-
Efficient Crystal Structure Prediction Using Genetic Algorithm and Universal Neural Network Potential
Authors:
Takuya Shibayama,
Hideaki Imamura,
Katsuhiko Nishimra,
Kohei Shinohara,
Chikashi Shinagawa,
So Takamoto,
Ju Li
Abstract:
Crystal structure prediction (CSP) is crucial for identifying stable crystal structures in given systems and is a prerequisite for computational atomistic simulations. Recent advances in neural network potentials (NNPs) have reduced the computational cost of CSP. However, searching for stable crystal structures across the entire composition space in multicomponent systems remains a significant cha…
▽ More
Crystal structure prediction (CSP) is crucial for identifying stable crystal structures in given systems and is a prerequisite for computational atomistic simulations. Recent advances in neural network potentials (NNPs) have reduced the computational cost of CSP. However, searching for stable crystal structures across the entire composition space in multicomponent systems remains a significant challenge. Here, we propose a novel genetic algorithm (GA) -based CSP method using a universal NNP. Our GA-based methods are designed to efficiently expand convex hull volumes while preserving the diversity of crystal structures. This approach draws inspiration from the similarity between convex hull updates and Pareto front evolution in multi-objective optimization. Our evaluation shows that the present method outperforms the symmetry-aware random structure generation, achieving a larger convex hull with fewer trials. We demonstrated that our approach, combined with the developed universal NNP (PFP), can accurately reproduce and explore phase diagrams obtained through DFT calculations; this indicates the validity of PFP across a wide range of crystal structures and element combinations. This study, which integrates a universal NNP with a GA-based CSP method, highlights the promise of these methods in materials discovery.
△ Less
Submitted 27 March, 2025;
originally announced March 2025.
-
A practical guide to machine learning interatomic potentials -- Status and future
Authors:
Ryan Jacobs,
Dane Morgan,
Siamak Attarian,
Jun Meng,
Chen Shen,
Zhenghao Wu,
Clare Yijia Xie,
Julia H. Yang,
Nongnuch Artrith,
Ben Blaiszik,
Gerbrand Ceder,
Kamal Choudhary,
Gabor Csanyi,
Ekin Dogus Cubuk,
Bowen Deng,
Ralf Drautz,
Xiang Fu,
Jonathan Godwin,
Vasant Honavar,
Olexandr Isayev,
Anders Johansson,
Boris Kozinsky,
Stefano Martiniani,
Shyue Ping Ong,
Igor Poltavsky
, et al. (5 additional authors not shown)
Abstract:
The rapid development and large body of literature on machine learning interatomic potentials (MLIPs) can make it difficult to know how to proceed for researchers who are not experts but wish to use these tools. The spirit of this review is to help such researchers by serving as a practical, accessible guide to the state-of-the-art in MLIPs. This review paper covers a broad range of topics related…
▽ More
The rapid development and large body of literature on machine learning interatomic potentials (MLIPs) can make it difficult to know how to proceed for researchers who are not experts but wish to use these tools. The spirit of this review is to help such researchers by serving as a practical, accessible guide to the state-of-the-art in MLIPs. This review paper covers a broad range of topics related to MLIPs, including (i) central aspects of how and why MLIPs are enablers of many exciting advancements in molecular modeling, (ii) the main underpinnings of different types of MLIPs, including their basic structure and formalism, (iii) the potentially transformative impact of universal MLIPs for both organic and inorganic systems, including an overview of the most recent advances, capabilities, downsides, and potential applications of this nascent class of MLIPs, (iv) a practical guide for estimating and understanding the execution speed of MLIPs, including guidance for users based on hardware availability, type of MLIP used, and prospective simulation size and time, (v) a manual for what MLIP a user should choose for a given application by considering hardware resources, speed requirements, energy and force accuracy requirements, as well as guidance for choosing pre-trained potentials or fitting a new potential from scratch, (vi) discussion around MLIP infrastructure, including sources of training data, pre-trained potentials, and hardware resources for training, (vii) summary of some key limitations of present MLIPs and current approaches to mitigate such limitations, including methods of including long-range interactions, handling magnetic systems, and treatment of excited states, and finally (viii) we finish with some more speculative thoughts on what the future holds for the development and application of MLIPs over the next 3-10+ years.
△ Less
Submitted 12 March, 2025;
originally announced March 2025.
-
Calculations of Real-System Nanoparticles Using Universal Neural Network Potential PFP
Authors:
Gerardo Valadez Huerta,
Yusuke Nanba,
Iori Kurata,
Kosuke Nakago,
So Takamoto,
Chikashi Shinagawa,
Michihisa Koyama
Abstract:
It is essential to explore the stability and activity of real-system nanoparticles theoretically. While applications of theoretical methods for this purpose can be found in literature, the expensive computational costs of conventional theoretical methods hinder their massive applications to practical materials design. With the recent development of neural network algorithms along with the advancem…
▽ More
It is essential to explore the stability and activity of real-system nanoparticles theoretically. While applications of theoretical methods for this purpose can be found in literature, the expensive computational costs of conventional theoretical methods hinder their massive applications to practical materials design. With the recent development of neural network algorithms along with the advancement of computer systems, neural network potentials have emerged as a promising candidate for the description of a wide range of materials, including metals and molecules, with a reasonable computational time. In this study, we successfully validate a universal neural network potential, PFP, for the description of monometallic Ru nanoparticles, PdRuCu ternary alloy nanoparticles, and the NO adsorption on Rh nanoparticles against first-principles calculations. We further conduct molecular dynamics simulations on the NO-Rh system and challenge the PFP to describe a large, supported Pt nanoparticle system.
△ Less
Submitted 2 July, 2021;
originally announced July 2021.
-
Towards Universal Neural Network Potential for Material Discovery Applicable to Arbitrary Combination of 45 Elements
Authors:
So Takamoto,
Chikashi Shinagawa,
Daisuke Motoki,
Kosuke Nakago,
Wenwen Li,
Iori Kurata,
Taku Watanabe,
Yoshihiro Yayama,
Hiroki Iriguchi,
Yusuke Asano,
Tasuku Onodera,
Takafumi Ishii,
Takao Kudo,
Hideki Ono,
Ryohto Sawada,
Ryuichiro Ishitani,
Marc Ong,
Taiki Yamaguchi,
Toshiki Kataoka,
Akihide Hayashi,
Nontawat Charoenphakdee,
Takeshi Ibuka
Abstract:
Computational material discovery is under intense study owing to its ability to explore the vast space of chemical systems. Neural network potentials (NNPs) have been shown to be particularly effective in conducting atomistic simulations for such purposes. However, existing NNPs are generally designed for narrow target materials, making them unsuitable for broader applications in material discover…
▽ More
Computational material discovery is under intense study owing to its ability to explore the vast space of chemical systems. Neural network potentials (NNPs) have been shown to be particularly effective in conducting atomistic simulations for such purposes. However, existing NNPs are generally designed for narrow target materials, making them unsuitable for broader applications in material discovery. To overcome this issue, we have developed a universal NNP called PreFerred Potential (PFP), which is able to handle any combination of 45 elements. Particular emphasis is placed on the datasets, which include a diverse set of virtual structures used to attain the universality. We demonstrated the applicability of PFP in selected domains: lithium diffusion in LiFeSO${}_4$F, molecular adsorption in metal-organic frameworks, an order-disorder transition of Cu-Au alloys, and material discovery for a Fischer-Tropsch catalyst. They showcase the power of PFP, and this technology provides a highly useful tool for material discovery.
△ Less
Submitted 1 April, 2022; v1 submitted 28 June, 2021;
originally announced June 2021.
-
TeaNet: universal neural network interatomic potential inspired by iterative electronic relaxations
Authors:
So Takamoto,
Satoshi Izumi,
Ju Li
Abstract:
A universal interatomic potential for an arbitrary set of chemical elements is urgently needed in computational materials science. Graph convolution neural network (GCN) has rich expressive power, but previously was mainly employed to transport scalars and vectors, not rank $\ge 2$ tensors. As classic interatomic potentials were inspired by tight-binding electronic relaxation framework, we want to…
▽ More
A universal interatomic potential for an arbitrary set of chemical elements is urgently needed in computational materials science. Graph convolution neural network (GCN) has rich expressive power, but previously was mainly employed to transport scalars and vectors, not rank $\ge 2$ tensors. As classic interatomic potentials were inspired by tight-binding electronic relaxation framework, we want to represent this iterative propagation of rank $\ge 2$ tensor information by GCN. Here we propose an architecture called the tensor embedded atom network (TeaNet) where angular interaction is translated into graph convolution through the incorporation of Euclidean tensors, vectors and scalars. By applying the residual network (ResNet) architecture and training with recurrent GCN weights initialization, a much deeper (16 layers) GCN was constructed, whose flow is similar to an iterative electronic relaxation. Our traning dataset is generated by density functional theory calculation of mostly chemically and structurally randomized configurations. We demonstrate that arbitrary structures and reactions involving the first 18 elements on the periodic table (H to Ar) can be realized satisfactorily by TeaNet, including C-H molecular structures, metals, amorphous SiO${}_2$, and water, showing surprisingly good performance (energy mean absolute error 19 meV/atom) and robustness for arbitrary chemistries involving elements from H to Ar.
△ Less
Submitted 10 October, 2021; v1 submitted 2 December, 2019;
originally announced December 2019.
-
Elucidation of the atomic-scale mechanism of the anisotropic oxidation rate of 4H-SiC between the ($0001$) Si-face and ($000\overline{1}$) C-face by using a new Si-O-C interatomic potential
Authors:
So Takamoto,
Takahiro Yamasaki,
Takahisa Ohno,
Chioko Kaneta,
Asuka Hatano,
Satoshi Izumi
Abstract:
Silicon carbide (SiC) is an attractive semiconductor material for applications in power electronic devices. However, fabrication of a high-quality SiC/SiO${}_2$ interface has been a challenge. It is well-known that there is a great difference in oxidation rate between the Si-face and C-face, and that the quality of oxide on the Si-face is greater than that on the C-face. However, the atomistic mec…
▽ More
Silicon carbide (SiC) is an attractive semiconductor material for applications in power electronic devices. However, fabrication of a high-quality SiC/SiO${}_2$ interface has been a challenge. It is well-known that there is a great difference in oxidation rate between the Si-face and C-face, and that the quality of oxide on the Si-face is greater than that on the C-face. However, the atomistic mechanism of the thermal oxidation of SiC remains to be solved. In this paper, a new Si-C-O interatomic potential was developed to reproduce the kinetics of the thermal oxidation of SiC. Using this newly developed potential, large-scale SiC oxidation simulations at various temperature were performed. The results showed that the activation energy of the Si-face is much larger than that of the C-face. In the case of the Si-face, a flat and aligned interface structure including Si${}^{1+}$ was created. Based on the estimated activation energies of the intermediate oxide states, it is proposed that the stability of the flat interface structure is the origin of the high activation energy of the oxidation of the Si-face. In contrast, in the case of the C-face, it is found that the Si atom at the interface are easily pulled up by the O atoms. This process generates the disordered interface and decreases the activation energy of the oxidation. It is also proposed that many excess C atoms are created in the case of the C-face.
△ Less
Submitted 24 April, 2018;
originally announced April 2018.
-
Atomistic mechanism of graphene growth on SiC substrate: Large-scale molecular dynamics simulation based on a new charge-transfer bond-order type potential
Authors:
So Takamoto,
Takahiro Yamasaki,
Jun Nara,
Takahisa Ohno,
Chioko Kaneta,
Asuka Hatano,
Satoshi Izumi
Abstract:
Thermal decomposition of silicon carbide is a promising approach for the fabrication of graphene. However, the atomistic growth mechanism of graphene remains unclear. This paper describes the development of a new charge-transfer interatomic potential. Carbon bonds with a wide variety of characteristics can be reproduced by the proposed vectorized bond-order term. Large-scale thermal decomposition…
▽ More
Thermal decomposition of silicon carbide is a promising approach for the fabrication of graphene. However, the atomistic growth mechanism of graphene remains unclear. This paper describes the development of a new charge-transfer interatomic potential. Carbon bonds with a wide variety of characteristics can be reproduced by the proposed vectorized bond-order term. Large-scale thermal decomposition simulation enables us to observe the continuous growth process of the multi-ring carbon structure. The annealing simulation reveals the atomistic process by which the multi-ring carbon structure is transformed to flat graphene involving only 6-membered rings. Also, it is found that the surface atoms of the silicon carbide substrate enhance the homogeneous graphene formation.
△ Less
Submitted 21 February, 2018;
originally announced February 2018.
