Close-packing of equal spheres (球の正規最密充填)

This repository contains the Python package cpes, which serves as a dependency for the commutazzio project. The cpes package focuses on constructing and manipulating FCC and HCP structures.

Documentation

The documentation for cpes can be found at https://commutativegrids.github.io/cpes/.

Key Features

Construction of face-centered cubic (fcc) and hexagonal close-packed (hcp) lattice structures layer by layer.
Load the fcc/hcp coordinates of atoms from ISAACS

Installation

To install the package, use:

pip install .

To install the package, use:

pip install . --upgrade

Development Mode Installation

For developers who wish to make changes to the package while using it:

pip install -e .

Optional Jupyter Extension

If you encounter issues with pyvista in Jupyter notebooks, try installing the following extension (ref):

jupyter labextension install @jupyter-widgets/jupyterlab-manager

Basic usage

Generate a fcc/hcp lattice layer by layer

Example: Creating a 10x10x10 hcp lattice with a sphere radius of 1.0:

from cpes import FaceCenteredCubic, HexagonalClosePacking
import numpy as np

fcc = FaceCenteredCubic(10, radius=1.0)

print(fcc.data)  # access the coordinates
np.savetxt('fcc.xyz', fcc.data)  # save the coordinates to a file
coords = np.loadtxt('fcc.xyz')  # load coordinates from a file

The generated lattice is nearly centered, has an atom at (0,0,0), with the normal vector of any added layer points upwards vertically.

Data Thinning Process

Example: Applying a thinning process with a 50% survival rate, and save the file in accordance with the format of homcloud:

fcc.thinning(survival_rate=0.5, save=True, style='homcloud')

Load data from ISAACS

Example: Loading Cartesian coordinates of Au:

from cpes import FccAuCart

fcc_au = FccAuCart(mode='online')
print(fcc_au.original)  # access the original cartesian coordinates
print(fcc_au.data)  # access the normalized coordinates (sphere radius normalized to 1.0)

Check that distance of the k-th nearest neighbor are the same for the generated data and the loaded data:

[i - j < 1e-3 for i, j in zip(fcc_au.distance_array(), fcc.distance_array())]

Methods

.plot: Visualize lattice structures.
.rdf_plot: Generate radial distribution function plots (susceptible to translation and orientation).
.thinning: compute the thinning of self.data with the provided survival_rate.
.sphere_confine: Filter atoms within a specified radius.

CommutativeGrid/cpes