gharib85/BinPo

BinPo: A code for electronic properties of 2D electron systems

BINPO: A CODE FOR ELECTRONIC PROPERTIES OF 2D ELECTRON SYSTEMS

Version 1.1.0

BinPo is a Python code to compute the electronic band structure and other properties in 2D electron systems (2DESs). At present, it could be used in cubic and hexagonal systems. For the cubic case, it is possible to analize the main confinement directions. There is an especial focus in the 2DESs with t2g manifold, like in SrTiO3 (STO) or KTaO3 (KTO), due to their increasing impact in complex oxide community. So that, in that cases further capabilities are available. BinPo solves the Schrödinger-Poisson scheme to obtain the self-consistent potential energy along the slab. The tight-binding Hamiltonian for this slab is created from the transfer integrals in the Maximally Localized Wannier Functions (MLWF) basis. Once the self-consistent (SC) solution is found, properties like projected band structure and Fermi surface, orbital decomposition of electron density and envelope wavefunctions can be computed in post-processing steps. BinPo gives priority to ease-of-use and efficiency in order to produce realistic simulations at low computational cost. High quality and customized figures can be obtained from the simulations. To see details of the methodology applied, please refer to our preprint on https://arxiv.org/abs/2203.11308. For a quick start about how to use this code, please see "Usage" section below.

Prerequisites

You will need to have installed a version 3.x of Python. If you don't have it, please, refer to the official Python website https://www.python.org/. Along with Python 3.x you will need the following libraries:

• NumPy >= 1.19 (https://numpy.org/)
• SciPy >= 1.5 (https://www.scipy.org/)
• Matplotlib >= 3.3 (https://matplotlib.org/)
• NumExpr >= 2.7 (https://numexpr.readthedocs.io/projects/NumExpr3/en/latest/)

We recommend to install a Python scientific distribution like Anaconda (https://www.anaconda.com/) or another one, in which the modules above mentioned are already included. Additionally, you will have to install the Atomic Simulation Enviroment (ASE) library, version >= 3.2 (https://wiki.fysik.dtu.dk/ase/).

Features

Currently, BinPo can be used for cubic and hexagonal systems. For the cubic case, a rotation algorithm allows you to analyze the main confinement directions ((001), (110) and (111)). On the other hand, for hexagonal systems the confinement direction of the 2DES must be along the c axis. We already included several W90 files in BinPo for the following systems:

• strontium titanate, STO
• potassium tantalate, KTO
• BiTeBr

Additionally, the code has modularity to append new materials (see ~/BPexamples/Adding_W90files.pdf).

BinPo allows for computing the following properties:

• The SC solution to the Schrödinger-Poisson scheme in the slab. It includes the SC-potential energy, electron density, electric field and relative permittivity values.
• Bandstructure of the 2DES along high-symmetry points within the irreducible Brillouin zone. The bands calculation can be total (without projections) or projected onto a set of planes. In the particular case of t2g 2DESs, projections onto the atomic orbitals are also available.
• Fermi surface and, more generally, energy slices. In the particular case of t2g 2DESs, projections onto the atomic orbitals are also available.
• For t2g 2DESs there are two more additional capabilities: decomposition of electron density into the atomic orbitals contributions and calculation of envelope wavefunctions at gamma point.

Furthermore, you can analyze the simulations results by Matplotlib interactive plots and generate customized high quality plots for publishing.

License

BinPo is Copyrighted by (C) 2021 BinPo Team. This code is distributed under the terms of the GNU General Public License v3 (GPLv3), see ~/COPYING file or http://www.gnu.org/copyleft/gpl.txt. So that, you are free to use, modify, improve and redistribute it.

Contact and help

For reporting bugs or asking questions about the code, please write to the following email: emanuelm@ucm.es. Request for adding features will be welcomed but without guarantees of future implementations.

What does each file/folder mean?

• WFolder: This folder holds the initial Wannier90 output files (W90 files). These files come from a DFT calculation + Wannierization, and can be seen as the real space Hamiltonian of the bulk unit cell. There is a specific file for each material, bands manifold, as well as for the characteristics of the original DFT calculation (e. g. XC-functional).

• BPexamples: This folder contains .pdf files with detailed descriptions about the calculation presented as examples in our main manuscript. Also a stepwise guide to add new W90 files can be found.

• config_files: This folder contains the configuration files .yaml for the BinPo components, named as BP-components.py. For each component (except for BP_preproc.py and BP_fast-plot.py), there is an associated component.yaml configuration file, where the values not set by terminal are taken from. Besides, in this folder there is a help_config.md file which explains the meaning of all parameters in the .yaml files.

• BP-preproc.py: Pre-processing component. It performs the tasks of filtering and rearrangement from the initial Wannier files.

• BPmodule.py: This file is the main module and contains classes, methods an attributes. It is only called by others programs, the file itself does not need to be executed.

• BPdatabase.py: This module contains a dictionary with information about the materials whose W90 files are included. More data could be added if the corresponding W90 files are available. This file is not runnable, because it is imported by other programs and the information is obtained through the 'material' keyword.

• BP-scp.py: Component for calculation of the SC potential energy along the slab (SC-potential).

• BP-fast_plot.py: Quick plotter for the output of BP-scp.py.

• BP-bands.py: Band structure calculator component. Bands calculation can include projections onto planes. For the particular case of t2g 2DESs, projections onto the orbital character is also available. You can pass the identifier for band calculation by terminal, and many other parameters.

• BP-energy_slices.py:
  Energy slices calculator component. The calculation can include orbital character for t2g 2DESs. You can pass the identifier for slice calculation by terminal, and many other parameters. At the end, the file is automatically saved for plotting.

• BP-energy_plot.py: Energy slices plotter component. Plot the output of BP-energy_slices.py.

• BP-envelope_wfs.py: At present, it is just available for t2g 2DESs. Envelope wavefunctions calculator at gamma point. It allows for obtaining and visualizing the envelope wavefunctions around the gamma point.

• BP-orb_density.py: At present, it is just available for t2g 2DESs. Orbital decomposition of electron density. It allows for decomposing and plotting the electron density according to the orbital character.

• COPYING.md: GPLv3 license file.

• README.md: This file.

Usage

To see the usage and updatable parameters from command-line, please type:

 $ python BP-component.py -h

where BP-component.py is any BinPo component. A list of the more frequently adjustable parameters will appear. If a parameter is ommited, its value will be taken by default from the corresponding configuration file.

In the following lines you will find a short description of how to use BinPo. For further details you can take a look at ~/BPexamples.

STEP 1: pre-processing step:

Run BP-preproc.py. This component has two mandatory arguments: material (mt) and confinement direction (cfd). Choose the combination you want, for example, STO and 111, as follows:

$ python BP-preproc.py -mt STO -cfd 111

It will generate a folder, named as "Hr" + "material" + "direction", that holds the files corresponding to the rearrangment and filtering of r-space Hamiltonian, discretized along normal direction according to the number of planes. You will find the folder ~/HrSTO111 after a succesful pre-processing.

NOTE1: It should be noticed that you need to execute this step just once for each W90 file/material/direction combination.

NOTE2: For hexagonal systems rotations are not allowed, so that the (001) is the confinement direction.

STEP 2: SC-potential calculation:

Run BP-scp.py component with a defined job identifier (id) as:

$ python BP-scp.py -id identifier -mt STO -cfd 111

In this simplified example, many other parameters are being ignored. These will be taken from ~/conf_files/scp.yaml file. The identifier will be unique for this calculation and can be recall later in any post-processing steps. After a successful SC calculation, you will obtain an output folder holding the SC solution, a .log file and a .yaml file with a dictionary of parameters used in the calculation. You can quickly check the output file by means of BP-fast_plot.py as:

 $ python BP-fast_plot.py -id identifier

STEP 3: post-processing step:

NOTE: Each of the post-processing components needs a previous SC-potential calculation whose solution will be accessed by the specific identifier.

Recall the SC solutions and their features from any of the post-processing components (BP-bands.py, BP-energy_slices.py, etc.). Check the parameters needed in each particular case. All new runs will be automatically logged in the file ~/identifier/identifier.log. For example, to compute the bandstructure along KGK path (ph), run the line below. In this case, the ommited parameters will be taken from ~/conf_files/bands.yaml file.

$ python BP-bands.py -id identifier -ph KGK

Further information:

• For learning about the different options to edit plots (colors, colormaps, formats, etc.), check the Matplotlib documentation at https://matplotlib.org/.
• For general information about .yaml files, visit https://yaml.org/.