Computational chemistry could only deal with small, isolated molecules. For these, local optimizations from guessed starting structures were sufficient. These starting structures could be taken from chemical intuition or from experiment. This chapter deals with Non-Deterministic Global Optimization (NDGO) approaches. In contrast to deterministic global optimization methods, NDGO approaches have to forgo any attempts to cover all of search space and to find the global optimum with certainty. Interestingly, NDGO methods share the aim of “getting across barriers more quickly than normally” with other areas of method development, mostly pertaining to calculations of free energies. In essence, machine learning combines suitable representations with brute-force big-data interpolation, employing interpolation functions that are intentionally very flexible. Monte Carlo-style “mutation” moves (but also “crossover”-style information exchange moves) in global structure optimization can easily move several atoms too close to each other, resulting in strongly repulsive forces.
This chapter provides the essential background to a novice modeler on the choice of simulation techniques used to model Deep Eutectic Solvents (DESs). It describes methods used to obtain important physical, thermodynamic, transport, and structural properties of bulk DES systems including an evaluation of the strengths and drawbacks of the current simulation models. The chapter discusses future directions for simulating DES-based systems. One of the major thrust areas of ab initio investigations on DESs has been to provide a physical explanation for the observed low melting point in these systems and its effects on their physicochemical properties. Molecular simulations have played a crucial role in conjunction with experimental investigations in elucidating the structure–property relationships of DESs. Classical molecular dynamics simulations that obey Newton's laws of motion use force fields to calculate the potential energy of a system as a function of their atomic coordinates.
This chapter details the different experimental microphase formers and provides a minimal theoretical framework to present the simulation challenges associated with studying model microphase formers. Block copolymers are by far the most studied microphase formers. The chapter focuses on the phenomenological field theory description of the universality of the microphase formation and of the nature of the order-disorder transition. The chapter describes molecular simulation methods that have been specifically designed to achieve equilibrium in the periodic microphase regime. It details the thermodynamic framework and a free energy integration simulation method, followed by a concrete introduction to the ghost particle/cluster switching method. The chapter discusses several classical Monte Carlo algorithms to enhance the efficiency of simulating disordered microphases. It presents three models for which quantitative results have been obtained: a one-dimensional, a lattice, and an off-lattice microphase former. Fine-tuning colloidal suspensions to allow the formation of periodic microphases thus remains an open experimental problem.
This review focuses on the application of Density Functional Tight Binding (DFTB) to electronic-excited states, which has attracted significant attention for extending the computationally efficient approach to the time domain. The chapter highlights the use of real-time time-dependent-DFTB to probe the electron dynamics of large systems in external electric fields where the nuclei are held fixed. Surface hopping is a general mixed quantum-classical nonadiabatic dynamics methodology with many variants, such as fewest-switches surface-hopping, decoherence induced surface hopping, independent electron surface hopping, and others. The nuclei are propagated according to classical mechanics, and the forces on the nuclei, at any given instant of time, arise from a single adiabatic potential energy surfaces. The review illustrates the charge transfer dynamics of Phenyl-C61-butyric acid methyl ester/polythiophene, which is a model system for understanding photo-induced charge transfer dynamics in organic photovoltaics.
This chapter discusses a brief history of the lattice Boltzmann method (LBM). It provides an introduction to both single-phase and a multiphase LBM. The chapter shows how the lattice Boltzmann equation can be systematically derived and clarifies the approximations involved which are needed to understand the limits in which the LBM is valid and stable. It focuses on the set of continuum equations that capture the dynamics of the binary fluid. Flows of fluid-fluid multicomponent systems also occur in a variety of natural as well as technologically relevant processes. The chapter provides few examples where the continuum equations are numerically integrated using the free-energy multiphase LBM in both two and three dimensions. It provides discussion on the implementation of the multiphase LBM and concludes with some remarks on a well-known issue in multiphase LBM, known as "spurious velocities", and possible strategies to minimize these spurious velocities.
The response of proteins to chemical reactions or impulsive excitation that occurs within the molecule has fascinated chemists for decades. In recent years ultrafast X-ray studies have provided ever more detailed information about the evolution of protein structural change following ligand photolysis, and time-resolved IR and Raman techniques, e.g., have provided detailed pictures of the nature and rate of energy transport in peptides and proteins, including recent advances in identifying transport through individual amino acids of several heme proteins. Computational tools to locate energy transport pathways in proteins have also been advancing. Energy transport pathways in proteins have since some time been identified by molecular dynamics (MD) simulations, and more recent efforts have focused on the development of coarse graining approaches, some of which have exploited analogies to thermal transport in other molecular materials. With the identification of pathways in proteins and protein complexes, network analysis has been applied to locate residues that control protein dynamics and possibly allostery, where chemical reactions at one binding site mediate reactions at distance sites of the protein. In this chapter we review approaches for locating computationally energy transport networks in proteins. We present background into energy and thermal transport in condensed phase and macromolecules that underlies the approaches we discuss before turning to a description of the approaches themselves. We also illustrate the application of the computational methods for locating energy transport networks and simulating energy dynamics in proteins with several examples.
The goals of this chapter are twofold. First, we wish to introduce molecular dynamics (MD) and uncertainty quantification (UQ) in a common setting in order to demonstrate how the latter can increase confidence in the former. In some cases, this discussion culminates in our providing practical, mathematical tools that can be used to answer the question, "is this simulation reliable?" However, many questions remain unanswered. Thus, a second goal of this work is to highlight open problems where progress would aid the larger community.
Chapter 5 Machine Learning, Quantum Chemistry, and Chemical Space Raghunathan Ramakrishnan, Raghunathan Ramakrishnan Institute of Physical Chemistry and National Center for Computational Design and Discovery of Novel Materials, Department of Chemistry, University of Basel, Basel, SwitzerlandSearch for more papers by this authorO. Anatole von Lilienfeld, O. Anatole von Lilienfeld Institute of Physical Chemistry and National Center for Computational Design and Discovery of Novel Materials, Department of Chemistry, University of Basel, Basel, Switzerland General Chemistry, Free University of Brussels, Brussels, BelgiumSearch for more papers by this author Raghunathan Ramakrishnan, Raghunathan Ramakrishnan Institute of Physical Chemistry and National Center for Computational Design and Discovery of Novel Materials, Department of Chemistry, University of Basel, Basel, SwitzerlandSearch for more papers by this authorO. Anatole von Lilienfeld, O. Anatole von Lilienfeld Institute of Physical Chemistry and National Center for Computational Design and Discovery of Novel Materials, Department of Chemistry, University of Basel, Basel, Switzerland General Chemistry, Free University of Brussels, Brussels, BelgiumSearch for more papers by this author Book Editor(s):Abby L. Parrill, Abby L. Parrill College of Arts and Sciences, The University of Memphis, Memphis, TN, U.S.ASearch for more papers by this authorKenny B. Lipkowitz, Kenny B. Lipkowitz Office of Naval Research, Arlington, VA, U.S.ASearch for more papers by this author First published: 03 April 2017 https://doi.org/10.1002/9781119356059.ch5Citations: 41Book Series:Reviews in Computational Chemistry AboutPDFPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShareShare a linkShare onFacebookTwitterLinked InRedditWechat Summary A number of machine learning (ML) studies have appeared with the commonality that quantum mechanical properties are being predicted based on regression models defined in chemical compound space (CCS). The quantum mechanical framework is crucial for the unbiased exploration of CCS since it enables, at least in principle, the free variation of nuclear charges, atomic weights, atomic configurations, and electron number. This chapter first gives a brief tutorial summary of the employed ML model in Kernel Ridge Regression. A discussion on the various representations (descriptors) used to encode molecular species, in particular the molecular Coulomb-matrix (CM), sorted or its eigenvalues follows. The chapter also reviews quantum chemistry data of 134k molecules. The local, linearly scaling ML models for atomic properties such as forces on atoms, nuclear magnetic resonance (NMR) shifts, core-electron ionization energies, as well as atomic charges, dipole-moments, and quadrupole-moments for force-field predictions are finally discussed. Citing Literature Reviews in Computational Chemistry, Volume 30 RelatedInformation
This chapter presents a review and tutorial on molecular dynamics (MD) simulation of shock loading of solids, under which materials are compressed at ultra-fast rates to extreme conditions of pressure and temperature. Due to the ultra-fast loading rates, shockwaves can reveal processes not accessible otherwise, including melting below the equilibrium melting temperature and chemical reactions away from equilibrium. The timescales involved in shock physics make MD an ideal tool for their study and these atomic-level simulations have and continue to play a critical role of our understanding of the physics and chemistry of materials under extreme mechanical loads. Such simulations, when done with care, have been particularly useful to resolve complex processes that occur at or right behind the shock front, where plasticity, phase transformations, and the initiation of chemical reactions are relevant. Flyer plate simulations, coarse grain dynamics, and shock-induced plasticity are also discussed.
Chapter 1 Chemical Bonding at High Pressure Andreas Hermann, Andreas Hermann School of Physics and Astronomy, The University of Edinburgh, Edinburgh, United KingdomSearch for more papers by this author Andreas Hermann, Andreas Hermann School of Physics and Astronomy, The University of Edinburgh, Edinburgh, United KingdomSearch for more papers by this author Book Editor(s):Abby L. Parrill, Abby L. Parrill College of Arts and Sciences, The University of Memphis, Memphis, TN, U.S.ASearch for more papers by this authorKenny B. Lipkowitz, Kenny B. Lipkowitz Office of Naval Research, Arlington, VA, U.S.ASearch for more papers by this author First published: 03 April 2017 https://doi.org/10.1002/9781119356059.ch1Citations: 1Book Series:Reviews in Computational Chemistry AboutPDFPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShareShare a linkShare onFacebookTwitterLinked InRedditWechat Summary This chapter commences with a note on the synthesis of diamonds in high-pressure chambers. External pressure leads to a variety of interesting and, at first sight, unexpected effects in materials, which oftentimes involve changes to the chemical bonding of their constituents. The chapter discusses briefly the different methods available to generate high pressure, and the consequences this has on making measurements; these, in turn, affect the requirements put on computational approaches. First-principles methods, which allow a parameter-free description of the electronic structure but at much higher computational cost, are usually the method of choice in computational descriptions of compressed materials. Analysis of the electronic structure of a compressed material is often at the heart of a first-principles calculation. The chapter talks about the structural choice, the character of the bonding, and how it influences the mechanical, electronic, and other properties. Citing Literature Reviews in Computational Chemistry, Volume 30 RelatedInformation
Chapter 4 The Quantum Chemistry of Open-Shell Species Anna I. Krylov, Anna I. Krylov Department of Chemistry, University of Southern California, Los Angeles, CA, United StatesSearch for more papers by this author Anna I. Krylov, Anna I. Krylov Department of Chemistry, University of Southern California, Los Angeles, CA, United StatesSearch for more papers by this author Book Editor(s):Abby L. Parrill, Abby L. Parrill College of Arts and Sciences, The University of Memphis, Memphis, TN, U.S.ASearch for more papers by this authorKenny B. Lipkowitz, Kenny B. Lipkowitz Office of Naval Research, Arlington, VA, U.S.ASearch for more papers by this author First published: 03 April 2017 https://doi.org/10.1002/9781119356059.ch4Citations: 38Book Series:Reviews in Computational Chemistry AboutPDFPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShareShare a linkShare onFacebookTwitterLinked InRedditWechat Summary Open-shell electronic structure is ubiquitous in chemistry and spectroscopy. This chapter provides a useful guide for navigation through the myriad problems posed by open-shell species. It highlights the computational aspects relevant to applications, such as modeling spectroscopy and nonadiabatic processes in open-shell species. The chapter begins by summarizing standard quantum chemistry methods, outlining their scope of applicability, and explaining why open-shell species require special treatment. It then discusses several topics relevant to open-shell systems: spin and spin contamination, Jahn-Teller (JT), and pseudo-Jahn-Teller (PJT) effects. This is followed by an overview on the electronic structure of high-spin states and contrast cases with and without electronic degeneracies. Methods designed to tackle various types of open-shell electronic structure (simple high-spin states, charge-transfer systems, diradicals and triradicals, excited states of open-shell molecules) are introduced. Finally, calculations of molecular properties relevant to spectroscopy and excited-state processes are discussed. Citing Literature Reviews in Computational Chemistry, Volume 30 RelatedInformation
