Machine learning interatomic potentials (MLIPs) can treat large-scale material systems with near first-principles accuracy and have been widely used to accelerate point-defect simulations. However, the training of MLIPs usually relies on large amounts of DFT data. This issue is particularly pronounced for charged defec...
Zhen-Xing Dai, Ming Ni, Xin-Peng Li et al.· 0 citations
The integration of two‐dimensional (2D) semiconductors into large‐scale digital circuits demands precise, localized, and damage‐free doping to achieve threshold voltage (VTH) matching—a critical challenge for existing technologies. Here, we present a non‑destructive and precisely controllable n‑type doping strategy bas...
Floating-gate memory, as one of the most widely used nonvolatile memories for practical deployments, plays a significant role in modern electronics. However, its inherent speed limitations make it challenging to keep pace with the explosive growth of data generation and the evolving demands of innovative hardware appli...
Zheng Zhang, Kang Wang, Jianguo Dong et al.· Nano letters (Print)· 0 citations
This work proposes a machine-learning charge density route for predicting defect formation energies with higher data efficiency and shows that charge-density learning enables more robust cross-size transfer than direct energy-force fitting and that mixed-size data design can substantially reduce the cost of defect pred...