technology, Lithuania. He holds PhD in Technological Sciences in 2003. His main research interests are heat transfer and hydrodynamics, and environmental pollution reduction. He has published nearly 100 articles in journals and proceedings.

Habil. Dr. Povilas Poškas is a Head of Nuclear Engineering Laboratory at Lithuanian Energy Institute. He received his PhD degree in 1978. His main research interests are radioactive waste management, heat transfer and hydrodynamics. He has published more than 300 articles in journals, books and proceedings. He is a co-author of two monographs and a member of the Lithuanian Academy of Sciences.

MSc. Kęstutis Račkaitis is Junior Researcher in Nuclear Engineering Laboratory at Lithuanian Energy Institute. His research interests are numerical modeling of heat transfer and hydrodynamics. He has published nearly 10 articles in journals and proceedings.

MSc. Renoldas Zujus is Junior Researcher in Nuclear Engineering Laboratory at Lithuanian Energy Institute. His research interests are numerical modeling of heat transfer and hydrodynamics. He has published nearly 20 articles in journals and proceedings.

View full text