Repository navigation
ACTINV data v1.2.0
ACTINV data v1.2.0
This release adds one artifact, tendl-2025-neutron-damage, and a notice for it. Every v1.1.0 artifact keeps its
identifier, path, size, SHA-256 and source URL, so catalog:<id> references in existing problem files resolve to
exactly the same bytes.
actinv data list
actinv data fetch
actinv data verifytendl-2025-neutron-damage is an actinv-damage-table-2 file on the 709-group structure: damage energy and
displacement cross sections (NRT and, where published constants exist, arc-dpa) with the helium production needed for
He/dpa. The four neutron bundles (full and patched, with and without covariance) reference it, so a neutron problem
gets dpa by naming catalog:tendl-2025-neutron-damage in its damage section with no displacement energies;
actinv new writes that section, and actinv import-fispact uses it for decks that give ALLDISPEN or ATDISPEN.
The proton, deuteron and alpha bundles are unchanged.
What it covers. Every TENDL-2025 neutron evaluation whose nuclide the catalog's decay data (ENDF/B-VIII.0, JEFF-3.3
where ENDF/B-VIII.0 has no record) lists as stable or with a half-life of at least one Julian year. Shorter-lived
products are not given damage; a run counts them in covered_atom_fraction and uncovered_targets.
How it was made. actinv build-damage DIR OUT --method kinematic --projectile neutron --groups fispact-709 --continuum-recoil evaluated --continue-on-error true, from the unmodified TENDL-2025 neutron files (the patched
library's ordinate repairs are not applied), except Pb-208: the distributed file carries a literal NaN, so the
table uses the repaired working-corpus file the activation library is built from (SHA-256 86788a14…). A negative
0 K resonance reconstruction is set to zero, as NJOY RECONR does; this happens only for Fe-54's elastic between 3,491
and 3,517 eV (at most −1e-4 b), and the row records it as zero_k_clipped. The build includes a correction to the
Reich–Moore reconstruction: resonances of opposite AJ sign (the two channel spins) are separate channels, as in NJOY
RECONR. This changes Al-27, Cl-37, Mn-55 and Si-29. The unchanged v1.1.0 activation libraries predate it; for those
nuclides their resolved-range capture differs by up to 16 % near isolated Al-27 resonances (345 keV), with thermal
capture identical and capture resonance integrals within 3e-5. Displacement energies are the NJOY2016 HEATR defaults when no energy is given; the table records them and
a run refuses a different value. Each evaluation file's SHA-256 is recorded in the table.
Fetching. A file already installed under an older version directory with the catalog's SHA-256 is copied into
v1.2.0/ and re-verified instead of downloaded, so an existing v1.1.0 installation fetches only the table and the
notice.
Limits. The table inherits TENDL-2025's evaluation defects (docs/DATA_LIMITATIONS.md) and the kinematic method's
limits: Lindhard-Robinson partition, no cascade simulation or temperature effects, arc-dpa only where published
constants exist, uncovered channels as upper bounds. The displacement energies are a convention, not a
material-specific measurement; rebuild with --displacement-energy for better values.
The table is a transformation of CC-BY-4.0 TENDL-2025 evaluations and is distributed under the same licence; see
ACTINV-DATA-NOTICE-v1.2.0.md for creators, citation and the transformation. SHA256SUMS and SIZES cover every
attached payload and the catalog.