Premium
A new selection principle for model compounds in quantitative structure–retention relationship by HPLC for the determination of n ‐octanol/water partition coefficients of bisphenols
Author(s) -
Han Shuying,
Song Yilin,
Guan Hanyu,
Chen Tao,
Chi Yumei,
Deng Haishan
Publication year - 2018
Publication title -
separation science plus
Language(s) - English
Resource type - Journals
ISSN - 2573-1815
DOI - 10.1002/sscp.201800041
Subject(s) - partition coefficient , logarithm , correlation coefficient , chemistry , octanol , high performance liquid chromatography , chromatography , quantitative structure–activity relationship , partition (number theory) , binary number , mathematics , analytical chemistry (journal) , statistics , stereochemistry , mathematical analysis , combinatorics , arithmetic
The n ‐octanol/water partition coefficient ( K ow ) is an important parameter for evaluating the migration rules of pollutants in food, as well as assessing their risks to environment and biology. A high‐performance liquid chromatography method based on the quantitative structure–retention relationship was used for K ow determination in this work. The log k w – S linear relationship was established between the slope S of the Snyder–Soczewinski equation, and its intercept, which represents the logarithm of the high‐performance liquid chromatography retention factor of the solute corresponding to 100% water as the mobile phase (log k w ) in methanol/water binary system. As a criterion of material structure similarity, the linearity of the log k w – S relationship was used to assess the rationality of the selected model compounds. Based on this principle, 18 analogues with reliable experimental K ow data were chosen as the quantitative structure–retention relationship model compounds for establishing linear relationship between the logarithm of apparent n ‐octanol/water partition coefficient (log K ow ″) and log k w . A good correlation coefficient ( R 2 = 0.976), as well as satisfactory results in internal and external validation (the cross‐validated correlation coefficient R 2 cv of 0.971, and relative error ≤8.8% for nine validation compounds) indicated the reliability and accuracy of the proposed quantitative structure–retention relationship model to predict the K ow . This model was applied to measure the K ow of 11 bisphenols, and the predicted data were compared with software‐calculated values. The result suggests that the method for K ow determination developed in this work is accurate and provides a new protocol to determine the K ow accurately. To the best of our knowledge, this is the first report on measured K ow data for these bisphenols.