When Machine Learning Meets 2D Materials: A Review.

Area of Science:

• Materials Science
• Condensed Matter Physics
• Computational Chemistry

Background:

• Two-dimensional (2D) materials offer tunable properties through their rich internal degrees of freedom and heterostructure design.
• The vast parameter space in 2D material research necessitates advanced methodologies beyond traditional experimentation.
• Precise control over stacking sequence and crystallographic alignment is key for designing novel 2D heterostructures.

Purpose of the Study:

• To review the application of machine learning (ML) in accelerating the discovery and design of 2D materials and heterostructures.
• To highlight the potential of ML as a cost-effective and efficient alternative to conventional research methods.
• To discuss the current progress, challenges, and future opportunities in ML-driven 2D materials research.

Main Methods:

• Review of recent literature on machine learning applications in 2D materials science.
• Analysis of data-driven approaches for exploring multi-dimensional parameter spaces.
• Discussion of autonomous experimentation paradigms enabled by artificial intelligence.

Main Results:

• Machine learning provides a powerful data-driven approach to overcome the limitations of traditional experimental and computational strategies.
• ML enables efficient exploration of vast datasets, facilitating the design of functional 2D materials and heterostructures.
• The integration of ML with autonomous experimentation promises accelerated discovery cycles.

Conclusions:

• Machine learning represents a paradigm shift in materials science, enabling 'materials by design' for 2D systems.
• ML-driven approaches are crucial for navigating the complexity and scale of modern 2D materials research.
• Future research should focus on further developing and integrating ML tools for autonomous discovery and design of advanced 2D materials.