Fragment approach to constrained density functional theory calculations using Daubechies wavelets.

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Materials Science

Background:

• Linear scaling Kohn-Sham density functional theory (DFT) codes are crucial for large molecular systems.
• Optimizing localized support functions in situ adapts basis sets to specific chemical properties.
• Daubechies wavelets offer a systematically controllable accuracy for basis sets.

Purpose of the Study:

• To demonstrate the reusability of a minimal basis set for charge-constrained DFT calculations.
• To showcase the application of this formalism within a fragment approach.
• To highlight its utility in preparing diabatic states and setting up complex systems.

Main Methods:

• Development of a linear scaling DFT code utilizing Daubechies wavelets.
• In situ optimization of a minimal set of localized support functions.
• Projection of the density matrix onto a natural subset of the optimized basis.
• Reuse of pre-optimized fragment support functions for charge-constrained calculations.

Main Results:

• The minimal basis set achieves accuracy comparable to uncontracted, cubic scaling approaches for energies and forces.
• The basis set can be reused without reoptimization for charge-constrained DFT.
• High numerical precision is maintained when transforming fragment support functions.
• The approach facilitates precise and efficient calculations for diabatic states and complex environments.

Conclusions:

• The minimal basis formalism offers a flexible and efficient approach for advanced DFT calculations.
• This method is well-suited for charge-constrained computations and fragment-based studies.
• The technique provides a robust framework for preparing diabatic states and modeling complex molecular systems.