dirac-wavepacket

Time-dependent wave-packet transport in 2D tilted Dirac/Weyl systems.

dirac_wavepacket solves the time-dependent Dirac equation for a two-component spinor on a 2D grid using a symmetric split-operator Fourier scheme. It is designed for continuum-limit transport calculations in Dirac and Weyl materials with anisotropic Fermi velocities and tilted cones, and supports arbitrary electrostatic barrier shapes, reflecting mass-wall confinement, absorbing drain contacts, source–drain bias, and pseudo-spin-preserving intervalley coupling between K and K'.

Physical model

For valley index τ = ±1 (K and K′):

H_τ = ℏ (v_x σ_x k_x + v_y σ_y k_y)
      + τ ℏ (w_x k_x + w_y k_y) I
      + V(x, y) I  +  M(y) σ_z

The anisotropic Dirac term carries the pseudo-spin texture; the tilt term is proportional to the identity and rigidly displaces the Fermi contour along w. V(x, y) is an arbitrary electrostatic barrier (rectangular / p-n / shaped / polygon / multi-barrier), and M(y) σ_z implements reflecting channel walls via a local gap — the correct confinement for Dirac fermions, which cannot be reflected by a scalar wall because of Klein tunneling.

A 4-component coupled propagator is available for simulations with spatially local, pseudo-spin-preserving intervalley coupling U_KK'(x, y) · I_2. At zero coupling it reproduces two independent single-valley propagations bit-for-bit.

Installation

pip install dirac-wavepacket

Optional extras:

pip install "dirac-wavepacket[fft]"    # pyFFTW acceleration (2–5× speedup)
pip install "dirac-wavepacket[dev]"    # pytest and test utilities
pip install "dirac-wavepacket[all]"    # both of the above

Requires Python ≥ 3.11. Platform-independent (Linux, macOS, Windows).

For contributors (editable install from source)

git clone https://github.com/can-yesilyurt/Dirac-Wavepacket
cd Dirac-Wavepacket
pip install -e ".[dev]"
python -m pytest tests/ -q

Quickstart

from dirac_wavepacket import SimConfig, load_config, run_simulation

cfg = load_config("examples/configs/reflecting_walls_w35_W50.yaml")
cfg.output_dir = "results/quickstart"
result = run_simulation(cfg, make_animation=False, verbose=True)
print(f"T = {result['T']:.4f}, R = {result['R']:.4f}")

Or from the command line:

dwp examples/configs/reflecting_walls_w35_W50.yaml \
    --output results/quickstart

Worked examples

Two self-contained scripts in examples/ reproduce the figures in the companion paper:

Script	What it shows
01_angled_barrier_valley_filter.py	All-electrostatic valley filter via barrier rotation
02_klein_angular_dependence.py	Angular dependence of Klein tunneling, T(θ) scan

Each script has a --quick mode for laptop-scale smoke testing (~1–3 min) and a default mode for the publication-quality figure (~10–30 min on a modern multi-core CPU or a single GPU).

Production drivers

For parameter sweeps over many V_0, geometry, or source–drain voltage values, use the checkpointed parallel drivers installed as console scripts:

dwp-sweep      --config examples/configs/reflecting_walls_w35_W50.yaml \
                   --v0-min 0 --v0-max 1.6 --v0-step 0.02 --workers 8 --no-anim
dwp-sweep-vsd  --config ... --v0 0.228 --vsd-min -0.02 --vsd-max 0.02 \
                   --vsd-step 0.004 --jobs 8

Sweeps are resumable (per-task JSON checkpoints) and graceful against worker crashes — see docs/user_guide.md.

Repository layout

dirac_wavepacket/     Python package (propagator, potentials, etc.)
  cli/                Console-script entry points (sweep drivers)
examples/             Self-contained worked examples
  configs/            YAML configurations
tests/                pytest suite
docs/                 User guide and theory notes

Citation

If dirac-wavepacket contributes to a publication, please cite the Zenodo archive. A software paper describing the package is under review at SoftwareX; citation details will be updated here once the paper is accepted. In the meantime, please cite as:

@software{yesilyurt_dirac_wavepacket_2026,
  author    = {Yesilyurt, Can},
  title     = {Dirac-Wavepacket: time-dependent wave-packet transport
               in two-dimensional tilted Dirac and Weyl systems},
  year      = {2026},
  version   = {1.0.1},
  publisher = {Zenodo},
  doi       = {10.5281/zenodo.19651771},
  url       = {https://doi.org/10.5281/zenodo.19651771}
}

License

MIT — see LICENSE.