piro: rational planning of solid-state synthesis routes for inorganics

piro is a recommendation system for navigation and planning of synthesis of inorganic materials based on classical nucleation theory and semi-empirical, data-driven approximations to its parts. Currently it works with Materials Project data via its Rester API.

• piro creates synthesis reaction planning plots for target polymorphs under a specific set of thermodynamic conditions and a precursor library, where favorable routes are those that are (nearly) Pareto optimal in terms of two metrics: nucleation barrier and phase-selection.

• piro allows retrosynthetic analysis of target inorganic materials to generate a synthesis reaction tree. (i.e. laying out the reaction pathways necessary to arrive at the target from practical/purchasable reagents/starting materials)

• piro supports generation of interactive plots and a web-UI for easy-navigation.

Tutorial jupyter notebooks showing how to use piro locally are provided in the notebooks folder.

Prerequisites

Python 3.9

• piro uses Python 3.9. If you wish to develop or run a local server, you can also following the Docker instructions below to create your own piro container instead of configuring your local environment.

Pymatgen API key

• piro has a dependency on pymatgen which requires you to generate an API key. Go here and follow the instructions to generate your API key.
• Use the API key by either
  • A) adding the environment variable MAPI_KEY to your current environment

  # for example in the terminal
  export MAPI_KEY=<your API key>
  

  • B) using pymatgen's command line to set a global key (note: it's PMG_MAPI_KEY instead of just MAPI_KEY)

  pmg config --add PMG_MAPI_KEY <your API key>
  

Access to MongoDB Cached Database (optional)

• Once you get the credential, set up the environment variable as follow:

  MONGODB_URI = mongodb://{username}:{password}@{host}:{port}/{database}
  

• If you use conda for your development, you can do like this:

  conda env config vars set MONGODB_URI="mongodb://{username}:{password}@{host}:{port}/{database}"
  

• You will need to reactive the conda environment so the change will take effective.

Setup

There are a few options to install and run piro on your machine:

A) Get piro from PyPI

The most recent stable version of piro can be installed from PyPI. We recommend installing piro in a dedicated environment to avoid any version conflicts for its dependencies.

pip install piro

B) For development, install the source from this repo in development mode

python setup.py develop

Piro Module Usage

• Try out a Jupyter notebook:

  jupyter notebook
  

  See the notebooks in piro/notebooks

Web app

README

Citation

If you use piro, we kindly ask you to cite the following publication:

• Aykol, M., Montoya, J.H., Hummelshøj, J. "Rational Solid-State Synthesis Routes for Inorganic Materials" J. Am. Chem. Soc. (2021): https://doi.org/10.1021/jacs.1c04888