Skip to main content

A phonon calculation package based on dflow

Project description

dflow-phonon

Overview

This workflow is a part of AI Square.I want to make the process of computing phonon more automatic.So I write this workflow based on dflow.The related computational tasks are submitted to the Bohrium platform.

This workflow supports phonon calculations based on both DFT and DP potential.More specifically, DFT calculations support VASP code and ABACUS code.The part of phonon computation is implemented by using Phonopy code and Phonolammps code.

Easy Install:

pip install --index-url https://pypi.org/simple/ dflow-phonon

Quick Start

You can go to the example folder . You can see that there are some examples for reference . You can go to one of them and fill in the global.json file . Then you can submit the workflow.

If you want to use VASP code to do the DFT calculation , like the folder vasp_dfpt_demo and vasp_displacement_demo . You need to prepare INCAR , POSCAR , POTCAR , global.json and param.json , then :

dflowphonon --vasp

If you want to use ABACUS code,like the folder abacus_demo.You need to prepare INPUT , STRU , *.UPF , global.json and param.json ( notice that *.orb and KPT is optional ) , then :

dflowphonon --abacus

If you want to use DP potential,like the folder dp_demo , you need to prepare POSCAR , frozen_model.pb , global.json and param.json , then :

dflowphonon --dp

Then you can monitor the workflow process on the website.

If you want to watch the phonon band structure , you can go to the folder where band.dat exists , then

dflowphonon-plot

The file named phonon_band_structure is the result picture.

Details of Workflow

If you want to use VASP code,you can choose from two calculation methods.One is linear response method and anothers is finite displacement method.

One advantage of the linear response method over the finite displacement method is that one does not need to create super-cells, which can be in some cases computationally more efficient. Moreover, in systems where the phonon dispersions behave in an anomalous way (like systems with Kohn anomalies) the linear response method is more suitable, because it is capable of calculating the exact phonons at the requested points.

One advantage of the finite displacement method is that it is an add on that can work with any code, also non-density functional theory codes. All is needed is the ability of the external code to compute forces (codes like ABACUS and wien2k for example do not include an implementation of the linear response method, but the finite displacement method implemented in phon can be used in conjunction with these codes).

You can read this paper[1] to get more information about these methods.

Unlike VASP code, ABACUS code does not support linear response method for now, so the finite displacement method is used by default when using ABACUS code.

In order to calculate phonon structure band using DP potential,I use phonolammps code to achieve it.

In order to handle force constants obtained by finite displacement method and linear response method.I use phonopy code which is a robust and easy-to-use open-source code for first principles phonon calculations.You can read this paper[2] for more information.

Reference

[1] Alfe, D. (2007). Tutorial on calculating phonons: comparing the linear response and the small displacement methods. Department of Earth Sciences and Department of Physics and Astronomy, University College, London.

[2]Togo, A., & Tanaka, I. (2015). First principles phonon calculations in materials science. Scripta Materialia, 108, 1-5.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

dflow-phonon-0.0.26.tar.gz (11.2 kB view details)

Uploaded Source

Built Distribution

dflow_phonon-0.0.26-py3-none-any.whl (17.9 kB view details)

Uploaded Python 3

File details

Details for the file dflow-phonon-0.0.26.tar.gz.

File metadata

  • Download URL: dflow-phonon-0.0.26.tar.gz
  • Upload date:
  • Size: 11.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.10.4

File hashes

Hashes for dflow-phonon-0.0.26.tar.gz
Algorithm Hash digest
SHA256 f378ae11fc9ae32cd25cfc1d6d3395e9a442e08c1c673e6b6c953648a0a5fa62
MD5 dc2d58c98097c4dc132f63c3325ba1f0
BLAKE2b-256 851a66a3bb90454caee9faa4adeb01859f590475b50dfe4ff906f02ec96d1585

See more details on using hashes here.

File details

Details for the file dflow_phonon-0.0.26-py3-none-any.whl.

File metadata

File hashes

Hashes for dflow_phonon-0.0.26-py3-none-any.whl
Algorithm Hash digest
SHA256 863fc332650d940e46e53afea62804776727908d29886f6bed1a3d3b16d72195
MD5 12213d9d1aa24361ab173af773d96f8d
BLAKE2b-256 b1fcf59fc82cd9121461287f108e138e25c00ffabfae2bd46faf77abe38bf223

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page