This is a Python library to compute tight binding models in different dimensionalities and based on a variety of different lattices.
A generic version of this library is called pyqula, and can be downloaded from https://github.com/joselado/pyqula
- 0d, 1d, 2d and 3d systems
- Band structures
- Density of states
- Include magnetism, spin-orbit coupling and superconductivity
- Selfconsistent mean field calculations
- Topological characterization of electronic structures
- Green's function formalism for semi-infinite systems
- Spectral functions
- Kernel polynomial techniques
- Quantum Transport
In the examples folder there are several examples of usage of the library. You will find among others:
- Quantum anomalous Hall and topological insulators
- Topological superconductors and Shiba lattices
- Magnetism in graphene materials
- Twisted bilayer graphene
- Nodal line semimetals
from pygra import geometry
g = geometry.honeycomb_lattice() # get the geometry object
h = g.get_hamiltonian() # get the Hamiltonian object
h.get_bands() # compute the band structure
from pygra import geometry
from pygra import scftypes
g = geometry.honeycomb_zigzag_ribbon(10) # create geometry of a zigzag ribbon
h = g.get_hamiltonian() # create hamiltonian of the system
mf = scftypes.guess(h,"ferro",fun=lambda r: [0.,0.,1.])
scf = scftypes.hubbardscf(h,nkp=30,filling=0.5,mf=mf)
h = scf.hamiltonian # get the Hamiltonian
h.get_bands(operator="sz") # calculate band structure
from pygra import specialgeometry
from pygra.specialhopping import twisted_matrix
g = specialgeometry.twisted_bilayer(3)
h = g.get_hamiltonian(mgenerator=twisted_matrix(ti=0.12))
h.get_bands(nk=100)
from pygra import geometry
from pygra import topology
g = geometry.honeycomb_lattice()
h = g.get_hamiltonian()
h.add_rashba(0.3) # Rashba spin-orbit coupling
h.add_zeeman([0.,0.,0.3]) # Zeeman field
c = topology.chern(h) # compute Chern number
print("Chern number is ",c)
from pygra import geometry
from pygra import films
g = geometry.diamond_lattice_minimal()
g = films.geometry_film(g,nz=20)
h = g.get_hamiltonian()
h.get_bands()
from pygra import geometry
from pygra import kdos
g = geometry.honeycomb_lattice()
h = g.get_hamiltonian()
h.add_haldane(0.05)
kdos.surface(h)
from pygra import geometry
g = geometry.honeycomb_zigzag_ribbon(10) # create geometry of a zigzag ribbon
h = g.get_hamiltonian(has_spin=True) # create hamiltonian of the system
h.add_antiferromagnetism(lambda r: (r[1]>0)*0.5) # add antiferromagnetism
h.add_swave(lambda r: (r[1]<0)*0.3) # add superconductivity
h.get_bands() # calculate band structure
from pygra import geometry
from pygra import spectrum
import numpy as np
g = geometry.kagome_lattice()
h = g.get_hamiltonian()
spectrum.multi_fermi_surface(h,nk=60,energies=np.linspace(-4,4,100),
delta=0.1,nsuper=1)
Parts of the code are written in Fortran for a matter of performance. To compile those functions you need to execute "install.sh" In case they are not compiled, the library will still work but certain parts will be substantially slower. Compiling the fortran routines requires having a fortran compiler, such as gfortran.
Parts of the code rely on Python libraries
- numpy
- scipy
- multiprocess
- numba