Abstract:The structures and spectroscopic properties of molecules and condensed-phase systems are usually experimentally characterized by X-ray, infrared (IR), Raman, nuclear magnetic resonance (NMR), and electronic absorption/emission spectra. Quantum mechanics (QM) calculations are critical in quantitatively understanding the relationship between the structure and physicochemical properties of various chemical systems. However, it is very challenging to apply traditional QM methods to large molecules and condensed-phase systems with large unit cells due to their steep computational scaling with the system size. To overcome this difficulty, theoretical chemists have developed various linear (or low) scaling QM methods, among which energy-based fragmentation methods have achieved great success for large molecules or clusters. One of the most popular energy-based fragmentation methods is the generalized energy-based fragmentation (GEBF) approach developed by us.In this approach, the ground-state energy of a large molecule can be evaluated from the ground-state energies of a series of embedded subsystems. In this Account, we focus on the recent developments and applicability of the GEBF approach for the structures and spectroscopic properties of complicated large molecules and condensed-phase systems. With new fragmentation schemes, the GEBF approach can now describe ionic liquid clusters and metal-containing supramolecular systems accurately and can provide accurate binding energies for host–guest complexes. In addition, the GEBF approach is now available for describing the localized excited states of large systems including a chromophore. More importantly, the GEBF approach under periodic boundary conditions (PBC-GEBF) has been developed to deal with periodic molecular crystals and liquids. Then, the ground-state energy (or property) per unit cell of a periodic condensed phase system can be predicted with QM calculations on nonperiodic embedded subsystems. This feature enables accurate electron correlation calculations on molecular crystals and liquids to be feasible on ordinary workstations. The PBC-GEBF approach has been applied to predict the crystal structures, lattice energies, and spectroscopic properties of some typical molecular crystals and solutions. By combining the GEBF method and machine learning (ML) method, a GEBF-ML force field has been developed for long normal alkanes, and the IR spectra of long alkanes can be obtained from the GEBF-ML molecular dynamics (MD) simulations. The GEBF and its periodic variant are expected to play increasingly important roles in investigating real-life chemical systems of broad interests at the <i>ab initio</i> levels.The Supporting Information is available free of charge at <a class="ext-link" href="/doi/10.1021/acs.accounts.0c00580?goto=supporting-info">https://pubs.acs.org/doi/10.1021/acs.accounts.0c00580</a>.Construction of subsystems in the GEBF approach, detailed procedure of the PBC-GEBF approach, brief introduction to the LSQC program, and average binding energy per ion pair in ionic liquid clusters (<a class="ext-link" href="/doi/suppl/10.1021/acs.accounts.0c00580/suppl_file/ar0c00580_si_001.pdf">PDF</a>)This article has not yet been cited by other publications.

Fragment-Based Deep Learning for Simultaneous Prediction of Polarizabilities and NMR Shieldings of Macromolecules and Their Aggregates

Accurate Prediction of NMR Chemical Shifts in Macromolecular and Condensed-Phase Systems with the Generalized Energy-Based Fragmentation Method

Combined Fragment-Based Machine Learning Force Field with Classical Force Field and Its Application in the NMR Calculations of Macromolecules in Solutions.

An On-the-Fly Approach to Construct Generalized Energy-Based Fragmentation Machine Learning Force Fields of Complex Systems

Generalized energy-based fragmentation approach for computing the ground-state energies and properties of large molecules.

Generalized Energy-Based Fragmentation Approach and Its Applications to Macromolecules and Molecular Aggregates

An Efficient Fragment-Based Approach for Predicting the Ground-State Energies and Structures of Large Molecules.

Quantum Fragment Based ab Initio Molecular Dynamics for Proteins.

Structures and Spectroscopic Properties of Large Molecules and Condensed-Phase Systems Predicted by Generalized Energy-Based Fragmentation Approach

Multitask Deep Ensemble Prediction of Molecular Energetics in Solution: From Quantum Mechanics to Experimental Properties

Accurate Prediction of the Structure and Vibrational Spectra of Ionic Liquid Clusters with the Generalized Energy-Based Fragmentation Approach: Critical Role of Ion-Pair-based Fragmentation

Efficient and accurate density-based prediction of macromolecular polarizabilities

Towards Predicting Equilibrium Distributions for Molecular Systems with Deep Learning

Predicting Fragment Binding Modes Using Customized Lennard-Jones Potentials in Short Molecular Dynamics Simulations

Highly Accurate Prediction of NMR Chemical Shifts from Low-Level Quantum Mechanics Calculations Using Machine Learning

Efficient interatomic descriptors for accurate machine learning force fields of extended molecules

Predicting equilibrium distributions for molecular systems with deep learning

Molecular Simulations with a Pretrained Neural Network and Universal Pairwise Force Fields

EL-MLFFs: Ensemble Learning of Machine Leaning Force Fields

An Efficient Implementation of the Generalized Energy-Based Fragmentation Approach for General Large Molecules.