Skip to content

6.2 From a q mesh to a phonon dispersion

These are unexecuted teaching inputs and starting models. Original diagrams are schematics, not calculated results. Validate version-specific syntax, licensed or authorized data, numerical convergence and the scientific model before using this workflow.

6.2.1 Model, units and provenance

PW cutoffs and energies use Ry, common force output uses Ry/bohr, and pressure uses kbar. Geometry cards state their coordinate units. Different executables have distinct grammars and time-unit conventions.

Shared inputs, conventions and evidence

Original schematic: From a q mesh to a phonon dispersion. No numerical results are claimed.
Original schematic: From a q mesh to a phonon dispersion. No numerical results are claimed.

6.2.2 Unexecuted inputs and explicit deltas

Use the accompanying instructions to identify the parent calculation and placement of every delta; a snippet is not automatically a standalone input. Preserve all blank-line and file-provenance requirements.

6.2.2.1 Input block 1

Si q mesh
&INPUTPH
 prefix='si', outdir='./scratch/si', tr2_ph=1.0d-14,
 ldisp=.true., nq1=4, nq2=4, nq3=4, fildyn='si.dyn'
/

6.2.2.2 Input block 2

&INPUT
 fildyn='si.dyn', flfrc='si.fc', zasr='crystal'
/

6.2.2.3 Input block 3

&INPUT
 flfrc='si.fc', flfrq='si.GL.freq', flvec='si.GL.modes',
 asr='crystal', q_in_cryst_coord=.true., q_in_band_form=.true.
/
2
0.0 0.0 0.0 40
0.5 0.5 0.5 1

6.2.3 Worked investigation

6.2.3.1 Intuition and prerequisites

DFPT evaluates dynamical matrices on a coarse reciprocal-space mesh. Fourier interpolation converts them into real-space interatomic force constants and reconstructs frequencies elsewhere. Interpolation cannot create missing long-range information from an underconverged mesh. Reuse the Si SCF and Γ test from 6.1, but give the mesh calculation a separate dynamical-matrix prefix.

6.2.3.2 Original three-stage input

See input block 1 above.

6.2.3.3 Checks, pitfalls and exercise

Compare an interpolated off-grid point with an independent direct PH calculation at that point. Increase the original q mesh, not merely the number of plotted path points. Check branch count, acoustic behavior and eigenvector continuity; apparent crossings can be real, and sorting frequencies alone can relabel modes. Exercise: compare 3³,4³ and a stricter mesh, judging a selected optical frequency and acoustic slope rather than smoothness of the line drawing.

6.2.5 Sources and further reading