Exploiting QM/MM Capabilities in Geometry Optimization:  A Microiterative Approach Using Electrostatic Embedding.

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Biochemistry

Background:

• Quantum mechanical/molecular mechanical (QM/MM) methods are crucial for studying large systems.
• Efficient energy minimization is essential for QM/MM simulations.
• Standard QM/MM optimization can be computationally expensive, especially for large molecular mechanics (MM) regions.

Purpose of the Study:

• To present a novel microiterative adiabatic scheme for QM/MM energy minimization.
• To improve the efficiency of QM/MM energy minimization, particularly for systems with large MM regions.
• To ensure accuracy and excellent convergence properties in QM/MM optimization.

Main Methods:

• A microiterative adiabatic scheme is introduced for QM/MM energy minimization.
• The scheme fully optimizes the MM part within each QM macroiteration.
• Electrostatic QM/MM interactions are computed using on-the-fly fitted electrostatic potential charges to the QM density.
• Corrections to energy and gradient expressions ensure consistency between macro- and microiterations.
• The method is tested on water clusters, enzymes, and surface models using various embedding schemes (mechanical, electrostatic, polarized).

Main Results:

• The proposed scheme demonstrates excellent convergence properties.
• No loss of accuracy is observed compared to standard QM/MM optimization methods.
• The computational cost is nearly independent of the MM system size, making it suitable for large MM regions.
• Microiterations typically reduce the number of required QM calculations by a factor of 2-10.

Conclusions:

• The microiterative adiabatic QM/MM scheme offers a significant computational advantage for large systems.
• The method provides an accurate and efficient approach for energy minimization in complex molecular systems.
• This technique is particularly beneficial for simulations involving extensive MM environments.