| 陈思源,虞育杰,韩清清,毕浩,吴平,黄睿.化学通报,2025,88(5):519-526,534. |
| 分子动力学在新型溶剂设计优化中的应用 |
| Application of molecular dynamics in the design and optimization of novel solvents |
| 投稿时间：2024-11-26 修订日期：2024-12-27 |
| DOI： |
| 中文关键词: 新型溶剂 分子动力学 机器学习 设计 萃取 |
| 英文关键词:novel solvents molecular dynamics simulations machine learning design extraction |
| 基金项目:国家自然科学基金项目(52266006,52366008)、贵州省科技计划项目(黔科合基础-ZK[2022]一般061,ZK[2022]一般139)、贵州省教育厅青年科技人才成长项目(黔教技[2022]108号)和贵州大学培育计划项目(贵大培育[2020]20号)资助 |
|
| 摘要点击次数: 78 |
| 全文下载次数: 56 |
| 中文摘要: |
| 工业生产中频繁使用石油基有机溶剂,对环境、人体健康等带来负面影响。新型绿色溶剂如离子液体(Ionic Liquids,ILs)、超临界流体(Supercritical Fluids,SCFs)和深共熔溶剂(Deep Eutectic Solvents,DESs),作为传统有机溶剂的替代品,逐渐受到人们的广泛关注。而在实际应用中,现有的表征技术如核磁共振、红外光谱等,尚无法完全揭示分子间的微观作用机理。分子动力学模拟作为一种能从微观层面观测分子行为的方法在新型溶剂设计优化中得到广泛应用。本文对分子动力学的基本流程、常用力场、积分算法和机器学习算法进行了简单概述,重点介绍了分子动力学模拟在设计新型溶剂萃取过程的微观机理研究,并通过耦合机器学习的方法快速优化溶剂萃取性能,达到高效萃取目标化合物目的。最后总结了分子动力学模拟在受力场精度和计算资源等因素的限制下,仍存在精度有待提升的问题,并对新型溶剂萃取的未来发展方向进行展望。 |
| 英文摘要: |
| The frequent use of petroleum-based organic solvents in industrial production has a negative impact on the environment and human health. New green solvents such as ionic liquids(ILs), supercritical fluids(SCFs) and deep eutectic solvents(DESs) are gaining widespread attention as alternatives to traditional organic solvents. In practical applications, the existing characterization techniques, such as nuclear magnetic resonance and infrared spectroscopy, are not yet able to fully reveal the intermolecular microscopic interaction mechanism. Molecular dynamics simulation, as a method that enables the observation of molecular behavior at the microscopic level, has been extensively utilized in the design and optimization of novel solvents. This article provides a brief overview of the basic procedures, commonly used force fields, and integration algorithms in molecular dynamics simulations, with a focus on the application of molecular dynamics in the study of the micro-mechanisms of novel solvent extraction processes. Additionally, it highlights the rapid optimization of solvent extraction performance by coupling molecular dynamics with machine learning methods, aiming to efficiently extract target compounds. Finally, it is summarized that the molecular dynamics simulation still has the problem of accuracy to be improved under the limitation of factors such as the accuracy of the force field and computational resources, and the future development direction of the novel solvent extraction is prospected. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|