CIF Bond Analyzer (CBA)

The CIF Bond Analyzer (CBA) is an interactive, command-line-based application designed for high-throughput extraction of bonding information from CIF (Crystallographic Information File) files. CBA offers Site Analysis, System Analysis for binary/ternary systems, and Coordination Analysis. The outputs are saved in .json, .xlsx, and .pngformats.

The current README.md serves as a tutorial and documentation - last update July 9, 2024

Demo

The code is designed for interactive use without the need to write any code.

Scope

Any .cif files.

Value

CBA simplifies crystal structure analysis by automating the extraction of minimum bond lengths, which are crucial for understanding geometric configurations and identifying irregularities. Histograms and figures assist in identifying distinct bond lengths and structural patterns.

Getting started

Copy each line into your command-line applications:

$ git clone https://github.com/bobleesj/cif-bond-analyzer.git
$ cd cif-bond-analyzer
$ pip install -r requirements.txt
$ python main.py

Once the code is executed using python main.py, the following prompt will appear, asking you to choose one of the three analysis options:

Welcome! Please choose an option to proceed:
[1] Conduct site analysis.
[2] Conduct system analysis.
[3] Conduct coordination analysis.
Enter your choice (1-3): 1

For any option, CBA will ask you to choose folders containing .cif files:


Folders with .cif files:
1. 20240623_ErCoIn_nested, 16 files, 136 nested files
2. 20240612_ternary_only, 2 files
3. 20240611_ternary_binary_combined, 5 files
4. 20240623_teranry_3_unique_elements, 3 files
5. 20240611_binary_2_unique_elements, 4 files

Would you like to process each folder above sequentially?
(Default: Y) [Y/n]:

You may then choose to process folders either sequentially or select specific folders by entering numbers associated with the folders prompted. For each folder, CBA generates site pair data saved in site_pairs.json or site_pairs.xlsx.

Preprocess

The following discusses formatting, supercell generation, and atomic mixing information.

1. Format files

CBA uses the CifEnsemble object from cifkit (https://github.com/bobleesj/cifkit) to conduct preprocessing automatically.

CBA standardizes the site labels in atom_site_label. Some site labels may contain a comma or a symbol such as M due to atomic mixing. CBA reformats each atom_site_label so it can be parsed into an element type that matches atom_site_type_symbol.
CBA removes the content of publ_author_address. This section often has an incorrect format that otherwise requires manual modifications.
CBA relocates any ill-formatted files, such as those with duplicate labels in atom_site_label, missing fractional coordinates, or files that require supercell generation.

2. Supercell generation

For each .cif file, a unit cell is generated by applying the symmetry operations. A supercell is generated by applying ±1 shifts from the unit cell.

3. Atomic mixing info

Each bonding pair is defined with one of four atomic mixing categories:

Full occupancy is assigned when a single atomic site occupies the fractional coordinate with an occupancy value of 1.
Full occupancy with mixing is assigned when multiple atomic sites collectively occupy the fractional coordinate to a sum of 1.
Deficiency without mixing is assigned when a single atomic site occupying the fractional coordinate with a sum less than 1.
Deficiency with atomic mixing is assigned when multiple atomic sites occupy the fractional coordinate with a sum less than 1.

Analysis Options

CBA provides three options for analysis.

Option 1. Site Analysis

Purpose: Site Analysis determines the shortest distance and its nearest neighbor for each label in atom_site_label.
Process: For each atom in the unit cell, Euclidean distances are calculated from the atom to all atoms in the supercell. The position of the atom in the unit cell for each site label is determined based on the atom with the greatest number of shortest distances to its neighbors.
Example: If a .cif file under atom_site_label contains four site labels: Er1, Er2, Er3, and Er4. The bonding pair from the site label Er4 and its nearest neighbor Er2 is unique and recorded. The bonding pair from Er3 to Er2 is also considered unique. However, the pairs Er4-Er2 and Er2-Er4 are considered identical. Out of the two pairs, the pair with the shorter distance is recorded below.

Output 1.1 Excel and JSON

Data for each folder is saved in site_pairs.json or site_pairs.xlsx. Below is an example of the JSON structure for bond pairs:

{
  "Co-Co": {
    "250361": [
      {
        "dist": 2.529,
        "mixing": "full_occupancy",
        "formula": "ErCo2",
        "tag": "rt",
        "structure": "MgCu2"
      }
    ],
    "1955204": [
      {
        "dist": 2.46,
        "mixing": "full_occupancy",
        "formula": "Er2Co17",
        "tag": "hex",
        "structure": "Th2Ni17"
      },
      {
        "dist": 2.274,
        "mixing": "full_occupancy",
        "formula": "Er2Co17",
        "tag": "hex",
        "structure": "Th2Ni17"
      }
    ]
  }
}

The minimum bond pair for each file is saved in element_pairs.json and element_pairs.xlsx.

{
  "Co-Co": {
    "250361": [
      {
        "dist": 2.529,
        "mixing": "full_occupancy",
        "formula": "ErCo2",
        "tag": "rt",
        "structure": "MgCu2"
      }
    ],
    "1955204": [
      {
        "dist": 2.274,
        "mixing": "full_occupancy",
        "formula": "Er2Co17",
        "tag": "hex",
        "structure": "Th2Ni17"
      }
    ]
  }
}

Here is a screenshot of element_pairs.xlsx.

Output 1.2 text summary

A summary text file, summary_element.txt, lists the shortest bonding pairs and identifies missing pairs across selected folders:

Summary:
Pair: In-In, Count: 4, Distances: 2.736, 2.782, 2.785, 2.793
Pair: Pd-Ge, Count: 4, Distances: 2.449, 2.455, 2.489, 2.672
Pair: Pd-Sb, Count: 4, Distances: 2.505, 2.700, 2.737, 2.793
Pair: Si-Si, Count: 4, Distances: 1.975, 2.289, 2.325, 2.533
Pair: Rh-Ge, Count: 2, Distances: 2.484, 2.495
Pair: Ru-Si, Count: 2, Distances: 2.394, 2.519
Pair: Sb-Sb, Count: 2, Distances: 2.573, 2.793
Pair: Co-Ga, Count: 1, Distances: 2.485
Pair: Co-Sb, Count: 1, Distances: 2.594
Pair: Co-Sn, Count: 1, Distances: 2.737

Missing pairs:
Co-In
Co-Ir
Co-Ni
Co-Pd
Co-Pt
Co-Rh
Co-Si
Fe-Co

Output 1.3 histograms

histogram_element_pair.png and histogram_site_pair.png are used visualize data, with colors indicating atomic mixing types.

To modify the x-axis, run python plot-histogram.py. This script allows you to interactively specify parameters such as the bin width and x-axis range:

Option 2. System Analysis

Purpose: System Analysis provides an overview of bond fractions acquired from Option 1: Site Analysis, or bond fractions in coordination number geometries.
Scope: System Analysis is applicable for folders containing either 2 or 3 unique elements.

4 types of folders are applicable for System Analysis.

Type 1. Binary files, 2 unique elements
Type 2. Binary files, 3 unique elements
Type 3. Ternary files, 3 unique elements
Type 4. Ternary and binary combined, 3 unique elements

Here is an example of CBA detecting folders containing 2 or 3 unique elements.

Available folders containing 2 or 3 unique elements:
1. 20240623_ErCoIn_nested, 3 elements (In, Er, Co), 152 files
2. 20240612_ternary_only, 3 elements (In, Er, Co), 2 files
3. 20240611_ternary_binary_combined, 3 elements (In, Er, Co), 5 files
4. 20240623_teranry_3_unique_elements, 2 elements (Er, Co), 3 files
5. 20240611_binary_2_unique_elements, 2 elements (Er, Co), 4 files````

Output 2.1 Binary/ternary figures

For Types 2, 3, and 4:

Customize legend position:

To adjust the legend position in the ternary diagram, modify the values of X_SHIFT = 0.0 and Y_SHIFT = 0.0 in core/configs/ternary.py.

Customize extra lines:

To add extra lines to the ternary diagram based on tags, edit TAGS_IN_FIRST_EXTRA_LINE = ["lt", "ht", "hp", "hp1", "hp2", "hp3"] and TAGS_IN_SECOND_EXTRA_LINE = ["lt", "ht", "hp", "hp1", "hp2", "hp3"] in core/configs/ternary.py.

For Type 1:

All of the individual hexagon figures also saved in order.

Output 2.2 Color map

For Types 2, 3, and 4, color maps for each bond type and overall are generated.

Output 2.3 Excel

Bond count per each cif file is recorded in system_analysis_files.xlsx.

Average bond lenghts, count, and statistical values are recorded in system_analysis_main.xlsx.

Option 3. Coordination Analysis

Purpose: This option determines the best coordination geometry using four methods provided in cifkit. Excel files and JSON are saved with nearest neighbor information.
Customization: The Excel contains Δ, which is defined as the interatomic distance subtracted by the sum of atomic radii. You may provide your radii values by modifying the radii.xlsx file.

Ouput 3.1 JSON

For each site, the nearest neighbors within the coordination number geometry are recorded in CN_connections.json.

{
    "250361": {
        "Co": [
            {
                "connected_label": "Co",
                "distance": 2.529,
                "delta": 1.16,
                "mixing": "full_occupancy",
                "neighbor": 1
            },
            {
                "connected_label": "Co",
                "distance": 2.529,
                "delta": 1.16,
                "mixing": "full_occupancy",
                "neighbor": 2
            },
            ...
            {
                "connected_label": "Er",
                "distance": 2.966,
                "delta": -0.603,
                "mixing": "full_occupancy",
                "neighbor": 11
            },
            {
                "connected_label": "Er",
                "distance": 2.966,
                "delta": -0.603,
                "mixing": "full_occupancy",
                "neighbor": 12
            }
        ]
    }
}

Output 3.2 Excel

For each .cif file, the nearest neighbor information is wrriten in each sheet within CN_connections.xlsx.

Installation

git clone https://github.com/bobleesj/cif-bond-analyzer.git
cd cif-bond-analyzer
pip install -r requirements.txt
python main.py

If you are interested in using Conda with a new environment run the following:

git clone https://github.com/bobleesj/cif-bond-analyzer.git
cd cif-bond-analyzer
conda create -n cif python=3.12
conda activate cif
pip install -r requirements.txt
python main.py

Contributors

Anton Oliynyk
Emil Jaffal
Sangjoon Bob Lee

How to ask for help

CBA is also designed for experimental materials scientists and chemists.

If you have any issues or questions, please feel free to reach out or leave an issue.

How to contribute

Here is how you can contribute to the CBA project if you found it helpful:

Star the repository on GitHub and recommend it to your colleagues who might find CBA helpful as well.
Fork the repository and consider contributing changes via a pull request.
If you have any suggestions or need further clarification on how to use CBA, please feel free to reach out to Sangjoon Bob Lee (@bobleesj).

Changelog

20240623 - Implement CN bond fractions, add GitHub CI. See Pull #22.
20240330 - Add sequential folder processing and customizable histogram generation. See Pull #16.
20240311 - Integrate PEP8 linting with black. See Pull #12.
20240310 - Enhance output options to include both element-based and label-based data for Excel, JSON, and histograms. See Pull #11.
20240301 - Display atom counts and execution time per file in Terminal; adds CSV logging.
20240229 - Expand file support to include all CIF files.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

README.md

README.md

CIF Bond Analyzer (CBA)

Demo

Scope

Value

Getting started

Preprocess

1. Format files

2. Supercell generation

3. Atomic mixing info

Analysis Options

Option 1. Site Analysis

Output 1.1 Excel and JSON

Output 1.2 text summary

Output 1.3 histograms

Option 2. System Analysis

Output 2.1 Binary/ternary figures

Output 2.2 Color map

Output 2.3 Excel

Option 3. Coordination Analysis

Ouput 3.1 JSON

Output 3.2 Excel

Installation

Contributors

How to ask for help

How to contribute

Changelog

Files

README.md

Latest commit

History

README.md

File metadata and controls

CIF Bond Analyzer (CBA)

Demo

Scope

Value

Getting started

Preprocess

1. Format files

2. Supercell generation

3. Atomic mixing info

Analysis Options

Option 1. Site Analysis

Output 1.1 Excel and JSON

Output 1.2 text summary

Output 1.3 histograms

Option 2. System Analysis

Output 2.1 Binary/ternary figures

Output 2.2 Color map

Output 2.3 Excel

Option 3. Coordination Analysis

Ouput 3.1 JSON

Output 3.2 Excel

Installation

Contributors

How to ask for help

How to contribute

Changelog