Cross Section Libraries for Nuclear Calculation Codes

Fundamentals of Nuclear Cross Sections

In nuclear physics, the cross section represents the probability of a specific nuclear interaction occurring when incident particles interact with target nuclei. Cross sections are energy-dependent and vary by several orders of magnitude across the energy spectrum, from thermal energies of approximately 0.025 eV to energies exceeding 14 MeV relevant to nuclear fusion.

Cross section libraries must account for these various reaction channels while also providing information on secondary particle distributions, angular distributions, and energy spectra. For a complete representation, the libraries include nuclear data such as resonance parameters, prompt neutron multiplicities, delayed neutron data, and gamma-ray production data.

Major International Cross Section Libraries

ENDF/B Series

The Evaluated Nuclear Data File (ENDF) library maintained by the National Nuclear Data Center (NNDC) in the United States is one of the most widely used nuclear data libraries worldwide. The current version, ENDF/B-VIII.0, contains evaluated data for isotopes from hydrogen to einsteinium, encompassing neutron-induced reactions, spontaneous fission, charged-particle-induced reactions, and decay data.

JENDL Series

The Japanese Evaluated Nuclear Data Library (JENDL) is developed and maintained by the Japanese Nuclear Data Committee. JENDL-5, the latest version as of 2023, provides enhanced data for reactor applications, particularly for light water reactors, fast reactors, and fusion systems. JENDL libraries are known for their emphasis on covariance data and extensive coverage of minor actinides.

JEFF Series

The Joint Evaluated Fission and Fusion (JEFF) project, coordinated by the OECD Nuclear Energy Agency, produces nuclear data libraries for fission and fusion applications. JEFF-3.3, the current version, contains comprehensive data for reactor physics, fuel cycle studies, and activation calculations, with particular attention to European reactor designs.

Rossendorf File Libraries

The Russian nuclear data libraries, including BROND and ROSFOND, provide alternative evaluated nuclear data used extensively in Russian-designed reactors and Eastern European nuclear facilities. These libraries often include distinct evaluation methodologies that prove valuable for uncertainty quantification through international comparisons.

Cross Section Processing

Raw evaluated nuclear data from libraries requires processing into forms suitable for use in nuclear calculation codes. This processing involves several sophisticated steps:

1. Resonance Reconstruction: Converting resonance parameters in the unresolved and resolved resonance ranges to pointwise cross sections.
2. Linearization: Converting continuous energy data to multigroup (energy-binned) structures appropriate for specific calculation methods.
3. Self-shielding Correction: Accounting for the depression of neutron flux in resonance regions due to heavy isotope presence.
4. Doppler Broadening: Adjusting cross sections for the thermal motion of target nuclei, which affects resonance widths and amplitudes.
5. Probability Table Generation: Creating probability tables for use in Monte Carlo calculations with unresolved resonances.
6. Fission Product Yield Data Processing: Converting fission product yield data for burnup calculations.

Several processing systems have been developed to perform these transformations:

• NJOY: The most widely used nuclear data processing system, capable of generating libraries for most calculation codes.
• PREPRO: A collection of modular programs for processing evaluated nuclear data.
• FRENDY: A Japanese-developed processing system compatible with JENDL libraries.
• GAIA: A relatively new processing system designed for efficiency and modern computing environments.

Applications in Nuclear Calculation Codes

Monte Carlo Codes

Monte Carlo codes such as MCNP, Serpent, OpenMC, and TRIPOLI utilize detailed continuous-energy cross sections to simulate particle transport through statistical sampling of individual particle histories. These codes typically require specially formatted libraries containing pointwise cross sections with energy resolution sufficient for accurate resonance treatment.

Deterministic Codes

Deterministic transport codes like PARTISN, DRAGON, APOLLO, and CASMO typically utilize multigroup cross sections where the energy domain is discretized into a finite number of energy groups. These libraries often contain group-averaged cross sections with additional factors for transport approximation and scattering matrix representations.

Depletion Codes

Codes that simulate fuel depletion such as ORIGEN, FISPACT, and VESTA rely on cross section data for neutron-induced activation, fission product yields, and radioactive decay properties. Specialized libraries focus on burnup-related data with particular attention to long-lived fission products and minor actinides.

Shielding Dosimetry Codes

Shielding calculations often require specialized cross section libraries focused on high-energy physics interactions above 20 MeV. Libraries such as LA150, TENDL, and FENDL provide extended energy ranges for these applications, including photon production and transport data essential for radiation protection assessments.

Validation and Uncertainty Quantification

The quality of cross section libraries is established through rigorous validation against critical experiments, integral benchmarks, and integral parameters. The International Criticality Safety Benchmark Evaluation Project (ICSBEP) and the International Reactor Physics Experiment Evaluation Project (IRPhEP) compile high-quality benchmark experiments that provide essential validation data for cross section libraries.

Uncertainty quantification in nuclear data has received increased attention in recent years. Modern libraries now include covariance data that quantify uncertainties and correlations between different energy points and reaction channels. This information enables propagation of nuclear data uncertainties through calculation codes to establish confidence bounds on simulation results.

Several sensitivity and uncertainty analysis codes such as TSUNAMI, SANDY, and Total Monte Carlo exploit these covariance data to assess the impact of nuclear data uncertainties on calculated integral parameters, helping to identify areas needing improved nuclear data evaluation.

Current Challenges and Future Developments

Advanced Reactor Systems

Emerging reactor technologies including Small Modular Reactors (SMRs), Molten Salt Reactors (MSRs), and Generation IV fast reactors pose new challenges for nuclear data. These systems often involve novel materials, higher burnup levels, and different neutron spectra that require extended validation and enhanced cross section data.

Covariance Data Development

While significant progress has been made in covariance data production, coverage remains incomplete, particularly for fission yields, minor actinides, and certain structural materials. Future nuclear data evaluation efforts increasingly prioritize comprehensive covariance data to enable robust uncertainty quantification.

Machine Learning Applications

Machine learning techniques are showing promise for nuclear data evaluation, interpolating between existing evaluations, and identifying inconsistencies in large datasets. These approaches may accelerate nuclear data development while improving quality control.

Multi-scale Data Integration

Future cross section libraries will likely incorporate data across multiple scales, from microscopic nuclear information to macroscopic material properties. This integration will support advanced multi-physics simulations that tightly couple nuclear processes with thermal-hydraulics, structural mechanics, and other phenomena.

Standardization Efforts

Ongoing efforts within the nuclear data community aim to standardize data formats, evaluation methodologies, and processing tools to facilitate data exchange and ensure reproducibility in nuclear calculations worldwide.

Conclusion

Cross section libraries constitute the fundamental physics foundation for all nuclear calculations. Their continual development represents an international collaborative effort involving nuclear data evaluators, experimentalists, and code developers. As nuclear technology evolves to meet emerging energy and non-proliferation challenges, these libraries will continue to adapt with enhanced data quality, extended coverage, improved uncertainty quantification, and advanced capabilities for supporting next-generation nuclear systems.

The accuracy of nuclear design and safety calculations ultimately rests on the quality of these cross section libraries, making their continued development and validation essential for nuclear science and engineering progress worldwide.