Benchmarking ionization potentials using the simple pCCD model.

Area of Science:

• Quantum Chemistry
• Computational Chemistry
• Molecular Spectroscopy

Background:

• Ionization potential (IP) is a key molecular electronic signature.
• Accurate IP prediction is vital for designing organic optoelectronic devices.
• Existing theoretical models require validation for diverse organic molecules.

Purpose of the Study:

• To benchmark the performance of the IP-EOM-pCCD model for predicting ionization potentials.
• To assess the accuracy of IP-EOM-pCCD against experimental data and higher-order coupled cluster methods.
• To investigate the impact of basis sets and particle-hole operators on IP predictions.

Main Methods:

• Benchmarking IP-EOM-pCCD against experimental ionization energies.
• Statistical assessment of 201 electron-detached states across 41 organic molecules.
• Evaluation using three molecular orbital basis sets and two particle-hole operator sets.

Main Results:

• IP-EOM-pCCD demonstrates a reasonable spread and skewness in ionization energy predictions.
• Mean error and standard deviation can deviate up to 1.5 eV from reference data.
• The study highlights limitations of pCCD reference functions without dynamical correlation.

Conclusions:

• Dynamical correlation is essential for reliable IP prediction in small organic molecules using pCCD-based methods.
• IP-EOM-pCCD requires further refinement to achieve high accuracy comparable to experimental values.
• This work provides critical insights into the accuracy and limitations of a new theoretical model for electronic structure calculations.