相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Materials Science, Ceramics
Collin J. Wilkinson et al.
Summary: Predicting the effects of ceramic microstructures on macroscopic properties, such as Knoop hardness, has been challenging, especially in glass-ceramics with overlapping crystalline and glassy phases. To address this, two computational techniques, computer vision algorithm and machine learning with convolutional neural networks, were employed to predict Knoop hardness based on scanning electron microscopy images. The computer vision algorithm provides physical insights by extracting features from images, while the machine learning method allows for more accurate predictions but lacks transparency. The relative merits of the models are discussed.
JOURNAL OF THE AMERICAN CERAMIC SOCIETY
(2023)
Review
Materials Science, Biomaterials
Oluwatosin David Abodunrin et al.
Summary: Due to their excellent biologically active qualities, bioactive glasses (BGs) are extensively used in the biomedical field to enhance tissue-implant interactions, bone regeneration, and wound healing. Researchers have recently become interested in a borate-based composition (BBG) instead of the traditional silicate-based bioactive glass due to its faster apatite synthesis and potential for tissue integration. However, the low chemical durability and fast degradation of BBG have raised concerns for their utilization in biological and biomedical applications.
JOURNAL OF MATERIALS CHEMISTRY B
(2023)
Review
Engineering, Biomedical
Adam Shearer et al.
Summary: At least 25 bioactive glass medical devices have been approved for clinical use worldwide. These devices have diverse applications, including implants, fillers, agents for dentin hypersensitivity, wound dressing, and cancer therapeutics. Bioactive glasses have evolved in morphology and delivery systems, allowing for greater flexibility and control. This article critically discusses the approved products, regulatory standards, and the future development, applications, and challenges of bioactive glasses.
ACTA BIOMATERIALIA
(2023)
Article
Chemistry, Multidisciplinary
Andrew J. Lew et al.
Summary: Biominerals, the hardest and toughest tissues in living organisms, are often polycrystalline with varying mesostructures. Marine biominerals such as coral skeletons and nacre, consisting of different calcium carbonate polymorphs, share a common characteristic of slight crystal misorientation. Nanoindentation and molecular dynamics simulations show that this slight misorientation can significantly increase fracture toughness. This phenomenon can be utilized in the synthesis of bioinspired materials without the need for multiple materials or specific top-down architecture.
ADVANCED MATERIALS
(2023)
Article
Materials Science, Multidisciplinary
Ravinder Bhattoo et al.
Summary: The researchers developed composition-property models for 25 glass properties using large-scale experimental data and machine learning, and interpreted them using game-theoretic concepts. They found that glass components play different roles in governing the optical, physical, electrical, and mechanical properties of glasses, and these components exhibit interdependence.
Article
Chemistry, Medicinal
Jiuyang Zhao et al.
Summary: Text mining in the optical-materials domain is increasingly important due to the growing number of scientific publications. The proposed OpticalBERT and OpticalPureBERT models outperform BERT and previous models in various text-mining tasks. The released OpticalTable-SQA model significantly outperforms Tapas-SQA on optical-materials-related tables.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2023)
Article
Materials Science, Ceramics
Amreen Jan et al.
Summary: In this study, the behavior of pristine and preirradiated glasses under aqueous corrosion conditions was investigated using experimental and simulation methods. The results showed that preirradiation significantly increased the initial dissolution rate and the depth of the alteration layer. Furthermore, radiation-induced structural changes and enhanced water diffusion were observed.
INTERNATIONAL JOURNAL OF APPLIED GLASS SCIENCE
(2023)
Article
Multidisciplinary Sciences
Tanjin He et al.
Summary: By text-mining 29,900 solid-state synthesis recipes from the scientific literature, we established a knowledge base that enables the recommendation of precursor materials for the synthesis of a novel target material. Our data-driven approach learns the chemical similarity of materials and mimics human synthesis design, achieving a success rate of at least 82% when proposing precursor sets for unseen test target materials.
Article
Materials Science, Ceramics
Daniel R. Cassar
Summary: GlassNet is a multitask deep neural network model trained on a large dataset of various glass compositions. It can accurately predict a wide range of glass properties, and its open-source nature allows for widespread use and improvement. In predicting viscosity, it outperformed another general-purpose model.
CERAMICS INTERNATIONAL
(2023)
Article
Computer Science, Artificial Intelligence
Ravinder Bhattoo et al.
Summary: Traditional approaches require knowledge of abstract quantities to infer particle dynamics, but the Lagrangian graph neural network (LGnn) can learn the Lagrangian directly from trajectories. LGnn outperforms baselines like Lnn in systems with constraints and drag, and has zero-shot generalizability to larger and unseen systems. The graph architecture of LGnn simplifies learning with better performance and less data.
MACHINE LEARNING-SCIENCE AND TECHNOLOGY
(2023)
Review
Chemistry, Multidisciplinary
Lothar Wondraczek et al.
Summary: This article discusses the impact of structure-property relations on the mechanical performance of glass materials, and introduces the methods of designing glass materials using structure-property relations, stress-strain modeling, and machine learning predictions. It also introduces the development direction and potential of novel glass materials.
ADVANCED MATERIALS
(2022)
Article
Materials Science, Ceramics
Mohd Zaki et al.
Summary: This paper demonstrates the use of machine learning models to predict and interpret the optical properties of oxide glasses, showing the application of ML models in compositional control and providing explanations of the model results through Shapely additive explanations (SHAP).
JOURNAL OF THE AMERICAN CERAMIC SOCIETY
(2022)
Article
Materials Science, Multidisciplinary
Indrajeet Mandal et al.
Summary: This study investigates the factors influencing the movement of sodium cations in glasses and their impact on the conductivity. The results show that the mobility of sodium cations plays a significant role in enhancing the ionic conductivity. Additionally, inter-ionic Coulombic interactions and structural modifications also have a significant influence on the mobility and conductivity of sodium cations.
FRONTIERS IN MATERIALS
(2022)
Article
Materials Science, Multidisciplinary
Shweta R. Keshri et al.
Summary: This study improves the crack resistance of glass as a solid electrolyte for sodium-ion batteries through Na+-Ag+ ion exchange while maintaining its ionic conductivity. The study evaluates the structure, crack resistance, and conductivity of the glass using various techniques and concludes through experiments and simulations.
Article
Energy & Fuels
Mohd Zaki et al.
Summary: Recent literature has shown an increasing use of artificial intelligence and machine learning in computational design and discovery of new materials. While most studies focus on inverse design or property prediction, the materials' processing route and testing conditions have received less attention. This paper presents a framework combining text-mining with machine learning to extract processing and testing conditions for improved property prediction. The proposed approach is demonstrated by predicting the hardness of inorganic glasses with composition, annealing temperature, and loading conditions as input. Results show that the combination of process parameters and composition yields superior predictions of hardness, and the processing and testing parameters play a significant role in controlling the hardness of glasses. The proposed approach also outperforms heuristic approaches for glass synthesis.
CHEMICAL ENGINEERING AND PROCESSING-PROCESS INTENSIFICATION
(2022)
Article
Chemistry, Medicinal
Shu Huang et al.
Summary: A large number of scientific papers in battery research pose challenges in extracting useful information efficiently. The BERT models, trained on a large dataset in an unsupervised manner, provide an automatic approach to process the scientific text. Six battery-specific BERT models were developed and outperformed the original BERT models on specific battery tasks.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Materials Science, Ceramics
Yueh-Ting Shih et al.
Summary: Physics and chemistry-informed machine learning models were trained using descriptors in the element physical and chemical properties domain. These models successfully predicted the elastic moduli and temperature dependence of electrical resistivity for oxide glasses, and revealed the relationships between predicted glass properties and elemental features.
JOURNAL OF NON-CRYSTALLINE SOLIDS
(2022)
Article
Multidisciplinary Sciences
Matthias Scheffler et al.
Summary: Achievements in condensed matter physics, chemistry, and materials science greatly influence the prosperity and lifestyle of our society, as new materials are crucial for various sectors. However, the value of the enormous amount of research data produced in these fields can only be realized through the establishment of a FAIR data infrastructure, allowing for comprehensive characterization and sharing of the data to advance science.
Article
Multidisciplinary Sciences
Simon Batzner et al.
Summary: This paper introduces an E(3)-equivariant deep learning method for accelerating molecular dynamics simulations. The method shows state-of-the-art accuracy and remarkable sample efficiency in faithfully describing the dynamics of complex systems. The Neural Equivariant Interatomic Potentials (NequIP) approach employs E(3)-equivariant convolutions to interact with geometric tensors, resulting in a more information-rich and faithful representation of atomic environments. NequIP outperforms existing models with significantly fewer training data, challenging the commonly held belief about the necessity of massive training sets for deep neural networks.
NATURE COMMUNICATIONS
(2022)
Article
Chemistry, Physical
Tanishq Gupta et al.
Summary: This article presents MatSciBERT, a materials-aware language model trained on a large corpus of peer-reviewed materials science publications. MatSciBERT outperforms SciBERT, a language model trained on a science corpus, and achieves state-of-the-art results on three downstream tasks: named entity recognition, relation classification, and abstract classification.
NPJ COMPUTATIONAL MATERIALS
(2022)
Article
Nanoscience & Nanotechnology
Mohd Zaki et al.
Summary: This study utilizes machine learning techniques to develop a model that can predict the sharpness of the atomic force microscope (AFM) probe directly from indentation images. Additionally, explainable machine learning models are employed to interpret the features learned by the model. The research demonstrates that machine learning approaches can accelerate experiments by providing complex information about instrument performance, thereby enhancing the quality of experiments.
SCRIPTA MATERIALIA
(2022)
Article
Computer Science, Artificial Intelligence
Amalie Trewartha et al.
Summary: Efficiently connecting new materials discoveries to established literature can be achieved by using named entity recognition (NER) to extract structured summary-level data from unstructured materials science text. In this study, the performance of four NER models on materials science datasets is compared, and it is found that domain-specific pre-training provides measurable advantages, with the bidirectional long short-term memory (BiLSTM) model consistently outperforming BERT.
Article
Materials Science, Ceramics
Daniel R. Cassar et al.
Summary: The engineering of new glass compositions has shifted from educated trial-and-error to data and simulation-driven strategies. A computer program was developed to combine data-driven predictive models with a genetic algorithm to design glass compositions with desired properties, using large datasets of glass compositions for inducing the predictive models. This new tool shows potential for accelerating the design of novel glasses through machine learning-guided techniques.
CERAMICS INTERNATIONAL
(2021)
Article
Materials Science, Ceramics
Mohammad Amin Golestani Fard et al.
Summary: In this study, a W-ZrC composite coating was successfully applied on a graphite substrate using a novel TIG-aided cladding process. The addition of TiC and Co powders significantly enhanced the mechanical performance of the coating.
CERAMICS INTERNATIONAL
(2021)
Article
Materials Science, Ceramics
Vineeth Venugopal et al.
Summary: Although glasses have been an integral part of human life for more than 2000 years, some fundamental and practical questions about glasses remain unanswered. Recent studies suggest that data-driven techniques such as artificial intelligence and machine learning can provide fresh perspectives to tackle these questions.
INTERNATIONAL JOURNAL OF APPLIED GLASS SCIENCE
(2021)
Article
Chemistry, Physical
Indrajeet Mandal et al.
Summary: The study investigates the effect of substituting NaF for Al2O3 on the conductivity of Na3Al2P3O12 (NAP) glass, revealing a significant impact of NaF concentration on the conductivity of NAP glass, which is highly dependent on a specific temperature range. Raman spectra and impedance spectra show changes in the coordination of Al3+ and the distribution of sodium cations, as well as variations in conductivity with NaF concentration. Time-temperature scaling analysis indicates different ion conduction mechanisms at varying NaF concentrations in the glass samples.
JOURNAL OF ALLOYS AND COMPOUNDS
(2021)
Article
Materials Science, Multidisciplinary
Daniel R. Cassar
Summary: The study developed a physics-informed machine learning model capable of predicting the temperature-dependence of the viscosity of oxide liquids with good extrapolation capabilities.
Article
Materials Science, Ceramics
Sebastien Kerisit et al.
Summary: By developing a patchy particle model of hydrated amorphous silica, researchers have successfully addressed the issue of reproducing the residual rate of glass corrosion in MC simulations. While the model cannot simultaneously replicate the structure and dynamics of the systems of interest with a single set of parameters, it effectively describes the connectivity of hydrated amorphous silica structures and the local coordination geometry of individual species.
JOURNAL OF NON-CRYSTALLINE SOLIDS
(2021)
Article
Materials Science, Ceramics
R. Daniel Cassar et al.
Summary: The study investigated the predictive performance of three machine learning algorithms for six different glass properties, showing that for five properties, the induced predictive models had a comparable uncertainty to the usual data spread, while glasses with extremely low or high values of these properties had significantly higher prediction uncertainty. Furthermore, glasses containing chemical elements poorly represented in the training set yielded higher prediction errors. The analysis of SHAP values indicated key elements affecting the modeled properties, which can help with empirical compositional tuning and computer-aided inverse design of glass formulations.
CERAMICS INTERNATIONAL
(2021)
Article
Materials Science, Multidisciplinary
J. Qi et al.
Summary: This study presents a strategy using moment tensor potentials (MTP5) to improve the accuracy of predicting physical properties of lithium superionic conductors (LSCs), leading to more accurate predictions of ionic conductivities and activation energies. NPT MD simulations of three LSC5 revealed a transition between two quasi-linear Arrhenius regimes at relatively low temperatures, attributed to an increase in the number and diversity of diffusion pathways.
MATERIALS TODAY PHYSICS
(2021)
Article
Materials Science, Ceramics
Rajesh Kumar et al.
Summary: Using molecular dynamics simulations, the study observed that irradiation increases disorder in short- and medium-range order in borosilicate glasses. The volumetric response of borosilicate glasses under irradiation shows a composition-dependent transition, with borate-rich compositions tending to swell and silica-rich glasses tending to densify. The increase in disorder in medium-range order plays a significant role in governing the volumetric changes in the irradiated structure.
JOURNAL OF THE AMERICAN CERAMIC SOCIETY
(2021)
Proceedings Paper
Materials Science, Multidisciplinary
Anshuman Raunak et al.
Summary: This paper deposited and optimized a ZnO layer using the sol-gel technique for piezoelectric applications, discussing the effects of solvents and substrate materials on the deposited piezoelectric ZnO film. The optimized ZnO layer was found useful for MEMS Acoustic Sensors and similar applications.
MATERIALS TODAY-PROCEEDINGS
(2021)
Article
Computer Science, Artificial Intelligence
Vineeth Venugopal et al.
Summary: This framework utilizes natural language processing to extract knowledge from inorganic glasses literature and presents a comprehensive summary of topics and elemental distribution through image and plot analysis.
Article
Computer Science, Artificial Intelligence
Ivan S. Novikov et al.
Summary: This paper focuses on the technology of constructing machine-learning interatomic potentials through active learning in the MLIP package, addressing efficient methods for automatically sampling training sets, the impact of expanding training sets on prediction errors, and cost-effective setup of ab initio calculations. The MLIP package can be downloaded at https://mlip.skoltech.ru/download/.
MACHINE LEARNING-SCIENCE AND TECHNOLOGY
(2021)
Article
Materials Science, Multidisciplinary
Suresh Bishnoi et al.
Summary: This study develops composition-property models for inorganic glasses using a scalable GPR algorithm. The models demonstrate superiority over existing machine learning models and are publicly shared to accelerate glass design.
MATERIALS ADVANCES
(2021)
Article
Engineering, Biomedical
Taihao Han et al.
ACTA BIOMATERIALIA
(2020)
Review
Materials Science, Multidisciplinary
Maziar Montazerian et al.
INTERNATIONAL MATERIALS REVIEWS
(2020)
Article
Chemistry, Medicinal
Edward Kim et al.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2020)
Article
Materials Science, Ceramics
Binghui Deng
JOURNAL OF NON-CRYSTALLINE SOLIDS
(2020)
Article
Materials Science, Multidisciplinary
Ying Shi et al.
Article
Chemistry, Physical
Chuhong Wang et al.
CHEMISTRY OF MATERIALS
(2020)
Article
Materials Science, Ceramics
Joseph N. P. Liilington et al.
JOURNAL OF NON-CRYSTALLINE SOLIDS
(2020)
Article
Physics, Multidisciplinary
V Bapst et al.
Article
Engineering, Mechanical
M. Kang et al.
EXTREME MECHANICS LETTERS
(2020)
Article
Materials Science, Multidisciplinary
Anamul Haq Mir et al.
NPJ MATERIALS DEGRADATION
(2020)
Article
Materials Science, Multidisciplinary
Jaekyun Hwang et al.
SCIENCE AND TECHNOLOGY OF ADVANCED MATERIALS
(2020)
Article
Chemistry, Multidisciplinary
R. Ravinder et al.
MATERIALS HORIZONS
(2020)
Article
Multidisciplinary Sciences
Vahe Tshitoyan et al.
Article
Multidisciplinary Sciences
Hua Tong et al.
NATURE COMMUNICATIONS
(2019)
Article
Materials Science, Multidisciplinary
Evgeny Podryabinkin et al.
Article
Multidisciplinary Sciences
Yue Liu et al.
ADVANCED THEORY AND SIMULATIONS
(2019)
Article
Materials Science, Multidisciplinary
Han Liu et al.
NPJ MATERIALS DEGRADATION
(2019)
Article
Engineering, Biomedical
Nicholas Stone-Weiss et al.
ACTA BIOMATERIALIA
(2018)
Review
Materials Science, Multidisciplinary
John C. Mauro
CURRENT OPINION IN SOLID STATE & MATERIALS SCIENCE
(2018)
Article
Materials Science, Ceramics
N. M. Anoop Krishnan et al.
JOURNAL OF NON-CRYSTALLINE SOLIDS
(2018)
Article
Materials Science, Ceramics
David L. Morse et al.
INTERNATIONAL JOURNAL OF APPLIED GLASS SCIENCE
(2016)
Article
Multidisciplinary Sciences
Mark D. Wilkinson et al.
Article
Construction & Building Technology
Ahmet B. Kizilkanat et al.
CONSTRUCTION AND BUILDING MATERIALS
(2015)
Article
Materials Science, Multidisciplinary
S. Gin et al.
Article
Materials Science, Multidisciplinary
Albert P. Bartok et al.
Article
Materials Science, Ceramics
John D. Vienna et al.
INTERNATIONAL JOURNAL OF APPLIED GLASS SCIENCE
(2013)
Article
Chemistry, Physical
Joerg Behler
JOURNAL OF CHEMICAL PHYSICS
(2011)
Article
Materials Science, Ceramics
Lothar Wondraczek et al.
JOURNAL OF THE EUROPEAN CERAMIC SOCIETY
(2009)
Article
Multidisciplinary Sciences
John C. Mauro et al.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2009)
Article
Engineering, Multidisciplinary
S. A. Silling et al.
JOURNAL OF ELASTICITY
(2007)
Article
Physics, Multidisciplinary
Joerg Behler et al.
PHYSICAL REVIEW LETTERS
(2007)
Review
Physics, Applied
U Ozgür et al.
JOURNAL OF APPLIED PHYSICS
(2005)
Letter
Materials Science, Ceramics
ED Zanotto et al.
JOURNAL OF NON-CRYSTALLINE SOLIDS
(2004)
Review
Materials Science, Ceramics
C Dreyfus et al.
JOURNAL OF NON-CRYSTALLINE SOLIDS
(2003)
