In materials science research, including the search for new materials for advanced computing, artificial intelligence (AI) has proven to be a valuable tool. Researchers can accelerate the discovery process and explore vast material design spaces more efficiently by leveraging AI techniques such as machine learning and data mining.
Using cutting-edge artificial intelligence (AI) tools, researchers discovered new van der Waals (vdW) magnets. Using semi-supervised learning, the team identified transition metal halide vdW materials with large magnetic moments that are predicted to be chemically stable. These two-dimensional (2D) vdW magnets could be used in data storage, spintronics, and even quantum computing.
A team of researchers led by Trevor David Rhone, assistant professor in the Department of Physics, Applied Physics, and Astronomy at Rensselaer Polytechnic Institute, identified novel van der Waals (vdW) magnets using cutting-edge artificial intelligence (AI) tools. Using semi-supervised learning, the team identified transition metal halide vdW materials with large magnetic moments that are predicted to be chemically stable. These two-dimensional (2D) vdW magnets could be used in data storage, spintronics, and even quantum computing.
Rhone is an expert in using materials informatics to discover new materials with unexpected properties that advance science and technology. Materials informatics is a new field of study that combines artificial intelligence and materials science. The cover of Advanced Theory and Simulations recently featured his team’s most recent research.
Our framework can easily be applied to investigate materials with different crystal structures. Mixed crystal structure prototypes, such as a data set of both transition metal halides and transition metal trichalcogenides, can also be explored with this framework.
Trevor David Rhone
2D materials, which can be as thin as a single atom, were discovered in 2004 and have piqued scientists’ interest due to their unexpected properties. 2D magnets are significant because their long-range magnetic ordering remains even when reduced to one or a few layers. Magnetic anisotropy is to blame for this. The interaction between magnetic anisotropy and low dimensionality may result in exotic spin degrees of freedom, such as spin textures, that can be used in the development of quantum computing architectures. 2D magnets have a wide range of electronic properties and can be used in high-performance, low-energy devices.
Rhone and colleagues combined high-throughput density functional theory (DFT) calculations to determine the properties of vdW materials with AI to implement a type of machine learning known as semi-supervised learning. Semi-supervised learning identifies patterns in data and makes predictions by combining labeled and unlabeled data. Semi-supervised learning addresses a major issue in machine learning: a lack of labeled data.
“Using AI saves time and money,” said Rhone. “The typical materials discovery process requires expensive simulations on a supercomputer that can take months. Lab experiments can take even longer and can be more expensive. An AI approach has the potential to speed up the materials discovery process.”
An AI model that could predict the properties of many thousands of material candidates in milliseconds on a laptop was trained using an initial subset of 700 DFT calculations on a supercomputer. The researchers then discovered promising candidate vdW materials with large magnetic moments and low formation energy. Low formation energy is an indication of chemical stability, which is essential for laboratory synthesis and subsequent industrial applications.
“Our framework can easily be applied to investigate materials with different crystal structures,” Rhone said. “Mixed crystal structure prototypes, such as a data set of both transition metal halides and transition metal trichalcogenides, can also be explored with this framework.”
“Dr. Rhone’s application of AI to the field of materials science continues to produce exciting results,” said Curt Breneman, dean of the School of Science at Rensselaer. “Not only has he accelerated our understanding of novel 2D materials, but his findings and methods are likely to contribute to new quantum computing technologies.”
