Two-dimensional ferroelectric materials

The dataset of ferroelectric materials mainly comes from high-throughput computing and literature mining, covering multi-level information from first principles calculations to experimental characterization: for example, hundreds of new ferroelectric materials automatically screened based on the Materials Project, nearly 3000 high-quality ferroelectric phase transition data extracted from more than 40000 papers, and fine experimental data for specific materials such as hafnium based, PZT ceramics, and two-dimensional multiferroics.

In recent years, these data have been deeply integrated with deep learning and interactive platforms, giving rise to tools such as FerroAI (generating phase diagrams in 20 seconds with prediction accuracy exceeding 80%) and PTST polarization topology databases, driving ferroelectric research from traditional trial and error to data-driven and intelligent design.