Quantum Alchemy for Atoms and Dimers

Requirements

Python3 with the following packages:

• NumPy
• Pandas
• SciPy
• findiff

Installation

In order to use, reproduce, and analyze quantum alchemy atom and dimer data you need to install the qa_tools package from the GitHub repository. You just have to clone and install the GitHub repo like so.

git clone https://github.com/keithgroup/qa-atoms-dimers
cd qa-atoms-dimers
pip install .

Contents

Note: Atom property predictions involving three-electron systems might slightly differ (± 0.001 eV) from values presented in the respective manuscript. Data here uses CCSD(T) calculations for three-electron ground state systems; whereas the manuscript does not include the perturbative triples corrections.

json-data

Cumulative JSON files for atoms and dimers containing quantum chemistry and quantum alchemy data for all species of interest. Individual JSON, log files, and explanations for atoms and dimers data can be found in their respective repositories.

qa_tools

A collection of modules that streamline the quantum alchemy analysis workflow.

• data.py: Manages parsing and converting JSON files into panda dataframes.
• prediction.py: Property predictions or querying of data frames.
• analysis.py: Automated functions to assist analysis.
• utils.py: Useful functions that support other modules.

scripts

Routines to check and analyze calculation JSON files and generate PySCF/quantum alchemy calculations.

notebooks

Jupyter notebook examples of how to query and analyze atom and dimer calculations for: ionization energies, electron affinities, multiplicity gaps, equilibrium bond lengths, and quantum alchemical potential energy surfaces.

Terminology

We will be using several terms throughout the repository.

• Quantum chemistry (QC): Data using straightforward wave function methods or density functional theory.
• Quantum alchemy (QA): Predictions of target systems through nuclear charge perturbations of reference systems.
• Target: A specific system, and its respective property, we are interested in predicting. For example, the ionization energy, electron affinity, or multiplicity gap of N.
• Reference: A system, used with quantum alchemy, to predict target energies and properties. Must have the same number of electrons as the target system. For example, to predict the energy of N we could use B^2–, C ^–, O⁺, and F²⁺.
• Quantum alchemy with Taylor series (QATS): Approximating the alchemical potential energy surface (PES) using a Taylor series centered at ∆Z = 0. Central finite difference using an h of 0.01 was used to calculate first through fourth derivatives.

Citation

If you use this software, please cite it as specified in CITATION.cff.

Related publications

This software was used to generate and analyze data for the following manuscripts.

• Eikey, E. A.; Maldonado, A. M.; Griego, C. D.; von Rudorff, G. F.; Keith, J. A. Evaluating quantum alchemy of atoms with thermodynamic cycles: Beyond ground electronic states. ChemRxiv 2021. DOI: 10.26434/chemrxiv-2021-3l4zh
• Eikey, E. A.; Maldonado, A. M.; Griego, C. D.; von Rudorff, G. F.; Keith, J. A. Quantum alchemy beyond singlets: Bonding in diatomic molecules with hydrogen. ChemRxiv 2021. DOI: 10.26434/chemrxiv-2021-pt5gd

License

Distributed under the MIT License. See LICENSE for more information.