Density-of-states similarity descriptor for unsupervised learning from materials data DOI Creative Commons
Martin Kubáň, Santiago Rigamonti, Markus Scheidgen

et al.

Scientific Data, Journal Year: 2022, Volume and Issue: 9(1)

Published: Oct. 22, 2022

We develop a materials descriptor based on the electronic density-of-states (DOS) and investigate similarity of it. As an application example, we study Computational 2D Materials Database (C2DB) that hosts thousands two-dimensional with their properties calculated by density-functional theory. Combining our clustering algorithm, identify groups similar structure. introduce additional descriptors to characterize these clusters in terms crystal structures, atomic compositions, configurations members. This allows us rationalize found (dis)similarities perform automated exploratory confirmatory analysis C2DB data. From this analysis, find majority consist isoelectronic sharing symmetry, but also outliers, i.e., whose cannot be explained way.

Language: Английский

When Machine Learning Meets 2D Materials: A Review DOI Creative Commons
Bin Lu, Yuze Xia,

Yuqian Ren

et al.

Advanced Science, Journal Year: 2024, Volume and Issue: 11(13)

Published: Jan. 26, 2024

Abstract The availability of an ever‐expanding portfolio 2D materials with rich internal degrees freedom (spin, excitonic, valley, sublattice, and layer pseudospin) together the unique ability to tailor heterostructures made by in a precisely chosen stacking sequence relative crystallographic alignments, offers unprecedented platform for realizing design. However, breadth multi‐dimensional parameter space massive data sets involved is emblematic complex, resource‐intensive experimentation, which not only challenges current state art but also renders exhaustive sampling untenable. To this end, machine learning, very powerful data‐driven approach subset artificial intelligence, potential game‐changer, enabling cheaper – yet more efficient alternative traditional computational strategies. It new paradigm autonomous experimentation accelerated discovery machine‐assisted design functional heterostructures. Here, study reviews recent progress such endeavors, highlight various emerging opportunities frontier research area.

Language: Английский

Citations

47
Application of valence-variable transition-metal-oxide-based nanomaterials in electrochemical analysis: A review DOI
Huan Xu, Qiuyu Wang, Min Jiang

et al.

Analytica Chimica Acta, Journal Year: 2024, Volume and Issue: 1295, P. 342270 - 342270

Published: Jan. 24, 2024

Language: Английский

Citations

20
Machine learning-accelerated discovery and design of electrode materials and electrolytes for lithium ion batteries DOI

Guangsheng Xu,

Mingxi Jiang, Jinliang Li

et al.

Energy storage materials, Journal Year: 2024, Volume and Issue: 72, P. 103710 - 103710

Published: Aug. 13, 2024

Language: Английский

Citations

18
Recent Advances in Machine Learning‐Assisted Multiscale Design of Energy Materials DOI Creative Commons
Bohayra Mortazavi

Advanced Energy Materials, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 10, 2024

Abstract This review highlights recent advances in machine learning (ML)‐assisted design of energy materials. Initially, ML algorithms were successfully applied to screen materials databases by establishing complex relationships between atomic structures and their resulting properties, thus accelerating the identification candidates with desirable properties. Recently, development highly accurate interatomic potentials generative models has not only improved robust prediction physical but also significantly accelerated discovery In past couple years, methods have enabled high‐precision first‐principles predictions electronic optical properties for large systems, providing unprecedented opportunities science. Furthermore, ML‐assisted microstructure reconstruction physics‐informed solutions partial differential equations facilitated understanding microstructure–property relationships. Most recently, seamless integration various platforms led emergence autonomous laboratories that combine quantum mechanical calculations, language models, experimental validations, fundamentally transforming traditional approach novel synthesis. While highlighting aforementioned advances, existing challenges are discussed. Ultimately, is expected fully integrate atomic‐scale simulations, reverse engineering, process optimization, device fabrication, empowering system design. will drive transformative innovations conversion, storage, harvesting technologies.

Language: Английский

Citations

17
Transfer learning enhanced water-enabled electricity generation in highly oriented graphene oxide nanochannels DOI Creative Commons
Ce Yang, Haiyan Wang, Jiaxin Bai

et al.

Nature Communications, Journal Year: 2022, Volume and Issue: 13(1)

Published: Nov. 10, 2022

Harvesting energy from spontaneous water flow within artificial nanochannels is a promising route to meet sustainable power requirements of the fast-growing human society. However, large-scale nanochannel integration and multi-parameter coupling restrictive influence on electric generation are still big challenges for macroscale applications. In this regard, long-range (1 20 cm) ordered graphene oxide assembled framework with integrated 2D have been fabricated by rotational freeze-casting method. The structure can promote absorption directional transmission inside channels generate considerable energy. A transfer learning strategy implemented address complicated multi-parameters problem under limited experimental data, which provides highly accurate performance optimization efficiently guides design enabled generators. generator unit produce ~2.9 V voltage or ~16.8 μA current in controllable manner. High output ~12 ~83 realized connecting several devices series parallel. Different electricity systems developed directly commercial electronics like LED arrays display screens, demonstrating material's potential development clean

Language: Английский

Citations

46
Computational Simulation for Breakdown and Energy Storage Performances with Optimization in Polymer Dielectrics DOI
Yue Dong, Jinghua Yin, Wenchao Zhang

et al.

Advanced Functional Materials, Journal Year: 2023, Volume and Issue: 33(30)

Published: April 28, 2023

Abstract The breakthrough of energy storage technology will enable distribution and adaptation across space‐time, which is revolutionary for the generation energy. Optimizing performance polymer dielectrics remains challenging via physical process electrical breakdown in solid hard to be intuitively obtained. In this review article, application computational simulation technologies summarized energy‐storage effect control variables design structures on material properties with an emphasis dielectric are highlighted. prediction evaluation by combining various data analysis methods reviewed. Finally, outlook challenges discussed based their current developments. This article covers not only overview state‐of‐the‐art advances modeling but also prospects that provide a new knob synthesize high research paradigm.

Language: Английский

Citations

37
MatGPT: A Vane of Materials Informatics from Past, Present, to Future DOI
Zhilong Wang, An Chen, Kehao Tao

et al.

Advanced Materials, Journal Year: 2023, Volume and Issue: 36(6)

Published: Oct. 10, 2023

Abstract Combining materials science, artificial intelligence (AI), physical chemistry, and other disciplines, informatics is continuously accelerating the vigorous development of new materials. The emergence “GPT (Generative Pre‐trained Transformer) AI” shows that scientific research field has entered era intelligent civilization with “data” as basic factor “algorithm + computing power” core productivity. continuous innovation AI will impact cognitive laws methods, reconstruct knowledge wisdom system. This leads to think more about informatics. Here, a comprehensive discussion models infrastructures provided, advances in discovery design are reviewed. With rise paradigms triggered by “AI for Science”, vane informatics: “MatGPT”, proposed technical path planning from aspects data, descriptors, generative models, pretraining directed collaborative training, experimental robots, well efforts preparations needed develop generation informatics, carried out. Finally, challenges constraints faced discussed, order achieve digital, intelligent, automated construction joint interdisciplinary scientists.

Language: Английский

Citations

30
Advanced theoretical modeling methodologies for electrocatalyst design in sustainable energy conversion DOI Creative Commons
Tianyi Wang, Qilong Wu, Yun Han

et al.

Applied Physics Reviews, Journal Year: 2025, Volume and Issue: 12(1)

Published: Feb. 6, 2025

Electrochemical reactions are pivotal for energy conversion and storage to achieve a carbon-neutral sustainable society, optimal electrocatalysts essential their industrial applications. Theoretical modeling methodologies, such as density functional theory (DFT) molecular dynamics (MD), efficiently assess electrochemical reaction mechanisms electrocatalyst performance at atomic levels. However, its intrinsic algorithm limitations high computational costs large-scale systems generate gaps between experimental observations calculation simulation, restricting the accuracy efficiency of design. Combining machine learning (ML) is promising strategy accelerate development electrocatalysts. The ML-DFT frameworks establish accurate property–structure–performance relations predict verify novel electrocatalysts' properties performance, providing deep understanding mechanisms. ML-based methods also solution MD DFT. Moreover, integrating ML experiment characterization techniques represents cutting-edge approach insights into structural, electronic, chemical changes under working conditions. This review will summarize DFT current application status design in various conversions. underlying physical fundaments, advancements, challenges be summarized. Finally, future research directions prospects proposed guide revolution.

Language: Английский

Citations

1
Machine Learning for Harnessing Thermal Energy: From Materials Discovery to System Optimization DOI
Man Li,

Lingyun Dai,

Yongjie Hu

et al.

ACS Energy Letters, Journal Year: 2022, Volume and Issue: 7(10), P. 3204 - 3226

Published: Sept. 2, 2022

Recent advances in machine learning (ML) have impacted research communities based on statistical perspectives and uncovered invisibles from conventional standpoints. Though the field is still early stage, this progress has driven thermal science engineering to apply such cutting-edge toolsets for analyzing complex data, unraveling abstruse patterns, discovering non-intuitive principles. In work, we present a holistic overview of applications future opportunities ML methods crucial topics energy research, bottom-up materials discovery top-down system design across atomistic levels multi-scales. particular, focus spectrum impressive endeavors investigating state-of-the-art transport modeling (density functional theory, molecular dynamics, Boltzmann equation), different families (semiconductors, polymers, alloys, composites), assorted aspects properties (conductivity, emissivity, stability, thermoelectricity), prediction optimization (devices systems). We discuss promises challenges current approaches provide directions new algorithms that could make further impacts research.

Language: Английский

Citations

33
Exploring and machine learning structural instabilities in 2D materials DOI Creative Commons
Simone Manti, Mark Kamper Svendsen,

Nikolaj R. Knøsgaard

et al.

npj Computational Materials, Journal Year: 2023, Volume and Issue: 9(1)

Published: March 4, 2023

Abstract We address the problem of predicting zero-temperature dynamical stability (DS) a periodic crystal without computing its full phonon band structure. Here we report evidence that DS can be inferred with good reliability from frequencies at center and boundary Brillouin zone (BZ). This analysis represents validation test employed by Computational 2D Materials Database (C2DB). For 137 dynamically unstable crystals, displace atoms along an mode relax procedure yields stable in 49 cases. The elementary properties these new structures are characterized using C2DB workflow, it is found their differ significantly those original e.g., gaps opened 0.3 eV on average. All available C2DB. Finally, train classification model data for 3295 materials representation encoding electronic structure crystal. obtain excellent receiver operating characteristic (ROC) curve area under (AUC) 0.90, showing drastically reduce computational efforts high-throughput studies.

Language: Английский

Citations

18