Skip to content

Latest commit

 

History

History
93 lines (73 loc) · 4.75 KB

CHANGELOG.md

File metadata and controls

93 lines (73 loc) · 4.75 KB
Raw

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[1.0.1] - 2024-07-11

Added

Fixed

  • Bug introduced with with numpy 2.0 when calculating the unique reciprocal lattice vector delta_G for the permittivity Fourier transform.
  • Bug introduced with with numpy 2.0 when creating the array with the effective permittivity of each layer.

[1.0.0] - 2024-06-04

Added

GuidedModeExp

  • kz_symmetry argument in GuidedModeExp.set_run_options to separate even and odd modes with respect the kz-plane, where k is the in-plane wavevector.

  • symm_thr argument in GuidedModeExp.set_run_options to check that the Hamiltonian is effectively seprated into even and odd blocks.

  • use_sparse argument in GuidedModeExp.set_run_options to use sparse matrices instead of dense when changing the basis to separate the Hamiltonian into even/odd blocks with respect to the kz symmetry plane.

  • delta_gx argument in GuidedModeExp.set_run_options, a small component added to the x-component of g-vectors to avoid problems at g=k+G=0.

  • delta_gx argument in GuidedModeExp.set_run_options, small shift of the absolute value of g=k+G to avoid problems at g=0.

  • only_gmodes argument in GuidedModeExp.set_run_options to calculate the guided modes of the effective slabs.

  • unbalance_sp property in GuidedModeExp to store unbalance of farfield coupling between s-polarized and p-ploarized component.

  • kz_symms property to GuidedModeExp to store the symmetry of the modes with respect the kz symmetry plane.

  • compute_rad_sp method to GuidedModeExp to calculate losses and unbalance_sp.

  • _square_an, _hex_an, _rec_an to GuidedModeExp to store in-plane angles defining possbile kz-planes for square, triangular (hexagonal) and rectangular lattices.

  • _construct_sym_mat method to GuidedModeExp to calculate the reflection matrix w.r.t. a plane parallel to z-axis and rotate by in-plane angle theta.

  • _ind_g method to GuidedModeExp to calculate the index of a vector in an array.

  • _separate_hamiltonian_dense and _separate_hamiltonian_sparse methods to GuidedModeExp to separate the Hamiltonian into even and odd blocks w.r.t. the kz symmetry plane.

  • _calculate_refl_mat method to GuidedModeExp to calculate all the reflection matrices neeed by _separate_hamiltonian_dense and _separate_hamiltonian_sparse.

ExcitonSchroedEq

  • ExcitonSchroedEq class for solving 2D Schroedinger equation of excitons in a periodic potential.

HopfieldPol

  • HopfieldPol class to calculate exciton-photon coupling giving rise to polariton eigenstates.

Photonic Crystal

  • QuantumWellLayer to layer module.

  • add_qw method to PhotCryst to add a QuantumWellLayer (2D active layer) to the photonic crystal.

  • Ellipse shape to shapes module.

  • angle key to lattice.bz_path to store in-plane angles of wavevectros (kx,ky).

  • k_indexes key to lattice.bz_path to store normalized indexes in units of wavevector (indexes are just integer indexes).

Autograd

  • round, shape, concatenate, size, full, eig, matmul, tan, eigsh_ag, inv_ag, sqrt_ag, extend_ag, eig_ag, spdot_ag functions added to numpy-autograd backend.

Visualization

  • eV and a arguments to viz.bands to plot photonic bands in [eV] unit.
  • k_units argument to viz.bands to plot x-axis in normalized unit proportional to the wavevector.
  • calculate_x and _calculate_LL methods to viz module to calculate x-coordinate and light line of bands to plot.

Miscellaneous

  • constants module with physical constants.

  • CHANGELOG file.

  • PrintBackend class and print_utils for printing options. The PrintBackend can be set to base or rich with set_print_backend. The backend can be set to rich only if the package rich is installed.

  • Examples to show new features usage.

  • Test files for polariton module and symmetry separation.

Changed

  • When truncate_abs=abs we check if gmax is equal to one of the |G| in the reciprocal lattice. If that is the case, we increase gmax+=1e-10 to avoid rounding problems.
  • When truncate_abs=abs we calculate only the unique and not-redundant Fourier components of the permittivity profile in each layer.
  • When truncate_abs=abs we calculate the Fourier transform of the permittivity only if the layer is patterned.
  • z_to_lind is now defined in utils.