A flexible implementation of frozen-density embedding for use in multilevel simulations

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Materials Science

Background:

• Density-functional theory (DFT) is a powerful quantum mechanical modeling method.
• Frozen-density embedding (FDE) offers a way to model large systems by dividing them into subsystems.
• Existing FDE implementations may lack flexibility for complex simulation scenarios.

Purpose of the Study:

• To present a new, flexible implementation of frozen-density embedding (FDE) within the Amsterdam Density Functional (ADF) program package.
• To enable various subsystem optimization strategies for enhanced computational efficiency and accuracy.
• To demonstrate the utility of this flexible FDE approach in multilevel simulations.

Main Methods:

• Developed a new FDE implementation based on a subsystem formulation of DFT.
• Coupled subsystems using an effective embedding potential.
• Implemented options for full optimization of all subsystems, active fragment calculation with a frozen environment, and intermediate optimization levels.

Main Results:

• The new FDE implementation allows for linear-scaling computational cost, an alternative to conventional DFT.
• It enables the calculation of a single active fragment interacting with a frozen environment.
• Demonstrated the flexibility of optimizing individual subsystems to varying degrees (fully, partially, or frozen).

Conclusions:

• The enhanced FDE implementation in ADF provides a versatile tool for computational chemistry.
• This flexibility facilitates the application of FDE in advanced multilevel simulation strategies.
• The method offers a computationally efficient and accurate approach for studying complex molecular systems.