-
A cost-effective strategy of enhancing machine learning potentials by transfer learning from a multicomponent dataset on ænet-PyTorch
Authors:
An Niza El Aisnadaa,
Kajjana Boonpalit Robin van der Kruit,
Koen M. Draijer,
Jon Lopez-Zorrilla,
Masahiro Miyauchi,
Akira Yamaguchi,
Nongnuch Artrith
Abstract:
Machine learning potentials (MLPs) offer efficient and accurate material simulations, but constructing the reference ab initio database remains a significant challenge, particularly for catalyst-adsorbate systems. Training an MLP with a small dataset can lead to overfitting, thus limiting its practical applications. This study explores the feasibility of developing computationally cost-effective a…
▽ More
Machine learning potentials (MLPs) offer efficient and accurate material simulations, but constructing the reference ab initio database remains a significant challenge, particularly for catalyst-adsorbate systems. Training an MLP with a small dataset can lead to overfitting, thus limiting its practical applications. This study explores the feasibility of developing computationally cost-effective and accurate MLPs for catalyst-adsorbate systems with a limited number of ab initio references by leveraging a transfer learning strategy from subsets of a comprehensive public database. Using the Open Catalyst Project 2020 (OC20) -- a dataset closely related to our system of interest -- we pre-trained MLP models on OC20 subsets using the ænet-PyTorch framework. We compared several strategies for database subset selection. Our findings indicate that MLPs constructed via transfer learning exhibit better generalizability than those constructed from scratch, as demonstrated by the consistency in the dynamics simulations. Remarkably, transfer learning enhances the stability and accuracy of MLPs for the CuAu/H2O system with approximately 600 reference data points. This approach achieved excellent extrapolation performance in molecular dynamics (MD) simulations for the larger CuAu/6H2O system, sustaining up to 250 ps, whereas MLPs without transfer learning lasted less than 50 ps. We also examine the potential limitations of this strategy. This work proposes an alternative, cost-effective approach for constructing MLPs for the challenging simulation of catalytic systems. Finally, we anticipate that this methodology will pave the way for broader applications in material science and catalysis research, facilitating more efficient and accurate simulations across various systems.
△ Less
Submitted 23 August, 2024;
originally announced August 2024.
-
Machine Learning Potential Powered Insights into the Mechanical Stability of Amorphous Li-Si Alloys
Authors:
Zixiong Wei,
Nongnuch Artrith
Abstract:
Understanding the mechanical properties of solid-state materials at the atomic scale is crucial for developing novel materials. For example, amorphous LiSi alloys are attractive anode materials for solid-state Li-ion batteries but face mechanical instabilities due to significant volume variations with changing Li content. A fundamental grasp of the mechanical behavior in such systems is essential…
▽ More
Understanding the mechanical properties of solid-state materials at the atomic scale is crucial for developing novel materials. For example, amorphous LiSi alloys are attractive anode materials for solid-state Li-ion batteries but face mechanical instabilities due to significant volume variations with changing Li content. A fundamental grasp of the mechanical behavior in such systems is essential to address their poor mechanical integrity. Experimental methods offer insufficient information to elaborate on dynamic mechanical degradation mechanisms at the atomic scale, and computationally demanding first-principles methods, like DFT, struggle to access the system sizes needed for modeling mechanical phenomena. Machine learning potentials (MLPs) can overcome the computational constraints of traditional DFT-based simulations, enabling large-scale, accurate, and efficient simulations. Here, we provide a concise tutorial on developing and applying MLPs to investigate mechanical properties in materials systems, ranging from bulk to nanoparticles, using Li-Si alloys as an example. Trained on a comprehensive dataset (~45,000 DFT structures) with the aenet package accelerated by PyTorch, a robust MLP is constructed to reproduce results consistent with previous experimental observations. We demonstrate applying the MLP to realistic structures to visualize the deformation mechanism and determine the origin of mechanical instabilities caused by fracturing. This work aims to establish MLP-based simulations as a tool to understand the atomic-scale mechanical behavior in different materials systems.
△ Less
Submitted 13 February, 2024;
originally announced February 2024.
-
Overcoming the Size Limit of First Principles Molecular Dynamics Simulations with an In-Distribution Substructure Embedding Active Learner
Authors:
Lingyu Kong,
Jielan Li,
Lixin Sun,
Han Yang,
Hongxia Hao,
Chi Chen,
Nongnuch Artrith,
Jose Antonio Garrido Torres,
Ziheng Lu,
Yichi Zhou
Abstract:
Large-scale first principles molecular dynamics are crucial for simulating complex processes in chemical, biomedical, and materials sciences. However, the unfavorable time complexity with respect to system sizes leads to prohibitive computational costs when the simulation contains over a few hundred atoms in practice. We present an In-Distribution substructure Embedding Active Learner (IDEAL) to e…
▽ More
Large-scale first principles molecular dynamics are crucial for simulating complex processes in chemical, biomedical, and materials sciences. However, the unfavorable time complexity with respect to system sizes leads to prohibitive computational costs when the simulation contains over a few hundred atoms in practice. We present an In-Distribution substructure Embedding Active Learner (IDEAL) to enable efficient simulation of large complex systems with quantum accuracy by maintaining a machine learning force field (MLFF) as an accurate surrogate to the first principles methods. By extracting high-uncertainty substructures into low-uncertainty atom environments, the active learner is allowed to concentrate on and learn from small substructures of interest rather than carrying out intractable quantum chemical computations on large structures. IDEAL is benchmarked on various systems and shows sub-linear complexity, accelerating the simulation thousands of times compared with conventional active learning and millions of times compared with pure first principles simulations. To demonstrate the capability of IDEAL in practical applications, we simulated a polycrystalline lithium system composed of one million atoms and the full ammonia formation process in a Haber-Bosch reaction on a 3-nm Iridium nanoparticle catalyst on a computing node comprising one single A100 GPU and 24 CPU cores.
△ Less
Submitted 14 November, 2023; v1 submitted 14 November, 2023;
originally announced November 2023.
-
Constructing and Compressing Global Moment Descriptors from Local Atomic Environments
Authors:
Vahe Gharakhanyan,
Max Aalto,
Aminah Alsoulah,
Nongnuch Artrith,
Alexander Urban
Abstract:
Local atomic environment descriptors (LAEDs) are used in the materials science and chemistry communities, for example, for the development of machine learning interatomic potentials. Despite the fact that LAEDs have been extensively studied and benchmarked for various applications, global structure descriptors (GSDs), i.e., descriptors for entire molecules or crystal structures, have been mostly d…
▽ More
Local atomic environment descriptors (LAEDs) are used in the materials science and chemistry communities, for example, for the development of machine learning interatomic potentials. Despite the fact that LAEDs have been extensively studied and benchmarked for various applications, global structure descriptors (GSDs), i.e., descriptors for entire molecules or crystal structures, have been mostly developed independently based on other approaches. Here, we propose a systematically improvable methodology for constructing a space of representations of GSDs from LAEDs by incorporating statistical information and information about chemical elements. We apply the method to construct GSDs of varying complexity for lithium thiophosphate structures that are of interest as solid electrolytes and use an information-theoretic approach to obtain an optimally compressed GSD. Finally, we report the performance of the compressed GSD for energy prediction tasks.
△ Less
Submitted 8 October, 2023;
originally announced October 2023.
-
Atomic Insights into the Oxidative Degradation Mechanisms of Sulfide Solid Electrolytes
Authors:
Chuntian Cao,
Matthew R. Carbone,
Cem Komurcuoglu,
Jagriti S. Shekhawat,
Kerry Sun,
Haoyue Guo,
Sizhan Liu,
Ke Chen,
Seong-Min Bak,
Yonghua Du,
Conan Weiland,
Xiao Tong,
Dan Steingart,
Shinjae Yoo,
Nongnuch Artrith,
Alexander Urban,
Deyu Lu,
Feng Wang
Abstract:
Electrochemical degradation of solid electrolytes is a major roadblock in the development of solid-state batteries, and the formed solid-solid interphase (SSI) plays a key role in the performance of solid-state batteries. In this study, by combining experimental X-ray absorption spectroscopy (XAS) measurements, first-principles simulations, and unsupervised machine learning, we have unraveled the…
▽ More
Electrochemical degradation of solid electrolytes is a major roadblock in the development of solid-state batteries, and the formed solid-solid interphase (SSI) plays a key role in the performance of solid-state batteries. In this study, by combining experimental X-ray absorption spectroscopy (XAS) measurements, first-principles simulations, and unsupervised machine learning, we have unraveled the atomic-scale oxidative degradation mechanisms of sulfide electrolytes at the interface using the baseline Li3PS4 (LPS) electrolyte as a model system. The degradation begins with a decrease of Li neighbor affinity to S atoms upon initial delithiation, followed by the formation of S-S bonds as the PS4 tetrahedron deforms. After the first delithiation cycle, the PS4 motifs become strongly distorted and PS3 motifs start to form. Spectral fingerprints of the local structural evolution are identified, which correspond to the main peak broadening and the peak shifting to a higher energy by about 2.5 eV in P K-edge XAS and a new peak emerging at 2473 eV in S K-edge XAS during delithiation. The spectral fingerprints serve as a proxy for the electrochemical stability of phosphorus sulfide solid electrolytes beyond LPS, as demonstrated in argyrodite Li6PS5Cl. We observed that the strong distortion and destruction of PS4 tetrahedra and the formation of S-S bonds are correlated with an increased interfacial impedance. To the best of our knowledge, this study showcases the first atomic-scale insights into the oxidative degradation mechanism of the LPS electrolyte, which can provide guidance for controlling macroscopic reactions through microstructural engineering and, more generally, can advance the rational design of sulfide electrolytes.
△ Less
Submitted 1 October, 2023;
originally announced October 2023.
-
ænet-PyTorch: a GPU-supported implementation for machine learning atomic potentials training
Authors:
Jon Lopez-Zorrilla,
Xabier M. Aretxabaleta,
Inwon Yue,
Inigo Etxebarria,
Hegoi Manzano,
Nongnuch Artrith
Abstract:
In this work, we present ænet-PyTorch, a PyTorch-based implementation for training artificial neural network-based machine learning interatomic potentials. Developed as an extension of the atomic energy network (ænet), ænet-PyTorch provides access to all the tools included in ænet for the application and usage of the potentials. The package has been designed as an alternative to the internal train…
▽ More
In this work, we present ænet-PyTorch, a PyTorch-based implementation for training artificial neural network-based machine learning interatomic potentials. Developed as an extension of the atomic energy network (ænet), ænet-PyTorch provides access to all the tools included in ænet for the application and usage of the potentials. The package has been designed as an alternative to the internal training capabilities of ænet, leveraging the power of graphic processing units to facilitate direct training on forces in addition to energies. This leads to a substantial reduction of the training time by one to two orders of magnitude compared to the CPU implementation, enabling direct training on forces for systems beyond small molecules. Here we demonstrate the main features of ænet-PyTorch and show its performance on open databases. Our results show that training on all the force information within a data set is not necessary, and including between 10% to 20% of the force information is sufficient to achieve optimally accurate interatomic potentials with the least computational resources.
△ Less
Submitted 18 February, 2023;
originally announced February 2023.
-
Simulated sulfur K-edge X-ray absorption spectroscopy database of lithium thiophosphate solid electrolytes
Authors:
Haoyue Guo,
Matthew R. Carbone,
Chuntian Cao,
Jianzhou Qu,
Yonghua Du,
Seong-Min Bak,
Conan Weiland,
Feng Wang,
Shinjae Yoo,
Nongnuch Artrith,
Alexander Urban,
Deyu Lu
Abstract:
X-ray absorption spectroscopy (XAS) is a premier technique for materials characterization, providing key information about the local chemical environment of the absorber atom. In this work, we develop a database of sulfur K-edge XAS spectra of crystalline and amorphous lithium thiophosphate materials based on the atomic structures reported in Chem. Mater., 34, 6702 (2022). The XAS database is base…
▽ More
X-ray absorption spectroscopy (XAS) is a premier technique for materials characterization, providing key information about the local chemical environment of the absorber atom. In this work, we develop a database of sulfur K-edge XAS spectra of crystalline and amorphous lithium thiophosphate materials based on the atomic structures reported in Chem. Mater., 34, 6702 (2022). The XAS database is based on simulations using the excited electron and core-hole pseudopotential approach implemented in the Vienna Ab initio Simulation Package. Our database contains 2681 S K-edge XAS spectra for 66 crystalline and glassy structure models, making it the largest collection of first-principles computational XAS spectra for glass/ceramic lithium thiophosphates to date. This database can be used to correlate S spectral features with distinct S species based on their local coordination and short-range ordering in sulfide-based solid electrolytes. The data is openly distributed via the Materials Cloud, allowing researchers to access it for free and use it for further analysis, such as spectral fingerprinting, matching with experiments, and developing machine learning models.
△ Less
Submitted 31 January, 2023;
originally announced February 2023.
-
Ultrafast X-ray imaging of the light-induced phase transition in VO2
Authors:
Allan S. Johnson,
Daniel Pérez-Salinas,
Khalid M. Siddiqui,
Sungwon Kim,
Sungwook Choi,
Klara Volckaert,
Paulina E. Majchrzak,
Søren Ulstrup,
Naman Agarwal,
Kent Hallman,
Richard F. Haglund Jr.,
Christian M. Günther,
Bastian Pfau,
Stefan Eisebitt,
Dirk Backes,
Francesco Maccherozzi,
Ann Fitzpatrick,
Sarnjeet Dhesi,
Pierluigi Gargiani,
Manuel Valvidares,
Nongnuch Artrith,
Frank de Groot,
Hyeongi Choi,
Dogeun Jang,
Abhishek Katoch
, et al. (4 additional authors not shown)
Abstract:
Using light to control transient phases in quantum materials is an emerging route to engineer new properties and functionality, with both thermal and non-thermal phases observed out of equilibrium. Transient phases are expected to be heterogeneous, either through photo-generated domain growth or by generating topological defects, and this impacts the dynamics of the system. However, this nanoscale…
▽ More
Using light to control transient phases in quantum materials is an emerging route to engineer new properties and functionality, with both thermal and non-thermal phases observed out of equilibrium. Transient phases are expected to be heterogeneous, either through photo-generated domain growth or by generating topological defects, and this impacts the dynamics of the system. However, this nanoscale heterogeneity has not been directly observed. Here we use time- and spectrally resolved coherent X-ray imaging to track the prototypical light induced insulator-to-metal phase transition in vanadium dioxide on the nanoscale with femtosecond time resolution. We show that the early-time dynamics are independent of the initial spatial heterogeneity and observe a 200 fs switch to the metallic phase. A heterogeneous response emerges only after hundreds of picoseconds. Through spectroscopic imaging, we reveal that the transient metallic phase is a highly orthorhombically strained rutile metallic phase, an interpretation that is in contrast to those based on spatially averaged probes. Our results demonstrate the critical importance of spatially and spectrally resolved measurements for understanding and interpreting the transient phases of quantum materials.
△ Less
Submitted 15 January, 2023; v1 submitted 17 February, 2022;
originally announced February 2022.
-
AI-Aided Mapping of the Structure-Composition-Conductivity Relationships of Glass-Ceramic Lithium Thiophosphate Electrolytes
Authors:
Haoyue Guo,
Qian Wang,
Alexander Urban,
Nongnuch Artrith
Abstract:
Lithium thiophosphates (LPS) with the composition (Li$_2$S)$_x$(P$_2$S$_5$)$_{1-x}$ are among the most promising prospective electrolyte materials for solid-state batteries (SSBs), owing to their superionic conductivity at room temperature ($>10^{-3}$ S cm$^{-1}$), soft mechanical properties, and low grain boundary resistance. Several glass-ceramic (gc) LPS with different compositions and good Li…
▽ More
Lithium thiophosphates (LPS) with the composition (Li$_2$S)$_x$(P$_2$S$_5$)$_{1-x}$ are among the most promising prospective electrolyte materials for solid-state batteries (SSBs), owing to their superionic conductivity at room temperature ($>10^{-3}$ S cm$^{-1}$), soft mechanical properties, and low grain boundary resistance. Several glass-ceramic (gc) LPS with different compositions and good Li conductivity have been previously reported, but the relationship between composition, atomic structure, stability, and Li conductivity remains unclear due to the challenges in characterizing non-crystalline phases in experiments or simulations. Here, we mapped the LPS phase diagram by combining first principles and artificial intelligence (AI) methods, integrating density functional theory, artificial neural network potentials, genetic-algorithm sampling, and ab initio molecular dynamics simulations. By means of an unsupervised structure-similarity analysis, the glassy/ceramic phases were correlated with the local structural motifs in the known LPS crystal structures, showing that the energetically most favorable Li environment varies with the composition. Based on the discovered trends in the LPS phase diagram, we propose a candidate solid-state electrolyte composition, (Li$_{2}$S)$_{x}$(P$_{2}$S$_{5}$)$_{1-x}$ ($x\sim{}0.725$), that exhibits high ionic conductivity ($>10^{-2}$ S cm$^{-1}$) in our simulations, thereby demonstrating a general design strategy for amorphous or glassy/ceramic solid electrolytes with enhanced conductivity and stability.
△ Less
Submitted 26 January, 2022;
originally announced January 2022.
-
Understanding the onset of surface degradation in LiNiO2 cathodes
Authors:
Xinhao Li,
Qian Wang,
Haoyue Guo,
Nongnuch Artrith,
Alexander Urban
Abstract:
Nickel-based layered oxides offer an attractive platform for the development of energy-dense cobalt-free cathodes for lithium-ion batteries but suffer from degradation via oxygen gas release during electrochemical cycling. While such degradation has previously been characterized phenomenologically with experiments, an atomic-scale understanding of the reactions that take place at the cathode surfa…
▽ More
Nickel-based layered oxides offer an attractive platform for the development of energy-dense cobalt-free cathodes for lithium-ion batteries but suffer from degradation via oxygen gas release during electrochemical cycling. While such degradation has previously been characterized phenomenologically with experiments, an atomic-scale understanding of the reactions that take place at the cathode surface has been lacking. Here, we develop a first-principles methodology for the prediction of the surface reconstructions of intercalation electrode particles as a function of the temperature and state of charge. We report the surface phase diagrams of the LiNiO2(001) and (104) surfaces and identify surface structures that are likely visited during the first charge and discharge. Our calculations indicate that both surfaces experience oxygen loss during the first charge, resulting in irreversible changes to the surface structures. At the end of charge, the surface Ni atoms migrate into tetrahedral sites, from which they further migrate into Li vacancies during discharge, leading to Li/Ni mixed discharged surface phases. Further, the impact of the temperature and voltage range during cycling on the charge/discharge mechanism is discussed. The present study thus provides insight into the initial stages of cathode surface degradation and lies the foundation for the computational design of cathode materials that are stable against oxygen release.
△ Less
Submitted 27 April, 2022; v1 submitted 8 December, 2021;
originally announced December 2021.
-
AENET-LAMMPS and AENET-TINKER: Interfaces for Accurate and Efficient Molecular Dynamics Simulations with Machine Learning Potentials
Authors:
Michael S. Chen,
Tobias Morawietz,
Hideki Mori,
Thomas E. Markland,
Nongnuch Artrith
Abstract:
Machine learning potentials (MLPs) trained on data from quantum-mechanics based first-principles methods can approach the accuracy of the reference method at a fraction of the computational cost. To facilitate efficient MLP-based molecular dynamics (MD) and Monte Carlo (MC) simulations, an integration of the MLPs with sampling software is needed. Here we develop two interfaces that link the Atomic…
▽ More
Machine learning potentials (MLPs) trained on data from quantum-mechanics based first-principles methods can approach the accuracy of the reference method at a fraction of the computational cost. To facilitate efficient MLP-based molecular dynamics (MD) and Monte Carlo (MC) simulations, an integration of the MLPs with sampling software is needed. Here we develop two interfaces that link the Atomic Energy Network (ænet) MLP package with the popular sampling packages TINKER and LAMMPS. The three packages, ænet, TINKER, and LAMMPS, are free and open-source software that enable, in combination, accurate simulations of large and complex systems with low computational cost that scales linearly with the number of atoms. Scaling tests show that the parallel efficiency of the ænet-TINKER interface is nearly optimal but is limited to shared-memory systems. The ænet-LAMMPS interface achieves excellent parallel efficiency on highly parallel distributed memory systems and benefits from the highly optimized neighbor list implemented in LAMMPS. We demonstrate the utility of the two MLP interfaces for two relevant example applications, the investigation of diffusion phenomena in liquid water and the equilibration of nanostructured amorphous battery materials.
△ Less
Submitted 23 July, 2021;
originally announced July 2021.
-
Data-driven Approach to Parameterize SCAN+U for an Accurate Description of 3d Transition Metal Oxide Thermochemistry
Authors:
Nongnuch Artrith,
José Antonio Garrido Torres,
Alexander Urban,
Mark S. Hybertsen
Abstract:
Semi-local DFT methods exhibit significant errors for the phase diagrams of transition-metal oxides that are caused by an incorrect description of molecular oxygen and the large self-interaction error in materials with strongly localized electronic orbitals. Empirical and semiempirical corrections based on the DFT+U method can reduce these errors, but the parameterization and validation of the cor…
▽ More
Semi-local DFT methods exhibit significant errors for the phase diagrams of transition-metal oxides that are caused by an incorrect description of molecular oxygen and the large self-interaction error in materials with strongly localized electronic orbitals. Empirical and semiempirical corrections based on the DFT+U method can reduce these errors, but the parameterization and validation of the correction terms remains an on-going challenge. We develop a systematic methodology to determine the parameters and to statistically assess the results by considering thermochemical data across a set of transition metal compounds. We consider three interconnected levels of correction terms: (1) a constant oxygen binding correction, (2) Hubbard-U correction, and (3) DFT/DFT+U compatibility correction. The parameterization is expressed as a unified optimization problem. We demonstrate this approach for 3d transition metal oxides, considering a target set of binary and ternary oxides. With a total of 37 measured formation enthalpies taken from the literature, the dataset is augmented by the reaction energies of 1,710 unique reactions that were derived from the formation energies by systematic enumeration. To ensure a balanced dataset across the available data, the reactions were grouped by their similarity using clustering and suitably weighted. The parameterization is validated using leave-one-out cross-validation (CV), a standard technique for the validation of statistical models. We apply the methodology to the SCAN density functional. Based on the CV score, the error of binary (ternary) oxide formation energies is reduced by 40% (75%) to 0.10 (0.03) eV/atom. The method and tools demonstrated here can be applied to other classes of materials or to parameterize the corrections to optimize DFT+U performance for other target physical properties.
△ Less
Submitted 12 September, 2021; v1 submitted 1 February, 2021;
originally announced February 2021.
-
Strategies for the Construction of Machine-Learning Potentials for Accurate and Efficient Atomic-Scale Simulations
Authors:
April M. Miksch,
Tobias Morawietz,
Johannes Kästner,
Alexander Urban,
Nongnuch Artrith
Abstract:
Recent advances in machine-learning interatomic potentials have enabled the efficient modeling of complex atomistic systems with an accuracy that is comparable to that of conventional quantum mechanics based methods. At the same time, the construction of new machine-learning potentials can seem a daunting task, as it involves data-science techniques that are not yet common in chemistry and materia…
▽ More
Recent advances in machine-learning interatomic potentials have enabled the efficient modeling of complex atomistic systems with an accuracy that is comparable to that of conventional quantum mechanics based methods. At the same time, the construction of new machine-learning potentials can seem a daunting task, as it involves data-science techniques that are not yet common in chemistry and materials science. Here, we provide a tutorial-style overview of strategies and best practices for the construction of artificial neural network (ANN) potentials. We illustrate the most important aspects of (i) data collection, (ii) model selection, (iii) training and validation, and (iv) testing and refinement of ANN potentials on the basis of practical examples. Current research in the areas of active learning and delta learning are also discussed in the context of ANN potentials. This tutorial review aims at equipping computational chemists and materials scientists with the required background knowledge for ANN potential construction and application, with the intention to accelerate the adoption of the method, so that it can facilitate exciting research that would otherwise be challenging with conventional strategies.
△ Less
Submitted 5 May, 2021; v1 submitted 25 January, 2021;
originally announced January 2021.
-
Predicting the Activity and Selectivity of Bimetallic Metal Catalysts for Ethanol Reforming using Machine Learning
Authors:
Nongnuch Artrith,
Zhexi Lin,
Jingguang G. Chen
Abstract:
Machine learning is ideally suited for the pattern detection in large uniform datasets, but consistent experimental datasets on catalyst studies are often small. Here we demonstrate how a combination of machine learning and first-principles calculations can be used to extract knowledge from a relatively small set of experimental data. The approach is based on combining a complex machine-learning m…
▽ More
Machine learning is ideally suited for the pattern detection in large uniform datasets, but consistent experimental datasets on catalyst studies are often small. Here we demonstrate how a combination of machine learning and first-principles calculations can be used to extract knowledge from a relatively small set of experimental data. The approach is based on combining a complex machine-learning model trained on a computational library of transition-state energies with simple linear regression models of experimental catalytic activities and selectivities from the literature. Using the combined model, we identify the key C-C bond scission reactions involved in ethanol reforming and perform a computational screening for ethanol reforming on monolayer bimetallic catalysts with architectures TM-Pt-Pt(111) and Pt-TM-Pt(111) (TM = 3d transition metals). The model also predicts four promising catalyst compositions for future experimental studies. The approach is not limited to ethanol reforming but is of general use for the interpretation of experimental observations as well as for the computational discovery of catalytic materials.
△ Less
Submitted 3 August, 2020;
originally announced August 2020.
-
Efficient Training of ANN Potentials by Including Atomic Forces via Taylor Expansion and Application to Water and a Transition-Metal Oxide
Authors:
April M. Cooper,
Johannes Kästner,
Alexander Urban,
Nongnuch Artrith
Abstract:
Artificial neural network (ANN) potentials enable the efficient large-scale atomistic modeling of complex materials with near first-principles accuracy. For molecular dynamics simulations, accurate energies and interatomic forces are a prerequisite, but training ANN potentials simultaneously on energies and forces from electronic structure calculations is computationally demanding. Here, we introd…
▽ More
Artificial neural network (ANN) potentials enable the efficient large-scale atomistic modeling of complex materials with near first-principles accuracy. For molecular dynamics simulations, accurate energies and interatomic forces are a prerequisite, but training ANN potentials simultaneously on energies and forces from electronic structure calculations is computationally demanding. Here, we introduce an efficient alternative method for the training of ANN potentials on energy and force information based on an extrapolation of the total energy via a Taylor expansion. By translating the force information to approximate energies, the quadratic scaling with the number of atoms exhibited by conventional force-training methods can be avoided, which enables the training on reference data sets containing complex atomic structures. We demonstrate for different materials systems, clusters of water molecules, bulk liquid water, and a lithium transition-metal oxide, that the proposed force-training approach provides substantial improvements over schemes that train on energies only. Including force information for training reduces the size of the reference data sets required for ANN potential construction, increases the transferability of the potential, and generally improves the force prediction accuracy. For a set of water clusters, the Taylor-expansion approach achieves around 50% of the force error improvement compared to the explicit training on all force components, at a much smaller computational cost. The alternative force training approach thus simplifies the construction of general ANN potentials for the prediction of accurate energies and interatomic forces for diverse types of materials, as demonstrated here for water and a transition-metal oxide.
△ Less
Submitted 1 May, 2020; v1 submitted 10 February, 2020;
originally announced February 2020.
-
Atomic-scale factors that control the rate capability of nanostructured amorphous Si for high-energy-density batteries
Authors:
Nongnuch Artrith,
Alexander Urban,
Yan Wang,
Gerbrand Ceder
Abstract:
Nanostructured Si is the most promising high-capacity anode material to substantially increase the energy density of Li-ion batteries. Among the remaining challenges is its low rate capability as compared to conventional materials. To understand better what controls the diffusion of Li in the amorphous Li-Si alloy, we use a novel machine-learning potential trained on more than 40,000 ab-initio cal…
▽ More
Nanostructured Si is the most promising high-capacity anode material to substantially increase the energy density of Li-ion batteries. Among the remaining challenges is its low rate capability as compared to conventional materials. To understand better what controls the diffusion of Li in the amorphous Li-Si alloy, we use a novel machine-learning potential trained on more than 40,000 ab-initio calculations and nanosecond-scale molecular dynamics simulations, to visualize for the first time the delithiation of entire LiSi nanoparticles. Our results show that the Si host is not static but undergoes a dynamic rearrangement from isolated atoms, to chains, and clusters, with the Li diffusion strongly governed by this Si rearrangement. We find that the Li diffusivity is highest when Si segregates into clusters, so that Li diffusion proceeds via hopping between the Si clusters. The average size of Si clusters and the concentration range over which Si clustering occurs can thus function as design criteria for the development of rate-improved anodes based on modified Si.
△ Less
Submitted 26 January, 2019;
originally announced January 2019.
-
An L$_0$L$_1$-norm compressive sensing paradigm for the construction of sparse predictive lattice models using mixed integer quadratic programming
Authors:
Wenxuan Huang,
Alexander Urban,
Penghao Xiao,
Ziqin Rong,
Hena Das,
Tina Chen,
Nongnuch Artrith,
Alexandra Toumar,
Gerbrand Ceder
Abstract:
First-principles based lattice models allow the modeling of ab initio thermodynamics of crystalline mixtures for applications such as the construction of phase diagrams and the identification of ground state atomic orderings. The recent development of compressive sensing approaches for the construction of lattice models has further enabled the systematic construction of sparse physical models with…
▽ More
First-principles based lattice models allow the modeling of ab initio thermodynamics of crystalline mixtures for applications such as the construction of phase diagrams and the identification of ground state atomic orderings. The recent development of compressive sensing approaches for the construction of lattice models has further enabled the systematic construction of sparse physical models without the need for human intuition other than requiring the compactness of effective cluster interactions. However, conventional compressive sensing based on L1-norm regularization is strictly only applicable to certain classes of optimization problems and is otherwise not guaranteed to generate optimally sparse and transferable results, so that the method can only be applied to some materials science applications. In this paper, we illustrate a more robust L0L1-norm compressive-sensing method that removes the limitations of conventional compressive sensing and generally results in sparser lattice models that are at least as predictive as those obtained from L1-norm compressive sensing. Apart from the theory, a practical implementation based on state-of-the-art mixed-integer quadratic programming (MIQP) is proposed. The robustness of our methodology is illustrated for four different transition-metal oxides with relevance as battery cathode materials: Li2xTi2(1-x)O2, Li2xNi2yO2, MgxCr2O4, and NaxCrO2. This method provides a practical and robust approach for the construction of sparser and more predictive lattice models, improving on the compressive sensing paradigm and making it applicable to a much broader range of applications.
△ Less
Submitted 16 July, 2018;
originally announced July 2018.
-
Hidden structural order controls Li-ion transport in cation-disordered oxides for rechargeable lithium batteries
Authors:
Huiwen Ji,
Alexander Urban,
Daniil A. Kitchaev,
Deok-Hwang Kwon,
Nongnuch Artrith,
Colin Ophus,
Wenxuan Huang,
Zijian Cai,
Tan Shi,
Jae Chul Kim,
Gerbrand Ceder
Abstract:
Crystal structures play a vital role in determining materials properties. In Li-ion cathodes, the crystal structure defines the dimensionality and connectivity of interstitial sites, thus determining Li-ion diffusion kinetics. While a perfect crystal has infinite structural coherence, a class of recently discovered high-capacity cathodes, Li-excess cation-disordered rocksalts, falls on the other e…
▽ More
Crystal structures play a vital role in determining materials properties. In Li-ion cathodes, the crystal structure defines the dimensionality and connectivity of interstitial sites, thus determining Li-ion diffusion kinetics. While a perfect crystal has infinite structural coherence, a class of recently discovered high-capacity cathodes, Li-excess cation-disordered rocksalts, falls on the other end of the spectrum: Their cation sublattices are assumed to be randomly populated by Li and transition metal ions with zero configurational coherence based on conventional X-ray diffraction, such that the Li transport is purely determined by statistical effects. In contrast to this prevailing view, we reveal that cation short-range order, hidden in diffraction, is ubiquitous in these long-range disordered materials and controls the local and macroscopic environments for Li-ion transport. Our work not only discovers a crucial property that has previously been overlooked, but also provides new guidelines for designing and engineering disordered rocksalts cathode materials.
△ Less
Submitted 15 June, 2018;
originally announced June 2018.
-
Constructing first-principles phase diagrams of amorphous LixSi using machine-learning-assisted sampling with an evolutionary algorithm
Authors:
Nongnuch Artrith,
Alexander Urban,
Gerbrand Ceder
Abstract:
The atomistic modeling of amorphous materials requires structure sizes and sampling statistics that are challenging to achieve with first-principles methods. Here, we propose a methodology to speed up the sampling of amorphous and disordered materials using a combination of a genetic algorithm and a specialized machine-learning potential based on artificial neural networks (ANN). We show for the e…
▽ More
The atomistic modeling of amorphous materials requires structure sizes and sampling statistics that are challenging to achieve with first-principles methods. Here, we propose a methodology to speed up the sampling of amorphous and disordered materials using a combination of a genetic algorithm and a specialized machine-learning potential based on artificial neural networks (ANN). We show for the example of the amorphous LiSi alloy that around 1,000 first-principles calculations are sufficient for the ANN potential assisted sampling of low-energy atomic configurations in the entire amorphous LixSi phase space. The obtained phase diagram is validated by comparison with the results from an extensive sampling of LixSi configurations using molecular dynamics simulations and a general ANN potential trained to ~45,000 first-principles calculations. This demonstrates the utility of the approach for the first-principles modeling of amorphous materials.
△ Less
Submitted 10 February, 2018;
originally announced February 2018.
-
The electronic-structure origin of cation disorder in transition-metal oxides
Authors:
Alexander Urban,
Aziz Abdellahi,
Stephen Dacek,
Nongnuch Artrith,
Gerbrand Ceder
Abstract:
Cation disorder is an important design criterion for technologically relevant transition-metal (TM) oxides, such as radiation-tolerant ceramics and Li-ion battery electrodes. In this letter, we use a combination of first-principles calculations, normal mode analysis, and band-structure arguments to pinpoint a specific electronic-structure effect that influences the stability of disordered phases.…
▽ More
Cation disorder is an important design criterion for technologically relevant transition-metal (TM) oxides, such as radiation-tolerant ceramics and Li-ion battery electrodes. In this letter, we use a combination of first-principles calculations, normal mode analysis, and band-structure arguments to pinpoint a specific electronic-structure effect that influences the stability of disordered phases. We find that the electronic configuration of a TM ion determines to which extent the structural energy is affected by site distortions. This mechanism explains the stability of disordered phases with large ionic radius differences and provides a concrete guideline for the discovery of novel disordered compositions.
△ Less
Submitted 13 October, 2017;
originally announced October 2017.
-
Efficient and Accurate Machine-Learning Interpolation of Atomic Energies in Compositions with Many Species
Authors:
Nongnuch Artrith,
Alexander Urban,
Gerbrand Ceder
Abstract:
Machine-learning potentials (MLPs) for atomistic simulations are a promising alternative to conventional classical potentials. Current approaches rely on descriptors of the local atomic environment with dimensions that increase quadratically with the number of chemical species. In this article, we demonstrate that such a scaling can be avoided in practice. We show that a mathematically simple and…
▽ More
Machine-learning potentials (MLPs) for atomistic simulations are a promising alternative to conventional classical potentials. Current approaches rely on descriptors of the local atomic environment with dimensions that increase quadratically with the number of chemical species. In this article, we demonstrate that such a scaling can be avoided in practice. We show that a mathematically simple and computationally efficient descriptor with constant complexity is sufficient to represent transition-metal oxide compositions and biomolecules containing 11 chemical species with a precision of around 3 meV/atom. This insight removes a perceived bound on the utility of MLPs and paves the way to investigate the physics of previously inaccessible materials with more than ten chemical species.
△ Less
Submitted 1 December, 2017; v1 submitted 20 June, 2017;
originally announced June 2017.