Physical quantites

Physical quantities such as Hamiltonian and density matrices are representated through specific classes enabling various handlings.

Spin

Spin

Spin class to determine configurations and spin components.

Matrices

EnergyDensityMatrix

Sparse energy density matrix object

DensityMatrix

Sparse density matrix object

Hamiltonian

Sparse Hamiltonian matrix object

DynamicalMatrix

Dynamical matrix of a geometry

Overlap

Sparse overlap matrix object

Tight-binding models

A tight-binding model from Slater-Koster integrals, used with H.construct(sk).

SlaterKoster

Two-center Slater-Koster parameterization of a tight-binding model

Models from DFTB parameter sets

A DFTB parameter set, for example mio from dftb.org, has one Slater-Koster file (.skf) for each ordered pair of elements. read_slater_koster reads the integrals of one pair into a SlaterKoster with the Hamiltonian and the overlap, interpolated as in DFTB+. It reads O-H.skf and H-O.skf in one call. SlaterKoster.merge joins the pairs into one model:

>>> import sisl
>>> sk = sisl.SlaterKoster.merge(
...     *(sisl.get_sile(f).read_slater_koster() for f in ["O-O.skf", "O-H.skf", "H-H.skf"])
... )
>>> O, H = sk.atoms["O"], sk.atoms["H"]
>>> water = sisl.Geometry(
...     [[0, 0, 0], [0.757, 0.586, 0], [-0.757, 0.586, 0]], [O, H, H], lattice=[10, 10, 10]
... )
>>> Hw = sisl.Hamiltonian(water, orthogonal=False)
>>> Hw.construct(sk)
>>> Hw.eigh().round(2)
array([-24.85, -12.6 , -10.39,  -9.04,  10.2 ,  15.46])
  • Tags. The atoms are matched by their tag, which must be the element of the files ("O" and "H" above). A file with a suffix is another model of the same pair: read N-H_short.skf in place of N-H.skf, with N-N.skf and H-H.skf.

  • Shells. By default, the shells of an element are the basis that its homonuclear file (O-O.skf) documents, here 2s 2p for O and 1s for H. The orbitals of the atoms must have these labels: SlaterKoster.atoms has the atoms with these orbitals, and a range that gives the supercell of the model. Use the shells argument (a shell list for each tag) to select other shells, or when the folder has no homonuclear file (for example the hyb set has no H-H.skf, thus Ga-H.skf needs the shells of H):

    >>> sk_OH = sisl.get_sile("O-H.skf").read_slater_koster({"O": ["2s", "2p"], "H": ["1s"]})
    
  • On-site energies come from the homonuclear files. Thus read O-O.skf and H-H.skf too.

  • Documentation. read_documentation returns the XML documentation at the end of a file: the author, the elements and the basis of the set.

  • Files with f shells (the extended format, first character @) are not supported. The repulsive energy (Spline) is not part of the Hamiltonian and is not read.

Self energies

Self-energies are specific physical quantities that enables integrating out semi-infinite regions.

SelfEnergy

Self-energy object able to calculate the dense self-energy for a given sparse matrix

WideBandSE

Self-energy object with a wide-band electronic structure

SemiInfinite

Self-energy object able to calculate the dense self-energy for a given SparseGeometry in a semi-infinite chain.

RecursiveSI

Self-energy object using the Lopez-Sancho Lopez-Sancho algorithm

RealSpaceSE

Bulk real-space self-energy (or Green function) for a given physical object with periodicity

RealSpaceSI

Surface real-space self-energy (or Green function) for a given physical object with limited periodicity

Bloch’s theorem

Bloch’s theorem is a very powerful procedure that enables one to utilize the periodicity of a given direction to describe the complete system.

Bloch

Bloch's theorem object containing unfolding factors and unfolding algorithms