高等学校化学学报 ›› 1998, Vol. 19 ›› Issue (8): 1292.

• 论文 • 上一篇    下一篇

磷酸铝分子筛骨架晶格能与骨架拓扑结构的关系

徐文国1, 李宝宗2, 裘式纶2, 庞文琴2, 徐如人2   

  1. 1. 北京理工大学化工与材料学院, 北京, 100081;
    2. 吉林大学化学系, 无机合成与制备化学开放研究实验室, 长春, 130023
  • 收稿日期:1997-05-13 出版日期:1998-08-24 发布日期:1998-08-24
  • 通讯作者: 庞文琴
  • 作者简介:徐文国,男,37岁,博士,教授.
  • 基金资助:

    国家攀登计划、国家自然科学基金(批准号:29571011)、国家教委博士点基金及吉林大学无机合成与制备化学开放实验室资助课题.

The Relationship Between Framework Topological Structures and Lattice Energies of Aluminophosphate Molecular Sieves

XU Wen-Guo1, LI Bao-Zong2, QIU Shi-Lun2, PANG Wen-Qin2, XU Ru-Ren2   

  1. 1. Institute of Chemical Engineering and Material Science, University of Beijing Technology, Beijing, 100081;
    2. Department of Chemistry, Key Laboratory of Inorg
  • Received:1997-05-13 Online:1998-08-24 Published:1998-08-24

摘要: 应用晶格能极小化技术计算了19种磷酸铝分子筛的骨架晶格能,采用一元线性回归分析、多元线性回归分析和人工神经网络误差反传算法(BP),将其骨架晶格能与骨架拓扑结构参数相关.预测结果与计算晶格能吻合,表明晶格能与配位序(N2~N3)具有良好的多元线性关系.

关键词: 磷酸铝分子筛, 配位序, 拓扑密度, 晶格能, 构效关系

Abstract: By using computer modelling based on the lattice energy minimization technique, the lattice energies of nineteen types of aluminophosphate molecular sieves have been calculated. The studies on the relationship between the topological structures of these molecular sieves and framework lattice energies have been performed by monovariate, multivariate regression analysis and standard back-propagation(BP) algorithm of artifical neural networks. The predicted results by using multiregression analysis are in agreement with the calculated lattice energies and show a good multivariate linear relationship between the coordination sequences(N2~N3) of these molecular sieves and framework lattice energies. The lattice energies of ten types of the topological structures have been predicted.

Key words: Aluminophosphate molecular sieves, Coordination sequences, Topological density, Lattice energy, Structure-activity relationship

中图分类号: 

TrendMD: