首 頁
滾動信息 更多 >>
npj Computational Materials獲得第一個SCI影響因... (2018-09-07)
英文刊《npj Computational Materials(計算材料學... (2017-05-15)
開放的數據庫 (2015-12-17)
上海硅酸鹽所與Nature出版集團聯合創辦的中國首本n... (2015-11-27)
Nature Publishing Group and Shanghai Institute ... (2015-09-24)
快捷服務
最新文章 研究綜述
過刊瀏覽 作者須知
期刊編輯 審稿須知
相關鏈接
· 在線投稿
會議信息
友情鏈接
  中國科學院上海硅酸鹽研究所
  無機材料學報
  OQMD數據庫
mm.jpg
期刊介紹
  《npj 計算材料學》是在線出版、完全開放獲取的國際學術期刊。發表結合計算模擬與設計的材料學一流的研究成果。本刊由中國科學院上海硅酸鹽研究所與英國自然出版集團(Nature Publishing Group,NPG)以伙伴關系合作出版。
  主編為陳龍慶博士,美國賓州大學材料科學與工程系、工程科學與力學系、數學系的杰出教授。
  共同主編為陳立東研究員,中國科學院上海硅酸鹽研究所研究員高性能陶瓷與超微結構國家重點實驗室主任。
  辦刊目的與報道范圍
  《npj 計算材料學》是在線出版、完全開放獲取的國際...
【查看詳細】
近期文章 更多 >>

Fast and interpretable classification of small X-ray diffraction datasets using data augmentation and deep neural networks(基于數據增強和深度神經網絡對X射線衍射小型數據集的快速和可解釋分類)
Felipe OviedoZekun RenShijing SunCharles SettensZhe LiuNoor Titan Putri HartonoSavitha RamasamyBrian L. De CostSiyu I. P. TianGiuseppe RomanoAaron Gilad Kusne & Tonio Buonassisi
npj Computational Materials 5:60 (2019)
doi:s41524-019-0196-x
Published online:17 May 2019
Abstract| Full Text | PDF OPEN

摘要:X射線衍射(XRD)數據采集與分析是新型薄膜材料研發周期中最耗時的步驟之一。本研究提出了一種基于機器學習的方法,用于從有限數量的薄膜XRD圖譜中預測晶體學維度和空間群。基于無機晶體結構數據庫(ICSD)和實驗數據的模擬數據,我們將監督機器學習方法與模型無關的、物理信息輸入的數據增強策略相結合,克服了新材料開發固有的稀缺數據問題。作為實測案例,本研究合成115種跨越3個維度和7個空間群的金屬鹵化物薄膜并對其進行了分類。在測試了各種算法之后,我們開發、實現了一個全卷積神經網絡,其交叉驗證的維數和空間群分類的準確度分別達到93%和89%。依據整體平均匯集層計算,我們提出了平均分類激活圖,以便允許人們對實驗模型結果作充分的解釋、對分類錯誤的根本原因作出闡明。最后,我們系統地評估了發生預測精度損失的最大XRD圖案步長(數據采集速率)為2θ=0.16°,因而僅需5.5分鐘甚至更短時間就可以得到一張XRD圖譜并對其分類   

Abstract:X-ray diffraction (XRD) data acquisition and analysis is among the most time-consuming steps in the development cycle of novel thin-film materials. We propose a machine learning-enabled approach to predict crystallographic dimensionality and space group from a limited number of thin-film XRD patterns. We overcome the scarce data problem intrinsic to novel materials development by coupling a supervised machine learning approach with a model-agnostic, physics-informed data augmentation strategy using simulated data from the Inorganic Crystal Structure Database (ICSD) and experimental data. As a test case, 115 thin-film metal-halides spanning three dimensionalities and seven space groups are synthesized and classified. After testing various algorithms, we develop and implement an all convolutional neural network, with cross-validated accuracies for dimensionality and space group classification of 93 and 89%, respectively. We propose average class activation maps, computed from a global average pooling layer, to allow high model interpretability by human experimentalists, elucidating the root causes of misclassification. Finally, we systematically evaluate the maximum XRD pattern step size (data acquisition rate) before loss of predictive accuracy occurs, and determine it to be 0.16° 2θ, which enables an XRD pattern to be obtained and classified in 5.5min or less. 

Editorial Summary

Small X-ray diffraction datasets: Fast and interpretable classificationX射線衍射小型數據集:快速和可解釋的分類

快速材料表征對于高通量新材料探索十分重要。X射線衍射(XRD)圖譜的獲取和分析是當前高通量材料實驗篩選的瓶頸之一。針對上述問題,來自麻省理工學院和新加坡的研究團隊發展了一種基于監督機器學習的框架用于快速獲得和識別新型薄膜材料的XRD圖譜。他們首先根據ICSD數據庫中164種薄膜鹵化物和115種實驗合成薄膜的XRD圖譜建立了一個數據庫。基于這個小型庫發展了一個與模型無關的、物理信息輸入的數據擴展方法用于構建訓練數據集。進而采用該數據集訓練了一個卷積神經網絡用于XRD圖譜分類,其維度和空間群分類準確率分別可達9389%。本研究提出的方法可以成功解決新材料探索固有的數據稀缺問題,能夠快速地(在5.5分鐘以內)得到一個新材料的XRD圖譜并對其進行分類

Rapid characterization are necessary ingredients for accelerated material discovery in high-throughput way. However, XRD characterization is currently a common bottleneck in such screening loops. A joint team from Massachusetts Institute of Technology and Singapore-MIT Alliance for Research and Technology proposed a supervised machine learning framework for rapid crystal structure identification of novel materials from thin-film XRD measurements. They first created a library including 164 XRD patterns of thin-film halide materials extracted from the ICSD and an additional 115 experimental experimental XRD patterns. With this small dataset, a model-agnostic, physics-informed data augmentation is proposed to generate a suitable and robust training dataset for thin-film materials. Then a convolutional neural network is trained as an accurate and interpretable classifier with cross-validated accuracies for dimensionality and space group classification of 93 and 89%, respectively. This approach successfully addresses the sparse/scarce data problem intrinsic to novel materials and enables rapid acquisition and analysis of XRD pattern, e.g. in 5.5 min or less.

Electric field tuning of the anomalous Hall effect at oxide interfaces(氧化物界面處反常霍爾效應的電場調諧)
Sayantika Bhowal & Sashi Satpathy
npj Computational Materials 5:61 (2019)
doi:s41524-019-0198-8
Published online:21 May 2019
Abstract| Full Text | PDF OPEN

摘要:反常霍爾效應是自旋極化電子的輸運特性受自旋軌道耦合控制的現象,自旋軌道耦合耦合了電子的軌道自由度和自旋自由度。本文證明了強自旋軌道耦合磁界面的反常霍爾效應可以通過外加電場進行調諧。通過改變反演對稱性破缺的強度,電場改變了Rashba相互作用,而Rashba相互作用反過來又改變了Berry曲率的大小,而Berry曲率是決定反常霍爾電導率的關鍵物理量。結果表明,在方點陣模型的小電場作用下,反常霍爾電導率呈二次相關關系。對新近生長的銥酸鹽界面,即(SrIrO3)1/(SrMnO3)1(001)結構進行了顯式密度泛函計算發現,該結構無論有無電場均表現出很強的電場依賴性。該效應在自旋電子學應用中具有很大的潛力   

Abstract:Anomalous Hall effect is the phenomenon where the transport properties of the spin-polarized electrons are governed by the spin-orbit coupling that couples the orbital and spin degrees of freedom of the electron. Here we show that the anomalous Hall effect at a magnetic interface with strong spin-orbit coupling can be tuned with an external electric field. By altering the strength of the inversion symmetry breaking, the electric field changes the Rashba interaction, which in turn modifies the magnitude of the Berry curvature, the central quantity in determining the anomalous Hall conductivity. The effect is illustrated with a square lattice model, which yields a quadratic dependence of the anomalous Hall conductivity for small electric fields. Explicit density-functional calculations were performed for the recently grown iridate interface, viz., the (SrIrO3)1/(SrMnO3)1 (001) structure, both with and without an electric field, which show a strong electric field dependence. The effect may be potentially useful in spintronics applications. 

Editorial Summary

Anomalous Hall effect: Electric field tuning at oxide interfaces反常霍爾效應:氧化物界面處的電場調諧

利用外加電場對Rashba自旋軌道相互作用進行修正,可以調節3d-5d界面處的反常霍爾效應。來自美國密蘇里大學的Sayantika Bhowal  Sashi Satpathy兩位研究人員,使用了一套普適參數以及密度泛函理論,計算了特定界面(SIO)1/(SMO)1結構的反常霍爾電導率電場依賴性的主要貢獻來源于靠近費米能量的能帶交叉點。此外,AHC可以通過摻雜來調控電子態,從而實現調整反常霍爾電導率。他們為說明該結果,使用了鐵磁晶格模型,并使用了最近生長的亞錳酸鹽-銥酸鹽界面 [(SIO)1/(SMO)1(001)]的密度泛函計算。實際上,最近的一些實驗已經發現了氧化物異質結構中反常霍爾電導率隨電場變化的證據。因此,從理論上和實驗上進一步發展這種效應,并著眼于潛在的自旋電子學應用將是極具價值的

The anomalous Hall effect at the 3d-5d interfaces can be tuned by modifying the Rashba spin-orbit interaction with the application of an external electric field. Prof. Sayantika Bhowal and Sashi Satpathy from University of Missouri, USAillustrated this method using general arguments as well as from density-functional calculations of the anomalous Hall conductivity for a specific interface structure (SIO)1/(SMO)1. The major contribution to the electric-field dependence comes from the band-crossing points close to the Fermi energy. In addition, the AHC can be tuned by manipulating the electron density with doping. They illustrated the results with a ferromagnetic square-lattice model as well as with density-functional calculations for the recently grown manganite-iridate interface, viz., (001) (SIO)1/(SMO)1. In fact, several recent experiments have found evidence for the electric field dependence of the anomalous Hall conductivity in the oxide heterostructures. Therefore, it would be valuable to develop this effect further, both theoretically and experimentally, with an eye towards potential spintronics applications.

Deep data analytics for genetic engineering of diatoms linking genotype to phenotype via machine learning (基于機器學習的硅藻基因型與表型對應聯系的硅藻基因工程深度數據分析)
Artem A. TrofimovAlison A. PawlickiNikolay BorodinovShovon MandalTeresa J. MathewsMark HildebrandMaxim A. ZiatdinovKatherine A. HausladenPaulina K. UrbanowiczChad A. SteedAnton V. IevlevAlex BelianinovJoshua K. MichenerRama Vasudevan & Olga S. Ovchinnikova
npj Computational Materials 5:67 (2019)
doi:s41524-019-0202-3
Published online:13 June 2019
Abstract| Full Text | PDF OPEN

摘要:用于材料合成的材料基因組工程,是在一定條件下制造出具有獨特性質的材料的一種有遠大前途的工程路徑。硅藻的生物礦化,是單細胞藻類使用二氧化硅構建的細胞壁,這種細胞壁雖是微米級的,但其諸多特征性結構是納米級的,是有望用于光學、傳感、過濾和藥物遞送等領域的合成功能材料,是這些領域引人注目的候選材料。因此,針對這些應用,通過定向遺傳修飾實現可控的硅藻結構修改,前途廣闊。本研究中,我們在硅藻(偽矮海鏈藻)中,采用基因敲除技術創建經過基因修飾的藻株,使藻體形態發生變化,應用監督機器學習實現了基因型變化與表型變化對應聯系我們開發了人工神經網絡(NN區分野生基因敲除型硅藻,NN可依據硅藻殼的SEM照片所展示的由特定蛋白質(Thaps3_21880)引起的表型變化進行辨別區分,辨別區分準確度94。類激活映射使物理變化可視化,允許NN區分硅藻藻株,隨后篩查找到控制孔的特定基因。進一步創建了另一個NN以批量處理圖像數據,自動識別剛毛毛孔,提取毛孔相關參數。使用多變量數據可視化(CrossVis)工具可視化所提取的參數之間的類相互關系,并允許直接將孔徑和分布的形態學表型變化,與基因型的變化對應聯系起來   

Abstract:Genome engineering for materials synthesis is a promising avenue for manufacturing materials with unique properties under ambient conditions. Biomineralization in diatoms, unicellular algae that use silica to construct micron-scale cell walls with nanoscale features, is an attractive candidate for functional synthesis of materials for applications including photonics, sensing, filtration, and drug delivery. Therefore, controllably modifying diatom structure through targeted genetic modifications for these applications is a very promising field. In this work, we used gene knockdown in Thalassiosira pseudonana diatoms to create modified strains with changes to structural morphology and linked genotype to phenotype using supervised machine learning. An artificial neural network (NN) was developed to distinguish wild and modified diatoms based on the SEM images of frustules exhibiting phenotypic changes caused by a specific protein (Thaps3_21880), resulting in 94% detection accuracy. Class activation maps visualized physical changes that allowed the NNs to separate diatom strains, subsequently establishing a specific gene that controls pores. A further NN was created to batch process image data, automatically recognize pores, and extract pore-related parameters. Class interrelationship of the extracted paraments was visualized using a multivariate data visualization tool, called CrossVis, and allowed to directly link changes in morphological diatom phenotype of pore size and distribution with changes in the genotype. 

Editorial Summary

Genotype modification linking changes of diatom frustule phenotype: machine learning硅藻基因型改變與硅藻殼變化:機器學習

該研究比較了一種野生型和基因敲除型硅藻,以揭示修改的基因型和表達的表型之間的相互作用,因為基因操作可以使這些生物體被直接用作特別定制的納米結構和微結構材料。來自美國橡樹林國家實驗的Olga S. Ovchinnikova領導的研究團隊,通過敲除懷疑可能與硅藻殼形成有關的基因來修改硅藻基因型,并通過掃描電鏡表征該基因引起的表型變化。他們使用圖像處理和機器學習分類算法(人工NN)來篩選影響硅藻表型的基因,并將野生型硅藻與基因修飾型區分開來。就控制毛孔形態的蛋白檢測來說,他們的識別野生型和基因修飾型硅藻的NN,檢測準確度為94%。為解釋基于NN的分類表觀準確率,他們用類激活圖(CAM)來突出顯示網絡使用的圖像區域,發現硅藻殼的孔是將野生型硅藻與一種特定的敲低基因表達的藻株分開的穩定特征。隨后,他們創建了另一個神經網絡,專門針對毛孔并提取其參數。這種自動化特征提取過程使人們能夠將遺傳修飾與硅藻形態對應關聯起來。這一方法確定了由給定的遺傳修飾產生的藻殼結構的變化,為生物礦化過程提供了生物學探測能力

Wild type and genetically modified diatoms is investigated to capture the interplay between the changing genotype and the expressed phenotype in diatom frustule, as gene manipulation could enable these organisms to be used as a direct source of specifically tailored nanostructured and microstructured materials.  A team led by Olga S. Ovchinnikova from the Oak Ridge National Laboratory, USA, modified the genotype by knocking down genes potentially involved with frustule formation and characterized the phenotype by scanning electron microscopy. They used image processing and machine learning classification algorithms (artificial NNs) to screen for genes that affect diatom phenotype and to distinguish diatoms with wild type and modified morphologies. With regard to inspecting a protein modification that controls pores in frustule, they demonstrated a NN that can identify wild and modified diatoms with 94% accuracy. To explain the apparent efficiency of NN-based classification, class activation maps (CAMs) were used to highlight the image regions used by the network, consolidating the defining features separating wild-type diatoms from one specific knockdown strain. They then created a separate neural net to focus specifically on pores and to extract their parameters. This automated feature extraction process could correlate the genetic modification with diatom morphology. This approach identifies the changes in frustule structure that result from a given genetic modification, offering biological insight into the biomineralization process.

Coordination corrected ab initio formation enthalpies (協同糾正的從頭算形成焓)
Rico FriedrichDemet UsanmazCorey OsesAndrew SupkaMarco FornariMarco Buongiorno NardelliCormac Toher & Stefano Curtarolo
npj Computational Materials 5:59 (2019)
doi:s41524-019-0192-1
Published online:15 May 2019
Abstract| Full Text | PDF OPEN

摘要:如何正確計算形成焓是從頭算材料設計的一個功能。使用標準密度泛函理論對幾類材料系統(如氧化物)計算時,會得到一些錯誤的結果。本研究基于最近鄰陽離子-陰離子鍵的數量,提出了“協調校正焓”的方法(CCE),不僅校正焓,還能校正多晶型的相對穩定性。CCE使用PerdewBurke-ErnzerhofPBE)、局部密度近似(LDA)和強約束和適當規范(SCAN)交換相關泛函,結合準諧波Debye模型來處理零點振動和熱效應。在二元和三元氧化物(鹵化物)上進行的基準測試結果顯示,所有函數的室溫結果都非常準確,用SCAN計算獲得的最小平均絕對誤差為27(24)meV/atom。這個誤差對形成焓的零點振動和熱貢獻很小,并且不同的誤差信號在很大程度上相互抵消   

Abstract:The correct calculation of formation enthalpy is one of the enablers of ab-initio computational materials design. For several classes of systems (e.g. oxides) standard density functional theory produces incorrect values. Here we propose the “coordination corrected enthalpies” method (CCE), based on the number of nearest neighbor cation–anion bonds, and also capable of correcting relative stability of polymorphs. CCE uses calculations employing the Perdew, Burke and Ernzerhof (PBE), local density approximation (LDA) and strongly constrained and appropriately normed (SCAN) exchange correlation functionals, in conjunction with a quasiharmonic Debye model to treat zero-point vibrational and thermal effects. The benchmark, performed on binary and ternary oxides (halides), shows very accurate room temperature results for all functionals, with the smallest mean absolute error of 27(24)meV/atom obtained with SCAN. The zero-point vibrational and thermal contributions to the formation enthalpies are small and with different signs—largely canceling each other. 

Editorial Summary

Coordination corrected ab initio formation enthalpies預測化合物穩定性的計算誤差:協同糾正

該研究基于最近鄰陽離子-陰離子鍵的數量引入了一種完全主動的校正方案:“協調校正焓(CCE)”方案,可以解決預測化合物的熱力學穩定性時的誤差。來自美國杜克大學的Stefano Curtarolo領導的團隊,使用三種校正計算方法:Perdew-Burke-ErnzerhofPBE)、局部密度近似(LDA)和強約束和適當規范(SCAN)交換相關泛函的計算,結合準諧波Debye模型來校正717)三元氧化物(鹵化物)的零點振動和熱效應,分別給出了3849),2974)和2724meV / atomMAE極為準確的校正形成焓。他們用準諧波Debye模型處理零點溫度和有限溫度的振動時,發現振動在很大程度上被消除了,誤差比以前的方法要小得多。CCE得到精確的形成焓,平均絕對誤差小至20-30 meV /原子。該方法簡單且易于擴展到其他體系如氮化物、磷化物或硫化物等材料。本方法可用于預測依賴于精確形成焓的各種性質,例如電池電壓、缺陷能量和高熵材料的形成。由于CCE考慮了化學鍵的連接和拓撲結構,因此它還可以糾正給定組分的不同結構的相對穩定性

A physically motivated correction scheme — coordination corrected enthalpies (CCE), based on the number of bonds between each cation and surrounding anions, is proposedwhich can minimize the error in predicting thermodynamic stability of compounds. A team led by Stefano Curtarolo from the Duke University, USA, employed the Perdew, Burke and Ernzerhof (PBE), local density approximation (LDA) and strongly constrained and appropriately normed (SCAN) exchange correlation functionals, in conjunction with a quasiharmonic Debye model to treat zero-point vibrational and thermal effects of 71(7) ternary oxides (halides), and gives very accurate corrected formation enthalpies with mean absolute errors of 38(49), 29(74) and 27(24)meV/atom, respectively. Zero-point and finite temperature vibrational contributions are treated within a quasiharmonic Debye model and are found to largely cancel out, with errors significantly smaller than previous approaches. CCE yields accurate formation enthalpies with an average absolute error as small as 20–30meV/atom. The method is simple and easy to extend to other materials classes, e.g. nitrides, phosphides, or sulfides. It can be used to predict a wide variety of properties relying on accurate formation enthalpies, such as battery voltages, defect energies, and the formation of high-entropy materials. Because CCE considers bonding connectivity and topology, it can also correct the relative stability of different structures at a given composition.

Computational strategies for design and discovery of nanostructured thermoelectrics (設計和發現納米結構熱電材料的計算策略)
Shiqiang HaoVinayak P. DravidMercouri G. Kanatzidis & Christopher Wolverton
npj Computational Materials 5:58 (2019)
doi:s41524-019-0197-9
Published online:14 May 2019
Abstract| Full Text | PDF OPEN

摘要:理論計算和預測在先進高性能熱電材料的發展中發揮越來越重要的貢獻,并成功地引導實驗理解并實現破紀錄的好結果。本文從理論計算的角度,綜述了近年來高性能納米結構體熱電材料的研究進展。提高熱電性能的一個有效的新興策略涉及多尺度調控的電子散射最小化、聲子散射的最大化。我們提出了幾個重要的策略和關鍵的例子,突出了基于第一性原理的計算在揭示熱電性能協同優化的復雜但易于處理的關系方面的貢獻。綜合優化方法為改進材料提供了四重設計策略:1)通過多尺度分層架構顯著降低晶格熱導率;2)通過本征矩陣的電子能帶簡并工程大幅提高塞貝克系數;3)通過主相和第二相之間的帶邊形狀調控載流子遷移率;4)通過最大化加強非諧振動和聲子Gruneisen參數來設計具有本征低導熱率的材料。這些組合效應可以在降低晶格熱導率的同時提高功率因子。本綜述對理論如何影響該領域的現狀提供了更好的理解,并有助于指導未來高性能熱電材料的研究   

Abstract:The contribution of theoretical calculations and predictions in the development of advanced high-performance thermoelectrics has been increasingly significant and has successfully guided experiments to understand as well as achieve record-breaking results. In this review, recent developments in high-performance nanostructured bulk thermoelectric materials are discussed from the viewpoint of theoretical calculations. An effective emerging strategy for boosting thermoelectric performance involves minimizing electron scattering while maximizing heat-carrying phonon scattering on many length scales. We present several important strategies and key examples that highlight the contributions of first-principles-based calculations in revealing the intricate but tractable relationships for this synergistic optimization of thermoelectric performance. The integrated optimization approach results in a fourfold design strategy for improved materials: (1) a significant reduction of the lattice thermal conductivity through multiscale hierarchical architecturing, (2) a large enhancement of the Seebeck coefficient through intramatrix electronic band convergence engineering, (3) control of the carrier mobility through band alignment between the host and second phases, and(4) design of intrinsically low-thermal-conductivity materials by maximizing vibrational anharmonicity and acoustic-mode Gruneisen parameters. These combined effects serve to enhance the power factor while reducing the lattice thermal conductivity. This review provides an improved understanding of how theory is impacting the current state of this field and helps to guide the future search for high-performance thermoelectric materials. 

Editorial Summary

Nanostructured thermoelectrics: design and discovery納米結構熱電材料:設計和發現

該綜述描述了四種典型的計算策略在提高納米結構體相熱電性能方面的應用。來自美國西北大學Christopher Wolverton領導的研究小組綜合了最近的重要研究進展,揭示了納米結構熱電體相材料設計和發現的計算策略的規律。到目前為止,已經用高ZT > 2證明了幾種體積熱電材料的優異熱電性能。所有這些高ZT優值的材料都優雅地體現了PGEC的概念。特別是,利用最小電子散射和最大限度地利用納米結構方法的全長尺度熱載流子散射的結合,實現了許多材料ZT優值的提高。納米結構方法集成了許多調用多尺度聲子散射的新思想:包括原子尺度合金化、內生納米結構和中尺度顆粒邊界控制,并以協同的方式結合了能帶對齊和簡并工程方法。這種綜合方法也是一種將ZT提高到3的最合理方法。在追求更高的ZT優值時,第一性原理計算對于提供理論解釋、材料選擇甚至ZT預測都是至關重要的

The use of four typical computational strategies to enhance the thermoelectric performance of nanostructured bulk materials is reviewed. A team led by Christopher Wolverton from the Northwestern University, USA, combined all the recent important research progress and revealed the trends in computational strategies for design and discovery of nanostructured thermoelectrics. Thus far, the extraordinary thermoelectric performance of several bulk thermoelectric materials has been demonstrated with a high ZT>2. All of these high-ZT materials elegantly reflect the PGEC concept. In particular, many of the enhanced figures of merit were achieved using a combination of minimizing electron scattering and maximizing all-length-scale heat-carrying phonon scattering using nanostructuring methods. The nanostructuring methods integrate many new concepts of invoking multiscale phonon scattering, including atomic-scale alloying, endotaxial nanostructuring, and mesoscale grain-boundary control, with band alignment and convergence engineering methods in a synergistic manner. This integrated methodology is also the most plausible approach to increase ZT to the next threshold of ZT=3. In the pursuit of higher ZT, first-principles calculations are critical to providing theory explanations, material selections, and even ZT predictions.

Bayesian inference of atomistic structure in functional materials (功能材料中原子結構的貝葉斯預測)
Milica TodorovicMichael U. GutmannJukka Corander & Patrick Rinke
npj Computational Materials 5:35 (2019)
doi:s41524-019-0175-2
Published online:18 March 2019
Abstract| Full Text | PDF OPEN

摘要:訂制先進有機/無機異質器件使其符合預期技術應用的功能特性,需要了解器件內部的微觀結構并能對其調控。原子尺度量子力學模擬方法可以針對具體材料結構給出精確預測的能量和性質,然而,通過計算的結構仍然比較困難。本研究提出了一種基于“構筑模塊”的貝葉斯優化結構搜索(BOSS)方法,用于解決擴展的有機/無機界面問題,并證明了其在分子表面吸附研究中的可行性。在BOSS中,貝葉斯模型通過主動學習采樣的原子構象快速確定材料的勢能面。這使我們能夠在TiO2 銳鈦礦相的(101)面上確定C60的幾種最有利的分子吸附結構,并闡明控制結構組裝的關鍵分子-表面相互作用。預測的結構與實驗圖像非常一致,證明了BOSS的良好預測能力,并為分子聚集體和薄膜的大尺度表面吸附研究開辟了道路   

Abstract: Tailoring the functional properties of advanced organic/inorganic heterogeneous devices to their intended technological applications requires knowledge and control of the microscopic structure inside the device. Atomistic quantum mechanical simulation methods deliver accurate energies and properties for individual configurations, however, finding the most favourable configurations remains computationally prohibitive. We propose a ‘building block’-based Bayesian Optimisation Structure Search (BOSS) approach for addressing extended organic/inorganic interface problems and demonstrate its feasibility in a molecular surface adsorption study. In BOSS, a Bayesian model identifies material energy landscapes in an accelerated fashion from atomistic configurations sampled during active learning. This allowed us to identify several most favourable molecular adsorption configurations for C60 on the (101) surface of TiO2 anatase and clarify the key molecule-surface interactions governing structural assembly. Inferred structures were in good agreement with detailed experimental images of this surface adsorbate, demonstrating good predictive power of BOSS and opening the route towards large-scale surface adsorption studies of molecular aggregates and films. 

Editorial Summary

Atomistic structure in functional materials: Bayesian inference功能材料中的原子結構:貝葉斯預測

該研究針對有機無機材料界面結構預測提出基于“結構塊”的貝葉斯結構搜索方案。由芬蘭阿爾托大學Milica Todorovic等領導的團隊,將人工智能采樣策略、自然“構建塊”表示與精確的量子力學計算相結合,提出了一種新穎的結構搜索方案。他們以C60團簇在二氧化鈦(101)表面的吸附結構研究為例證明了該方法的準確性。其預測的吸附結構與實驗觀測很好的吻合。不僅如此他們還通過上述方法得到分子與表面的作用機理,理解了穩定吸附結構的成因。該研究提出的方法可以進一步推廣用于分子聚集體和薄膜等大尺度表面吸附結構的研究

Applicability of PS algorithm can now restore full spectral and full spatial resolution AFM-IR dataset. A team led by Olga S. Ovchinnikova from the Center for Nanophase Materials Science, Oak Ridge National Laboratory, USA, applied a coupled non-negative matrix factorization (CNMF) pan-sharpening (PS) algorithm for AFM-IR data to enable rapid reconstruction of high spatial resolution hyperspectral chemical imaging data. They discussed the influence of the parameter affecting the result such as downsampling rate, number of components used for decomposition as well as number of fixed wavenumber maps involved in dataset restoration. Finally, they showcased the application of PS CNMF algorithm for the correlative analysis of plant cell walls in identifying the relationship between local mechanical properties and chemical composition. Their method drastically decreases time required to acquire spectral images while simultaneously providing multicomponent analysis capability. Such approaches can be readily adopted for other spectral imaging techniques utilized in chemical imaging of complex materials.

Application of pan-sharpening algorithm for correlative multimodal imaging using AFM-IR (全色銳化算法在AFM-IR相關多模態成像中的應用)
Nikolay BorodinovNatasha BilkeyMarcus FostonAnton V. IevlevAlex BelianinovStephen JesseRama K. VasudevanSergei V. Kalinin & Olga S. Ovchinnikova 
npj Computational Materials 5:49 (2019)
doi:s41524-019-0186-z
Published online:16 April 2019
Abstract| Full Text | PDF OPEN

摘要:原子力顯微鏡與紅外光譜(AFM-IR)的耦合提供了獨特的能力,可對各種材料的局部化學和物理組成作納米分辨的表征。然而,為了充分利用AFM-IR的測量能力,需要取得3D數據集(具有光譜維度的2D圖),常規的AFM掃描要達到相同的空間分辨率會非常耗時。本研究提出了一種基于多組分全色銳化算法來處理光譜AFM-IR數據的新方法。該方法僅需要低空間分辨率光譜和有限數量的高空間分辨率單波數化學圖,即可產生高空間分辨率的高光譜圖像,可極大地減少數據采集時間。基于此,我們能夠得到高分辨率的成分分布圖,在光譜范圍內的任何波數處生成化學圖,并可對樣品的物理和化學性質進行相關分析。本研究以植物細胞壁成像作為模型系統來突顯本方法的作用,并顯示樣品的力學剛度與其化學成分之間的相互作用。我們相信我們的全色銳化方法可以更廣泛地應用于不同類別的材料,從而更深入地研究納米尺度的結構-性能關系   

Abstract: The coupling of atomic force microscopy with infrared spectroscopy (AFM-IR) offers the unique capability to characterize the local chemical and physical makeup of a broad variety of materials with nanoscale resolution. However, in order to fully utilize the measurement capability of AFM-IR, a three-dimensional dataset (2D map with a spectroscopic dimension) needs to be acquired, which is prohibitively time-consuming at the same spatial resolution of a regular AFM scan. In this paper, we provide a new approach to process spectral AFM-IR data based on a multicomponent pan-sharpening algorithm. This approach requires only a low spatial resolution spectral and a limited number of high spatial resolution single wavenumber chemical maps to generate a high spatial resolution hyperspectral image, greatly reducing data acquisition time. As a result, we are able to generate high-resolution maps of component distribution, produce chemical maps at any wavenumber available in the spectral range, and perform correlative analysis of the physical and chemical properties of the samples. We highlight our approach via imaging of plant cell walls as a model system and showcase the interplay between mechanical stiffness of the sample and its chemical composition. We believe our pan-sharpening approach can be more generally applied to different material classes to enable deeper understanding of that structure-property relationship at the nanoscale. 

Editorial Summary

Multimodal imaging using AFM-IR: Pan-sharpening algorithmAFM-IR相關多模態成像:全色銳化算法

本研究證明了全色銳化算法在恢復全光譜和全空間分辨率AFM-IR數據集中的適用性。來自美國橡樹嶺國家實驗室納米材料科學中心的Olga S. Ovchinnikova教授應用AFM-IR數據的耦合非負矩陣分解(CNMF)全色銳化(PS)算法,實現了高空間分辨率、高光譜化學成像數據的快速重建。他們討論了諸如下采樣率(downsampling rate)、用于分解的組分數量、數據集恢復所涉及的固定波數圖數量等因素對結果的影響。最后,該研究展示了全色銳化-非負矩陣分解算法在植物細胞壁相關分析中的應用,確定了局部力學性質與化學組分之間的關系。這一方法極大地減少了獲取光譜圖像所需的時間,同時提供了多組分分析能力。使用這些方法即可借助其他光譜成像技術很容易地實現復雜材料的化學成像

Applicability of PS algorithm can now restore full spectral and full spatial resolution AFM-IR dataset. A team led by Olga S. Ovchinnikova from the Center for Nanophase Materials Science, Oak Ridge National Laboratory, USA, applied a coupled non-negative matrix factorization (CNMF) pan-sharpening (PS) algorithm for AFM-IR data to enable rapid reconstruction of high spatial resolution hyperspectral chemical imaging data. They discussed the influence of the parameter affecting the result such as downsampling rate, number of components used for decomposition as well as number of fixed wavenumber maps involved in dataset restoration. Finally, they showcased the application of PS CNMF algorithm for the correlative analysis of plant cell walls in identifying the relationship between local mechanical properties and chemical composition. Their method drastically decreases time required to acquire spectral images while simultaneously providing multicomponent analysis capability. Such approaches can be readily adopted for other spectral imaging techniques utilized in chemical imaging of complex materials.

Analyzing machine learning models to accelerate generation of fundamental materials insights (分析機器學習模型以加速對基礎材料的認識)
Mitsutaro UmeharaHelge S. SteinDan GuevarraPaul F. NewhouseDavid A. Boyd & John M. Gregoire 
npj Computational Materials 5:34 (2019)
doi:s41524-019-0172-5
Published online:8 March 2019
Abstract| Full Text | PDF OPEN

摘要:材料科學的機器學習設想通過自動識別關鍵數據之間的關系來擴充人類對于規律的解釋,獲得科學的認知,從而加速基礎科學研究。科學家的主要作用是從數據中提取基礎知識,我們證明,通過分析訓練的神經網絡模型本身,而非將其作為預測工具應用,可以加速這種提取。卷積神經網絡在多維參數空間中復雜數據關系(如通過組合材料科學實驗得到的復雜數據)的建模方面具有優勢。測量一種給定材料空間中的性能指標,可提供有關(局部)最佳材料的直接信息,但不會給出引起性能變化背后的機理。通過建立模型基于材料參數(如本文中組合物和拉曼信號)來預測材料性能(太陽能燃料光陽極的光電化學發電),進而對訓練模型的梯度分析,我們揭示了人工觀察或傳統統計分析不易識別的關鍵數據關系。并通過對這些關鍵關系的闡釋進一步了獲取本質的理解,由此展示了通過機器學習結合人類科學家的分析來加速數據解釋的一種框架。我們還演示了使用神經網絡梯度分析來自動預測參數空間中的優化方向(如添加特定的合金元素),其可突破數據限制來提高材料的性能   

Abstract: Machine learning for materials science envisions the acceleration of basic science research through automated identification of key data relationships to augment human interpretation and gain scientific understanding. A primary role of scientists is extraction of fundamental knowledge from data, and we demonstrate that this extraction can be accelerated using neural networks via analysis of the trained data model itself rather than its application as a prediction tool. Convolutional neural networks excel at modeling complex data relationships in multi-dimensional parameter spaces, such as that mapped by a combinatorial materials science experiment. Measuring a performance metric in a given materials space provides direct information about (locally) optimal materials but not the underlying materials science that gives rise to the variation in performance. By building a model that predicts performance (in this case photoelectrochemical power generation of a solar fuels photoanode) from materials parameters (in this case composition and Raman signal), subsequent analysis of gradients in the trained model reveals key data relationships that are not readily identified by human inspection or traditional statistical analyses. Human interpretation of these key relationships produces the desired fundamental understanding, demonstrating a framework in which machine learning accelerates data interpretation by leveraging the expertize of the human scientist. We also demonstrate the use of neural network gradient analysis to automate prediction of the directions in parameter space, such as the addition of specific alloying elements, that may increase performance by moving beyond the confines of existing dat. 

Editorial Summary

Analyzing machine learning models to accelerate generation of fundamental materials insights分析機器學習模型加速材料的基礎認識

研究訓練了一種卷積神經網絡模型,以模擬高維材料參數空間中復雜數據關系。來自美國加州理工學院的John M. Gregoire領導的團隊,使用他們訓練的卷積神經網絡預測了BiVO4基光陽極的光電化學性能。他們利用高通量實驗獲得的1379個光陽極樣品的組成和拉曼光譜來訓練神經網絡模型。該模型的梯度能有效地可視化材料參數空間中特定區域的數據規律,以及整個數據集的數據規律。梯度自動分析為材料研究提供了指導,包括如何超越現有數據集的限制,以進一步提高材料性能。這種解釋機器學習模型的方法加速了人們對材料科學的認識,并揭示了科學發現的自動化途徑

A convolutional neural networks model is trained to model complex data relationships in high-dimensional materials parameter spaces. A team led by John M. Gregoire from the California Institute of Technology predicted photoelectrochemical performance of BiVO4-based photoanodes using their trained convolutional neural networks. The composition and Raman spectrum of 1379 photoanode samples obtained from high-throughput measurements were used to train the model. Gradients from this model enabled effective visualization of data trends at specific locations in the materials parameter space as well as collectively for the entire dataset. Automated analysis of the gradients provides guidance for research, including how to move beyond the confines of the present dataset to further improve performance. This approach to interpreting machine learning models accelerates scientific understanding and illustrates avenues for continued automation of scientific discovery.

Unlocking the potential of weberite-type metal fluorides in electrochemical energy storage (釋放氟鋁鎂鈉石型金屬氟化物在電化學儲能中的潛力)
Holger EuchnerOliver Clemens & M. Anji Reddy 
npj Computational Materials 5:31 (2019)
doi:s41524-019-0166-3
Published online:6 March 2019
Abstract| Full Text | PDF OPEN

摘要:鈉離子電池(NIBs)是有望取代鋰離子電池(LIB)的替代電池技術中的先行者,然而鈉離子電池的比能量明顯低于鋰離子電池,這主要是由于鈉嵌入型正極材料具有較低的反應電位和較高的分子量。NIB要想與LIB的高能量密度競爭,它就需要高電壓的正極材料。本研究報告了對Weberite型鈉金屬氟化物(SMF)的理論研究,該氟化物是一種新型的高電壓和高能量密度的材料,迄今為止尚未作為NIB的正極材料而被研究。Weberite型結構對于含鈉過渡金屬氟化物非常有利,其中多種過渡金屬組合(MM')均屬于相應的Na2MM'F7結構。本工作通過計算研究了一系列具有Weberite型結構的已知和假設的化合物,以評估它們作為NIB正極材料的潛力。WeberiteSMF顯示出Na+擴散的二維路徑,具有異常低的活化能壘。高能量密度與Na+的低擴散勢壘結合,使得這種類型的化合物有望成為NIB正極材料的候選   

Abstract:Sodium-ion batteries (NIBs) are a front-runner among the alternative battery technologies suggested for substituting the state-of-the-art lithium-ion batteries (LIBs). The specific energy of Na-ion batteries is significantly lower than that of LIBs, which is mainly due to the lower operating potentials and higher molecular weight of sodium insertion cathode materials. To compete with the high energy density of LIBs, high voltage cathode materials are required for NIBs. Here we report a theoretical investigation on weberite-type sodium metal fluorides (SMFs), a new class of high voltage and high energy density materials which are so far unexplored as cathode materials for NIBs. The weberite structure type is highly favorable for sodium-containing transition metal fluorides, with a large variety of transition metal combinations (M, M’) adopting the corresponding Na2MM’F7 structure.. A series of known and hypothetical compounds with weberite-type structure were computationally investigated to evaluate their potential as cathode materials for NIBs. Weberite-type SMFs show two-dimensional pathways for Na+ diffusion with surprisingly low activation barriers. The high energy density combined with low diffusion barriers for Na+ makes this type of compounds promising candidates for cathode materials in NIBs. 

Editorial Summary

New hope of sodium-ion batteries: Weberite-type metal fluoridesNa離子電池的新希望:weberite型金屬氟化物

該研究考查了一系列擬作為NIB正極材料的weberite鈉金屬氟化物。來自德國烏爾姆亥姆霍茲研究所M. Anji Reddy領導的研究小組,篩查了一些真實和虛擬的化合物,以揭示weberite金屬氟化物作為NIB正極材料的潛力。雖然他們將研究限定于考查僅一定數量的化合物,但這些材料的范圍及對它們的各種修飾的可能性將非常大。除了不同元素組合外,通過多種物種填充每個金屬亞晶格也可能是有意義的,這些策略可促進更快的擴散路徑的形成同時又保持高的能量密度,以實現化合物的進一步優化。按照這一策略,他們建議將一些高能量密度的材料與一定量的Ti合金化,以產生快速擴散通道。他們的研究從理論角度證明了這些材料具有作為NIB正極的潛力,作者希望未來的研究會開啟這些化合物的合成和實驗測試

A series of weberite-type sodium metal fluorides as cathode materials for NIBs have investigated. A group led by M. Anji Reddy from the Helmholtz Institute Ulm, Germany, screened real and virtual compounds revealing the potential of weberite-type metal fluorides as cathode materials for NIBs. They limited their study to the investigation of only a certain number of compounds, but the playground for these materials in combination with their variety of possible modifications might be even larger. Apart from other element combination, they highlighted that it may also be of interest to populate each of the metal sublattices by more than one species, which could allow for further optimization of the compounds by facilitating faster diffusion pathways while maintaining high energy density. Following this strategy, they suggested to alloy some high-energy density materials with a certain amount of Ti to create fast diffusion channels. With the potential of these materials being demonstrated from the theoretical viewpoint, the authors aim to trigger the synthesis and experimental testing of these compounds in future studies.

Topological superconducting phase in high-Tc superconductor MgB2 with Dirac–nodal-line fermions (Tc超導體MgB2中的拓撲超導相具有Dirac節點線費米子)
Kyung-Hwan JinHuaqing HuangJia-Wei MeiZheng LiuLih-King Lim & Feng Liu 
npj Computational Materials 5:57 (2019)
doi:s41524-019-0191-2
Published online:3 March 2019
Abstract| Full Text | PDF OPEN

摘要:拓撲超導體是一種有趣且難以捉摸的量子相,具有拓撲保護的無帶隙表面/邊緣態特征,存在于體材超導帶隙中,包含了Majorana費米子。不幸的是,所有目前已知的拓撲超導體轉變溫度都非常低,限制了Majorana費米子的實驗測量。本研究發現,在眾所周知的傳統高溫超導體MgB2中存在拓撲狄拉克節線態。第一性原理計算表明,受空間反演和時間反演對稱性保護的Dirac節點線結構具有獨特的一維色散特征,連接著電子和空位Dirac態。最重要的是,我們用傳統的s波超導帶隙實現了拓撲超導相,用MgB2薄膜的拓撲邊緣模式證明了手性邊緣狀態。我們的這一發現可以在高溫下實現對Majorana費米子的實驗測量   

Abstract:Topological superconductors are an intriguing and elusive quantum phase, characterized by topologically protected gapless surface/edge states residing in a bulk superconducting gap, which hosts Majorana fermions. Unfortunately, all currently known topological superconductors have a very low transition temperature, limiting experimental measurements of Majorana fermions. Here we discover the existence of a topological Dirac–nodal-line state in a well-known conventional high-temperature superconductor, MgB2. First-principles calculations show that the Dirac–nodal-line structure exhibits a unique one-dimensional dispersive Dirac–nodal line, protected by both spatial-inversion and time-reversal symmetry, which connects the electron and hole Dirac states. Most importantly, we show that the topological superconducting phase can be realized with a conventional s-wave superconducting gap, evidenced by the topological edge mode of the MgB2 thin films showing chiral edge states. Our discovery may enable the experimental measurement of Majorana fermions at high temperature. 

Editorial Summary

Topological superconducting phase in high-Tc superconductor MgB2 with Dirac–nodal-line fermionsTc超導體MgB2中的拓撲超導相具有Dirac節點線費米子

本研究在高溫超導體MgB2中揭示了一種有趣的反演和時間反演對稱保護的Dirac節點線態。來自由美國猶他大學和中國量子物質協同創新中心的劉鋒領導的團隊,使用第一性原理計算和模型分析,揭示了這種Dirac節點線態。最重要的是,他們用傳統的s波超導帶隙實現了拓撲超導相,用MgB2薄膜的拓撲邊緣模式證明了手性邊緣狀態。他們的發現為在前所未有的高溫下研究拓撲超導相提供了一個令人興奮的機會,并可能為構建新型量子和自旋電子器件,提供有前途的材料平臺。有可能在高溫下實現對Majorana費米子的實驗測量,將激發未來更廣泛的超導材料拓撲相(如蜂窩狀層狀晶格結構)研究

An intriguing inversion and time-reversal symmetry- protected Dirac nodal line state is revealed in a high-temperature superconductor MgB2. A team led by Feng Liu from the University of Utah, USA, and Collaborative Innovation Center of Quantum Matter, China, performed first-principles calculations to discover the existence of a topological Dirac–nodal-line state in a well-known conventional high-temperature superconductor, MgB2. Most importantly, they showed that the topological superconducting phase can be realized with a conventional s-wave superconducting gap, evidenced by the topological edge mode of the MgB2 thin films showing chiral edge states. Their finding provokes an exciting opportunity to study a topological superconducting phase in an unprecedented high temperature and may offer a promising material platform to building novel quantum and spintronics devices. The authors’ discovery may enable the experimental measurement of Majorana fermions at high temperature. And it will stimulate future studies of topological phases in a broader range of superconducting materials, such as a honeycomb lattice layered structure.

版權所有 © 中國科學院上海硅酸鹽研究所
地址:上海市長寧區定西路1295號 郵政編碼:200050
小悟空返水