pyNJOY2GENDF

A Python package to automate NJOY execution for generating multigroup cross-section libraries (GENDF format) and extracting macroscopic cross-sections for reactor physics applications.

Features

• NJOY Input Generation: Automated creation of NJOY input files for RECONR, BROADR, UNRESR/PURR, HEATR, and GROUPR modules
• GENDF Parsing: Robust parser for reading multigroup cross-section data from GENDF files
• Macroscopic XS Extraction: Calculate material-level cross-sections with temperature and sigma0 interpolation
• Parallel Execution: Multi-threaded NJOY processing for large isotope libraries
• Flexible API: Direct material specification or legacy MONTE input file support

Installation

Install this python package in edit mode with the command in this folder:

pip3 install -e .

Quick Start

Generating GENDF Files

import pyNJOY2GENDF.njoy2gendf

# Create NJOY input generator
wr = pyNJOY2GENDF.njoy2gendf.njoy2gendf(
    savedir='gendf_library',
    path2endf='path/to/endf/files',
    njoy_exec='njoy21',
    nuclide_list=['U235', 'U238', 'Pu239'],
    filename_convention='ENDF-B'
)

# Configure and write NJOY inputs
wr.write_njoy_gendf_inputs(
    energybin_edges=[1e-5, 1, 100, 1000, 2e6],
    library_name='ENDF-B-VIII.0',
    temperatures=[300, 600, 900],
    sig0s=[1e10, 1e5, 1e4, 1000, 100, 10, 1],
    groupr_iwt=8  # Thermal reactor spectrum
)

# Run NJOY (parallel execution)
wr.run_njoy_inputs(nthreads=4)

Extracting Macroscopic Cross-Sections

from pyNJOY2GENDF.MatXS import MatXS, MaterialSpec
from pathlib import Path

# Define material composition
fuel = MaterialSpec(
    name='UO2_FUEL',
    nuclides=['U235.gendf', 'U238.gendf', 'O16.gendf'],
    number_densities=[4.5e-3, 2.0e-2, 4.2e-2],  # atoms/barn-cm
    temperature=900.0,  # Kelvin
    gendf_directory=Path('gendf_library/GENDF')
)

# Create processor and extract macroscopic XS
processor = MatXS([fuel], printmicro=False)
processor.process(output_directory='macro_xs')

API Reference

MatXS Class

Extract macroscopic cross-sections from GENDF files with temperature and sigma0 interpolation.

Public Methods

__init__(materials: List[MaterialSpec], printmicro: bool = False)

• Initialize with material specifications
• materials: List of MaterialSpec objects defining materials to process
• printmicro: If True, write microscopic XS per isotope alongside macro XS

from_monte_input(input_path: str | Path, printmicro: bool = False) -> MatXS (classmethod)

• Create from a MONTE input deck file
• input_path: Path to MONTE input file with XSP/SIGNAL/MAT cards
• printmicro: If True, write microscopic XS per isotope
• Returns: MatXS instance ready to process()

process(output_directory: Optional[str | Path] = None) -> None

• Process all materials and write output files
• output_directory: Where to write material_xs/ outputs (default: current directory)

Private Methods

_parse_monte_input(path: Path) -> List[MaterialSpec] (static)

• Parse MONTE input file format
• Returns list of MaterialSpec objects

_load_isotope(gendf_directory: Path, filename: str) -> dict

• Load and cache all data from a GENDF file
• Returns dictionary with cross-section arrays and metadata

_sigma0_iteration(iso_t_list, numdens: np.ndarray, tol: float, maxiter: int) (static)

• Perform sigma0 iteration for multi-isotope materials
• Returns: (sig0 array, list of interpolated isotope data)

_compute_material(mat: MaterialSpec, outdir: Path) -> None

• Compute and write macroscopic cross-sections for one material
• Generates CSV files for total, fission, scattering, etc.

Helper Functions (Module-level)

_write_csv_1d(path: Path, arr: np.ndarray) -> None

• Write 1D cross-section array to CSV file
• Format: "GroupIndex,Value"

_write_sparse_scatter(path: Path, SigS: np.ndarray, threshold: float) -> None

• Write sparse scattering matrix to CSV
• Format: "LegendreIndex,FromIndex,ToIndex,Value"

_ensure_material_dirs(base: Path, matname: str, iso_names: List[str], printmicro: bool) -> None

• Create directory structure for material outputs

MaterialSpec Dataclass

Specification for a single material to process.

Attributes:

• name: str - Material name
• nuclides: List[str] - GENDF filenames (e.g., ["U235.gendf", "U238.gendf"])
• number_densities: List[float] - Atom densities in atoms/barn-cm
• temperature: float - Material temperature in Kelvin
• gendf_directory: Path - Path to directory containing GENDF files

Output Files

The MatXS.process() method generates the following directory structure:

output_directory/
└── material_xs/
    ├── material_list           # List of all materials
    └── <material_name>/
        ├── isolist.csv         # Isotope list with temperatures and densities
        ├── sig0.csv            # Background cross-sections per isotope and group
        └── macro/
            ├── total.csv       # Total cross-section
            ├── fission.csv     # Fission cross-section
            ├── nu_fission.csv  # Nu*fission cross-section
            ├── chi.csv         # Fission spectrum
            ├── absorption.csv  # Absorption cross-section
            ├── kerma.csv       # Kinetic energy release per mass
            └── scatter.csv     # Scattering matrix (sparse format)

Examples

See the examples/ directory for complete working examples:

• endfb8_example/ - ENDF/B-VIII.0 library generation
• jeff4_example/ - JEFF-4.0 library generation
• matxs_example/ - Macroscopic cross-section extraction

Dependencies

• numpy >= 2.3.3
• tabulate
• Python >= 3.13.9
• NJOY21 executable (for GENDF generation)

Citation

If you use this package in your research, please cite the relevant nuclear data libraries and NJOY code.