Bulletin of the American Physical Society
APS March Meeting 2022
Volume 67, Number 3
Monday–Friday, March 14–18, 2022; Chicago
Session S46: Emerging Trends in Molecular Dynamics Simulations and Machine Learning IFocus Recordings Available
|
Hide Abstracts |
Sponsoring Units: DCOMP GDS DSOFT DPOLY Chair: Priya Vashishta, University of Southern California Room: McCormick Place W-470A |
Thursday, March 17, 2022 8:00AM - 8:36AM |
S46.00001: AI-driven modeling of quantum materials architectures Invited Speaker: Rajiv K Kalia
|
Thursday, March 17, 2022 8:36AM - 8:48AM |
S46.00002: Deep potential molecular dynamics of water self-ionization Marcos C Andrade, Roberto Car, Annabella Selloni The chemical equilibrium between self-ionized and molecular water dictates the acid-base chemistry in aqueous solutions. Despite its relevance, determining the extent of water self ionization reaction remains an outstanding problem from a theoretical perspective. The lack of computationally efficient and reactive interatomic potentials and the challenging description of the water self-ionization prevented a direct quantification of this reaction from first principles. In this work, we combined deep neural network potentials with enhanced sampling techniques to perform extensive atomistic simulations of water self-ionization at finite temperature. The free energy as a function of hydroxide and hydronium separation is computed, evidencing a finite-size dependence of the free energy curves and allowing us to evaluate the equilibrium pH of water from first principles. Our computed equilibrium pH of 7.3 ± 0.1 is close to the experimental value of 7.0. The methodology developed in this work can be readily generalized to compute acid-base equilibrium constants of other condensed phase systems. |
Thursday, March 17, 2022 8:48AM - 9:00AM |
S46.00003: Dynamics of Wrinkle-Ridge Transition in Graphene Supported on a Polymer: Quantum Molecular Dynamics Simulations Anikeya Aditya, Shogo Fukushima, Ankit Mishra, Ken-ichi Nomura, Fuyuki Shimojo, Aiichiro Nakano, Priya Vashishta, Rajiv K Kalia, Mark J Stevens Miura-ori pattern is seen in nature be it the unfolding of a leaf or opening and closing of insect wings. Similar patterns in the form of wrinkles and ridges are also observed in biaxially compressed rigid thin films supported on soft substrates. We have investigated the formation of wrinkles and ridges in graphene placed on a polyethylene substrate using quantum molecular dynamics (QXMD) simulation. QXMD is a massively parallel quantum molecular dynamics software with various “eXtensions” such as non-adiabatic dynamics for the study of light-induced electronic excitations. In our QXMD simulations, we have seen the formation of wrinkles and ridges in biaxially compressed graphene sheets and calculated the electronic density of states and stress and strain distributions in wrinkle and ridge states. We will also present results of large-scale classical molecular dynamics simulations of wrinkle and ridge states of graphene on the polyethylene substrate. |
Thursday, March 17, 2022 9:00AM - 9:12AM |
S46.00004: A Critical Assessment of Neural Network Potentials for Water and the Role of Nuclear Quantum Effects through the Van Hove Correlation Function Murali Gopal Muraleedharan, Paul Kent High accuracy studies of the properties of liquid water should combine the accuracies achievable by ab initio molecular dynamics (AIMD) with the long time and large length scales achievable by classical MD. Here, we assess two different neural network potential (NNP) based modeling strategies within classical MD in predicting the spatiotemporal correlations in liquid water i.e., the Van Hove correlation function (VHF). In principle, the NNP’s can deliver ab initio quality results at significantly larger time and length scales than would be directly accessible using ab initio methods. We apply the DeepMD [1] and NequIP [2] approaches and critically assess their efficacy, particularly regarding the size of training set and accuracy of the final predictions. By varying the training sets and using path integral approaches, we analyze the role of nuclear quantum effect on the VHF. These results are also contrasted with recent inelastic X-ray scattering data [3]. |
Thursday, March 17, 2022 9:12AM - 9:24AM |
S46.00005: Many-body interatomic potential with Bayesian active learning, an application ofSiC Yu Xie Machine learning interatomic potentials (MLIPs) have high efficiency and quantum accuracy to model atomic interactions and simulate atomic level processes. Active learning methods have been developed to train MLIPs efficiently. Among them, Bayesian active learning (BAL) utilizes uncertainty quantification as an acquisition threshold. In this work, we present a highly efficient BAL workflow, where MLIPs is constructed using Gaussian process (GP) kernels based on the atomic cluster expansion (ACE) descriptors which is trained efficiently with MPI parallelization. A high-performance mapping of the potential and an approximation of the uncertainty of the trained GP are developed. We demonstrate that our workflow is orders faster compared to BAL with exact GPs. |
Thursday, March 17, 2022 9:24AM - 10:00AM |
S46.00006: Structural optimization using learned optimizers and graph neural networks Invited Speaker: Ekin D Cubuk
|
Thursday, March 17, 2022 10:00AM - 10:12AM |
S46.00007: Opto-Electro-Mechanical control of Ferroelectric Topological Structures for Ultralow Power Topotronic Devices using Hybrid Neural Network Quantum Molecular Dynamics and Molecular Mechanics Simulations Thomas M Linker, Ken-ichi Nomurra, Shogo Fukshima, Rajiv K Kalia, Aravind Krishnamoorthy, Aiichiro Nakano, Kohei Shimamura, Fuyuki Shimojo, Priya Vashishta The recent discovery of the formation of complex polar topological structures such as skyrmions and merons in paraelectric(PE)/ferroelectric(FE) oxide heterostructures has opened the field of ferroelectric topotronics for development of ultralow power electronic devices. To understand how to electrically and mechanically manipulate these complex structures we have developed a hybrid Neural-Network (NN) and Molecular Mechanics (MM) approach. Analogous to modern Quantum Mechanical (QM)/MM forcefield approaches, a highly scalable neural network is trained on ab-intio molecular dynamics (MD) simulations to encapsulate the complex polarization dynamics within the ferroelectric layer and a classical forcefield is tuned to correctly describe the strain and polarization fields applied by the paraelectric layer. We examine the effects of strain, shear, electric field, and light induced electronic excitation to manipulate topological structures in PE/FE oxide heterostructures composed of SrTiO3(STO) and PbTiO3(PTO). The hybrid NN/MM approach developed here can perform up to billion atom MD simulations, offering an exciting new avenue for exploring opto-electro-mechanical control ultralow power topological based devices. |
Thursday, March 17, 2022 10:12AM - 10:24AM |
S46.00008: Active learning of reactive Bayesian force fields: Application to heterogeneous catalysis dynamics of H/Pt Jonathan P Vandermause, Yu Xie, Jin Soo Lim, Cameron J Owen, Boris Kozinsky Atomistic modeling of chemically reactive systems has traditionally relied on either expensive ab initio methods or bond-order force fields requiring arduous parametrization. Here, we introduce FLARE++, a Bayesian active learning method for training reactive many-body force fields on the fly during molecular dynamics (MD) simulations. At each timestep, the predictive uncertainties of a sparse Gaussian process (SGP) are evaluated to automatically determine whether additional ab initio data are needed. The resulting SGP is mapped onto a polynomial model whose prediction cost is independent of the training set size. Using this method, we perform large-scale MD simulations of a prototypical system in heterogeneous catalysis---H2 chemisorption on Pt(111)---at chemical accuracy. The model, trained within three days of wall time, performs at twice the simulation speed of an available ReaxFF model while maintaining ab initio accuracy to a much higher fidelity. Our method enables efficient automated training of fast and accurate reactive force fields for chemically complex systems. |
Thursday, March 17, 2022 10:24AM - 10:36AM |
S46.00009: Committee Disagreement Biased Active Learning of Interatomic Potentials Michael J Waters, James M Rondinelli Committee models are well known to improve generalizability of machine-learned models and neural-network models in general. Moreover, the disagreement between the predictions of the individual models can be used as a proxy for overall model uncertainty quantification. By exploiting the differentiability of interatomic potential models, atomic structures can be driven into regions of high uncertainty to find new training structures as part of an adversarial attack scheme. We explore several ways of incorporating this adversarial attack scheme into practical structure generation schemes like molecular dynamics and our contour exploration scheme [1] for efficient active learning of interatomic potentials. We showcase the performance of this approach using transition metals and their oxides as benchmark systems. |
Thursday, March 17, 2022 10:36AM - 10:48AM |
S46.00010: Ab Initio Thermodynamics of Ferroelectrics: The case of PbTiO3 Pinchen Xie, Yixiao Chen, Weinan E, Roberto Car We report a molecular dynamics study of the ferroelectric phase transition in PbTiO3 using deep neural networks, trained on first-principles SCAN-DFT data, to represent potential and polarization surfaces [1,2]. Our approach includes anharmonic effects beyond the limitations of reduced models and of the linear approximation for the polarization. An external isotropic pressure of ~28 kbar was applied to correct for the overestimate of the tetragonality by the SCAN functional. The calculated enthalpy, spontaneous polarization, specific heat and dielectric susceptibility agree well with experiments on single domain crystals. In addition, we studied how the free energy depends on the polarization with enhanced sampling methods, further supporting the first-order character of the transition. |
Thursday, March 17, 2022 10:48AM - 11:00AM |
S46.00011: Designing Machine Learning Surrogates using Outputs of Molecular Dynamics Simulations as Soft Labels Jayanath Chamindu Sandanuwan K Kadupitige, Nasim Anousheh, Vikram Jadhao Many outputs of molecular dynamics simulations of soft materials are associated with statistical uncertainties. We show that these uncertainties can be utilized to informate the training of deep neural networks for designing machine learning surrogates aimed at predicting the relationship between input variables and simulation outputs. The approach is illustrated with the design of a surrogate for molecular dynamics simulations of confined electrolytes to predict the complex relationship between the input electrolyte attributes and the output ionic structure. We demonstrate that the prediction error for samples in the unseen test data can be significantly reduced by utilizing a modified loss function that leverages the uncertainties in the output ionic distributions during training. Using such soft labels for the ground truth facilitates a sampling mechanism that implicitly expands the dataset with more samples as the model undergoes training over many epochs, yielding a surrogate with higher generalizability. The surrogate predictions for the ionic density profiles are found to be in excellent agreement with the ground truth results produced using molecular dynamics simulations. |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700