Introduction

unfolding maps band structures computed in a supercell back onto the Brillouin zone of the primitive cell — the standard “band unfolding” analysis for defect, distorted, or commensurately-modulated calculations.

It works for electrons and phonons, and reads the output of several popular DFT and force-constant codes:

• SIESTA Hamiltonians and wavefunctions (LCAO, spinors supported)
• ABINIT planewave wavefunctions (ETSF netCDF WFK) and phonon DDB files
• phonopy force constants
• Wannier90 tight-binding Hamiltonians

Weights follow the Popescu–Zunger / Lee et al. formulation; every adapter returns a matplotlib Axes with weight-coded (alpha) bands.

Where to go next

• Installation — pip extras per code backend
• Example gallery — figures with entry points to each walkthrough
• API reference — consumer entry points and building blocks

The one idea behind unfolding

A supercell calculation folds several primitive-cell momenta $\mathbf{k}_m = \mathbf{K} + \mathbf{G}_m$ onto each supercell momentum $\mathbf{K}$. The unfolding weight

$$ W_n(\mathbf{k}m) = \sum{\mathbf{G}s \in m} \left|c{n\mathbf{K}}(\mathbf{G}_s)\right|^2 $$

measures how much of supercell band $n$ lives in fold sector $\mathbf{k}_m$. For a pristine crystal the weights are 0 or 1; defects and distortures turn them fractional, which is exactly the diagnostic power of the method.