Machine-learning accelerated prediction of two-dimensional conventional superconductors.

Area of Science:

• Materials Science
• Condensed Matter Physics
• Computational Chemistry

Background:

• Two-dimensional (2D) materials are a frontier in condensed matter physics.
• Superconductivity in 2D systems is of great scientific interest.
• Discovering new superconductors is crucial for technological advancement.

Purpose of the Study:

• To conduct a large-scale search for novel two-dimensional (2D) superconductors.
• To leverage computational methods and machine learning for accelerated discovery.
• To identify 2D materials with high critical temperatures (Tc) and thermodynamic stability.

Main Methods:

• Electron-phonon calculations using density-functional perturbation theory.
• Machine learning models for predictive screening.
• High-throughput screening of over 140,000 2D compounds from the Alexandria database.

Main Results:

• Identified a multitude of 2D superconductors with diverse chemistries and structures.
• Found that 2D materials generally have stronger electron-phonon coupling but lower average phonon frequencies than 3D counterparts.
• Discovered 105 2D systems with critical temperatures (Tc) > 5 K, including CuH2, NbN, and V2NS2.

Conclusions:

• Computational databases and machine learning are vital for accelerating superconductor discovery.
• Several novel 2D superconductors exhibit high Tc and thermodynamic stability, suitable for experimental synthesis and applications.
• The study expands the landscape of known 2D superconductors and provides candidates for further research.