Premium
LogP measurement of a highly hydrophobic properfume: evaluation of extrapolation of RP‐HPLC results and impact of chemical moieties on accuracy
Author(s) -
Begnaud F.,
Larcinese J.P.,
Fankhauser P.,
Maddalena U.
Publication year - 2016
Publication title -
flavour and fragrance journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.393
H-Index - 70
eISSN - 1099-1026
pISSN - 0882-5734
DOI - 10.1002/ffj.3309
Subject(s) - chemistry , partition coefficient , alkyl , octanol , extrapolation , moiety , thioether , linear regression , high performance liquid chromatography , reversed phase chromatography , chromatography , computational chemistry , stereochemistry , organic chemistry , mathematical analysis , mathematics , machine learning , computer science
Abstract The n‐octanol/water partition coefficient (LogP) of a chemical substance is a key parameter for assessing its potential (eco)‐toxicological effects and environmental fate. To measure the LogP of properfume Haloscent®D (expected LogP > 9), no direct measurement technique can be applied and the reliability of in silico models is impaired by the absence of robust datasets in the high LogP range. Indirect method using reversed‐phase‐HPLC (RP‐HPLC) is promising but we demonstrate that the lack of suitable standards lead to significant discrepancy and poor accuracy. Extrapolation of the correlation between the RP‐HPLC retention factor (logk) of low‐carbon chain homologues and their LogP measured by slow‐stir method has been studied. To select the most suitable standards, the impact of specific moiety representative of Haloscent®D (linear alkyl chain, delta‐damascone moiety, thioether function) on the logk vs. estimated LogP has been checked on different molecules and it was demonstrated that the alkyl chain had the most impact on this correlation. The stability in the molecular conformation of Haloscent®D homologue series was also verified to ensure that no folding occurred as from a certain alkyl chain length, and that consequently extrapolation is feasible up to Haloscent®D. Accuracy of the extrapolated LogP value was calculated from the prediction interval of the logk vs. LogP regression. The LogP of the Haloscent®D was concluded to be 9.7 ± 0.5. Copyright © 2016 John Wiley & Sons, Ltd.