Eigen Vector Following Trasition State Search Algorithm

Eigen vector following transition state search method based on PRFO method and Bofill's Hessian estimation using GAUSSIAN as calculation engine.

Installation:

to install, clone the repository

git clone https://github.com/arminshzd/EF_TSS.git

and install from inside the EF_TSS directory using pip

pip install .

Usage:

Create a CM_TSS object and pass the path to the settings.json file and the path to an initial structure. The settings.json file is formatted as (default values in parentheses):

{
    "N": {number of atoms} (required),
    "charge": {charge of the system} (0),
    "spin": {spin of the system} (1),
    "N-procs": {number of processors to use} (8),
    "conv-radius":{atom position convergence radius} (1e-1),
    "conv-grad":{gradient convergence radius} (1e-6),
    "trust-radius": {trust radius for maximum displacement} (0.2)
    "max-iter": {max number of iterations} (10),
    "reset-H-every": {how many iterations between recalculating Hessian exactly from force calculations} (20)
    "working-dir": {path to directory with all the files} (required),
    "basis-f-name": {name of the file containing basis information for GAUSSIAN calculations} ("" skipped)
    "history-f-name": {name of the file to record optimization history}("history.xyz")
    "final-f-name": {name of the file to record optimized structure}("final.xyz")
    "submit-f-dir": {path to GAUSSIAN submission file} (required),
    "gaussian-f-name": {name for GAUSSIAN input file} ("in"), 
    "energy-header-calc": {GAUSSIAN header for energy calculations} ("#P wB97XD/6-31G** nosymm force"),
    "hess-header-calc": {GAUSSIAN header for Hessian calculations } ("#P b3lyp/6-31G** nosymm freq"),
}

A GAUSSIAN submission script is necessary. This is a system dependent file.

basis-f-name is the name of the file containing basis specifications if necessary and will be added to the bottom of the GAUSSIAN input file.

Input coordinates file should have the format:

{atomic number} {-1 for frozen 0 otherwise} {x} {y} {z}

There's an example calculation for HCN available under the example directory. The EF_TSS.py should either be added to PATH or copied to the same directory as the test script.

Unit tests are available under tests directory and can be run from the EF_TSS directory using

pytest tests/

(requires pytest to be installed in your environment)