Source-linked AI summary
Towards a More Complete and Accurate Experimental Nuclear Reaction Data Library (EXFOR): International Collaboration Between Nuclear Reaction Data Centres (NRDC)
N. Otuka, E. Dupont, V. Semkova, B. Pritychenko, A. I. Blokhin, M. Aikawa, S. Babykina, M. Bossant, G. Chen, S. Dunaeva, R. A. Forrest, T. Fukahori, N. Furutachi, S. Ganesan, Z. Ge, O. O. Gritzay, M. Herman, S. Hlavač, K. Katō, B. Lalremruata, Y. O. Lee, A. Makinaga, K. Matsumoto, M. Mikhaylyukova, G. Pikulina, V. G. Pronyaev, A. Saxena, O. Schwerer, S. P. Simakov, N. Soppera, R. Suzuki, S. Takacs, X. Tao, S. Taova, F. Tarkanyi, V. V. Varlamov, J. Wang, S. C. Yang, V. Zerkin, Y. Zhuang
TL;DR
EXFOR must remain complete and accurate as the scope of published nuclear reaction data expands. The paper describes how the NRDC and IAEA NDS maintain and improve the library through collaborative compilation, systematic quality assurance, and format extensions. These efforts include recovering missed datasets, detecting erroneous entries, and supporting machine-readable uncertainties and covariances.
Problem
Maintaining completeness and accuracy is increasingly difficult as EXFOR’s published-data scope expands, while light charged-particle coverage remains weaker than neutron coverage.
Method
The NRDC collaborates under IAEA NDS coordination through distributed compilation, user feedback, systematic error detection, and extensions to EXFOR’s machine-readable format.
Results
Targeted efforts compiled missed therapeutic radioisotope, proton-induced total reaction, and nuclear resonance fluorescence datasets, while inspections identified specific erroneous data sets.
Takeaways & Limitations
EXFOR maintenance depends on combining compilation, quality assurance, user feedback, and richer machine-readable uncertainty and covariance information.
Abstract
from arXiv · showhide
The International Network of Nuclear Reaction Data Centres (NRDC) coordinated by the IAEA Nuclear Data Section (NDS) is successfully collaborating in the maintenance and development of the EXFOR library. As the scope of published data expands (e.g., to higher energy, to heavier projectile) to meet the needs from the frontier of sciences and applications, it becomes nowadays a hard and challenging task to maintain both completeness and accuracy of the whole EXFOR library. The paper describes evolution of the library with highlights on recent developments.
I. INTRODUCTION
EXFOR is a comprehensive library of experimental nuclear reaction data maintained through international collaboration among fourteen NRDC data centres under IAEA NDS coordination.
- EXFOR contains cross sections and other nuclear reaction quantities induced by neutron, charged-particle, and photon beams.
- Compilation is mandatory for low- and intermediate-energy neutron and light charged-particle reactions up to 1 GeV and A ≤12.
- Heavy-ion reactions with A ≥13 and photon-induced reactions are additionally compiled on a voluntary basis.
- Fourteen data centres collaborate in the NRDC mainly to compile and exchange experimental data using the common EXFOR format.
II. COMPILATION
EXFOR compilation combines literature scanning, distributed centre responsibilities, review-based transmissions, and shared retrieval infrastructure, while its scope and entry count have expanded over time.
- Literature scanning identifies experimental articles, which are registered for assignment of EXFOR entry numbers to responsible data centres.NDS regularly scans more than 60 journals after CINDA readers became unavailable.
- Originating centres transmit new or revised entries preliminarily, receive comments for at least one month, and then issue corrected final transmissions.
- Since 2005, NDS has maintained the EXFOR Master File, updating it monthly and providing its contents through a web retrieval service.
- Neutron data were compiled first, while charged-particle and photon-induced data began in the middle of the 1970s; neutron and charged-particle contents are now comparable.
- More than 20,000 experimental works have accumulated in EXFOR, and compilation tools, digitizers, and compiler workshops support data entry and extraction.
III. COMPLETENESS
EXFOR is expected to be complete for low- and intermediate-energy neutron and light charged-particle data, but coverage gaps remain and targeted efforts have recovered missed measurements.
- Light charged-particle coverage, especially differential cross sections, is weaker than neutron coverage because compilation began later.
- Completeness depends strongly on data type and availability, with conference proceedings, arbitrary-unit raw data, and unavailable author data near the scope boundary.
IV. QUALITY ASSURANCE
EXFOR quality assurance addresses unavoidable manual-entry errors through user feedback, systematic detection, checking codes, and monitored corrections.
- Manual typing of EXFOR entries, including hundreds of numerical lines, means compilation cannot eliminate all errors.
- WPEC SG30 systematically detected and corrected errors after EXFOR contents were translated into the extended Computational Format.
- Additional inspections detected 29 energy-unit errors, 59 implausible level energies, 17 conservation violations, and 288 partial datasets lacking excitation levels.
- Checking codes support compilers in eliminating format and physical errors, while user comments are registered and corrections monitored by NDS.
V. OTHER IMPROVEMENTS
EXFOR improvements focus on making uncertainty information more complete and machine-readable for evaluation. The effort also preserves primary spectra needed to derive covariances between resonance parameters.
- EXFOR now accommodates correlation properties and covariance matrices in computer-readable form to support evaluations with fewer assumptions.Guides encourage experimentalists to submit more detailed uncertainty information because article-level error propagation is often insufficient.
- NDS is compiling and documenting time-of-flight spectra with EC-JRC IRMM to support covariance evaluation for resonance parameters.The spectra enable error propagation from the primary measurable when covariances between resonance parameters are needed.
VI. CONCLUSIONS
NRDC faces growing demand for experimental reaction data while increasingly supporting machine-readable processing. It therefore uses multiple approaches and user feedback to maintain EXFOR’s completeness and accuracy.
- Growing demand for experimental reaction data and machine-readable information increases the challenge of maintaining EXFOR.
- NRDC applies varied approaches to maintain EXFOR as a complete and accurate library.
- Feedback from EXFOR users is important for achieving the library’s maintenance goal.