Jim Gulliford

An Overview of the International Reactor Physics Experiment Evaluation Project

Abstract Interest in high-quality integral benchmark data is increasing as efforts to quantify and reduce calculational uncertainties associated with advanced modeling and simulation accelerate to meet the demands of next-generation reactor and advanced fuel cycle concepts. Two Organisation for Economic Co-operation and Development (OECD)/Nuclear Energy Agency (NEA) activities, the International Criticality Safety Benchmark Evaluation Project (ICSBEP), initiated in 1992, and the International Reactor Physics Experiment Evaluation Project (IRPhEP), initiated in 2003, have been identifying existing integral experiment data, evaluating those data, and providing integral benchmark specifications for methods and data validation for nearly two decades. Data provided by those two projects will be of use to the international reactor physics, criticality safety, and nuclear data communities for future decades. An overview of the IRPhEP and a brief update of the ICSBEP are provided in this paper.

Use of International Criticality Safety Benchmark Evaluation Project Data to Benchmark a JEF2.2-Based Library for MONK

Abstract Modern nuclear criticality safety analysis places great reliance on calculations performed using computer codes, in particular, those employing the Monte Carlo method of solution. In the United Kingdom the acknowledged standard Monte Carlo code for criticality safety assessment is MONK. The accuracy achievable with MONK is ultimately governed by the accuracy of the nuclear data employed and their representation within the code nuclear data library. The U.K. industry uses JEFF-based libraries, taking advantage of modern nuclear data evaluations. Following the release of a frozen version of the library (JEF2.2), a program of work was undertaken in the United Kingdom to develop nuclear data libraries for use in reactor physics, shielding, and criticality application codes and to provide benchmark evidence to support their use. For criticality, this involved developing a hyper-fine-group energy library for the MONK code and undertaking a large program of comparison calculations for selected international experiments. A significant contribution to this validation effort has been the high-quality experimental data from the International Criticality Safety Benchmark Evaluation Project (ICSBEP) International Handbook of Evaluated Critical Safety Benchmark Experiments. This paper summarizes the work involved in arriving at the current stage whereby the use of MONK in conjunction with a JEF2.2-based library is accepted within the U.K. nuclear industry. Specific examples are given, where ICSBEP has provided experimental evaluations for application areas previously unsupported by more traditional experimental data sources.

Feedback on 239Pu and 240Pu nuclear data and associated covariances through the CERES integral experiments

Benchmark measurements of irradiated and un-irradiated fuel samples were performed in the framework of the CERES collaborative program between AEA (UK Atomic Energy Agency) and CEA (French Atomic Energy Commission). These experiments provide relevant data for the validation of fuel burn-up and criticality-safety calculations for the whole fuel cycle. As part of this program, pile-oscillation measurements were carried out on a range of mixed-oxide samples with plutonium of various mass and isotopic contents, both in the MINERVE and DIMPLE reactors. Four core configurations, two over-thermalized situations and two pressurized water reactor (PWR)-type situations, were constituted with different forward and adjoint flux spectra, emphasizing fission and/or capture contributions. The experiments were analyzed using reference TRIPOLI4 calculations with the JEFF-3.1.1 library, using exact three-dimensional models of the core configurations. In a first step, calculations of each DIMPLE configuration were performed and compared with the experiment, showing very good agreements with a maximum C-E of −230 pcm. In the second step, reactivity worth experiments were analyzed, using recently developed exact-perturbation capabilities in TRIPOLI4. A consistent assimilation of calculation over experiment discrepancies was performed with the CONRAD code, using the integral data assimilation method. Covariance matrix on multigroup neutron cross sections and multiplicities were generated and significant trends were identified, especially on the 239Pu and 240Pu capture cross sections in the thermal energy range (E < 0.1 eV). Further investigations should be required to confirm these conclusions, due to the strong dependence of these trends and of posterior covariances to prior covariances.

Preface: Special Issue on the International Reactor Physics Experiment Evaluation Project

The Nuclear Energy Agency (NEA) is a specialized agency within the Organisation for Economic Co-operation and Development (OECD), an intergovernmental organization based in Paris, France. As part of its mission, the NEA assists member countries in maintaining and furthering the scientific and technological knowledge required for safe, economical, and peaceful uses of nuclear energy. The state of this knowledge is often encoded in the increasingly complex algorithms that exist within our predictive computer codes. However, these codes are not developed in isolation; they are anchored to real-world applications through validation. Historically, and at significant expense, countries embarked on experimental programs to generate the experimental data needed for this validation. While much of this work was carried out decades ago, the experimental data generated during these campaigns still underpins our confidence in modern-day simulations of nuclear systems. Furthermore, the gaps identified during validation will continue to shape the direction of future research and development. The NEA Nuclear Science Committee has recognized the need to preserve and share information from programs of integral experiments covering a wide variety of phenomena and has, in close collaboration with the NEA Data Bank, taken initiatives in a number of areas to safeguard such data. These areas include fuel behavior [the International Fuel Performance Experiments (IFPE) database], radiation shielding [the Shielding Integral Benchmark Archive and Database (SINBAD) project], criticality safety [International Criticality Safety Benchmark Evaluation Project (ICSBEP) and Spent Fuel Isotopic Composition (SFCOMPO) database], and reactor physics [International Reactor Physics Experiment Evaluation Project (IRPhEP)]. Fifteen years ago the IRPhEP Handbook was initiated to capture the key physics data required for validation. Since that time hundreds of experimental measurements from 48 different reactor facilities, performed in 19 countries, have been collected, evaluated, and approved. This special issue of Nuclear Science and Engineering attempts to provide an overview of the output from that program by highlighting a small subset of the experiments contributed to the IRPhEP Handbook to date. A key part of the success of the IRPhEP activity has been the development of a formal process for creating experimental reactor physics benchmarks adapted from the method previously established within the ICSBEP. This process involves an international community of experts who have dedicated countless hours to creating and scrutinizing the experimental data and uncertainties associated with the creation of these benchmark models. While time-consuming, the evaluation process frequently turns up valuable information not previously published or publicly known. This information may come from discussions with the experimentalists, discussion of the contents of logbooks or other documents, or feedback gained during sensitivity calculations done during the uncertainty analysis. Besides the specific benefits derived from the validation of our current suite of computer code packages, the IRPhEP activity has also proven to be a very powerful means of transferring important knowledge to a new generation of nuclear engineers and scientists. In particular, the inclusion of younger participants in the review process, working alongside experienced specialists, provides the opportunity for much ‘‘tacit’’ knowledge to be passed on, which would otherwise be lost. We hope that these future specialists will continue to engage with this important activity and successfully apply the knowledge and understanding gained. Finally, we recognize, with thanks, the very high-quality body of work that has been established by the preceding generation of experimentalists.