CrystOD for phonopy users#
crystod-phonon runs on the files a phonopy calculation leaves behind and
adds the symmetry layer: which irrep every mode belongs to, in the labels of
the space-group tables, and which structures an imaginary mode can lower the
crystal to. Nothing has to be recomputed on the phonopy side, and the phonopy
API objects can be handed to CrystOD directly from Python.
What you already have#
phonopy file |
how |
|---|---|
unit cell ( |
|
|
read by default from the current directory, as phonopy does ( |
|
|
the supercell ( |
|
|
supplies the supercell to |
|
|
|
read only with |
The unit cell is reduced to its primitive cell (primitive_matrix="auto"),
so q points are given in primitive reciprocal coordinates, either as
three numbers or as a special-point label (GM, R, X, M, …).
A unit cell plus 4x4x4 FORCE_SETS of cubic SrTiO3 is bundled with the
package, and --example copies them into the current directory and runs the
ordinary command line, so everything on this page can be tried before your own
data is at hand:
crystod-phonon --example
Examples bundled with crystod-phonon (run one with --example NAME):
SrTiO3 ISO-IR irrep labels of the phonon modes of cubic SrTiO3 (writes phonon_irreps.yaml)
= crystod-phonon --irreps --dim '4 4 4' -c 221_PPOSCAR_SrTiO3
SrTiO3_subgroup space groups the imaginary R-point phonon of cubic SrTiO3 can condense into
= crystod-phonon --subgroup --dim '4 4 4' -c 221_PPOSCAR_SrTiO3 --qpoint R
More examples, a directory per feature with its README and inputs: crystod-example
Irreps at every special point (--irreps)#
phonopy’s --irreps (the IRREPS tag) works one q point per run, and
assigns Mulliken labels of the point group (T1u, Eg, …) at the Gamma
point only; elsewhere it prints the characters. crystod-phonon --irreps
labels every degenerate level at every special q point of the space group in
one run, with the ISO-IR labels (GM4-, R5-, X5+, … - the Miller-Love
convention of the ISOTROPY Software Suite, GM4- being T1u in m-3m). These
are the labels the isotropy-subgroup tables and the direct-product tables use,
so a level can be carried straight into crystod-group --parent or
--product.
mkdir phonopy-guide && cd phonopy-guide
crystod-phonon --example SrTiO3
Wrote 221_PPOSCAR_SrTiO3 (bundled example input)
Wrote FORCE_SETS (bundled example input)
Running: crystod-phonon --irreps --dim '4 4 4' -c 221_PPOSCAR_SrTiO3
* Gamma-point activity *
GM4- [T1u] x4 (3 IR, 1 acoustic)
GM5- [T2u] x1 (silent)
* Band path *
k path (seekpath): GM-X-M-GM-R-X | R-M (306 q points)
Labeled points: X, GM, R, M
* Output files *
Phonon irreps written to: phonon_irreps.yaml
Phonon dispersion with irrep labels written to: phonon_irreps.pdf
The result is the file, phonon_irreps.yaml (-o FILE writes it to another
path; the dispersion with the labels, phonon_irreps.pdf, is described
below), a YAML file: the special points of the space group, then for each of
them the modes grouped into degenerate
sets (band_sets), with their irrep and frequency in THz, and at Gamma the
Mulliken symbol and the infrared/Raman activity:
space_group: Pm-3m
nac: false
special_points:
- q_label: GM
q_position: [0.0, 0.0, 0.0]
- q_label: R
q_position: [0.5, 0.5, 0.5]
- q_label: X
q_position: [0.0, 0.5, 0.0]
- q_label: M
q_position: [0.5, 0.5, 0.0]
irreps:
- q_label: GM
q_position: [0.0, 0.0, 0.0]
activity_summary: GM4- x4 (3 IR, 1 acoustic), GM5- x1 (silent)
band_sets:
- band_indices: [1, 2, 3]
irrep_label: ['GM4-(3)']
mulliken: T1u
frequency: 0.0000003685
activity: [acoustic]
- band_indices: [4, 5, 6]
irrep_label: ['GM4-(3)']
mulliken: T1u
frequency: 2.6629186664
activity: [IR]
- band_indices: [7, 8, 9]
irrep_label: ['GM4-(3)']
mulliken: T1u
frequency: 4.7144147359
activity: [IR]
- band_indices: [10, 11, 12]
irrep_label: ['GM5-(3)']
mulliken: T2u
frequency: 7.3464877515
activity: [silent]
- band_indices: [13, 14, 15]
irrep_label: ['GM4-(3)']
mulliken: T1u
frequency: 15.3054117938
activity: [IR]
- q_label: R
q_position: [0.5, 0.5, 0.5]
band_sets:
- band_indices: [1, 2, 3]
irrep_label: ['R5-(3)']
frequency: -1.0867090338
- band_indices: [4, 5, 6]
irrep_label: ['R4-(3)']
frequency: 3.9890931052
...
At Gamma the 15 modes of the five-atom cell are 4 x GM4- (T1u: the acoustic
triplet plus the three infrared-active modes) + 1 x GM5- (T2u, the silent
mode), the textbook count for a cubic perovskite. At R the lowest level is a
negative-frequency R5- triplet: the antiferrodistortive octahedral-rotation
instability of cubic SrTiO3, which the rest of this page follows.
band_indices are 1-based, the numbering of phonopy’s band.yaml and the
one --vector --mode and --modulation --mode take. Any YAML library reads
the file; crystod.phonon.read_phonon_irreps("phonon_irreps.yaml") returns it
as a dictionary (the keys are listed in
04-02. Phonon irreps).
--all-irreps also labels the midpoint of every seekpath path segment, i.e.
the symmetry lines (DT, Z, SM, LD, S, T for Pm-3m), and writes
phonon_irreps_all.yaml instead, so the two surveys coexist:
crystod-phonon --irreps --dim "4 4 4" -c 221_PPOSCAR_SrTiO3 --all-irreps
# -> phonon_irreps_all.yaml
k_path: GM-X-M-GM-R-X | R-M # seekpath
path_midpoints: # midpoints of the k-path segments, ISO-IR k-vector types
- q_label: DT
segment: GM-X
q_position: [0.0, 0.25, 0.0]
- q_label: Z
segment: X-M
q_position: [0.25, 0.5, 0.0]
...
- q_label: DT
segment: GM-X
q_position: [0.0, 0.25, 0.0]
band_sets:
- band_indices: [1, 2]
irrep_label: ['DT5(2)']
...
This is what tells two branches of the same line apart (which DT5 is which)
and takes a few times longer than the special-points-only default (about 10 s
against 2 s for this example). --readfc reads FORCE_CONSTANTS in place of
FORCE_SETS in every mode.
Every --irreps run also draws the picture: the dispersion along the
seekpath path (or --band), as phonopy -p draws it, with the label of
every level at the special points and the imaginary levels in red, in
phonon_irreps.pdf next to the YAML file (the YAML name with .pdf):
crystod-phonon --irreps --dim "4 4 4" -c 221_PPOSCAR_SrTiO3
# -> phonon_irreps.yaml, phonon_irreps.pdf
Documentation: 04-02. Phonon irreps, Symmetry lines as well, The dispersion with the labels.
Imaginary modes: which structures? (--subgroup)#
An imaginary mode says the structure is unstable; its irrep says what it can
turn into. --subgroup labels the imaginary levels and lists the isotropy
subgroups of their irreps, i.e. every order-parameter direction of the
degenerate level and the space group it condenses into:
crystod-phonon --example SrTiO3_subgroup
Kept 221_PPOSCAR_SrTiO3 (identical to the bundled example input)
Kept FORCE_SETS (identical to the bundled example input)
Running: crystod-phonon --subgroup --dim '4 4 4' -c 221_PPOSCAR_SrTiO3 --qpoint R
* Parent structure *
Pm-3m (No. 221)
* Imaginary mode at q = (0.5, 0.5, 0.5) (R) *
mode 1, 2, 3: -1.086709 THz, irrep R5- (degeneracy 3)
* Order parameter directions and isotropy subgroups *
irrep subgroup size index
R5-(0,0,a) 140 I4/mcm 2 6
R5-(a,a,a) 167 R-3c 2 8
R5-(0,a,a) 74 Imma 2 12
R5-(0,a,b) 12 C2/m 2 24
R5-(a,a,b) 15 C2/c 2 24
R5-(a,b,c) 2 P-1 2 48
The distortion of each order-parameter direction can be generated with
crystod-phonon --modulation, or by adding --modulate here.
This is the R-point story of SrTiO3 in one table. The R5- triplet is the
rotation of the TiO6 octahedra about one, two or three cubic axes: rotating
about one axis, (0,0,a), gives the tetragonal I4/mcm phase that SrTiO3
actually adopts below 105 K; equal rotations about all three, (a,a,a), give
R-3c (the LaAlO3 structure); two axes, (0,a,a), give Imma; and the
unequal combinations (0,a,b), (a,a,b), (a,b,c) lower the symmetry to
C2/m, C2/c and P-1. size is the primitive cell of the daughter phase
in units of the parent one, index the index of the subgroup. Freezing in a
single eigenvector, the way a MODULATION run does, explores one of these six
directions, and which one depends on the eigenvector the diagonalizer happened
to return for the degenerate level.
Without --qpoint every q point commensurate with the supercell is scanned
(64 of them for 4x4x4), most unstable level first, so the unstable q points do
not have to be known beforehand; for this force set only the R triplet turns
up. --threshold sets what counts as imaginary (default -0.1 THz), and
--yaml phonopy_params.yaml replaces -c + --dim.
--modulate writes the distorted structure of every direction as well
(--example passes further options on):
crystod-phonon --example SrTiO3_subgroup --modulate
* Distorted structures (amplitude 0.3 A) *
R5-(0,0,a) I4/mcm -> MPOSCAR_R_R5-_0-0-a_I4mcm
crystod-phonon --modulation -c 221_PPOSCAR_SrTiO3 --dim "4 4 4" --qpoint 0.5 0.5 0.5 --mode 1 --amplitude 0.3
R5-(0,a,a) Imma -> MPOSCAR_R_R5-_0-a-a_Imma
crystod-phonon --modulation -c 221_PPOSCAR_SrTiO3 --dim "4 4 4" --qpoint 0.5 0.5 0.5 --mode 1 2 --amplitude 0.3 0.3
R5-(0,a,b) C2/m -> MPOSCAR_R_R5-_0-a-b_C2m
crystod-phonon --modulation -c 221_PPOSCAR_SrTiO3 --dim "4 4 4" --qpoint 0.5 0.5 0.5 --mode 1 2 --amplitude 0.3 0.16434
R5-(a,a,a) R-3c -> MPOSCAR_R_R5-_a-a-a_R-3c
crystod-phonon --modulation -c 221_PPOSCAR_SrTiO3 --dim "4 4 4" --qpoint 0.5 0.5 0.5 --mode 1 2 3 --amplitude 0.3 0.3 0.3
R5-(a,a,b) C2/c -> MPOSCAR_R_R5-_a-a-b_C2c
crystod-phonon --modulation -c 221_PPOSCAR_SrTiO3 --dim "4 4 4" --qpoint 0.5 0.5 0.5 --mode 1 2 3 --amplitude 0.3 0.3 0.16434
R5-(a,b,c) P-1 -> MPOSCAR_R_R5-_a-b-c_P-1
crystod-phonon --modulation -c 221_PPOSCAR_SrTiO3 --dim "4 4 4" --qpoint 0.5 0.5 0.5 --mode 1 2 3 --amplitude 0.3 0.16434 0.08913
Six POSCAR files, one per direction, named after the subgroup, each with the
--modulation command that reproduces it; the space group of every file is
measured with spglib, not assumed. These are the candidate structures to relax
in a structure search. --amplitude scales all of them (default 0.3 A).
Documentation: 04-11. Subgroups from imaginary modes,
Generating the daughter structures,
and the theory in Isotropy subgroups.
--modulation versus phonopy’s MODULATION#
phonopy’s MODULATION tag (MODULATION = 2 2 2, 1/2 1/2 1/2 1 0.3:
supercell, q point, band index, amplitude) freezes one eigenvector of the
dynamical matrix into MPOSCAR. For a non-degenerate mode that is all there
is to it. For a degenerate level, the eigenvector behind a band index is
whatever the diagonalizer returned, an arbitrary combination within the level,
and the structure it produces has no defined symmetry. crystod-phonon --modulation block-diagonalizes the dynamical matrix in the irrep-projected
basis first, so that each mode number is one component of the order parameter:
--mode 1 of the R triplet is the (0,0,a) rotation, --mode 1 2 3 with
equal amplitudes is (a,a,a). It also prints the irrep and degeneracy of every
mode, the star of q, and the space group of the structure it wrote, which it
names after that space group.
What is frozen in is the same physical displacement: phonopy’s eigenvector
divided by the square root of the atomic mass, with its Bloch phase, so for a
non-degenerate mode at a time-reversal-invariant q the pattern is the one
MODULATION writes, up to its sign. Only the amplitude is counted differently.
phonopy multiplies that field by its amplitude divided by the square root of
the number of supercell atoms (a quantity in Å·√amu); --amplitude is in Å,
the norm of the displacement of one primitive cell. See
What is frozen in.
Without --mode, the mode table is printed so the modes can be chosen:
crystod-phonon --modulation -c 221_PPOSCAR_SrTiO3 --qpoint 0.5 0.5 0.5
* Phonon source *
Supercell 4x4x4 inferred from the 320 atoms of FORCE_SETS; pass --dim if that is not the supercell of your force calculation. Primitive cell: 5 atoms of the 5-atom input cell (primitive_matrix auto).
* Selected Q point *
Loading '221_PPOSCAR_SrTiO3 + FORCE_SETS' at q = [0.5, 0.5, 0.5]...
* Phonon modes at q = [0.5 0.5 0.5] *
Mode Freq (THz) Irrep Degeneracy
--------------------------------------------------
1 -1.0867 R5-(3) 3
2 -1.0867 R5-(3) 3
3 -1.0867 R5-(3) 3
4 3.9891 R4-(3) 3
5 3.9891 R4-(3) 3
6 3.9891 R4-(3) 3
7 11.6887 R5+(3) 3
...
15 23.2302 R2-(1) 1
* Star of q (arms related by the space-group rotations) *
|G| = 48, |G_k| = 48, |star of k| = 1
arm 1: k = [+0.5, +0.5, +0.5]
No --mode given. Choose mode number(s) from the table above and rerun with
--mode (and optionally --amplitude) to generate a modulated structure.
Then with the modes:
crystod-phonon --modulation -c 221_PPOSCAR_SrTiO3 --qpoint 0.5 0.5 0.5 --mode 1 2 3 --amplitude 0.3
...
* Modulation *
q-point: [0.5, 0.5, 0.5]
Modes: [1, 2, 3]
Amplitudes (A): [0.3, 0.3, 0.3]
* Symmetry of the generated structure *
Space group: R-3c (#167)
Hall symbol: -R 3 2"c
* Output files *
Modulated structure written to: MPOSCAR_R_mode1+2+3_R5-_R-3c
Note that --dim was not given: --modulation takes the supercell from
--dim if present, else from the supercell_matrix of a phonopy_disp.yaml
(or phonopy_params.yaml) in the current directory, else from the atom count
of FORCE_SETS, and always prints which. --yaml phonopy_params.yaml in
place of -c writes the same file byte for byte. Several q points combine
through --qpoint1/--mode1/--amplitude1, --qpoint2/..., which is how the
arms of a multi-arm star (the three M points of a perovskite) are frozen in
together. Any arm, written as any q + G, is accepted: -0.5 0.5 0 and
0.5 0.5 0 differ by a reciprocal lattice vector and give one structure, and
--qpoint and --qpointN also take special-point labels (--qpoint R).
Where MODULATION takes the supercell you give it, --modulation builds the
supercell of the modulated structure from q, with at most 12 cells per axis:
give q as fractions with denominators up to 12; a q such as 0.15 0 0
(3/20) is refused with an ERROR that names the smallest commensurate
supercell (20x1x1).
Documentation: 04-07. Phonon modulation,
Where the supercell comes from.
Fatbands, L/T character and eigenvectors#
--fatband draws the dispersion along the seekpath path with the
element-projected weight of every mode as the dot size, one PDF per element in
the VESTA element colors, straight from -c + FORCE_SETS; no band.conf
or band.yaml is needed. --element O restricts it to one element,
--band/--band-labels give a manual path, and --nac reads BORN for the
LO/TO splitting (the files are written as fatband_nac_<El>.pdf and
phonon_band_LT_nac.pdf, so both versions coexist).
Documentation: 04-04. Phonon fatbands.
crystod-phonon --fatband --dim 4 4 4 -c 221_PPOSCAR_SrTiO3
* Structure *
Space group: Pm-3m (#221)
* Band path *
k-path (seekpath): $\Gamma$-X-M-$\Gamma$-R-X R-M
Computing phonon band structure with eigenvectors (306 q-points)...
* Output files *
Fatband for Sr written to: fatband_Sr.pdf
Fatband for Ti written to: fatband_Ti.pdf
Fatband for O written to: fatband_O.pdf
--lt draws the same dispersion colored by the longitudinal/transverse
character of each mode, the norm of the eigenvector projected onto the
propagation direction (red = longitudinal, blue = transverse, white = mixed),
valid along diagonal path segments as well; with --nac the split-off LO
branches come out purely red.
Documentation: 04-05. Longitudinal/transverse bands.
crystod-phonon --lt --dim 4 4 4 -c 221_PPOSCAR_SrTiO3
# -> phonon_band_LT.pdf
--vector diagonalizes the dynamical matrix at one q point, prints the mode
table with irrep labels, and exports the chosen modes as .vesta files with
displacement arrows; --mode takes several numbers and sums them, degenerate
levels are exported in the same symmetry-adapted form as --modulation, and
--conventional writes the conventional cell. For a zone-boundary q point the
commensurate supercell and the Bloch phases are applied automatically:
Documentation: 04-06. Phonon eigenvectors.
crystod-phonon --vector --dim "4 4 4" -c 221_PPOSCAR_SrTiO3 --qpoint R --mode 1 2 3
...
* Selected Q point *
Selected q-point: R = [0.5, 0.5, 0.5]
* Phonon modes at q = R *
Mode Freq (THz) Irrep
----------------------------------------
1 -1.0867 R5-(3)
2 -1.0867 R5-(3)
3 -1.0867 R5-(3)
4 3.9891 R4-(3)
...
* Displacement patterns *
Commensurate supercell for visualization: 2x2x2 primitive cells
+ mode 1: R5-(3), -1.0867 THz
+ mode 2: R5-(3), -1.0867 THz
+ mode 3: R5-(3), -1.0867 THz
Arrows are scaled so the largest displacement is 1.5 A; adjust arrow size in VESTA via Edit > Vectors or Properties > Vectors if needed.
* Output files *
Mode table written to: phonon_modes_SrTiO3_R.txt
Sum of modes 1+2+3 written to: POSCAR_SrTiO3_R_mode01+02+03_R5-.vesta
--vibration needs no forces at all: it lists the
irrep-grouped vibration spaces at a q point from the crystal symmetry alone
(04-08. Vibration bases).
--visualize is the interactive counterpart of --vector, and of the
band.yaml that phonopy writes with --eigvecs: one run writes that file
along the --irreps path, as band_vector.yaml (phonopy’s format, an
eigenvector of [re, im] pairs per atom for every band, labels GM, X,
…), and a standalone HTML page that shows every mode at the special points
of the path: the dispersion with a dot on every level, the mode table with
the ISO-IR set labels of phonon_irreps.yaml (Mulliken symbol and activity
at GM; imaginary levels in red) and a 3D scene with the symmetry-adapted
displacement pattern of the selected mode as arrows, on the commensurate
supercell, with an animation of the vibration and, with --bond Ti O 2.1,
the coordination polyhedra.
Documentation: 04-10. Phonon mode viewer.
crystod-phonon --visualize --dim 4 4 4 -c 221_PPOSCAR_SrTiO3 --bond Ti O 2.1
# -> phonon_visualize.html, band_vector.yaml
--mode-resolved-fc turns phonopy’s eigenvalues and eigenvectors into
stiffnesses: for every degenerate set at the special points, the force
constant k = u^+ Phi(q) u^ = mu w of its displacement pattern (eV/A^2 per
primitive cell, the pattern normalized to 1 A over the primitive cell; k < 0
for an imaginary mode) and its effective mass mu, and for every irrep the
force-constant matrix restricted to the subspace of all its modes (the lowest
eigenvalue is the stability of the whole irrep subspace). It reads the same
FORCE_SETS (or FORCE_CONSTANTS with --readfc) and writes
mode-resolved-fc.yaml; in Python, crystod.phonon.mode_resolved_force_constants
takes the phonopy object.
Documentation: 04-12. Mode-resolved force constants.
crystod-phonon --mode-resolved-fc --dim 4 4 4 -c 221_PPOSCAR_SrTiO3
# -> mode-resolved-fc.yaml (R5- of SrTiO3: k = -0.077309 eV/A^2, mu = 15.9994 amu)
Band unfolding (--unfold) and phonopy’s Unfolding#
phonopy’s Python API unfolds bands with phonopy.unfolding.core.Unfolding;
crystod-phonon --unfold computes the same projection (P. B. Allen et al.,
Phys. Rev. B 87, 085322 (2013)) from the files of two phonopy runs, one for
the distorted child and one for its high-symmetry parent, each FORCE_SETS
(or FORCE_CONSTANTS, with --child-readfc / --parent-readfc) read next to
its POSCAR:
crystod-phonon --unfold --child-poscar child/62_PPOSCAR_CaTiO3 --child-dim 2 2 2 \
--parent-poscar parent/221_PPOSCAR_CaTiO3 --parent-dim 2 2 2
# -> phonon_unfold.pdf, phonon_unfold.dat
It differs from Unfolding in four ways:
The child eigenvectors are converted to the ideal-site phase before the projection.
Unfoldingkeeps the Bloch phase of the actual atomic positions, so for a distorted child its weights depend on the supercell the force constants are written in: the force constants of the example (Pnma CaTiO3 and R-3c AlF3, 2x2x2FORCE_SETS), written in a 3x3x3 instead of the 2x2x2 supercell (Phonopy.ph2ph), change its weights by up to 0.10 and 0.22, and those of--unfoldby 6e-12 at most (weights of the degenerate groups at the 306 k of the example path). On the 2x2x2 supercell its weights differ from those of--unfoldby up to 0.07 and 0.14; with the ideal-site phase added they agree to 1e-11.The transformation matrix and the origin shift are not inputs: they are the setting
crystod-group --supergroup-cif PARENT --subgroup-cif CHILDfinds for the two structures, and--trans-mat(rows, child = T @ parent) picks another of the equivalent settings.The child is diagonalized in its own cell (60 x 60 for CaTiO3) instead of the force-constant supercell, so the 306 k of the example take 0.07 s rather than about 22 s (
Unfoldingon the 2x2x2 supercell; about 700 s on the 3x3x3 one).Every unfolded level gets its parent irrep, and the run adds the arms of the stars of the path vertices, the average over the orientation domains, and the parent k and irreps the child Γ modes come from.
Documentation: 04-09. Phonon band unfolding.
The Python API on a live Phonopy object#
Everything above is also a function of crystod.phonon, taking the
Phonopy object you already have. label_phonon_modes is --irreps without
the YAML file, imaginary_mode_subgroups is --subgroup at one q point, and
scan_imaginary_modes is the full scan:
import phonopy
from crystod.phonon import label_phonon_modes, imaginary_mode_subgroups
ph = phonopy.load(supercell_matrix=[4, 4, 4], primitive_matrix="auto",
unitcell_filename="221_PPOSCAR_SrTiO3",
force_sets_filename="FORCE_SETS")
for mode in label_phonon_modes(ph, [0.5, 0.5, 0.5]):
print(mode)
for result in imaginary_mode_subgroups(ph, [0.5, 0.5, 0.5]):
mode = result.mode
print(f"{mode.frequency:.4f} THz {'+'.join(mode.labels)} (degeneracy {mode.degeneracy})")
for sub in result.subgroups:
print(" ", sub.label, "->", sub.symbol)
modes 1,2,3: -1.0867 THz R5-
modes 4,5,6: 3.9891 THz R4-
modes 7,8,9: 11.6887 THz R5+
modes 10,11,12: 12.6621 THz R4-
modes 13,14: 14.8133 THz R3-
modes 15: 23.2302 THz R2-
-1.0867 THz R5- (degeneracy 3)
R5-(0,0,a) -> I4/mcm
R5-(a,a,a) -> R-3c
R5-(0,a,a) -> Imma
R5-(0,a,b) -> C2/m
R5-(a,a,b) -> C2/c
R5-(a,b,c) -> P-1
Each PhononMode carries band_indices (1-based), frequency, labels,
qpoint, qpoint_label, degeneracy and is_imaginary; a q point given as
another arm of the star ([-0.5, 0.5, 0.5]) is mapped onto the tabulated one.
scan_imaginary_modes(ph) runs the same over every commensurate q point
(commensurate_qpoints(ph), 64 here), most unstable level first, and a level
whose irrep has no tabulated subgroups comes back with the reason in
result.errors instead of stopping the scan:
from crystod.phonon import scan_imaginary_modes, commensurate_qpoints
len(commensurate_qpoints(ph)) # 64
for result in scan_imaginary_modes(ph):
print(result.mode, [sub.symbol for sub in result.subgroups], result.errors)
modes 1,2,3: -1.0867 THz R5- ['I4/mcm', 'R-3c', 'Imma', 'C2/m', 'C2/c', 'P-1'] {}
Note
Build the Phonopy object with primitive_matrix="auto", as above. With an
identity primitive matrix on a supercell-sized unit cell, a zone-boundary
instability folds onto the supercell Gamma point, where no irrep of the parent
space group can be attached to it. The -c route of the command line always
uses "auto".
Documentation: Phonon modes and their irreps, From imaginary phonons to subgroups.
Where next#
Python API: the full API page, including
crystod.group.isotropy_subgroupsfor the subgroups of any irrep without a phonon calculation.Tutorials: notebook 01 (phonon irrep labeling, with the labels drawn on a matplotlib dispersion) and notebook 02 (the isotropy subgroup search, including the experiment that shows what freezing one eigenvector at a time misses).
crystod-phonon: the reference of every mode flag, and 02-10. Isotropy subgroups for the same tables from
crystod-group --parent Pm-3m --irrep R5-.