Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userAtomic-scale representation and statistical learning of tensorial properties
80
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
This chapter discusses the importance of incorporating three-dimensional symmetries in the context of statistical learning models geared towards the interpolation of the tensorial properties of atomic-scale structures. We focus on Gaussian process regression, and in particular on the construction of structural representations, and the associated kernel functions, that are endowed with the geometric covariance properties compatible with those of the learning targets. We summarize the general formulation of such a symmetry-adapted Gaussian process regression model, and how it can be implemented based on a scheme that generalizes the popular smooth overlap of atomic positions representation. We give examples of the performance of this framework when learning the polarizability and the ground-state electron density of a molecule.
rate research
Read More
Atomistic modeling of energetic disorder in organic semiconductors (OSCs) and its effects on the optoelectronic properties of OSCs requires a large number of excited-state electronic-structure calculations, a computationally daunting task for many OSC applications. In this work, we advocate the use of deep learning to address this challenge and demonstrate that state-of-the-art deep neural networks (DNNs) are capable of predicting the electronic properties of OSCs at an accuracy comparable with the quantum chemistry methods used for generating training data. We extensively investigate the performances of four recent DNNs (deep tensor neural network, SchNet, message passing neural network, and multilevel graph convolutional neural network) in predicting various electronic properties of an important class of OSCs, i.e., oligothiophenes (OTs), including their HOMO and LUMO energies, excited-state energies and associated transition dipole moments. We find that SchNet shows the best performance for OTs of different sizes (from bithiophene to sexithiophene), achieving average prediction errors in the range of 20-80meV compared to the results from (time-dependent) density functional theory. We show that SchNet also consistently outperforms shallow feed-forward neural networks, especially in difficult cases with large molecules or limited training data. We further show that SchNet could predict the transition dipole moment accurately, a task previously known to be difficult for feed-forward neural networks, and we ascribe the relatively large errors in transition dipole prediction seen for some OT configurations to the charge-transfer character of their excited states. Finally, we demonstrate the effectiveness of SchNet by modeling the UV-Vis absorption spectra of OTs in dichloromethane and a good agreement is observed between the calculated and experimental spectra.
The atomic cluster expansion (Drautz, Phys. Rev. B 99, 014104 (2019)) is extended in two ways, the modelling of vectorial and tensorial atomic properties and the inclusion of atomic degrees of freedom in addition to the positions of the atoms. In particular, atomic species, magnetic moments and charges are attached to the atomic positions and an atomic cluster expansion that includes the different degrees of freedom on equal footing is derived. Expressions for the efficient evaluation of forces and torques are given. Relations to other methods are discussed.
Electronic nearsightedness is one of the fundamental principles governing the behavior of condensed matter and supporting its description in terms of local entities such as chemical bonds. Locality also underlies the tremendous success of machine-learning schemes that predict quantum mechanical observables -- such as the cohesive energy, the electron density, or a variety of response properties -- as a sum of atom-centred contributions, based on a short-range representation of atomic environments. One of the main shortcomings of these approaches is their inability to capture physical effects, ranging from electrostatic interactions to quantum delocalization, which have a long-range nature. Here we show how to build a multi-scale scheme that combines in the same framework local and non-local information, overcoming such limitations. We show that the simplest version of such features can be put in formal correspondence with a multipole expansion of permanent electrostatics. The data-driven nature of the model construction, however, makes this simple form suitable to tackle also different types of delocalized and collective effects. We present several examples that range from molecular physics, to surface science and biophysics, demonstrating the ability of this multi-scale approach to model interactions driven by electrostatics, polarization and dispersion, as well as the cooperative behavior of dielectric response functions.
Net atomic charges (NACs) are widely used in all chemical sciences to concisely summarize key information about the partitioning of electrons among atoms in materials. Although widely used, there is currently no atomic population analysis method suitable for being used as a default method in quantum chemistry programs. To address this challenge, we introduce a new atoms-in-materials method with the following nine properties: (1) exactly one electron distribution is assigned to each atom, (2) core electrons are assigned to the correct host atom, (3) NACs are formally independent of the basis set type because they are functionals of the total electron distribution, (4) the assigned atomic electron distributions give an efficiently converging polyatomic multipole expansion, (5) the assigned NACs usually follow Pauling scale electronegativity trends, (6) NACs for a particular element have good transferability among different conformations that are equivalently bonded, (7) the assigned NACs are chemically consistent with the assigned atomic spin moments, (8) the method has predictably rapid and robust convergence to a unique solution, and (9) the computational cost of charge partitioning scales linearly with increasing system size. Across a broad range of material types, the DDEC6 NACs reproduced electron transfer trends, core electron binding energy shift trends, and electrostatic potentials across multiple system conformations with excellent accuracy compared to other charge assignment methods. Due to non-nuclear attractors, Baders quantum chemical topology could not assign NACs for some of these materials. The DDEC6 method alleviates the bifurcation or runaway charges problem exhibited by earlier DDEC variants and the Iterative Hirshfeld method. These characteristics make the DDEC6 method ideally suited for use as a default charge assignment method in quantum chemistry programs.
Quantum-chemical processes in liquid environments impact broad areas of science, from molecular biology to geology to electrochemistry. While density-functional theory (DFT) has enabled efficient quantum-mechanical calculations which profoundly impact understanding of atomic-scale phenomena, realistic description of the liquid remains a challenge. Here, we present an approach based on joint density-functional theory (JDFT) which addresses this challenge by leveraging the DFT approach not only for the quantum mechanics of the electrons in a solute, but also simultaneously for the statistical mechanics of the molecules in a surrounding equilibrium liquid solvent. Specifically, we develop a new universal description for the interaction of electrons with an arbitrary liquid, providing the missing link to finally transform JDFT into a practical tool for the realistic description of chemical processes in solution. This approach predicts accurate solvation free energies and surrounding atomic-scale liquid structure for molecules and surfaces in multiple solvents without refitting, all at a fraction of the computational cost of methods of comparable detail and accuracy. To demonstrate the potential impact of this method, we determine the structure of the solid/liquid interface, offering compelling agreement with more accurate (but much more computationally intensive) theories and with X-ray reflectivity measurements.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
