Researchers from Finland's University of Jyväskylä have combined quantum-mechanical modeling and universal machine learning to reveal how geometry determines the stability of graphene - metallene interfaces - an important step toward bringing these promising 2D materials into real-world technologies.

Integrating density-functional theory and machine-learning to assess the stability of lateral graphene–metallene interfaces. Image credit: University of Jyväskylä
Metallenes are atomically thin, nonlayered metallic sheets with great potential for applications ranging from next-generation electronics to energy storage and catalysis. Yet their strong metallic bonding makes them unstable in isolation, often requiring confinement within the pores of 2D templates such as graphene. To tackle this challenge, Professor Pekka Koskinen’s team conducted a large-scale computational study of 1,080 graphene–metallene interfaces, covering 45 different metals and four interface geometries. Using density-functional theory (DFT) alongside the MatterSim machine-learning interatomic potential, the researchers optimized interface structures, analyzed their electronic properties, and tested their stabilities under strain, defects, and thermal motion.