Basis set convergence of molecular correlation energy differences within the random phase approximation.

Area of Science:

• Computational chemistry
• Quantum chemistry
• Theoretical chemistry

Background:

• The random phase approximation (RPA) is a method for calculating correlation energy.
• Accurate calculation of molecular interactions requires careful consideration of basis set convergence.
• Dispersion-bound systems present unique challenges for basis set convergence.

Purpose of the Study:

• To investigate the basis set convergence of RPA energy differences for various molecular interactions.
• To determine the necessary basis set size and extrapolation techniques for reliable binding energy calculations.
• To assess the impact of core-valence correlations and diffuse augmentation on convergence.

Main Methods:

• Calculations of energy differences using the random phase approximation (RPA).
• Systematic investigation of basis set convergence, including quadruple-zeta and larger basis sets.
• Extrapolation to the complete basis set (CBS) limit.
• Analysis of counterpoise correction effects.

Main Results:

• Basis set incompleteness error is most significant for dispersion-bound systems (e.g., S22 benchmark).
• Very large basis sets or CBS extrapolation are required for accurate binding energies in dispersion-bound systems.
• Quadruple-zeta basis sets provide converged results for medium- and short-range correlations in various organic molecules.
• Diffuse augmentation generally slows down convergence, except for weakly bound systems.
• Core-valence correlations have a minor impact.

Conclusions:

• For dispersion-bound systems, achieving reliable RPA binding energies necessitates very large basis sets or CBS extrapolation.
• Quadruple-zeta basis sets offer a practical balance of accuracy and computational cost for most RPA applications.
• Standard counterpoise correction alone is insufficient without extrapolation for these systems.