A simple but fully nonlocal correction to the random phase approximation.

Area of Science:

• Quantum Chemistry
• Computational Materials Science
• Theoretical Physics

Background:

• The random phase approximation (RPA) is a leading method for ground-state density functional approximations.
• Direct RPA accurately predicts isoelectronic energy differences but underestimates molecular atomization energies.
• Semilocal corrections improve total and ionization energies but do not fix the atomization energy underestimation.

Purpose of the Study:

• To develop a novel correction to the RPA that addresses the underestimation of molecular atomization energies.
• To incorporate missing middle-range multicenter nonlocality in correlation energy calculations.
• To improve the accuracy of density functional approximations for molecular systems.

Main Methods:

• Proposed a fully nonlocal, hybrid-functional-like addition to existing semilocal RPA corrections.
• Investigated the behavior of the nonlocal correction under uniform-density scaling.
• Evaluated the method's performance using atomization energies of ten molecules.

Main Results:

• The fully nonlocal correction is significant for molecules but not for atoms.
• The correction scales appropriately with density changes, similar to key correlation energy components.
• Achieved significantly better performance for molecular atomization energies compared to second-order screened exchange corrections, using only one fit parameter.

Conclusions:

• The proposed fully nonlocal correction effectively improves RPA's accuracy for molecular atomization energies.
• This approach captures crucial nonlocal correlation effects missed by direct RPA and semilocal corrections.
• The method offers a more accurate and efficient alternative for molecular electronic structure calculations.