Improving models for dry storage

TO DATE, MORE THAN 3000 dry storage systems have been safely loaded and emplaced at over 75 domestic independent spent fuel storage installations (ISFSIs) at nuclear power plant sites. Some of these dry storage systems have been in service for more than 30 years and their initial 20-year licences have been extended to cover an additional 40 years of operation.

Dry storage systems are typically designed and licensed to ensure that their structures, systems and components fulfil their intended safety functions under the US Nuclear Regulatory Commission’s NUREG-2215 Standard Review Plan for Spent Fuel Dry Storage Systems and Facilities and Title 10 Code of Federal Regulations 72.24 and 72.236.

The systems are categorised as follows:

Confinement: reasonably maintain confinement of radioactive material;

Subcriticality: maintain the stored spent nuclear fuel in a subcritical state;

Retrievability: designed to allow retrieval of the stored spent fuel;

Shielding: protect workers and public against direct radiation doses and releases of radioactive material and minimise the dose from normal operations and from any off-normal or accident conditions;

Structural integrity: able to accommodate combined normal, off-normal, and accident loads while preserving recoverability;

Thermal: heat-removal systems ensure the spent fuel and other systems remain functional during loading, short- term operations and storage.

Stakeholders across the globe have conducted research to advance the state of knowledge in these functional areas to modernise dry cask storage system designs. They have made design modifications for greater heat rejection, leading to better loading efficiencies with higher thermal loadouts; improved shielding characteristics; made canister inspection easier; and produced corrosion- resistant designs for longer storage. For example, the latest designs use corrosion resistant alloys that minimise chloride-induced stress corrosion cracking; higher-density concrete overpacks that result in a smaller footprint; and underground systems that provide better seismic response and enhance natural shielding from the soil.

Recent approved canister designs require just two years of minimum cooling time for spent fuel wet storage, down from 3-5 years. This enhances safety and operational flexibility at nuclear plants, as moving fuel from spent fuel pools and into passively-cooled dry storage systems reduces the load required for active cooling in the spent fuel pool and increases coping times in the event of an accident. Efforts to further reduce the minimum required cooling times — to one year — have been informed by a recent Electric Power Research Institute (EPRI) thermal modelling benchmarking study, which could further increase safety and operational flexibility benefits.

EPRI coordinates many of these research activities through the Extended Storage Collaboration Program (ESCP), a consortium of over 600 nuclear industry members from 21 participating countries, including utilities, vendors, regulators, government research organisations, national laboratories and universities.

The ESCP thermal modelling subcommittee (one of six in the programme) was established in 2017 to assess and increase the accuracy of best-estimate thermal models through advanced experimental and model benchmarking research efforts. The first phase of its work acquired data for model validation in a single-assembly BWR dry storage system simulator. The second phase was a double-blind thermal model benchmarking study to compare the models with real data from the US Department of Energy (DOE) High-Burnup Demonstration Project for spent fuel with > 45GWd/MTU burnups.

Advanced thermal analyses of dry storage systems use margins in design-basis input assumptions to ensure the peak cladding temperature does not exceed regulatory limits. Maximum temperature limits are set to minimise any concerns with spent fuel integrity during loading, 

drying, and transfer, and during storage. But these conservative design margins mean there is no best- estimate understanding of the thermal behaviour of the dry storage system.

Developing accurate best-estimate thermal models should allow for more efficient use of storage, transport and disposal systems.

EPRI led the DOE HBU project, which loaded 32 high- burnup spent fuel assemblies into an Orano TN-32B dry storage cask at North Anna (in Virginia). The project aims to evaluate the performance of HBU fuel under actual dry storage conditions after 10 years of storage. The TN-32B cask was instrumented with internal thermocouples to collect temperature data during short-term operations and up to 10 years of dry storage. Custom thermocouple lances with nine thermocouples per lance were inserted in selected guide tubes in seven penetrations drilled into the system lid (Figure 1), for 63 thermocouples. The cask was then transferred and placed onto the North Anna ISFSI pad where temperature data will be collected during the storage period.

The maximum temperature near the fuel cladding surface was 237°C during the drying operations and 229°C after steady-state temperatures were reached after 13-15 days of storage. The measured temperature was not the cladding itself but very near the it, in an empty guide tube. The cladding temperature was expected to be 3-5°C higher.

The original licensing basis calculations for the peak cladding temperature in the TN-32B cask was 348°C. The design licensing basis thermal analysis was updated specifically for the HBU TN-32B demonstration cask with a peak cladding temperature calculated to be 318°C.

Using the best-estimate approach, modellers from Pacific Northwest National Laboratory, Orano, and NRC predicted 254-288°C using more realistic input assumptions. These results indicate that the bounding assumptions used in design licensing significantly bias the results to the high side.

The HBU Demonstration Project provided substantial insights into the thermal behaviour of HBU spent fuel in dry storage casks much sooner than anticipated. The results significantly reduce the concerns for the potential embrittling effects of hydride reorientation in HBU

fuel during short-term operations and dry storage, as temperatures were found to be lower than predicted and much lower than the 400°C regulatory limit.

Furthermore, the project identified potential benefits associated with more accurate thermal models:

• More efficient loading, which means fewer casks loaded;
• Shorter minimum required cooling times, enabling faster decommissioning;
• Improved operational flexibilities that could reduce worker doses by allowing additional shielding blankets to be placed during loadings;
• Additional fuel loading patterns to accommodate more diverse fuel assemblies, like higher burnup fuel;
• A smaller repository footprint, allowing for more efficient use of geologic repositories and saving money.

For external cask temperatures, the models predicted the external cask temperature measurements accurately. These temperatures are key for assessing and managing long-term ageing effects associated with extended dry-storage time periods.

The thermal modelling efforts for the HBU Demonstration Project provided additional insights into the importance of identifying, quantifying and ranking uncertainties associated with input parameters and methodology to create accurate thermal models and fuel performance assessments.

As a follow-on activity, subject matter experts from the NRC, DOE, national laboratories, EPRI, vendors and utilities will participate in an expert elicitation using the Phenomena Identification and Ranking Table (PIRT) process to identify, quantify and rank the uncertainties for thermal modelling and regulatory limits on fuel performance during storage operations under the auspices of ESCP. Regulators have used this for an independent expert assessment to support their decision-making, and any technology and/or methodology developments that could further focus research and development such as additional testing or model benchmarking. This process has been an effective tool to accelerate the integration
of new methodologies and advanced technologies in the nuclear industry.

The PIRT results could provide a forward path in using improved best-estimate thermal models, leading to better spent fuel management decision-making. The goal is to provide the technical basis from internationally recognised subject matter experts to develop thermal modelling results with appropriate safety margins that would enhance safety, improve efficiencies and operational flexibilities, and reduce costs. Ultimately, the PIRT report would inform future regulatory licensing activities, such as in the short-term vendor licence amendment submittals, and in the long term topical report that could pave the way for acceptance and implementation of best-estimate thermal models.

Acknowledgements: Multiple organisations collaborated and have provided invaluable contributions to improving thermal modelling. Including the U.S. Department of Energy, Oak Ridge National Laboratory, Pacific Northwest National Laboratory, Sandia National Laboratories, the US Nuclear Regulatory Commission, the Nuclear Energy Institute, Orano TN, and Dominion Energy.

Author information: Aladar A Csontos is Technical executive in the Fuel, Chemistry, Low-Level Waste and High-Level Waste group at the Electric Power Research Institute