Physical security analyses for nuclear reactors have historically sought to ensure that there is an acceptably low probability of success for a "design basis" adversary to accomplish a theft or sabotage objective, even for the adversary's most advantageous path. While some have used probabilistic risk assessment to characterize these risks, the lack of a validated attack frequency, among other things, has made this difficult. Recent work at Sandia National Laboratories (SNL) characterizes a facility's security risk for a scenario in terms of level of difficulty an adversary would encounter in order to be reasonably sure of success (the Risk Informed Management of Enterprise Security (RIMES) methodology). Scenarios with lower levels of difficulty can then be addressed through design changes or improvements to the physical protection system. This work evaluates the level of difficulty of a number of attack scenarios for Small Modular Reactors (SMRs), and provides insight to help designers optimize the protection of their facilities. The methodology and general insights are described here.
This report gives an overview of expected design characteristics, concepts, and procedures for small modular reactors. The purpose of this report is to provide those who are interested in reducing the cost and improving the safety of advanced nuclear power plants with a generic design that possesses enough detail in a non-sensitive manner to give merit to their conclusions. The report is focused on light water reactor technology, but does add details on what could be different in a more advanced design (see Appendix). Numerous reactor and facility concepts were used for inspiration (documented in the bibliography). The final design described here is conceptual and does not reflect any proposed concept or sub-systems, thus any details given here are only relevant within this report. This report does not include any design or engineering calculations.
Nuclear fuel reprocessing plants contain a wealth of plant monitoring data including material measurements, process monitoring, administrative procedures, and physical protection elements. Future facilities are moving in the direction of highly-integrated plant monitoring systems that make efficient use of the plant data to improve monitoring and reduce costs. The Separations and Safeguards Performance Model (SSPM) is an analysis tool that is used for modeling advanced monitoring systems and to determine system response under diversion scenarios. This report both describes the architecture for such a future monitoring system and present results under various diversion scenarios. Improvements made in the past year include the development of statistical tests for detecting material loss, the integration of material balance alarms to improve physical protection, and the integration of administrative procedures. The SSPM has been used to demonstrate how advanced instrumentation (as developed in the Material Protection, Accounting, and Control Technologies campaign) can benefit the overall safeguards system as well as how all instrumentation is tied into the physical protection system. This concept has the potential to greatly improve the probability of detection for both abrupt and protracted diversion of nuclear material.
The future of reprocessing in the United States is strongly driven by plant economics. With increasing safeguards, security, and safety requirements, future plant monitoring systems must be able to demonstrate more efficient operations while improving the current state of the art. The goal of this work was to design and examine the incorporation of advanced plant monitoring technologies into safeguards systems with attention to the burden on the operator. The technologies examined include micro-fluidic sampling for more rapid analytical measurements and spectroscopy-based techniques for on-line process monitoring. The Separations and Safeguards Performance Model was used to design the layout and test the effect of adding these technologies to reprocessing. The results here show that both technologies fill key gaps in existing materials accountability that provide detection of diversion events that may not be detected in a timely manner in existing plants. The plant architecture and results under diversion scenarios are described. As a tangent to this work, both the AMUSE and SEPHIS solvent extraction codes were examined for integration in the model to improve the reality of diversion scenarios. The AMUSE integration was found to be the most successful and provided useful results. The SEPHIS integration is still a work in progress and may provide an alternative option.