Applied Artificial Intelligence in Materials Science and Material Design
Прикладной искусственный интеллект в материаловедении и дизайне материалов
2025-03-02
SCID: 54.1/h5jhsv82
Discuss with AI
artificial intelligencemachine learningmaterial designmaterials discoverymaterials science
Figures from the paper
Abstract (AI)
Materials science has traditionally relied on a combination of experimental techniques and theoretical modeling to discover and develop new materials with desired properties. However, these processes can be time‐consuming, resource‐intensive, and often limited by the complexity of material systems. The advent of artificial intelligence (AI), particularly machine learning, has revolutionized materials science by offering powerful tools to accelerate the discovery, design, and characterization of novel materials. AI not only enhances the predictive modeling of material properties but also streamlines data analysis in techniques like X‐Ray diffraction, Raman spectroscopy, scanning probe microscopy, and electron microscopy. By leveraging large datasets, AI algorithms can identify patterns, reduce noise, and predict material behavior with unprecedented accuracy. In this review, recent advancements in AI applications across various domains of materials science, including spectroscopy, synchrotron studies, scanning probe and electron microscopies, metamaterials, atomistic modeling, molecular design, and drug discovery, are highlighted. It is discussed how AI‐driven methods are reshaping the field, making material discovery more efficient, and paving the way for breakthroughs in material design and real‐time experimental analysis.
Key Findings
1
AI and machine learning accelerate materials discovery, design, and characterization compared with conventional experimental and theoretical workflows.
2
AI improves prediction of material properties by extracting patterns from large datasets and modeling complex material systems.
3
AI streamlines analysis of X-ray diffraction, Raman spectroscopy, scanning probe microscopy, and electron microscopy by identifying patterns and reducing noise.
4
AI-driven methods enable more efficient material discovery and support real-time experimental analysis, potentially advancing material design.
5
Applications span spectroscopy, synchrotron studies, microscopy, metamaterials, atomistic modeling, molecular design, and drug discovery.
Research Object
materials science and material design systems, including novel materials and their characterization data
Research Subject
AI-driven discovery, design, characterization, property prediction, and real-time analysis of materials
Publication Details
Publication Date
2025-03-02
Journal
Publisher
ISSN
Cited by
75
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai12
ImageNet classification with deep convolutional neural networks2017
Highly accurate protein structure prediction with AlphaFold2021
"Why Should I Trust You?": Explaining the Predictions of Any Classifier2016
Accurate structure prediction of biomolecular interactions with AlphaFold 32024
A Comprehensive Survey on Graph Neural Networks2020
Graph neural networks: A review of methods and applications2020
Machine learning in materials informatics: recent applications and prospects2017
Perspective: Fifty years of density-functional theory in chemical physics2014
An autonomous laboratory for the accelerated synthesis of inorganic materials2023
Deep-Learning-Enabled On-Demand Design of Chiral Metamaterials2018
Augmenting large language models with chemistry tools2024
Symmetry prediction and knowledge discovery from X-ray diffraction patterns using an interpretable machine learning approach2020