Modeling and Simulation of Gas Centrifuge Cascades for Enhancing the Efficiency of IAEA Safeguards

Since its inception in 1957, the International Atomic Energy Agency (IAEA) has been tasked with ensuring the peaceful uses of nuclear facilities in nation-states that have ratified the Treaty on the Non-Proliferation of Nuclear Weapons (NPT). To achieve this goal, the IAEA employs safeguards techniques to verify the declared use of facilities related to nuclear power. Recently, there has been an increasing interest and demand for nuclear power throughout the world while the budget of the IAEA has remained fairly stagnant. In order for the IAEA to continue to meet their verification goals, there has been an emphasis to enhance the effectiveness and efficiency of safeguards, most notably at gas centrifuge enrichment plants. Most efforts have focused on reducing the level of on-site inspection activities through the development of new technologies for unattended process monitoring. In this research, computational models are developed to study the dynamics and capabilities of gas centrifuge cascades and address the verification goals of the IAEA in the following areas:

One safeguards tool that IAEA inspectors employ during on-site inspections is environmental sampling – analyzing swipe samples from a plant to determine enrichment levels. Discrepancies between measured and expected values of enrichment may not always indicate misuse of a cascade. Here, a method to quantify reasonable levels of enrichment in a cascade due to normal operating transients is developed. This knowledge can help reduce the number of false alarms created by anomalies and the number of on-site inspection activities.

When analyzing the capabilities of a cascade, it is necessary to know the separative performance of a gas centrifuge over a range of operating conditions. Typically, the knowledge of centrifuge geometry and operational parameters are required to characterize a machine. Due to the sensitive nature of gas centrifuge research and development, it is often the case that this information is not known. A novel, semi-empirical method for calculating the separative power and separation factor of a gas centrifuge is developed. The method is verified through a comparative study with results from the Pancake code.

In the event that a nation decides to withdraw from the NPT, it is important to understand the capabilities of an enrichment program to develop enough enriched material for a nuclear weapon. A method to study the cascade interconnection scenario is developed and is used to analyze the Fuel Enrichment Plant in Iran. The importance of including inefficiencies in a capability study is shown.

Unattended monitoring systems are being developed to allow the IAEA to draw safeguards conclusions with less effort. Computational modeling can offer insight into the dynamics of a cascade allowing these technologies to be used in a more efficient way. A transient fluid dynamics and isotope separation model is developed to study signatures of misuse and time-frames associated with the transition between normal and off-normal operating states. Illustrative results show potential indicators of intentional misuse and the time that detectable phenomena remain in the cascade.

The approaches and ideas developed in this dissertation are generic and can be applied to any cascade and enrichment plant.