Density-functional-theory approach to the Hamiltonian adaptive resolution simulation method.

Area of Science:

• Computational chemistry
• Statistical mechanics
• Multiscale modeling

Background:

• The Hamiltonian adaptive resolution simulation method (H-AdResS) allows simulating systems with atomistic and continuum regions.
• An apparent dependence of the H-AdResS Hamiltonian on thermodynamic state challenges its statistical mechanics consistency.
• Bridging potentials and external forces are used to manage transitions between regions.

Purpose of the Study:

• To establish the statistical mechanics consistency of H-AdResS.
• To demonstrate the equivalence between H-AdResS and density functional theory (DFT).
• To assess the convergence and efficiency of the H-AdResS approach.

Main Methods:

• Explicitly including an external potential dependent on the switching function in H-AdResS.
• Constructing a grand canonical potential for inhomogeneous systems.
• Verifying the equivalence between the external potential and the system's excess chemical potential.
• Computing excess chemical potentials for water, urea solutions, and Lennard-Jones mixtures.

Main Results:

• The external potential inducing a constant density profile was found to equal the system's excess chemical potential.
• A Hamiltonian description of the system is compatible with enforcing reference density.
• The H-AdResS approach demonstrated convincing convergence and accuracy for various systems.
• The method's robustness and capabilities were emphasized.

Conclusions:

• H-AdResS is statistically consistent and equivalent to DFT.
• The numerical implementation of H-AdResS is validated.
• The method is robust and efficient for diverse chemical systems.