Overview:
This project explores the properties of approximately 2,000 one-dimensional materials using density functional theory (DFT). The focus is on calculating their structural, thermodynamic, electronic, magnetic, and optical properties, aiming to discover novel stable materials with potential applications.
Data Sources:
Materials were identified through extensive screening of experimental databases, creating a "core" set of 1D materials. This was expanded by substituting atoms to form "shell" materials, with further new structures generated using a generative neural network model .
Tools and Technologies Used:
We used DFT for property calculations and employed machine learning to generate new materials. A generative neural network model was trained on core and shell materials to produce additional structures and predict their stability.
Key Findings:
Hundreds of stable one-dimensional materials were identified, including novel classes generated by machine learning. The materials showed potential in applications such as Majorana-bound states, photovoltaics, and gas atom binding.
Learn More:
For a detailed exploration of this project—including methodologies, results, and implications—please refer to the database, first paper, and second paper.