Search

Scholarly Works (6 results)

Sort By:

Thesis
Peer Reviewed

Machine Learning Towards Large-scale Atomistic Simulation and Materials Discovery

Zuo, Yunxing
Advisor(s): Ong, Shyue Ping

UC San Diego Electronic Theses and Dissertations (2021)

In materials science, the first principles modeling, especially density functional theory (DFT), serves as the de facto tool in studying physical phenomena and properties of materials from the atomistic level. However, the high computational cost and poor scaling of DFT has limited its applications in two important scientific problems — large-scale atomistic simulations and high-throughput screening for materials discovery. This thesis demonstrates how the machine learning (ML) techniques enable atomistic simulations in large size and time scale with DFT-accuracy and accelerate materials discovery with the state-of-the-art graph neural network models. This thesis is divided into two topics.In the first topic (Chapters 2 and 3), we will investigate how the machine learning interatomic potentials (ML-IAPs) are trained and provide a systematic assessment of the cost and accuracy performances for several major ML-IAPs. We have also implemented high-level Python interfaces for ML-IAPs development and materials properties calculators using a molecular dynamic (MD) engine. This toolkit enabled us to develop a highly accurate and efficient ML-IAP for refractory high-entropy alloy NbMoTaW, an important alloy system yielding exceptional mechanical properties under high temperature. We will demonstrate how the ML-IAP driven atomistic simulations help us understand the mobility of edge/screw dislocations with the presence of short-range order (SRO). In the second topic (Chapter 4), we developed a Bayesian Optimization With Symmetry Relaxation (BOWSR) algorithm using MatErials Graph Network (MEGNet) energy model to obtain equilibrium crystal structures, bypassing the high-cost DFT relaxations. The BOWSR algorithm enabled us to screen ∼ 400,000 transition metal borides and carbides for ultra-incompressible hard materials. Attempts were made to synthesize the top ten candidates with the highest computed bulk modulus with eight unique compositions, and two new crystals yielding ultra-incompressibility were successfully synthesized.

Cover page: Machine Learning Towards Large-scale Atomistic Simulation and Materials Discovery

Article
Peer Reviewed

Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals

UC San Diego Previously Published Works (2019)

Graph networks are a new machine learning (ML) paradigm that supports both relational reasoning and combinatorial generalization. Here, we develop universal MatErials Graph Network (MEGNet) models for accurate property prediction in both molecules and crystals. We demonstrate that the MEGNet models outperform prior ML models such as the SchNet in 11 out of 13 properties of the QM9 molecule data set. Similarly, we show that MEGNet models trained on ?60 000 crystals in the Materials Project substantially outperform prior ML models in the prediction of the formation energies, band gaps, and elastic moduli of crystals, achieving better than density functional theory accuracy over a much larger data set. We present two new strategies to address data limitations common in materials science and chemistry. First, we demonstrate a physically intuitive approach to unify four separate molecular MEGNet models for the internal energy at 0 K and room temperature, enthalpy, and Gibbs free energy into a single free energy MEGNet model by incorporating the temperature, pressure, and entropy as global state inputs. Second, we show that the learned element embeddings in MEGNet models encode periodic chemical trends and can be transfer-learned from a property model trained on a larger data set (formation energies) to improve property models with smaller amounts of data (band gaps and elastic moduli).

Cover page: Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals

Article
Peer Reviewed

Bridging the Gap Between Simulated and Experimental Ionic Conductivities in Lithium Superionic Conductors

UC San Diego Previously Published Works (2021)

Lithium superionic conductors (LSCs) are of major importance as solid electrolytes for next-generation all-solid-state lithium-ion batteries. While $ab$ $initio$ molecular dynamics have been extensively applied to study these materials, there are often large discrepancies between predicted and experimentally measured ionic conductivities and activation energies due to the high temperatures and short time scales of such simulations. Here, we present a strategy to bridge this gap using moment tensor potentials (MTPs). We show that MTPs trained on energies and forces computed using the van der Waals optB88 functional yield much more accurate lattice parameters, which in turn leads to accurate prediction of ionic conductivities and activation energies for the Li$_{0.33}$La$_{0.56}$TiO$_3$, Li$_3$YCl$_6$ and Li$_7$P$_3$S$_{11}$ LSCs. NPT MD simulations using the optB88 MTPs also reveal that all three LSCs undergo a transition between two quasi-linear Arrhenius regimes at relatively low temperatures. This transition can be traced to an expansion in the number and diversity of diffusion pathways, in some cases with a change in the dimensionality of diffusion. This work presents not only an approach to develop high accuracy MTPs, but also outlines the diffusion characteristics for LSCs which is otherwise inaccessible through $ab$ $initio$ computation.

Cover page: Bridging the Gap Between Simulated and Experimental Ionic Conductivities in Lithium Superionic Conductors

Article
Peer Reviewed

Performance and Cost Assessment of Machine Learning Interatomic Potentials

UC San Diego Previously Published Works (2020)

Machine learning of the quantitative relationship between local environment descriptors and the potential energy surface of a system of atoms has emerged as a new frontier in the development of interatomic potentials (IAPs). Here, we present a comprehensive evaluation of machine learning IAPs (ML-IAPs) based on four local environment descriptors-atom-centered symmetry functions (ACSF), smooth overlap of atomic positions (SOAP), the spectral neighbor analysis potential (SNAP) bispectrum components, and moment tensors-using a diverse data set generated using high-throughput density functional theory (DFT) calculations. The data set comprising bcc (Li, Mo) and fcc (Cu, Ni) metals and diamond group IV semiconductors (Si, Ge) is chosen to span a range of crystal structures and bonding. All descriptors studied show excellent performance in predicting energies and forces far surpassing that of classical IAPs, as well as predicting properties such as elastic constants and phonon dispersion curves. We observe a general trade-off between accuracy and the degrees of freedom of each model and, consequently, computational cost. We will discuss these trade-offs in the context of model selection for molecular dynamics and other applications.

Cover page: Performance and Cost Assessment of Machine Learning Interatomic Potentials

Article
Peer Reviewed

Atomistic simulations of dislocation mobility in refractory high-entropy alloys and the effect of chemical short-range order

UC Berkeley Previously Published Works (2021)

Refractory high-entropy alloys (RHEAs) are designed for high elevated-temperature strength, with both edge and screw dislocations playing an important role for plastic deformation. However, they can also display a significant energetic driving force for chemical short-range ordering (SRO). Here, we investigate mechanisms underlying the mobilities of screw and edge dislocations in the body-centered cubic MoNbTaW RHEA over a wide temperature range using extensive molecular dynamics simulations based on a highly-accurate machine-learning interatomic potential. Further, we specifically evaluate how these mechanisms are affected by the presence of SRO. The mobility of edge dislocations is found to be enhanced by the presence of SRO, whereas the rate of double-kink nucleation in the motion of screw dislocations is reduced, although this influence of SRO appears to be attenuated at increasing temperature. Independent of the presence of SRO, a cross-slip locking mechanism is observed for the motion of screws, which provides for extra strengthening for refractory high-entropy alloy system.

$Cover page: Atomistic simulations of dislocation mobility in refractory high-entropy alloys and the effect of chemical short-range order$

Article
Peer Reviewed

Atomic-scale origin of the low grain-boundary resistance in perovskite solid electrolyte Li_0.375Sr_0.4375Ta_0.75Zr_0.25O₃.

UC San Diego Previously Published Works (2023)

Oxide solid electrolytes (OSEs) have the potential to achieve improved safety and energy density for lithium-ion batteries, but their high grain-boundary (GB) resistance generally is a bottleneck. In the well-studied perovskite oxide solid electrolyte, Li_3xLa_2/3-xTiO₃ (LLTO), the ionic conductivity of grain boundaries is about three orders of magnitude lower than that of the bulk. In contrast, the related Li_0.375Sr_0.4375Ta_0.75Zr_0.25O₃ (LSTZ0.75) perovskite exhibits low grain boundary resistance for reasons yet unknown. Here, we use aberration-corrected scanning transmission electron microscopy and spectroscopy, along with an active learning moment tensor potential, to reveal the atomic scale structure and composition of LSTZ0.75 grain boundaries. Vibrational electron energy loss spectroscopy is applied for the first time to reveal atomically resolved vibrations at grain boundaries of LSTZ0.75 and to characterize the otherwise unmeasurable Li distribution therein. We find that Li depletion, which is a major reason for the low grain boundary ionic conductivity of LLTO, is absent for the grain boundaries of LSTZ0.75. Instead, the low grain boundary resistivity of LSTZ0.75 is attributed to the formation of a nanoscale defective cubic perovskite interfacial structure that contained abundant vacancies. Our study provides new insights into the atomic scale mechanisms of low grain boundary resistivity.