Journals
  Publication Years
  Keywords
Search within results Open Search
Please wait a minute...
For Selected: Toggle Thumbnails
Revised Homogeneous Nucleation Mechanism of KCl in Aqueous Solutions
ZHANG Meixi, SUN Qiang
Acta Scientiarum Naturalium Universitatis Pekinensis    2025, 61 (4): 687-696.   DOI: 10.13209/j.0479-8023.2025.053
Abstract633)   HTML    PDF(pc) (6080KB)(3448)       Save
Molecular dynamics (MD) simulations on supersaturated KCl solutions are conducted to understand the homogeneous nucleation mechanism of ions in aqueous solution. According to the MD simulation results, dissolved K+ and Cl ions tend to aggregate in solution. When the aggregates reach a critical size, KCl grains may crystallize inside the aggregates. The free energy surface (FES) during the nucleation process, calculated by the METAD method, demonstrates that the process of ion aggregation does not require the overcoming of a potential barrier, which is different from the two-step nucleation theory. Based on the structural analysis, the formation of aggregates may be closely related to hydrophobic effects. Therefore, the revised classical nucleation theory (rev-CNT) is proposed to understand the KCl homogeneous nucleation mechanism in aqueous solution.
Related Articles | Metrics | Comments0
N3LDG: A Lightweight Neural Network Library for Natural Language Processing
WANG Qiansheng, YU Nan, ZHANG Meishan, HAN Zijia, FU Guohong
Acta Scientiarum Naturalium Universitatis Pekinensis    2019, 55 (1): 113-119.   DOI: 10.13209/j.0479-8023.2018.065
Abstract3861)   HTML    PDF(pc) (1161KB)(929)       Save

The authors propose a neural network library N3LDG for natural language processing. N3LDG supports constructing computation graphs dynamically, and organizing executions into batches automatically. Experiments show that N3LDG can efficiently construct and execute computation graphs when training CNN, Bi-LSTM, and Tree-LSTM. When using CPU to train above models, the training speed of N3LDG is better than that of PyTorch. When using GPU to train CNN and Tree-LSTM, N3LDG is better than PyTorch.

Related Articles | Metrics | Comments0
Analysis of Particle Characteristics and Differentiation Processes of Glacial Sediment in Tianshan and Eolian Sediment in the Taklimakan Desert
PAN Renyi,LI Chuanchuan,ZHANG Mei,SUN Caiqi,LIU Gengnian
Acta Scientiarum Naturalium Universitatis Pekinensis