Featured
-
-
Article
| Open Access
Towards a transferable fermionic neural wavefunction for molecules
Neural wavefunctions have become a highly accurate approach to solve the Schrödinger equation. Here, the authors propose an approach to optimize for a generalized wavefunction across compounds, which can help developing a foundation wavefunction model.
- Michael Scherbela
- , Leon Gerard
- & Philipp Grohs
-
Article
| Open Access
Fast free energy estimates from λ-dynamics with bias-updated Gibbs sampling
Calculations of relative binding free energy are crucial for lead optimization in structure-based drug design, but classical methods are computationally expensive. Here, the authors describe a more efficient method for calculating the free energy that is as accurate as thermodynamic integration.
- Michael T. Robo
- , Ryan L. Hayes
- & Jonah Z. Vilseck
-
Article
| Open Access
Data-driven discovery of electrocatalysts for CO2 reduction using active motifs-based machine learning
Conventional ab initio calculations and machine learning provide limited information on catalytic activity and selectivity and often show discrepancy with experimental results. Here, the authors report a high-throughput virtual screening strategy to identify active and selective catalysts, leading to the discovery of Cu-Ga and Cu-Pd catalysts for CO2 electroreduction.
- Dong Hyeon Mok
- , Hong Li
- & Seoin Back
-
Article
| Open Access
Knowledge-driven design of solid-electrolyte interphases on lithium metal via multiscale modelling
The application of Li metal electrodes in rechargeable batteries is limited by inherent high reactivity. Here, the authors provide model-based insights into the composition and formation mechanisms of the solid-electrolyte interphase on the µs-scale and suggest design strategies for the interphase.
- Janika Wagner-Henke
- , Dacheng Kuai
- & Ulrike Krewer
-
Article
| Open Access
Machine learning electronic structure methods based on the one-electron reduced density matrix
Electronic structure methods are vital, yet they are often too computationally expensive. Here, the authors develop machine learned density matrices to fully represent electronic structures in a computationally cheap and accurate way.
- Xuecheng Shao
- , Lukas Paetow
- & Michele Pavanello
-
Article
| Open Access
Minimizing non-radiative decay in molecular aggregates through control of excitonic coupling
Exciton delocalization in molecular aggregates is suggested to counteract the Energy Gap Law. Here, authors reveal the underlying physical picture and find the optimal excitonic coupling that minimizes nonradiative decay by nearly exact simulations.
- Yuanheng Wang
- , Jiajun Ren
- & Zhigang Shuai
-
Article
| Open Access
Bayesian-optimization-assisted discovery of stereoselective aluminum complexes for ring-opening polymerization of racemic lactide
Stereoselective catalysts impact polymer’s properties, but discovering such catalysts is expensive and based on trial-and-error. Here, the authors develop a machine-learning tool to guide catalyst discovery and reveal mechanistic features affecting stereoselectivity.
- Xiaoqian Wang
- , Yang Huang
- & Rong Tong
-
Article
| Open Access
Reaction performance prediction with an extrapolative and interpretable graph model based on chemical knowledge
Predictive modelling remains a key challenge for designing synthetic transformations. Here, the authors develop a knowledge-based graph model to predict reaction yield and stereoselectivity, offering an extrapolative and interpretable approach for evaluating reaction performance.
- Shu-Wen Li
- , Li-Cheng Xu
- & Xin Hong
-
Article
| Open Access
Retrosynthesis prediction using an end-to-end graph generative architecture for molecular graph editing
Retrosynthesis prediction is a fundamental problem in organic synthesis. Here, inspired by simplified arrow-pushing reaction mechanisms, the authors develop a graph-to-edits framework, Graph2Edits, based on graph neural network for retrosynthesis prediction.
- Weihe Zhong
- , Ziduo Yang
- & Calvin Yu-Chian Chen
-
Article
| Open Access
Single-step retrosynthesis prediction by leveraging commonly preserved substructures
Retrosynthesis is a critical task for organic chemistry with numerous industrial applications. Here, the authors build a machine learning model to learn the concept of substructures from a large reaction dataset to achieve chemist-like intuitions.
- Lei Fang
- , Junren Li
- & Jian-Guang Lou
-
Article
| Open Access
Prediction of transition state structures of gas-phase chemical reactions via machine learning
Obtaining good initial structures is the main challenge for the computational study of transition states. Here, fast and accurate predictions for transition state of gas phase reactions are achieved by machine learning based on interatomic distances.
- Sunghwan Choi
-
Article
| Open Access
Electronic excited states in deep variational Monte Carlo
Deep neural networks can learn and represent nearly exact electronic ground states. Here, the authors advance this approach to excited states, achieving high accuracy across a range of atoms and molecules, opening up the possibility to model many excited-state processes.
- M. T. Entwistle
- , Z. Schätzle
- & F. Noé
-
Article
| Open Access
Machine learning the Hohenberg-Kohn map for molecular excited states
Density functional theory provides a formal map from the electron density to all observables of interest of a many-body system; however, maps for electronic excited states are unknown. Here, the authors demonstrate a data-driven machine learning approach for constructing multistate functionals.
- Yuanming Bai
- , Leslie Vogt-Maranto
- & William J. Glover
-
Article
| Open Access
Allotropy in ultra high strength materials
Here the authors propose a crystal thermodynamics framework describing the tensor stress induced phase transformations in solids based on nonlinear elasticity and first principles calculations. The proposed approach enables balanced design of high-strength, high-ductility materials.
- A. S. L. Subrahmanyam Pattamatta
- & David J. Srolovitz
-
Article
| Open Access
Language models can learn complex molecular distributions
Generative models for the novo molecular design attract enormous interest for exploring the chemical space. Here the authors investigate the application of chemical language models to challenging modeling tasks demonstrating their capability of learning complex molecular distributions.
- Daniel Flam-Shepherd
- , Kevin Zhu
- & Alán Aspuru-Guzik
-
Article
| Open Access
Machine learning the metastable phase diagram of covalently bonded carbon
Exploration of metastable phases of a given elemental composition is a data-intensive task. Here the authors integrate first-principles atomistic simulations with machine learning and high-performance computing to allow a rapid exploration of the metastable phases of carbon.
- Srilok Srinivasan
- , Rohit Batra
- & Subramanian K.R.S. Sankaranarayanan
-
Article
| Open Access
Towards universal neural network potential for material discovery applicable to arbitrary combination of 45 elements
Existing neural network potentials are generally designed for narrow target materials. Here the authors develop a neural network potential which is able to handle any combination of 45 elements and show its applicability in multiple domains.
- So Takamoto
- , Chikashi Shinagawa
- & Takeshi Ibuka
-
Article
| Open Access
Exact simulation of pigment-protein complexes unveils vibronic renormalization of electronic parameters in ultrafast spectroscopy
Multimode vibronic mixing in model photosynthetic systems revealed by numerically exact simulations is shown to strongly modify linear and non-linear optical responses and facilitate the persistence of coherent dynamics.
- F. Caycedo-Soler
- , A. Mattioni
- & M. B. Plenio
-
Article
| Open Access
Automated exploitation of the big configuration space of large adsorbates on transition metals reveals chemistry feasibility
The discovery of heterogeneous catalysts for large molecule conversion has been lagging due to the combinatorial inventory of intermediates. Here, the author presents an automated framework to explore the chemical space of reaction intermediates.
- Geun Ho Gu
- , Miriam Lee
- & Dionisios G. Vlachos
-
Article
| Open Access
Predicting the future of excitation energy transfer in light-harvesting complex with artificial intelligence-based quantum dynamics
Simulations of energy transfer in light-harvesting complexes are computationally very demanding. Here the authors apply an artificial intelligence quantum dissipative algorithm to study the excited state energy transfer dynamics in a light-harvesting complex.
- Arif Ullah
- & Pavlo O. Dral
-
Article
| Open Access
Self-consistent determination of long-range electrostatics in neural network potentials
Machine learning-based neural network potentials often cannot describe long-range interactions. Here the authors present an approach for building neural network potentials that can describe the electronic and nuclear response of molecular systems to long-range electrostatics.
- Ang Gao
- & Richard C. Remsing
-
Article
| Open Access
Crystal structure prediction by combining graph network and optimization algorithm
Predicting crystal structure prior to experimental synthesis is highly desirable. Here the authors propose a machine-learning framework combining graph network and optimization algorithms for crystal structure prediction, which is about three orders of magnitude faster than DFT-based approach.
- Guanjian Cheng
- , Xin-Gao Gong
- & Wan-Jian Yin
-
Article
| Open Access
Nonlocal pseudopotential energy density functional for orbital-free density functional theory
Orbital-free density functional theory is an electronic structure method with a low computational cost enabling large-scale material simulations. Here the authors present a novel protocol which allows for the application of nonlocal pseudopotentials to orbital-free density functional theory.
- Qiang Xu
- , Cheng Ma
- & Yanming Ma
-
Article
| Open Access
Molecular orbital theory in cavity QED environments
Theoretical description of light-matter coupling in the strong-coupling regime is challenging. Here the authors introduce a fully consistent ab-initio method of molecular orbital theory applicable to material systems in quantum electrodynamics environments.
- Rosario R. Riso
- , Tor S. Haugland
- & Henrik Koch
-
Article
| Open Access
Optical van-der-Waals forces in molecules: from electronic Bethe-Salpeter calculations to the many-body dispersion model
The authors devise an efficient quantum approach to address the van der Waals interactions due to photoexcitations by approximating the Bethe-Salpeter equation. Both attractive/repulsive forces can arise, that could couple to collective protein dynamics.
- Alberto Ambrosetti
- , Paolo Umari
- & Alexandre Tkatchenko
-
Article
| Open Access
SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects
Current machine-learned force fields typically ignore electronic degrees of freedom. SpookyNet is a deep neural network that explicitly treats electronic degrees of freedom, closing an important remaining gap for models in quantum chemistry.
- Oliver T. Unke
- , Stefan Chmiela
- & Klaus-Robert Müller
-
Article
| Open Access
Artificial intelligence-enhanced quantum chemical method with broad applicability
Artificial intelligence is combined with quantum mechanics to break the limitations of traditional methods and create a new general-purpose method for computational chemistry simulations with high accuracy, speed and transferability.
- Peikun Zheng
- , Roman Zubatyuk
- & Pavlo O. Dral
-
Article
| Open Access
Augmenting zero-Kelvin quantum mechanics with machine learning for the prediction of chemical reactions at high temperatures
Computational material design often does not account for temperature effects. The present manuscript combines quantum-mechanics based calculations with a machine-learned correction to establish a unified thermodynamics framework for accurate prediction of high temperature reaction free energies in oxides.
- Jose Antonio Garrido Torres
- , Vahe Gharakhanyan
- & Alexander Urban
-
Article
| Open Access
Elevating density functional theory to chemical accuracy for water simulations through a density-corrected many-body formalism
No existing density functional correctly describes the properties of water across the entire phase diagram. The authors report a data-driven many-body potential energy function based on density-corrected SCAN functional that quantitatively reproduces the energetics of gas-phase water clusters, and correctly predicts the properties of liquid water.
- Saswata Dasgupta
- , Eleftherios Lambros
- & Francesco Paesani
-
Article
| Open Access
Robust recognition and exploratory analysis of crystal structures via Bayesian deep learning
The present manuscript reports a Bayesian deep-learning approach for the automatic, robust classification of polycrystalline systems of both synthetic and experimental origin. The unsupervised analysis of the internal neural-network representations reveals physically understandable patterns.
- Andreas Leitherer
- , Angelo Ziletti
- & Luca M. Ghiringhelli
-
Article
| Open Access
Differentiable sampling of molecular geometries with uncertainty-based adversarial attacks
Neural Networks are known to perform poorly outside of their training domain. Here the authors propose an inverse sampling strategy to train neural network potentials enabling to drive atomistic systems towards high-likelihood and high-uncertainty configurations without the need for molecular dynamics simulations.
- Daniel Schwalbe-Koda
- , Aik Rui Tan
- & Rafael Gómez-Bombarelli
-
Article
| Open Access
Real space electron delocalization, resonance, and aromaticity in chemistry
The concept of delocalization, resonance and aromaticity are commonly discussed within electronic structure frameworks relying on specific wave function expansions. Here the authors propose a redefinition of these concepts from first-principles by investigating saddle points of the all-electron probability density.
- Leonard Reuter
- & Arne Lüchow
-
Article
| Open Access
Data-driven reaction coordinate discovery in overdamped and non-conservative systems: application to optical matter structural isomerization
Optical matter consisting of nanoparticle constituents in solution is of key interest due to the exhibited self-assembling mechanisms. The authors propose a principal components analysis based data-driven approach to determine the collective modes of colloidal clusters mimicking optical binding used in colloidal self-assembly.
- Shiqi Chen
- , Curtis W. Peterson
- & Norbert F. Scherer
-
Article
| Open Access
Magnetically induced currents and aromaticity in ligand-stabilized Au and AuPt superatoms
Efficient methods to calculate magnetically induced currents in metallic nanostructures are currently lacking. Here, the authors propose a theoretical method to compute and analyze magnetically induced currents in nanostructures validated for experimentally synthesized gold-based, hydrogen-containing ligand-protected clusters.
- Omar López-Estrada
- , Bernardo Zuniga-Gutierrez
- & Hannu Häkkinen
-
Article
| Open Access
Thermochemical electronegativities of the elements
Pauling’s electronegativity scale has a fundamental value and uses accessible thermochemical data, but fails at predicting the bonding behavior for several elements. The authors propose their thermochemical scale based on experimental dissociation energies that provides dimensionless values for the electronegativity and recovers the correct trends throughout the periodic table.
- Christian Tantardini
- & Artem R. Oganov
-
Article
| Open Access
Automated discovery of a robust interatomic potential for aluminum
The accuracy of a machine-learned potential is limited by the quality and diversity of the training dataset. Here the authors propose an active learning approach to automatically construct general purpose machine-learning potentials here demonstrated for the aluminum case.
- Justin S. Smith
- , Benjamin Nebgen
- & Kipton Barros
-
Article
| Open Access
Bridging scales in disordered porous media by mapping molecular dynamics onto intermittent Brownian motion
The diffusion of fluids in complex nanoporous geometries represents a challenge for modelling approaches. Here, the authors describe the macroscopic diffusivity of a simple fluid in disordered nanoporous materials by bridging microscopic and mesoscopic dynamics with parameters obtained from simple physical laws.
- Colin Bousige
- , Pierre Levitz
- & Benoit Coasne
-
Article
| Open Access
A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer
Machine learning potentials do not account for long-range charge transfer. Here the authors introduce a fourth-generation high-dimensional neural network potential including non-local information of charge populations that is able to provide forces, charges and energies in excellent agreement with DFT data.
- Tsz Wai Ko
- , Jonas A. Finkler
- & Jörg Behler
-
Article
| Open Access
Pure non-local machine-learned density functional theory for electron correlation
Semilocal density functionals, while computationally efficient, do not account for non-local correlation. Here, the authors propose a machine-learning approach to DFT that leads to non-local and transferable functionals applicable to non-covalent, ionic and covalent interactions across system of different sizes.
- Johannes T. Margraf
- & Karsten Reuter
-
Article
| Open Access
Coulomb interactions between dipolar quantum fluctuations in van der Waals bound molecules and materials
High-level methods to describe van der Waals interactions are limited due to their computational cost. This work introduces a new theoretical approach, that extends the dipolar many-body dispersion formalism to higher-order contributions, demonstrated to be applicable to practically-relevant systems and nano-environments.
- Martin Stöhr
- , Mainak Sadhukhan
- & Alexandre Tkatchenko
-
Article
| Open Access
Bayesian learning of chemisorption for bridging the complexity of electronic descriptors
Developing a generalizable model to describe adsorption processes at metal surfaces can be extremely challenging due to complex phenomena involved. Here the authors introduce a Bayesian learning approach based on ab initio data and the d-band model to capture the essential physics of adsorbate–substrate interactions.
- Siwen Wang
- , Hemanth Somarajan Pillai
- & Hongliang Xin
-
Article
| Open Access
Machine learning in chemical reaction space
Application of machine-learning approaches to exploring chemical reaction networks is challenging due to need of including open-shell reaction intermediates. Here the authors introduce a density functional theory database of closed and open-shell molecules for machine-learning predictions of reaction energies.
- Sina Stocker
- , Gábor Csányi
- & Johannes T. Margraf
-
Comment
| Open Access
Retrospective on a decade of machine learning for chemical discovery
Over the last decade, we have witnessed the emergence of ever more machine learning applications in all aspects of the chemical sciences. Here, we highlight specific achievements of machine learning models in the field of computational chemistry by considering selected studies of electronic structure, interatomic potentials, and chemical compound space in chronological order.
- O. Anatole von Lilienfeld
- & Kieron Burke
-
Article
| Open Access
Anharmonic quantum nuclear densities from full dimensional vibrational eigenfunctions with application to protonated glycine
Accurate interpretation of molecular vibrational spectroscopic signals is key to understand chemical processes. Here the authors introduce a new computational approach to represent vibrational modes in terms of nuclear densities that captures anharmonic effects in protonated glycine.
- Chiara Aieta
- , Marco Micciarelli
- & Michele Ceotto
-
Article
| Open Access
Multisecond ligand dissociation dynamics from atomistic simulations
Protein-ligand unbinding processes are out of reach for atomistic simulations due to time-scale involved. Here the authors demonstrate an approach relying on dissipation-corrected targeted molecular dynamics that enables to provide binding and unbinding rates with a speed-up of several orders of magnitude.
- Steffen Wolf
- , Benjamin Lickert
- & Gerhard Stock
-
Article
| Open Access
From quantum to continuum mechanics in the delamination of atomically-thin layers from substrates
The unexpectedly long-ranged interface stress observed in recent delamination experiments is yet to be clarified. Here, the authors develop an analytical approach to show the wavelike atomic deformation as the origin for the observed ultra long-range stress in delamination of graphene from various substrates.
- Paul Hauseux
- , Thanh-Tung Nguyen
- & Alexandre Tkatchenko
-
Article
| Open Access
The electronic structure of benzene from a tiling of the correlated 126-dimensional wavefunction
The electronic structure of benzene has been a test bed for competing theories along the years. Here the authors show via quantum chemistry calculations that the wavefunction of benzene can be partitioned into tiles which show that the two electron spins exhibit staggered Kekulé structures.
- Yu Liu
- , Phil Kilby
- & Timothy W. Schmidt
-
Article
| Open Access
Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions
Machine learning models can accurately predict atomistic chemical properties but do not provide access to the molecular electronic structure. Here the authors use a deep learning approach to predict the quantum mechanical wavefunction at high efficiency from which other ground-state properties can be derived.
- K. T. Schütt
- , M. Gastegger
- & R. J. Maurer
-
Article
| Open Access
Hydrogen bonding structure of confined water templated by a metal-organic framework with open metal sites
The properties of water under confinement are significantly altered with respect to the bulk phase. Here the authors use infrared spectroscopy and many-body molecular dynamics simulations to show the structure and dynamics of confined water as a function of relative humidity within a metal-organic framework.
- Adam J. Rieth
- , Kelly M. Hunter
- & Francesco Paesani
Combining stochastic resetting with Metadynamics to speed-up molecular dynamics simulations