Development of nuclear basis sets for multicomponent quantum chemistry methods.

Area of Science:

• Quantum Chemistry
• Computational Chemistry

Background:

• The nuclear-electronic orbital (NEO) framework enables direct inclusion of nuclear quantum and non-Born-Oppenheimer effects.
• Accurate quantum chemistry calculations require appropriate electronic and nuclear basis sets.
• Systematically developed nuclear basis sets balancing accuracy and efficiency have been lacking.

Purpose of the Study:

• Develop and validate a series of systematically improved Gaussian-type nuclear basis sets.
• Enhance the accuracy and efficiency of multicomponent wave function based methods, such as NEO coupled cluster and NEO-DFT.
• Provide a user-selectable balance between accuracy and efficiency for various chemical properties.

Main Methods:

• Development of three series of Gaussian-type nuclear basis sets (PB4, PB5, PB6) with varying angular momentum.
• Utilization of a machine-learning optimization procedure employing Gaussian process regression.
• Validation through predictions of ground state energies, proton densities, proton affinities, and proton vibrational excitation energies.

Main Results:

• The developed nuclear basis sets demonstrate accuracy and efficiency for multicomponent systems.
• PB4, PB5, and PB6 basis sets effectively describe both ground and excited state properties.
• The machine-learning approach accelerated the optimization of nuclear basis sets.

Conclusions:

• The new nuclear basis sets significantly enhance the tractability of NEO methods.
• These basis sets are crucial for accurate quantum mechanical treatment of electrons and protons in chemical systems.
• The findings facilitate broader applications of advanced quantum chemistry calculations to diverse chemical problems.