Size-dependent error of the density functional theory ionization potential in vacuum and solution

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Theoretical Chemistry

Background:

• Density functional theory (DFT) is crucial for modeling large molecular systems and solvation.
• Accurate modeling requires methods to be size-intensive, treating systems of varying sizes consistently.
• Prior research indicates approximate DFT methods have size-dependent errors in ionization potential (IP) calculations.

Purpose of the Study:

• To investigate the lack of size-intensivity of ionization potential (IP) in approximate DFT methods.
• To analyze these errors in both vacuum and solution environments.
• To explore the impact of explicit solvent and exact exchange on IP calculations.

Main Methods:

• Calculations using approximate density functionals in vacuum and with solvation models (polarizable continuum model and explicit solvent).
• Analysis of the ionization potential for increasing numbers of identical isolated molecules.
• Investigation of the effect of varying amounts of exact exchange in hybrid functionals.

Main Results:

• Local and semi-local DFT approximations show a decreasing IP with an increasing number of molecules in vacuum.
• This size-dependent error persists in solution, and explicit solvent can worsen the delocalization error.
• Increasing exact exchange alters solvent polarization; optimal tuning of long-range corrected hybrid functionals achieves size-intensivity.

Conclusions:

• Approximate DFT methods suffer from size-dependent errors in IP calculations, impacting their reliability for large systems.
• Solvation effects, particularly with explicit solvent, can exacerbate these errors.
• Long-range corrected hybrid functionals offer a viable route to achieving size-intensive IP calculations.