摘要
根据溶剂的５个选择性参数，ｘｅ，ｘｄ，ｘｎ，ｘｔ，ｘｍ，对常用的８１种溶剂应用化学模式识别方法进行了系统分析。各种模式识别方法（包括谱系聚类方法、非谱系聚类方法、主成分分析法和线性判别分析）的分类结果基本一致，且符合Ｓｎｙｄｅｒ应用其中３个参数以三角图形式进行溶剂分类的方法。本文结论有助于色谱分析、萃取和其它分离技术的溶剂选择。
Pattern recognition approach has been used to study the classification of the solvents properties. 81 common solvents have been classified according to their five selectivity parameters,xε, xd, xn, xt and xm. The present work confirms the Snyder's solvent selectivity triangle may be best for intuitive classification of solvents that may convienently be used in chromatography. The pattern recognition technology is another way for this purpose and perhaps is more reasonable. This classification may then be used to aid in the selection of solvents in chromatography, extraction and other separation techniques.
出处
《计算机与应用化学》
CAS
CSCD
1994年第2期150-153,共4页
Computers and Applied Chemistry
关键词
溶剂分类
多变量统计
化学模式识别
Classification of solvents
Multivariate statistical analysis
Chemical pattern recognition
