pyRDTP

* [The 5+1 W](#the-51-w)
* [Installation](#installation)
	* [Dependencies](#dependencies)
	* [Manual installation](#manual-installation)
	* [Automatic installation](#automatic-installation)
	* [Jupyter-Lab integration](#jupyter-lab-integration)
* [Usage](#usage)
	* [Basic usage](#basic-usage)
		* [Reading VASP files](#reading-vasp-files)
		* [Transformations](#transformations)
		* [Merging different molecules](#merging-different-molecules)
		* [Writing VASP files](#writing-vasp-files)
	* [Advanced usage](#advanced-usage)
		* [Graphs](#graphs)
		* [Notebook integration](#notebook-integration)
		* [Scripts](#scripts)

• Author

The 5+1 W

1. What?

• pyRDTP is a Python library designed to manipulate and analyze chemical geometries mainly for heterogeneous catalysis.

2. Why?

• Some computational chemistry projects involve a large number of species that are connected between them. The main purpose of pyRDTP is to ease the generation of these structures to calculate them with VASP or another software.

3. How?

• pyRDTP contains different packages to transform different formats into molecules and write these objects back into different formats. It also includes different modules that compare, manipulate and analyze these objects.

4. When?

• Right now! pyRDTP is still in development, but some modules work perfectly at this moment. Moreover, we will tag stable releases for daily usage.

5. Where?

• At Núria López group (ICIQ, Spain). PyRDTP is designed for the daily tasks that every computational chemist in heterogeneous catalysis has to face.

6. Who?

• The weird guy with the blue keyboard. I'm the main developer of these libraries and I HIGHLY appreciate the feedback. If you find a bug, or you need to implement some weird algorithm with these libraries, don't hesitate to contact us.

Installation

Pip

pip install pyrdtp

Dependencies

pyRDTP need the following libraries to run correctly:

• networkx
• shapely
• numpy
• scipy
• matplotlib

Additional libraries are required to visualize molecules with Jupyter-Lab:

• py3Dmol

Jupyter-Lab integration

3Dmol is used to see the molecules in Jupyter-Labs, you will need to install

Usage

Basic usage

Reading VASP files

To simply read a molecule from a CONTCAR file you will need to use the geomio module from pyRDTP.

from pyRDTP import geomio

mol_1 = geomio.file_to_mol('./CONTCAR', 'contcar')

Transformations

Then mol_1 will store all the information from the CONTCAR file into a Molecule object, different methods of Molecule can be used, for example you can remove all the hydrogen atoms from the molecule:

mol_1.atom_remove_by_element('H')

You can also specify that you are reading a Bulk which has special methods.

For example, imagine that you have an unit cell and want to perform a 3x3x4 expansion and add vacuum:

bulk_1 = geomio.file_to_mol('./POSCAR', 'contcar', bulk=True)
bulk_1.expand(x=3, y=3, z=4)
bulk_1.vacuum_add(distance=12, dimension='z')

Merging different molecules

Another useful object is the CatalyticSystem that allows to set a Bulk as surface and multiple Molecules as an unique system:

from pyRDTP.molecule import CatalyticSystem


caty = CatalyticSystem('dummy')
caty.surface_set(bulk_1)
caty.molecule_add(mol_1)

CatalyticSystem has his own methods, as for example move a molecule over 2A at the center of 3 atoms:

atoms = caty.surface[[3,6,7]]
caty.move_over_multiple_atoms(mol_1, atoms, distance=3)

Writing VASP files

Once you finish you can write the new molecule to a POSCAR file:

geomio.mol_to_file(caty, './POSCAR', 'contcar')

Advanced usage

Graphs

pyRDTP uses NetworkX to work with graphs. You can transform any molecule into a graph using the graph module, voronoi option can be used to automatically detect the bonds:

from pyRDTP.operations import graph

grp_1 = graph.generate(mol_1, voronoi=True)

You can also plot your graph:

graph.plot(grp)

Notebook integration

You can use 3Dmol to visualize your molecules in Jupyter-Lab. To do so you need to import the jupyter module.

from pyRDTP.tools import jupyter

jupyter.show_molecule(mol_1)

Scripts

A set of scripts is packed inside the libraries. These scripts are prepared to perform concrete tasks using the command line. They are located inside the scripts folder. Using:

~/pyrdtp/scripts/somescript --help

You can see the purpose and the usage of the script.

Author

Dr. Sergio P. García Carrillo, PostDoc (sp.garcia@utoronto.ca)

Mantainers

Santiago Morandi, PhD Student (smorandi@iciq.es), Oliver Loveday, PhD Student (oloveday@iciq.es)