使用图形神经网络预测和解释离子液体的点
Haijun Feng1, Lanlan Qin2, Bingxuan Zhang1
1School of Computer Sciences, Shenzhen Institute of Information Technology, Shenzhen, Guangdong 518172, China.
ACS omega
|April 15, 2024
概括
预测离子液体的点对于应用至关重要. 图形神经网络 (GNN) 准确地预测这些特性,识别原子的贡献,以指导向的离子液体设计.
科学领域:
- 材料科学 材料科学 材料科学
- 计算化学计算化学
- 化学工程是化学工程的重要组成部分.
背景情况:
- 离子液体 (ILs) 是多功能材料,在化学工程,能源和环境科学中广泛应用.
- 点 (MPs) 是关键的IL属性,但实验性确定是昂贵和耗时的.
- 准确预测IL MPs对于设计具有所需特征的新型ILs至关重要.
研究的目的:
- 开发和比较用于预测离子液体点的机器学习模型.
- 研究基于描述符的机器学习 (DBML) 和图形神经网络 (GNN) 方法的有效性.
- 建立一个可解释的GNN模型,以了解对IL国会议员的原子论贡献.
主要方法:
- 提出了使用分子指纹的三种基于描述器的机器学习 (DBML) 模型.
- 开发了八个使用分子图表表示的图形神经网络 (GNN) 模型.
- 使用诸如根-平均-平方误差 (RMSE),平均绝对误差 (MAE) 和相关系数 (R) 等指标评估模型性能.
主要成果:
- 在预测IL点方面,GNN模型的表现优于DBML模型.
- 图形卷积模型取得了最好的表现 (RMSE = 37.06,MAE = 28.79,R = 0.76).
- 一个可解释的GNN模型确定了增加MP的特定功能组 (例如,氨基,S+,N+,P+),以及减少MP的其他功能组 (例如,素,S-,N-).
结论:
- 机器学习,特别是GNN,为预测离子液体点提供了一种高效和准确的方法.
- 开发的可解释模型为ILs提供了对结构-财产关系的宝贵见解.
- 这种数据驱动的方法有助于快速选和合成具有所需化点的向离子液体.
更多相关视频
相关概念视频
Molecular and Ionic Solids
17.1K
Crystalline solids are divided into four types: molecular, ionic, metallic, and covalent network based on the type of constituent units and their interparticle interactions.
Molecular Solids
Molecular crystalline solids, such as ice, sucrose (table sugar), and iodine, are solids that are composed of neutral molecules as their constituent units. These molecules are held together by weak intermolecular forces such as London dispersion forces, dipole-dipole interactions, or hydrogen bonds, which...
Molecular Solids
Molecular crystalline solids, such as ice, sucrose (table sugar), and iodine, are solids that are composed of neutral molecules as their constituent units. These molecules are held together by weak intermolecular forces such as London dispersion forces, dipole-dipole interactions, or hydrogen bonds, which...
17.1K
Comparing Intermolecular Forces: Melting Point, Boiling Point, and Miscibility
44.3K
Intermolecular forces are attractive forces that exist between molecules. They dictate several bulk properties, such as melting points, boiling points, and solubilities (miscibilities) of substances. Molar mass, molecular shape, and polarity affect the strength of different intermolecular forces, which influence the magnitude of physical properties across a family of molecules.
Temporary attractive forces like dispersion are present in all molecules, whether they are polar or nonpolar. They...
Temporary attractive forces like dispersion are present in all molecules, whether they are polar or nonpolar. They...
44.3K
Phase Transitions: Melting and Freezing
12.4K
Heating a crystalline solid increases the average energy of its atoms, molecules, or ions, and the solid gets hotter. At some point, the added energy becomes large enough to partially overcome the forces holding the molecules or ions of the solid in their fixed positions, and the solid begins the process of transitioning to the liquid state or melting. At this point, the temperature of the solid stops rising, despite the continual input of heat, and it remains constant until all of the solid is...
12.4K
Types of Chemical Bonds
75.8K
Chemical bonding theories were pioneered by American chemist Gilbert N. Lewis. He developed a model called the Lewis model to explain the type and formation of different bonds. Chemical bonding is central to chemistry; it explains how atoms or ions bond together to form molecules. It explains why some bonds are strong and others are weak, or why one carbon bonds with two oxygens and not three; why water is H2O and not H4O.
75.8K
Bonding in Metals
47.3K
Metallic bonds are formed between two metal atoms. A simplified model to describe metallic bonding has been developed by Paul Drüde called the “Electron Sea Model”.
47.3K
Intermolecular Forces and Physical Properties
20.8K
20.8K


