materialsvirtuallab / matgenb

Jupyter notebooks demonstrating the utilization of open-source codes for the study of materials science.

Visit the Github Pages for a nicely formatted HTML page and notebook search functionality.

Introduction

This repo is started by the Materials Virtual Lab as a useful collection of Jupyter notebooks that demonstrate the utilization of open-source codes for the study of materials science.

We frequently get requests (from students, postdocs, collaborators, or just general users) for example codes that demonstrate various capabilities in the open-source software we maintain and contribute to, such as the Materials Project software stack comprising Python Materials Genomics (pymatgen), Custodian, and Fireworks. This repo is a start at building a more sustainable path towards sharing of code examples.

It is not limited to the codes we develop - any use of open source software for materials analysis is welcome. Also, anyone is welcome to contribute.

Running the examples from a browser

2013-01-01-Bandstructure of NiO
2013-01-01-Basic functionality
2013-01-01-Calculating Reaction Energies with the Materials API
2013-01-01-Calculating XRD patterns
2013-01-01-Getting crystal structures from online sources
2013-01-01-Ordering Disordered Structures
2013-01-01-Plotting and Analyzing a Phase Diagram using the Materials API
2013-01-01-Plotting the electronic structure of Fe
2016-09-08-Data-driven First Principles Methods for the Study and Design of
2016-09-08-Data-driven First Principles Methods for the Study
2016-09-08-Data-driven First Principles Methods for the Study
2016-09-25-Plotting phonon bandstructure and dos
2017-03-02-Getting data from Materials Project
2017-04-03-Slab generation and Wulff shape
2017-04-14-Inputs and Analysis of VASP runs
2017-05-11-Running Jupyter Notebook on clusters
2017-09-03-Analyze and plot band structures
2017-12-15-Plotting a Pourbaix Diagram
2018-01-01-ChemEnv - How to automatically identify coordination environments in a structure
2018-03-09-Computing the Reaction Diagram between Two Compounds
2018-03-14-Plotting COHP from LOBSTER
2018-07-24-Adsorption on solid surfaces
2018-09-25-Structure Prediction using Pymatgen and the Materials API
2018-11-6-Dopant suggestions using Pymatgen
2019-01-04-How to use Boltztrap2 interface
2019-01-11-How to plot and evaluate output files from Lobster
2019-03-11-Interface Reactions
2020-07-15-How to plot a Fermi surface
2021-08-26-Magnetic Structure Generation as Input for Initial DFT Calculations
2021-5-12-Explanation of Corrections
2022-07-23 Interactive Crystal Toolkit Structure Viewer

Contributing

1. Fork this repo and clone.

   git clone [email protected]:<your_github_username>/matgenb
   cd matgenb

2. Write a new notebook in the notebooks folder.

   cd notebooks
   jupyter notebook

3. Notebooks should be well-documented and simple. The idea here is to be pedagogical. A newcomer to the software (with the right materials science background) should be able to follow the logic without too much difficulty. Feel free to add authorship and contact information, as well as works to cite and acknowledge your contributions. In view that scientific codes tend to be continuously being updated, please put in a list of the key pinned dependencies so that other users can install the exact version of software to run the notebook if needed. The best practice is to put a section that provides a commented out pip install command that can be used in Google Colab. For example,

   # Uncomment the subsequent lines in this cell to install dependencies for Google Colab.
   # !pip install pymatgen==2022.2.27

4. Ideally, please update notebooks as needed to use more modern versions of the codes, and you may update the date of the notebook as needed.

5. Notebooks should be placed in the notebooks folder, and the name should start with the date in YYYY-MM-DD-<intuitive title> format. See existing examples. Remember to add it to the above table too.

6. In the root folder of the repo, convert the jupyter notebooks to html.

   jupyter nbconvert --to html notebooks/*.ipynb --output-dir docs/_posts

7. Commit and push.

   git add .
   git commit -a -m "Describe your contribution"
   git push

8. Submit a pull request from Github.

matgenb's Issues

Import Error when using "get_pourbaix_entries"

Using python version 3.7.1 I am trying to use the example for producing pourbaix diagrams.
The first errors I receive when importing the packages and calling "get_pourbaix_entries" are resolved upon recompiling.
ValueError: numpy.ufunc has the wrong size, try recompiling. Expected 192, got 216

I then get the following error upon recompiling the code below:
# Get all pourbaix entries corresponding to the Cu-O-H chemical system.
entries = mpr.get_pourbaix_entries(["Cu"])

ImportError: cannot import name 'LinearAssignment' from 'pymatgen.optimization.linear_assignment' (unknown location)

Dependabot couldn't authenticate with https://pypi.python.org/simple/

Dependabot couldn't authenticate with https://pypi.python.org/simple/.

You can provide authentication details in your Dependabot dashboard by clicking into the account menu (in the top right) and selecting 'Config variables'.

View the update logs.

`pymacy` package?

The 2018-03-09-Reaction Diagram Construction notebook has the line

from pymacy.bim.thermo import ReactionAnalysis

in the first code cell, but I couldn't find a pymacy package on PyPI. Could you amend the notebook with instructions on how to install that package?

Can developers update files?

Hi, I found that pymatgen has changed a lot change-log, which some modules can not be found in the lastest version. May I request developers update the notebook files for convenience?

Notebook for using adsorption.py

Add the tutorial for using adsorption.py from the supplementary material of:

Montoya, J. H., & Persson, K. A. (2017). A high-throughput framework for determining adsorption energies on solid surfaces. Npj Computational Materials, 3(1), 14. https://doi.org/10.1038/s41524-017-0017-z

to matgenb. Will take a pedantic approach to discussing the nuances of modelling adsorption.

Request: Notebook demonstrating pymatgen transformations capabillities

Requesting an example notebook of using the transformations framework in pymatgen to create structures with full provenance of how they were generated.

• using various transformations
• chaining transformations
• the one to many concept
• converting to a SNL
• embedding provenance for original structure into SNL ( e.g. ICSD, MP, OQMD, etc. IDs and info)

'BandStructure' object has no attribute 'branches'

On the 2013-01-01-Bandstructure of NiO.ipynb notebook 5) Plot Bands and DOS section

bsdosplot = BSDOSPlotter(
    bs_projection="elements",
    dos_projection="elements",
    vb_energy_range=22,
    egrid_interval=2.5,
)
plt = bsdosplot.get_plot(bs, dos=dos)  # error of  'BandStructure' object has no attribute 'branches'