Scripts and data to derive parameters for various potentials for diatomic compounds. Some of the scripts sport a -h option to get some help and information.
-
curve_fit.py will perform curve fitting of 28 potentials to 71 diatomic energy curves from either MP2 or CCSD(T) with a basis set of aug-cc-pvtz. Alternatively, a fit can be done to experimental data to try and reproduce the results from a paper by Royappa et al. https://doi.org/10.1016/j.molstruc.2005.11.008 Output is produced in a csv file, a pdf with plots of the fits (if requested) etc. Includes -h flag for help.
-
frequencies.py will estimate the ground state vibrational frequency of the quantum chemistry calculations or the fits and produce plots and tables. Includes -h flag for help.
-
analyse_z.py will summarize the results of curve fitting by averaging over potentials and produce latex tables used in the submitted manuscript.
-
data_graphs.py will plot potentials from quantum chemistry and, where available, experimental data.
-
make_fitting_bounds.py will read the json files containing the fitted parameters and generate a new potentials.json that can be used to limit the search in fitting to a minimum and a maximum.
-
diatomics.py is a utility code to read data files.
-
potentials.py is the code to compute potentials.
Data for diatomic molecules were taken from The Diatomic Molecular Spectroscopy Database. Please note though, that some errors were detected and some omissions were found that we have addressed in an additional file in the data directory. The NIST WebBook also contains useful information on diatomic molecules.
For running the scripts a working Python interpreter (version 3.8 or higher) is needed and the following libraries:
- numpy
- scipy
- pandas
- importlib
- matplotlib
- psi4
A manuscript about this the code and results from this repository, entitled Quantitative Evaluation of Anharmonic Bond Potentials for Molecular Simulations, by Paul J. van Maaren and David van der Spoel has been submitted for publication and is under revision.
The script and data have a permanent DOI through a Zenodo link.