About this Research Topic
Traditionally, the methods used for modeling molecular crystals, surfaces, and nanomaterials have been classical Molecular Mechanical (MM) force field methods. Owing to their low computational cost, force field methods are widely used and often the method of choice for modeling biological systems. However, based on many approximations and derived from different types of experimental data, conventional force fields are susceptible to deficiencies and numerous limitations. When the difference between polymorph structures is in the range of few kJ/mol and there are strong intramolecular interactions, hydrogen bonding, large dispersion, or anisotropic electrostatics, force fields do not perform well. The development of polarizable force fields may provide better treatment of intramolecular flexibility but inevitably become computationally expensive. Thus a compromise between accuracy and computational cost has to be found.
Quantum Mechanical (QM) studies of electronic structure currently rely on periodic density functional theory (DFT). Periodic DFT calculations with either atom-centered Gaussian or plane-wave basis sets are inexpensive and fairly accurate if the dispersion corrections are included. Nevertheless, the prediction of molecular crystal structures is sensitive to the choice of a DFT functional and type of the dispersion correction.
A large group of methods classified as fragment-based electronic structure methods provides an alternative approach. The main idea is to divide the crystal into individual molecules or clusters of molecules. The majority of fragment methods are based on the embedding of high-level treatment of intermolecular interactions of fragments with a low-level treatment of the whole periodic structure. One can distinguish between pure QM:QM approaches and hybrid QM:MM embedding. In the first case, post-Hartree-Fock methods such as second order perturbation theory (MP2) or even coupled cluster with single, double, and perturbative triple [CCSD(T)] excitation level of theory can be used at the high-level. Meanwhile DFT-D is used for the lower-tier method in contrast to the QM:MM approach which combines classically the QM method for high-level and Molecular Mechanics for periodic and low-level calculations.
However, breaking the whole crystal into fragments entirely destroys the hydrogen bond cooperativity and affects many-body interactions such as induction and dispersion. To minimize the related error, the interaction between fragments can be described by DFT embedding formalism. The embedding scheme is a quite challenging approach which assumes a common embedding potential shared by all subsystems, thus representing the influence of the surrounding parts of the system.
This Research Topic aims at providing a comprehensive overview of the recent advances in the development and performance of fragment-based methods. Themes to be addressed may include, but are not limited to:
• QM-based methods
• Density functional embedding
• Divide-and-Conquer Methods
• Many-body molecular interactions
• Symmetry-adapted perturbation theory
Quantum Mechanical (QM) studies of electronic structure currently rely on periodic density functional theory (DFT). Periodic DFT calculations with either atom-centered Gaussian or plane-wave basis sets are inexpensive and fairly accurate if the dispersion corrections are included. Nevertheless, the prediction of molecular crystal structures is sensitive to the choice of a DFT functional and type of the dispersion correction.
A large group of methods classified as fragment-based electronic structure methods provides an alternative approach. The main idea is to divide the crystal into individual molecules or clusters of molecules. The majority of fragment methods are based on the embedding of high-level treatment of intermolecular interactions of fragments with a low-level treatment of the whole periodic structure. One can distinguish between pure QM:QM approaches and hybrid QM:MM embedding. In the first case, post-Hartree-Fock methods such as second order perturbation theory (MP2) or even coupled cluster with single, double, and perturbative triple [CCSD(T)] excitation level of theory can be used at the high-level. Meanwhile DFT-D is used for the lower-tier method in contrast to the QM:MM approach which combines classically the QM method for high-level and Molecular Mechanics for periodic and low-level calculations.
However, breaking the whole crystal into fragments entirely destroys the hydrogen bond cooperativity and affects many-body interactions such as induction and dispersion. To minimize the related error, the interaction between fragments can be described by DFT embedding formalism. The embedding scheme is a quite challenging approach which assumes a common embedding potential shared by all subsystems, thus representing the influence of the surrounding parts of the system.
This Research Topic aims at providing a comprehensive overview of the recent advances in the development and performance of fragment-based methods. Themes to be addressed may include, but are not limited to:
• QM-based methods
• Density functional embedding
• Divide-and-Conquer Methods
• Many-body molecular interactions
• Symmetry-adapted perturbation theory
Keywords: DFT-D, embedding models, molecular crystals, nanomaterials, linear scaling
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
