Predicting the Surface Tension of Liquids: Comparison of Four Modeling Approaches and Application to Cosmetic Oils

The efficiency of four modeling approaches, namely, group contributions, corresponding-states principle, σ-moment-based neural networks, and graph machines, are compared for the estimation of the surface tension (ST) of 269 pure liquid compounds at 25 °C from their molecular structure. This study fo...

Full description

Saved in:
Bibliographic Details
Published inJournal of chemical information and modeling Vol. 57; no. 12; pp. 2986 - 2995
Main Authors Goussard, Valentin, Duprat, François, Gerbaud, Vincent, Ploix, Jean-Luc, Dreyfus, Gérard, Nardello-Rataj, Véronique, Aubry, Jean-Marie
Format Journal Article
LanguageEnglish
Published United States American Chemical Society 26.12.2017
Subjects
Chemical and Process Engineering
Chemical engineering
Chemical Sciences
Downloading
Engineering Sciences
Fluids
Liquids
Mathematical models
Modelling
Molecular structure
Neural networks
Oils & fats
Predictions
Surface tension
QSPR
Group contribution-based methods
Neural networks
QSAR
Ab initio modeling
Machine learning
Surface tension
Graph machines
COSMO-RS
Online AccessGet full text

Cover

More Information
Summary:The efficiency of four modeling approaches, namely, group contributions, corresponding-states principle, σ-moment-based neural networks, and graph machines, are compared for the estimation of the surface tension (ST) of 269 pure liquid compounds at 25 °C from their molecular structure. This study focuses on liquids containing only carbon, oxygen, hydrogen, or silicon atoms since our purpose is to predict the surface tension of cosmetic oils. Neural network estimations are performed from σ-moment descriptors as defined in the COSMO-RS model, while methods based on group contributions, corresponding-states principle, and graph machines use 2D molecular information (SMILES codes). The graph machine approach provides the best results, estimating the surface tensions of 23 cosmetic oils, such as hemisqualane, isopropyl myristate, or decamethylcyclopentasiloxane (D5), with accuracy better than 1 mN·m–1. A demonstration of the graph machine model using the recent Docker technology is available for download in the Supporting Information.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1549-9596
1549-960X
DOI:10.1021/acs.jcim.7b00512