spgrep-modulation¶
Collective atomic modulation analysis with irreducible space-group representation
Document (develop): https://phonopy.github.io/spgrep-modulation/develop/
Features¶
Calculate representation matrices and irreps formed by phonon eigenmodes
Calculate isotropy subgroups of irreps of space groups on the fly
Generate modulated structures in selected order-parameter directions of isotropy subgroups
Usage¶
from pathlib import Path
import phonopy
from phonopy.structure.symmetry import Symmetry
from spgrep_modulation.modulation import Modulation
# Load Phonopy object
path = Path(__file__).resolve().parent.parent / "tests" / "phonopy_mp-2998.yaml.xz"
ph = phonopy.load(path)
# Prepare Modulation class
qpoint = [0.5, 0, 0] # X point
md = Modulation.with_supercell_and_symmetry_search(
dynamical_matrix=ph.dynamical_matrix,
supercell_matrix=[2, 2, 2],
qpoint=qpoint,
factor=ph.unit_conversion_factor,
)
# Degenerated imaginary mode
frequency_index = 0
print(f"Frequency (THz): {md.eigvals_to_frequencies(md.eigenspaces[frequency_index][0]):.2f}")
# -> Frequency (THz): -4.88
print(f"Irrep shape: {md.eigenspaces[frequency_index][2].shape}")
# -> Irrep shape: (16, 2, 2)
# Modulated cells corresponding to one-dimensional order-parameter directions of isotropy subgroup
cells = md.get_high_symmetry_modulated_supercells(frequency_index)
for cell in cells:
symmetry = Symmetry(cell)
print(f"{symmetry.dataset['international']} (No. {symmetry.dataset['number']})")
# -> Pmma (No. 51) and Cmcm (No. 63)
Installation¶
pip install spgrep-modulation
conda create -n modulation python=3.10 pip
conda activate modulation
git clone git@github.com:lan496/spgrep-modulation.git
cd spgrep-modulation
pip install -e .
# pip install -e ".[dev,docs,vis]"
# pre-commit install
License¶
spgrep-modulation is released under a BSD 3-clause license.
Development¶
Document
sphinx-autobuild docs docs_build
# open localhost:8000 in your browser
Acknowledgements¶
Some test files tests/phonopy_mp-*.yaml.xz are adapted from phonondb under CC BY 4.0.