首 页
滚动信息 更多 >>
npj Computational Materials获得第一个SCI影响因... (2018-09-07)
英文刊《npj Computational Materials(计算材料学... (2017-05-15)
开放的数据库 (2015-12-17)
上海硅酸盐所与Nature出版集团联合创办的中国首本n... (2015-11-27)
Nature Publishing Group and Shanghai Institute ... (2015-09-24)
快捷服务
最新文章 研究综述
过刊浏览 作者须知
期刊编辑 审稿须知
相关链接
· 在线投稿
会议信息
友情链接
  中国科学院上海硅酸盐研究所
  无机材料学报
  OQMD数据库
mm.jpg
期刊介绍
  《npj 计算材料学》是在线出版、完全开放获取的国际学术期刊。发表结合计算模拟与设计的材料学一流的研究成果。本刊由中国科学院上海硅酸盐研究所与英国自然出版集团(Nature Publishing Group,NPG)以伙伴关系合作出版。
  主编为陈龙庆博士,美国宾州大学材料科学与工程系、工程科学与力学系、数学系的杰出教授。
  共同主编为陈立东研究员,中国科学院上海硅酸盐研究所研究员高性能陶瓷与超微结构国家重点实验室主任。
  办刊目的与报道范围
  《npj 计算材料学》是在线出版、完全开放获取的国际...
【查看详细】
近期文章 更多 >>

Theoretical analysis of spectral lineshapes from molecular dynamics (谱线形的分子动力学理论分析)
Andrew CupoDamien TristantKyle Rego & Vincent Meunier
npj Computational Materials 5:82(2019)
doi:s41524-019-0220-1
Published online:07 August 2019

Abstract| Full Text | PDF OPEN

摘要:用于计算非谐声子特性的传统方法,计算起来十分昂贵。为解决这个问题,本研究开发了一种理论方法,用于加速从有限时间分子动力学获得的波普振动线形的计算。该方法提供了研究非谐诱导的频移和寿命影响的方法,及其模拟扩展。我们的研究证明,采用简约模型(Toy model)获得收敛的振动特性所需的模拟步骤数量,与标准提取程序相比,几乎在所有情况下都减少了至少一个数量级。对于石墨烯、六方氮化硼和硅,本研究还从理论上证明了,在密度泛函理论水平上,在振动频率和延时中,达到收敛所需的模拟时间,均减少了近9倍。一般来说,当非谐性足够弱时,我们期望新开发的方法优于标准程序,从而得出明确定义的重正化声子准粒子。我们将信号分析扩展到材料振动,代表了计算与温度相关的声子特性的最新研究进展,并且可以在计算材料发现工具包中实现诸如热电材料的搜索,因为热导率对ZT的贡献强烈依赖于这些特征   

Abstract:Conventional methods for calculating anharmonic phonon properties are computationally expensive. To address this issue, a theoretical approach was developed for the accelerated calculation of vibrational lineshapes for spectra obtained from finite-time molecular dynamics. The method gives access to the effect of anharmonicity-induced frequency shift and lifetime, as well as simulation broadening. For a toy model we demonstrate at least an order of magnitude reduction in the number of simulation steps needed to obtain converged vibrational properties in nearly all cases considered as compared to the standard extraction procedure. The theory is also illustrated for graphene, hexagonal boron nitride, and silicon at the density functional theory level, with up to nearly a factor of 9 reduction in the required simulation time to reach convergence in the vibrational frequencies and lifetimes. In general, we expect the newly developed method to outperform the standard procedure when the anharmonicity is sufficiently weak so that well-defined renormalized phonon quasiparticles emerge. Our extension of signal analysis to material vibrations represents a state-of-the-art advance in calculating temperature-dependent phonon properties and could be implemented in computational materials discovery packages that search for thermoelectric materials for instance, since the thermal conductivity contribution to ZT depends strongly on these characteristics. 

Editorial Summary

Phononic anharmonicity: captured by molecular dynamics声子非谐性:独立特行但却被分子动力学拿下

从分子动力学计算的波普推导出了一种新方法,用来研究声子振动线形,包括非谐性引起的频移效应、寿命效应,以及模拟展宽。来自美国伦斯勒理工学院物理、应用物理和天文学系Vincent Meunier教授领导的团队,由分子动力学推导获得了波普精确解析表达式,以及速度的简单扰动正态模式的表达式。与标准提取程序相比,我们证明了简约模型(Toy model)在几乎所有情况下,所获得收敛振动特性的模拟步骤数量,至少降低了一个数量级。在50 K时,使用这两种方法可以收敛寿命,新方法将所需的模拟时间缩短了大约1.4倍。他们将推导获得的拟合函数应用于简单模型,石墨烯、六方氮化硼(hBN)和硅,以研究振动频率和寿命与模拟时间的收敛性。他们所提出的方法,在强相关系统到生物材料的各领域,都将对分子动力学声子特性的确定产生深远影响

A new approach to understand phononic vibrational lineshapes is derived for spectra calculated from molecular dynamics and include the effect of anharmonicity-induced frequency shift and lifetime, as well as the simulation broadening. A team led by Vincent Meunier from the Department of Physics, Applied Physics, and Astronomy, Rensselaer Polytechnic Institute, USA, derived exact analytical expressions for spectra obtained from molecular dynamics starting with a simple perturbed normal mode expression for the velocities. They applied the derived fitting functions to simple models, graphene, hexagonal boron nitride (hBN), and silicon to study the convergence of the vibrational frequency and lifetime with simulation time. For a toy model they demonstrated at least an order of magnitude reduction in the number of simulation steps to obtain converged vibrational properties in nearly all cases considered as compared to the standard extraction procedure. At 50K the lifetime can be converged using both methods, with the new method reducing the required simulation time by a factor of about 1.4. Their proposed method could have far reaching impact on the determination of phononic properties from MD, in areas ranging from strongly correlated systems to biological materials.

Recent advances and applications of machine learning in solid-state materials science机器学习在固体材料科学中的新进展和新应用
Jonathan SchmidtMário R. G. MarquesSilvana Botti & Miguel A. L. Marques
npj Computational Materials 5:83(2019)
doi:s41524-019-0221-0
Published online:08 August 2019

Abstract| Full Text | PDF OPEN

摘要:近年来材料科学中最激动人心的工具手段之一便是机器学习。事实证明,这种基于统计学的研究方法能够大大加快基础研究和应用研究的速度。目前,我们目睹了大量的最新研究进展,这些进展将机器学习开发、应用于固态系统。就机器学习在固体材料研究和应用中的作用这一主题,我们提供了新近研究的全面概述和分析。作为起点,我们介绍了应用于材料科学的机器学习原理、算法、描述符和数据库。接着,我们描述了用以发现稳定材料和预测其晶体结构的各种不同的机器学习方法,继而讨论了许多定量结构-属性关系的研究,以及通过机器学习取代第一原理计算的各种方法。我们综述了如何应用主动学习和如何基于代理优化,来改进理性设计过程和相关应用的实例。本文始终以1)机器学习模型的可解释性和2)机器学习模型的物理认识,作为两个主题。因此,我们关注了可解释性的不同方面,及其在材料科学中的重要作用。最后,我们为计算材料科学中的各种挑战提出了解决方案和未来研究途径   

Abstract:One of the most exciting tools that have entered the material science toolbox in recent years is machine learning. This collection of statistical methods has already proved to be capable of considerably speeding up both fundamental and applied research. At present, we are witnessing an explosion of works that develop and apply machine learning to solid-state systems. We provide a comprehensive overview and analysis of the most recent research in this topic. As a starting point, we introduce machine learning principles, algorithms, descriptors, and databases in materials science. We continue with the description of different machine learning approaches for the discovery of stable materials and the prediction of their crystal structure. Then we discuss research in numerous quantitative structure–property relationships and various approaches for the replacement of first-principle methods by machine learning. We review how active learning and surrogate-based optimization can be applied to improve the rational design process and related examples of applications. Two major questions are always the interpretability of and the physical understanding gained from machine learning models. We consider therefore the different facets of interpretability and their importance in materials science. Finally, we propose solutions and future research paths for various challenges in computational materials science. 

Editorial Summary

Review on the whole machine learning: becoming superman, human, application, and ideal机器学习:是神、是人、是应用、是理想

该文综述了机器学习在材料科学领域的最新应用。来自德国马丁路德大学物理研究所的Miguel A. L. Marques教授专注于详细讨论和分析固态材料科学(特别是最新的固态材料科学)机器学习的各种应用。由于机器学习算法在几个不同的科学和技术领域中取得了无与伦比的成功(神一般的成功),这些应用在过去几年中一直在蓬勃发展。该综述首先介绍了机器学习,特别是材料科学中的机器学习原理、算法、描述符和数据库(人的理论贡献)。然后,介绍了固态材料科学中机器学习的众多应用(应用是目的和推动力):新稳定材料的发现及其结构的预测、材料特性的机器学习计算、材料科学模拟的机器学习力场的发展、通过机器学习方法构建DFT功能、通过主动学习优化自适应设计过程,以及机器学习模型的可解释性和物理认识。最后,讨论了机器学习在材料科学中面临的挑战和局限,并提出了一些克服或规避它们的研究策略。作者坚信,这一系列高效的统计工具确实能够大大加快基础研究和应用研究的速度(理想)。 因此,它们显然不仅仅是一种短暂作用于材料科学的方式,而肯定一直是未来几年塑造材料科学的力量

The latest applications of machine learning in the field of materials science is reviewed. A team led by Miguel A. L. Marques from the Institut für Physik, Martin-Luther-Universitat, Germany, concentrated on the various applications of machine learning in solid-state materials science (especially the most recent ones) and discussed and analyzed them in detail. These applications have been mushrooming in the past couple of years, fueled by the unparalleled success that machine learning algorithms have found in several different fields of science and technology. As a starting point, they provide an introduction to machine learning, and in particular to machine learning principles, algorithms, descriptors, and databases in materials science. They then review numerous applications of machine learning in solid-state materials science: the discovery of new stable materials and the prediction of their structure, the machine learning calculation of material properties, the development of machine learning force fields for simulations in material science, the construction of DFT functionals by machine learning methods, the optimization of the adaptive design process by active learning, and the interpretability of, and the physical understanding gained from, machine learning models. Finally, they discuss the challenges and limitations machine learning faces in materials science and suggest a few research strategies to overcome or circumvent them. It is authors’ firm conviction that this collection of efficient statistical tools are indeed capable of speeding up considerably both fundamental and applied research. As such, they are clearly more than a temporary fashion and will certainly shape materials science for the years to come.

Machine learning enables polymer cloud-point engineering via inverse design (机器学习通过逆向设计实现聚合物浊点工程)
Jatin N. KumarQianxiao LiKaren Y. T. TangTonio BuonassisiAnibal L. Gonzalez-Oyarce & Jun Ye
npj Computational Materials 5:73(2019)
doi:s41524-019-0209-9
Published online:12 July 2019

Abstract| Full Text | PDF OPEN

摘要:诸如聚合物多种尺度无序系统的逆向设计是个重大挑战,在设计具有所需相行为的聚合物时,挑战尤为重大。本研究通过机器学习证实了聚(2-恶唑啉)浊点的高精度调整。我们在四个重复单元和一系列分子质量的设计空间中,采用梯度增强决策树于24~90范围内,实现了4均方根误差(RMSE)的精度。RMSE比线性和多项式回归好3倍。通过粒子群优化进行逆向设计,我们在37~804个目标浊点下预测和合成了17种具有约束设计的聚合物。本方法通过机器学习算法优于现有的聚合物设计方法,相应的算法能够快速、系统地发现新聚合物   

Abstract:Inverse design is an outstanding challenge in disordered systems with multiple length scales such as polymers, particularly when designing polymers with desired phase behavior. Here we demonstrate high-accuracy tuning of poly(2-oxazoline) cloud point via machine learning. With a design space of four repeating units and a range of molecular masses, we achieve an accuracy of 4°C root mean squared error (RMSE) in a temperature range of 24–90°C, employing gradient boosting with decision trees. The RMSE is >3x better than linear and polynomial regression. We perform inverse design via particle-swarm optimization, predicting and synthesizing 17 polymers with constrained design at 4 target cloud points from 37 to 80°C. Our approach challenges the status quo in polymer design with a machine learning algorithm that is capable of fast and systematic discovery of new polymers. 

Editorial Summary

Machine learning: polymer inverse design机器学习:逆向设计聚合物

该研究报道了通过合理应用机器学习方法,在聚合物设计方面实现了概念上的重大进步。由新加坡材料研究与工程研究所的Jatin N. Kumar教授领导的团队,通过三个重要步骤实现了聚合物的设计:首先,策划并分类历史数据和新数据;其次,选择并微调基于梯度增强回归和决策树的机器学习模型,从而得到3.9的浊点预测精度RMSE),该模型能够很好地规范化明确定义的历史数据集,以及新合成的分子量呈不对称分布的聚合物;第三步,经粒子群优化的聚合物逆向设计,该步骤基于所训练数据(37℃、45℃、60℃、80℃)的浊点范围扩展的期望浊点,对新聚合物的设计进行预测。作者讨论了逆向设计方法如何扩展用于多个目标函数,演示了如何通过神经网络集合作为交叉验证技术,以便推断超出训练集之外的数据,从而筛选出17种具有最低预测方差的聚合物。预测聚合物的RMSE与正向模型的RMSE相似。该方法提供了前所未有的对聚合物主动设计的实例,有望显著加速依据浊点等多种目标性能的聚合物设计

A significant conceptual advance in polymer design via judicious application of machine-learning methods is reported. A team led by Prof. Jatin N. Kumar from the Institute of Materials Research & Engineering, Singapore, achieved polymer design in three important steps: firstly, curated and categorized historical and new data; secondly, selected and fine-tuned a machine-learning model based on gradient boosting regression with decision trees, resulting in a cloud point predictive accuracy of 3.9°C (RMSE), which was able to generalize well with both well-defined historic data sets as well as newly synthesized polymers of unsymmetrical molecular weight distributions; and thirdly, polymer inverse design by particle-swarm optimization which predicted the design of new polymers based on desired cloud points spread over the range of the cloud points of the training data (37, 45, 60, 80°C). They discussed how the inverse-design methodology is scalable to more than one objective function. It could extrapolate beyond the training set via an ensemble of neural networks as a cross-validation technique to downselect 17 polymers with the lowest variance across predictions. The RMSE of predicted polymers were similar to those of the forward model. This methodology offers unprecedented control of polymer design, which may significantly accelerate polymer design for one or more objective properties well beyond cloud points.

High-throughput prediction of the ground-state collinear magnetic order of inorganic materials using Density Functional Theory(用密度泛函理论高通量预测无机材料的基态共线磁序)
Matthew Kristofer HortonJoseph Harold Montoya, Miao Liu & Kristin Aslaug Persson
npj Computational Materials 5:64(2019)
doi:s41524-019-0199-7
Published online:06 June 2019

Abstract| Full Text | PDF OPEN

摘要:在共线自旋极化密度泛函理论的框架下,我们提出了一种鲁棒的、自动化的高通量工作流程,用于计算固态无机晶体(无论是铁磁性、反铁磁性还是亚铁磁性)的磁性基态和相关磁矩。这是通过一个计算效率很高的方案来实现的,其中首先列举所有合理的磁序,并根据对称性进行优先排序,然后通过传统的DFT + U计算来弛豫结构和计算能量。这种自动化工作流程使用atomatecode进行形式化,以适用于数千种材料以可靠、系统的方式使用,也可完全自定义。该工作流程以64种实验已知的非平凡磁序离子磁性材料作为基准,通过计算500多种不同的磁序,对工作流程的性能进行了评估。该流程正确预测了基准材料中95%的非铁磁基态,正确预测了60%的实验测定的基态排序。对基态磁序的大规模预测为基于磁性能的高通量筛选研究开辟了可能性,从而加速了新功能材料的发现和认识   

Abstract:We present a robust, automatic high-throughput workflow for the calculation of magnetic ground state of solid-state inorganic crystals, whether ferromagnetic, antiferromagnetic or ferrimagnetic, and their associated magnetic moments within the framework of collinear spin-polarized Density Functional Theory. This is done through a computationally efficient scheme whereby plausible magnetic orderings are first enumerated and prioritized based on symmetry, and then relaxed and their energies determined through conventional DFT+U calculations. This automated workflow is formalized using the atomatecode for reliable, systematic use at a scale appropriate for thousands of materials and is fully customizable. The performance of the workflow is evaluated against a benchmark of 64 experimentally known mostly ionic magnetic materials of non-trivial magnetic order and by the calculation of over 500 distinct magnetic orderings. A non-ferromagnetic ground state is correctly predicted in 95% of the benchmark materials, with the experimentally determined ground state ordering found exactly in over 60% of cases. Knowledge of the ground state magnetic order at scale opens up the possibility of high-throughput screening studies based on magnetic properties, thereby accelerating discovery and understanding of new functional materials. 

Editorial Summary

High-throughput prediction: ground-state collinear magnetic order高通量预测:基态共线磁序

该研究报道了一种工作流程和支持分析,为基于磁性能的筛选开辟了道路,并为以前未被实验研究的磁性材料开展新的研究提供了一个起点。来自美国劳伦斯伯克利国家实验室和加州科学大学伯克利分校的Kristin Aslaug Persson教授领导的团队,提出了一种基于纯共线DFT模拟的工作流程,包括在高通量条件下确定材料是否为铁磁性的判据并试图找到在0 K温度下给定材料的基态磁序。作者讨论了两个关键成果:1)提出并实现了一种对给定材料进行合理的磁性排序的方案,并确定了计算的优先级;2)使用基于一组成熟磁性材料的传统DFT + U工作流来评估这些生成的顺序,计算和保存不同磁序造成的能量差,从而确定DFT预测的基态顺序。基于磁性能的高通量筛选研究有助于加速新功能材料的发现和理解

A workflow and supporting analyses opened up opportunities for screening based on magnetic properties, providing a starting point for the investigation of magnetic materials previously unstudied by experimental techniques is reported. A team led by Kristin Aslaug Persson from the Lawrence Berkeley National Laboratory, Berkeley, and the Science University of California Berkeley, USA, presented a workflow based on purely collinear DFT simulations with the modest but crucial goal of determining whether a material is ferromagnetic or not in a high-throughput context, and then of attempting to find the ground state magnetic order of a given material at 0K, presupposing that such a ground state exhibits collinear spin. There are two key advances addressed in this paper. Firstly, they proposed and implemented a scheme for enumerating plausible magnetic orderings for a given material, and decided on a ranking for prioritizing calculations. Secondly, they evaluated these generated orderings using a workflow based on conventional DFT+U for a set of well-established magnetic materials, stored the differences in energy between the calculated orderings and thus determine the ground-state ordering predicted by DFT. So now high-throughput screening studies based on magnetic properties may accelerate discovery and understanding of new functional materials.

Atom table convolutional neural networks for an accurate prediction of compounds properties (原子表卷积神经网络准确预测化合物性质)
Shuming Zeng, Yinchang Zhao, Geng Li, Ruirui Wang, Xinming Wang&Jun Ni 
Abstract| Full Text | PDF OPEN

npj Computational Materials 5:84(2019)
doi:s41524-019-0223-y
Published online:08 August 2019

摘要:机器学习技术在材料科学中具有广泛的应用。然而,大多数的机器学习模型需要很多先验知识来手动构建特征向量。本研究提出了一种原子表卷积神经网络模型,它仅需要组分信息就能自动构建特征向量,进而学习化合物的实验性质。对于带隙和形成能的预测,该模型的精度超过了标准DFT计算的结果。通过数据增强的方法,这种模型不仅能够准确预测超导体的超导转变温度,还能够区分超导体和非超导体。利用这种模型,我们从数据库中筛选出了20种可能具有高超导转变温度的材料。此外,从模型学习到的特征向量中,我们提取了主族元素的性质并重现了它们的化学趋势。此框架对于高通量材料筛选以及挖掘潜在的物理提供了有效手段   

Abstract:Machine learning techniques are widely used in materials science. However, most of the machine learning models require a lot of prior knowledge to manually construct feature vectors. Here, we develop an atom table convolutional neural networks that only requires the component information to directly learn the experimental properties from the features constructed by itself. For band
gap and formation energy prediction, the accuracy of our model exceeds the standard DFT calculations. Besides, through data enhanced technology, our model not only accurately predicts superconducting transition temperatures, but also distinguishes superconductors and non-superconductors. Utilizing the trained model, we have screened 20 compounds that are potential superconductors with high superconducting transition temperature from the existing database. In addition, from the learned features, we extract the properties of the elements and reproduce the chemical trends. This framework is valuable for high throughput screening and helpful to understand the underlying physics
.
 

Editorial Summary

Machine Learning: Accurate Prediction of Compounds Properties with only composition input机器学习:仅需组分信息就能准确预测化合物性质

该研究提出了一种称为原子表卷积神经网络的机器学习方法,可以在训练中不断学习合适的特征来预测化合物的形成能、带隙和超导转变温度。来自清华大学物理系倪军教授领导的团队,报道了一种端对端的机器学习方案,可以在材料结构数据缺乏的情况下,有效地预测材料的实验性质。利用化合物的组分信息,构造出一张与之对应的的原子表。材料的特征向量通过一个卷积网络来学习并直接用于性质的预测。整个网络是同步训练的,既避免了构造特征的麻烦,又搜索了更大的参数空间,能够提高准确率。这种模型在预测超导转变温度、带隙和形成能上相对于手动精心构造特征向量的方法准确率更高。模型还能够区别超导体和非超导体。同时,对学习到的特征向量的分析表明,模型捕捉到了相关性质内在的物理机制。该研究方法提出的模型可用于材料的高通量筛选和解释内在的物理机制

In this study, a machine learning framework called atom table convolutional neural networks is proposed, which can learn appropriate features in training to predict the formation energy, band gap and superconducting transition temperature of compounds. A team led by Professor Ni Jun, Department of Physics, Tsinghua University, reported an end-to-end machine learning scheme that can effectively predict the experimental properties of materials in the absence of structural data. A corresponding atomic table is constructed by using the component information of compounds. Material features are learned through a convolution network and are directly used for prediction. The whole network is trained synchronously, which not only avoids the difficulties of constructing features, but also searches for a larger parameter space, and can improve the accuracy. This model is more accurate in predicting superconducting transition temperature, band gap and formation energy than the common method of carefully constructing features manually. The model can also distinguish the superconductors and non-superconductors. At the same time, the analysis of the learned features shows that the model captures the intrinsic physical mechanism of the related properties. The proposed model can be used for high throughput screening and explaining the intrinsic physical mechanism of materials.

An electrostatic spectral neighbor analysis potential for lithium nitride (氮化锂的静电谱相邻分析电位)
Zhi DengChi ChenXiang-Guo Li & Shyue Ping Ong
npj Computational Materials 5:75(2019)
doi:s41524-019-0212-1
Published online:16 July 2019

Abstract| Full Text | PDF OPEN

摘要:基于局部环境描述符的机器学习原子间势,在预测精度上较基于刚性函数形式的传统势有了革命性的飞跃。然而,它们在离子体系中的应用面临的一个挑战是长程静电相互作用的处理。本研究提出了利用离子α-Li3N的高精度静电光谱邻域分析电位(eSNAP)来解决这一问题,离子α-Li3N是一种经典的锂离子快离子导体,可作为可充电锂离子电池的固体电解质或涂层。研究表明,优化后的eSNAP模型在能量和力的预测以及晶格常数、弹性常数、声子色散关系等各种性质的预测方面,明显优于传统的Coulomb–Buckingham势模型。该研究还展示了eSNAPLi3N中长时、大尺度的Li扩散研究中的应用,为测量协同离子运动(例如Haven比率)和晶界扩散提供了原子层次的认识。这项工作旨在提供一种方法,以发展量子精确力场的多组分离子导体体系下的SNAP形式,从而实现这种系统的大尺度模拟   

Abstract:Machine-learned interatomic potentials based on local environment descriptors represent a transformative leap over traditional potentials based on rigid functional forms in terms of prediction accuracy. However, a challenge in their application to ionic systems is the treatment of long-ranged electrostatics. Here, we present a highly accurate electrostatic Spectral Neighbor Analysis Potential (eSNAP) for ionic α-Li3N, a prototypical lithium superionic conductor of interest as a solid electrolyte or coating for rechargeable lithium-ion batteries. We show that the optimized eSNAP model substantially outperforms traditional Coulomb–Buckingham potential in the prediction of energies and forces, as well as various properties, such as lattice constants, elastic constants, and phonon dispersion curves. We also demonstrate the application of eSNAP in long-time, large-scale Li diffusion studies in Li3N, providing atomistic insights into measures of concerted ionic motion (e.g., the Haven ratio) and grain boundary diffusion. This work aims at providing an approach to developing quantum-accurate force fields for multi-component ionic systems under the SNAP formalism, enabling large-scale atomistic simulations for such systems. 

Editorial Summary

Lithium nitride: electrostatic spectral neighbor analysis potential氮化锂:做好电池的新算法

利用局部环境描述符,如光谱邻域分析势(SNAP),本研究证明了通过引入长程静电相互作用可适用于快离子体系的最新势函数。来自美国加州大学圣迭戈分校的Shyue Ping Ong教授领导的团队发现,静电SNAP (eSNAP)模型在预测能量和力,以及晶格常数、弹性常数和声子色散曲线等各种性质方面,显著优于传统的Coulomb-Buckingham势模型。他们应用eSNAP模型对复杂的α-Li3N模型(500~5000个原子)进行了长时间(~1 ns)的模拟。他们发现通过直接计算电荷扩散系数和颗粒边界得到的α-Li3NHaven比率在晶界处具有更快的扩散通道(相对于块体材料)。电荷扩散率的计算在从头算分子动力学模拟中非常难以收敛,而该参数的计算可使他们能够估算出更加可靠的α-Li3N各向异性扩散系数。有趣的是,尽管他们发现c方向的电导率通常比ab平面的电导率低,但与单晶的测量结果相比,电导率仅低一个数量级。该研究提供了一种在SNAP形式下发展多组分离子体系的量子精确力场的方法,为此类体系的大规模原子模拟提供了可能

Modern potentials based on local environment descriptors such as the Spectral Neighbor Analysis Potential (SNAP) that can be adapted for ionic systems by incorporating long-range electrostatics is demonstrated. A team led by Prof. Shyue Ping Ong from the University of California San Diego, USA, showed that the electrostatic SNAP (eSNAP) model substantially outperforms traditional Coulomb–Buckingham potential in the prediction of energies and forces, as well as various properties, such as lattice constants, elastic constants, and phonon dispersion curves. They applied the eSNAP model to conduct long-time-scale (~1ns) simulations of complex models (500–5000 atoms) of α-Li3N. They found the Haven ratio of α-Li3N by directly calculating charge diffusivity and the grain boundaries providing faster diffusion pathways (relative to bulk). The calculation of charge diffusivity, which is difficult to converge in ab initio molecular dynamics simulations, enables them to compute much more reliable estimates of the anisotropic diffusivities of α-Li3N. Interestingly, though they find that conductivity in the c-crystallographic direction is in general slower than the ab plane, the value is only one order of magnitude lower, contrary to single crystal measurements. This study provides an approach to developing quantum-accurate force fields for multi-component ionic systems under the SNAP formalism, enabling large-scale atomistic simulations for such systems.

Ab initio vibrational free energies including anharmonicity for multicomponent alloys (多组分合金非谐性振动自由能的从头算)
Blazej GrabowskiYuji IkedaPrashanth SrinivasanFritz KormannChristoph FreysoldtAndrew Ian DuffAlexander Shapeev & Jorg Neugebauer
npj Computational Materials 5:80(2019)
doi:s41524-019-0218-8
Published online:26 July 2019

Abstract| Full Text | PDF OPEN

摘要:多种主成分合金的独特和意想不到的特性使合金设计重新焕发活力,并引起了科学界的浓厚兴趣。通过计算设计,巨大的成分参数空间使这些合金成为一个独特的探索领域。然而,截至目前,还没有一种计算效率高、精度高的合金热力学性质的方法。其中一个根本原因是缺乏精确和有效的方法来计算这些多组分、化学上复杂的合金的振动自由能(包括其非谐性)。在这项工作中,通过密度泛函理论的方法来解决这个问题,其原理是热力学积分和机器学习势的结合使用。通过计算典型五组分VNbMoTaW难熔高熵合金的非谐自由能,证明了该方法的有效性   

Abstract:The unique and unanticipated properties of multiple principal component alloys have reinvigorated the field of alloy design and drawn strong interest across scientific disciplines. The vast compositional parameter space makes these alloys a unique area of exploration by means of computational design. However, as of now a method to compute efficiently, yet with high accuracy the thermodynamic properties of such alloys has been missing. One of the underlying reasons is the lack of accurate and efficient approaches to compute vibrational free energies—including anharmonicity—for these chemically complex multicomponent alloys. In this work, a density-functional-theory based approach to overcome this issue is developed based on a combination of thermodynamic integration and a machine-learning potential. We demonstrate the performance of the approach by computing the anharmonic free energy of the prototypical five-component VNbMoTaW refractory high entropy alloy. 

Editorial Summary

Multicomponent alloys: Ab initio vibrational free energies多组分合金:非谐性振动自由能

本文提出了一种新的算法,结合TU-TILD方法与力矩张量势(MTPs),这是目前计算化学复杂合金振动自由能贡献最有效的方法。由德国斯图加特大学的Blazej Grabowski教授领导的团队,证明了TU-TILD + MTP组合是一种理想的协同互作组合,可有效、准确地计算无序多组分合金的完全振动自由能。他们在TU-TILD中将MTP作为参考电位,用于化学复杂的无序VNbMoTaW高熵合金中,结果表明MTP明显优于其它参考电位。TU-TILD + MTP组合的优异性能的内在物理机制是,振动自由能是由相空间中一个定义明确、足够平滑且局部(虽然严格来说是非谐的)相空间的一部分来确定。本研究表明,该组合方法不仅适用于本研究中涉及的等原子比例的组成,也同样适用于任意非等原子比例的组成,且还适用于不同的晶格类型,如hcpfcc等,甚至适用于液相

A new algorithm combining the TU-TILD method with moment tensor potentials (MTPs), a presently most efficient combination to compute the vibrational free energy contribution of chemically complex alloys, is developed. A team led by Prof. Blazej Grabowski from the University of Stuttgart, Germany, demonstrated that the TU-TILD+MTP combination is an ideal symbiosis for an efficient and accurate calculation of the full vibrational free energy of disordered multicomponent alloys. In particular, they applied an MTP as a reference potential within TU-TILD for the chemically complex disordered VNbMoTaW HEA and showed that it is clearly superior to alternative reference potentials. The underlying physical reason for the excellent performance of the TU-TILD+MTP combination is the fact that the vibrational free energy is determined by a rather well-defined, sufficiently smooth, and local—although strictly anharmonic—part of the phase space. The present study indicates that this applies not only to equiatomic compositions such as the one studied in the present study, but likewise to arbitrary nonequiatomic compositions, and further also to different crystallographic lattice types such as hcp or fcc and even to the liquid phase.

3D non-isothermal phase-field simulation of microstructure evolution during selective laser sintering (选区激光烧结微结构演化的三维非等温相场模拟)
Yangyiwei Yang, Olav Ragnvaldsen, Yang Bai, Min Yi & Bai-Xiang Xu
npj Computational Materials 5:81(2019)
doi:s41524-019-0219-7
Published online:06 August 2019

Abstract| Full Text | PDF OPEN

摘要:选区激光烧结(SLS)增材制造过程中,微结构演化极度依赖于局部温度的急剧变化,故而常规的等温相场模型很难适用于SLS的模拟。本研究报道了一种新的非等温相场模型,该模型从熵出发,热力学自洽地推导出了控制微结构序参量演化的非等温动力学方程,以及耦合微结构演化的热传导方程,并考虑了SLS局部极高温导致的局部熔化以及激光-粉末相互作用。该模型经三维有限元数值化后,被用于模拟单次扫描的SLS。为了减小计算量并加快计算速度,提出了一种类似于求解最小着色数问题的新算法,结合晶粒追踪方法,该算法可仅用8个序参量来模拟具有多达200个晶粒的系统。特别地,将该非等温相场模型用于SLS处理316L不锈钢粉末的研究,揭示了激光功率和扫描速度对孔隙率、表面形貌、温度分布、晶粒几何形状以及致密度等微观结构特征的影响规律。此外,模拟结果验证了致密化过程中孔隙率变化的一阶动力学特征,并证实了该模型可用于预测SLS过程中致密化因子与激光比能量之间的关联   

Abstract:During selective laser sintering (SLS), the microstructure evolution and local temperature variation interact mutually. Application of conventional isothermal sintering model is thereby insufficient to describe SLS. In this work, we construct our model from entropy level, andderive the non-isothermal kinetics for order parameters along with the heat transfer equation coupled with microstructure evolution. Influences from partial melting and laser-powder interaction are also addressed. We then perform 3D finite element non-isothermal phase-field simulations of the SLS single scan. To confront the high computation cost, we propose a novel algorithm analogy to minimum coloring problem and manage to simulate a system of 200 grains with grain tracking algorithm using as low as 8 non-conserved order parameters. Specifically, applying the model to SLS of the stainless steel 316L powder, we identify the influences of laser power and scan speed on microstructural features, including the porosity, surface morphology, temperature profile, grain geometry, and densification. We further validate the first-order kinetics of the transient porosity during densification, and demonstrate the applicability of the developed model in predicting the linkage of densification factor to the specific energy input during SLS. 

Editorial Summary

Additive Manufacturing: Sino-German cooperation predicts complex microstructures增材制造:中德合作预测复杂微结构

该研究提出了一种热力学自洽的非等温相场模型以及相应的三维高效数值方法,可以模拟选区激光烧结(SLS)增材制造中复杂微结构的演化过程。德国达姆施塔特工业大学的终身教授胥柏香领导的团队,与南京航空航天大学的青年千人易敏教授合作,报道了一种热力学自洽的非等温相场模型,考虑微结构与热传导的强耦合、SLS局部极高温导致的局部熔化以及激光-粉末相互作用。他们提出了一种类似于求解最小着色数问题的新解决方案,结合晶粒追踪方法,该方案可仅用8个序参量来模拟具有多达200个晶粒的系统。研究人员还使用了基于LM算法的非线性优化方法,同时拟合模型与实验中表面能、晶界能随温度变化的趋势,以获取用于非恒温相场的模型参数。特别地,将该非等温相场模型用于SLS处理316L不锈钢粉末的研究,揭示了激光功率和扫描速度对孔隙率、表面形貌、温度分布、晶粒几何形状以及致密度等微观结构特征的影响规律,并证实了该模型可用于预测SLS过程中致密化因子与激光比能量之间的关联。他们的研究为基于SLS的增材制造的建模及计算模拟提供了有效方法或工具

An appropriate consideration of complex temperature profile and its extreme gradient which are the most prevailing feature of selective laser sintering (SLS)-additive manufacturing (AM) process for simulating the microstructure evolution during the SLS based AM is reported. A team led by Bai-Xiang Xu from Technical University of Darmstadt, Germany, cooperating with Min Yi from Nanjing University of Aeronautics and Astronautics (NUAA), developed a thermodynamically consistent non-isothermal phase-field model to simulate the microstructure evolution during SLS-AM. In order to save the high computation cost, the authors proposed a novel algorithm analogy to minimum coloring problem and manage to simulate a system of 200 grains with grain tracking algorithm using as low as 8 non-conserved order parameters. After applying the model to SLS of the stainless steel 316L powder, the influences of laser power and scan speed on microstructural features (i.e. the porosity, surface morphology, temperature profile, grain geometry, and densification) are successfully identified. Their work provides a phase-field model and the associated numeric scheme which are promising for the large-scale simulation of SLS-AM process.

Unconventional topological phase transition in non-symmorphic material KHgX (X=As, Sb, Bi)(非对称材料中的非常规拓扑相变KHgXX = AsSbBi)
Chin-Shen KuoTay-Rong ChangSu-Yang Xu & Horng-Tay Jeng
npj Computational Materials 5:65(2019)
doi:s41524-019-0201-4
Published online:06 June 2019

Abstract| Full Text | PDF OPEN

摘要:传统的拓扑相变描述了从拓扑平凡到拓扑非平凡态的演化。我们在这项工作中提出了由Dirac无间隙态介导的两个拓扑非平凡绝缘态之间的非常规拓扑相变体系,源于非对称型晶体对称性,不同于传统的拓扑相变。KHgXX = AsSbBi)族是第一个实验上实现的拓扑非同态晶体绝缘体(TNCI),其中拓扑表面态以Mobius扭曲连通性为特征。基于第一性原理计算,我们通过在KHgX上施加外部压力,提出了从TNCI到狄拉克半金属(DSM)的拓扑绝缘体-金属转变。我们发现在非同态晶体结构中KHgXDSM相具有不寻常的镜面ChernCm=-3,其在拓扑上不同于传统的DSM,例如Na3BiCd3As2。此外,我们通过对称性破坏预测KHgX中的新TNCI相。这个新的TNCI相的拓扑表面状态显示锯齿形连通性,不同于无应力的连通性。我们的研究结果为理解拓扑表面状态如何从量子演化提供了全面的研究   

Abstract:Traditionally topological phase transition describes an evolution from topological trivial to topological nontrivial state. Originated from the non-symmorphic crystalline symmetry, we propose in this work an unconventional topological phase transition scheme between two topological nontrivial insulating states mediated by a Dirac gapless state, differing from the traditional topological phase transition. The KHgX (X=As, Sb, Bi) family is the first experimentally realized topological non-symmorphic crystalline insulator (TNCI), where the topological surface states are characterized by the Mobius-twisted connectivity. Based on first-principles calculations, we present a topological insulator–metal transition from TNCI into a Dirac semimetal (DSM) via applying an external pressure on KHgX. We find an unusual mirror Chern number Cm=-3 for the DSM phase of KHgX in the non-symmorphic crystal structure, which is topologically distinct from the traditional DSM such as Na3Bi and Cd3As2. Furthermore, we predict a new TNCI phase in KHgX via symmetry breaking. The topological surface states in this new TNCI phase display zigzag connectivity, different from the unstressed one. Our results offer a comprehensive study for understanding how the topological surface states evolve from a quantum. 

Editorial Summary

Non-symmorphic material KHgX: Unconventional topological phase transition非对称材料KHgX:非常规拓扑相变

该研究提出了由Dirac无间隙态介导的两个拓扑非平凡绝缘态之间的非常规拓扑相变体系,该体系源于非对称型晶体对称性,不同于传统的拓扑相变。来自中国台湾两所大学的Tay-Rong ChangHorng-Tay Jeng等,基于第一行原理计算提出了非常规拓扑相转变。KHgXX = AsSbBi)族是第一个实验上实现的拓扑非同态晶体绝缘体,其中拓扑表面态以莫比乌斯扭曲连接为特征。他们基于第一原理计算,通过引入两个新相来使KHgX的拓扑相图多样化。通过施加应力,KHgX经历拓扑绝缘体-金属的转变,从拓扑非同态晶体绝缘体相转变为Cm = -2DSM相,在非对称晶体结构中的非平凡镜ChernCm = -3。通过对称性破坏,DSM相转换为另一个新的拓扑非同态晶体绝缘体相,其中Cm = -3主导着QSH效应。表面能带的连通性的变化,提供了拓扑相变的直接证明,而且要实现这些预测的新拓扑相,操纵带隙是其关键

An unconventional topological phase transition scheme between two topological non-trivial insulating states mediated by a Dirac gapless state, originated from non-symmorphic crystalline symmetry, differing from traditional topological phase transitions. A team co-led by Tay-Rong Chang and Horng-Tay Jeng from universities in Taiwan, China, proposed the unconventional topological phase transition based on the first-principles calculations. A KHgX (X = As, Sb, Bi) family which they studied is the first experimentally realized topologically non-homomorphic crystal insulator (TNCI) where the topological surface states are characterized by Mobius-twisted connectivity. Based on the first principles calculations, they diversify the topological phase diagram of KHgX by introducing two new phases. By applying stress, KHgX undergoes a topological insulator-metal transition from the TNCI phase with Cm = -2 into the DSM phase with a non-trivial mirror Chern number Cm = -3 in the non-symmorphic crystal structure. Through symmetry breakong, the DSM phase transforms into another new TNCI phase with Cm = -3 hosting the QSH effect. The change in the connectivity of the surface bands provides a direct justification of the topological phase transition, and manipulating the band gap is the key to realize these predicted new topological phases.

Tunable ferromagnetic Weyl fermions from a hybrid nodal ring (源于杂化节点环的可调铁磁外尔费米子)
Baobing Zheng, Bowen Xia, Rui Wang, Jinzhu Zhao, Zhongjia Chen, Yujun Zhao Hu Xu
npj Computational Materials 5:74(2019)
doi:s41524-019-0214-z
Published online:15 July 2019

Abstract| Full Text | PDF OPEN

摘要:近年来,实现非平庸的能带拓扑结构是凝聚态系统中一个极受关注的热点。基于第一性原理计算和对称性分析,本研究报道了在铁磁半金属氧化物CrP2O7中的可调外尔费米子的拓扑相。忽略自旋轨道耦合的情况下,CrP2O7能带中不同类型的节点形成杂化的节点环。考虑自旋轨道耦合的情况下,体系的自旋翻转对称性破缺,因此,杂化的节点环缩减为离散的节点,形成了不同类型的外尔点。该体系投影在(100)面的费米弧清晰可见,有助于在实验上研究CrP2O7的拓扑性质。此外,计算得到的准粒子干涉图样对实验研究也很有帮助。本工作提供了一种良好的铁磁外尔半金属候选材料,并有望应用于拓扑相关领域   

Abstract:Realization of nontrivial band topology in condensed matter systems is of great interest in recent years. Using first-principles calculations and symmetry analysis, we propose an exotic topological phase with tunable ferromagnetic Weyl fermions in a half-metallic oxide CrP2O7. In the absence of spin–orbit coupling (SOC), we reveal that CrP2O7 possesses a hybrid nodal ring. When SOC is present, the spin-rotation symmetry is broken. As a result, the hybrid nodal ring shrinks to discrete nodal points and forms different types of Weyl points. The Fermi arcs projected on the (100) surface are clearly visible, which can contribute to the experimental study for the topological properties of CrP2O7. In addition, the calculated quasiparticle interference patterns are also highly desirable for the experimental study of CrP2O7. Our findings provide a good candidate of ferromagnetic Weyl semimetals, and are expected to realize related topological applications with their attracted features. 

Editorial Summary

Ferromagnetic Weyl Semimetal CrP2O7: From a hybrid nodal ring to tunable Weyl fermions新型铁磁外尔半金属:来自南科大慢悠悠的老虎

该研究提出了CrP2O7 是一种第一类和第二类外尔点共存的铁磁外尔半金属材料,共存的不同类型外尔点来源于没有自旋轨道耦合时的杂化节点环。来自南方科技大学的徐虎教授领导的团队(简称慢悠悠老虎团),基于第一性原理计算和对称性分析,研究了铁磁材料CrP2O7 的拓扑能带结构。在考虑自旋轨道耦合的情况下,体系的杂化节点环缩减为不同类型的外尔点。通过外加磁场改变磁化方向,可以调节外尔点的数量和类型。此外,计算得到的费米弧和准粒子干涉图样非常有助于实验的进一步观测。该研究结果为深入研究磁性和拓扑之间的相互作用提供了一种理想候选材料,并且加深了人们对铁磁拓扑半金属材料的认识

CrP2O7 is demonstrated to be a ferromagnetic Weyl semimetal with the coexistence of type-I and type-II Weyl fermions, which originates from a hybrid nodal ring without spin-orbital coupling (SOC). A team led by Hu Xu from the Southern University of Science and Technology, reported the nontrivial band topology of CrP2O7 by using first-principles calculations and symmetry analysis. The hybrid nodal ring of CrP2O7 without SOC shrinks to different types of Weyl points when SOC is included, and the numbers and types of Weyl points can be tuned by external magnetic field. In addition, the calculated Fermi arcs and quasiparticle interference patterns facilitate the experimental study of the topological properties of CrP2O7. Their findings provide a good candidate of studying the interplay between magnetism and topology physics, and deepen the understanding of ferromagnetic Weyl semimetal.

版权所有 © 中国科学院上海硅酸盐研究所
地址:上海市长宁区定西路1295号 邮政编码:200050