Hybrid DFT Quality Thermochemistry and Environment Effects at GGA Cost via Local Quantum Embedding

Area of Science:

• Computational Chemistry
• Quantum Mechanics
• Chemical Physics

Background:

• Accurate thermochemical modeling of reaction mechanisms necessitates high-level computational methods (e.g., hybrid DFT) and consideration of environmental factors.
• Quantum embedding methods offer a way to reduce computational cost by focusing high-level treatments on critical regions.
• Existing embedding methods require further acceleration for practical application in complex systems.

Purpose of the Study:

• To develop and validate an accelerated quantum embedding framework for thermochemical modeling.
• To achieve hybrid DFT accuracy in calculations at a significantly reduced computational cost.
• To enable more accessible and predictive modeling of reaction mechanisms, including environmental effects.

Main Methods:

• Development of the local embedded subsystem (LESS) framework combining local approximations in hybrid DFT with multilayer DFT.
• Implementation of an in-core density fitting approach optimized for multilayer DFT.
• Application to reaction and activation energies in homogeneous, heterogeneous, and enzymatic catalysis.

Main Results:

• The LESS framework achieves asymptotically constant computational cost for the hybrid DFT layer with increasing environment size.
• Calculations using LESS retain the intrinsic accuracy of hybrid DFT, with errors of only a few tenths of a kcal/mol.
• LESS hybrid DFT-in-GGA computations are 30-90 times faster than complete density-fitted hybrid DFT on large systems (171-238 atoms).

Conclusions:

• The LESS framework provides a significant advancement, offering hybrid DFT-level energetics at GGA computational cost.
• This acceleration makes predictive thermochemistry, including dynamics and quantum environment effects, more computationally feasible.
• The approach is validated for diverse catalytic reaction systems, demonstrating its broad applicability.