Correction of residual errors in configuration interaction electronic structure calculations.

Area of Science:

• Computational chemistry
• Quantum chemistry
• Theoretical chemistry

Background:

• Configuration interaction (CI) calculations often suffer from residual energy errors.
• Accurate prediction of molecular energies is crucial for understanding chemical properties and reactions.

Purpose of the Study:

• To develop and validate a method for correcting residual energy errors in CI calculations.
• To improve the accuracy of calculated energies by mapping energy defects to atomic regions.

Main Methods:

• An atomic defect energy per electron is defined, independent of excitation details.
• Defect energies are derived from diatomic and hydride molecules and applied to larger systems.
• The method is tested on organic molecules with truncated virtual spaces and coupled cluster calculations.

Main Results:

• Defect corrections bring initially large CI calculation errors into agreement with experimental values.
• The method shows excellent consistency with coupled cluster calculations and experimental data.
• Applied to transition metal dmsn states, it significantly improves agreement with spectroscopic values.

Conclusions:

• Atomic defect energy correction is an effective strategy for enhancing the accuracy of CI and other electronic structure calculations.
• The method provides reliable energy predictions across diverse chemical systems, including transition metals.
• This approach offers a pathway to more precise computational chemistry results.