COEFFICIENT PARTITION PREDICTION OF SATURATED MONOCARBOXYLIC ACIDS USING THE MOLECULAR DESCRIPTORS
Carboxylic acids have clearly been absent from the quantitative structure-property relationship literature. The studies of the quantitative structure–property relationships (QSPR) involve various chemometric methods in which the physico-chemical behavior of a compound is correlated with its structure represented by the structural indices. For example, QSPR methods are applied for the prediction of octanol-water partition coefficient of an organic compound. In this study, the relationship between the octanol/water coefficient partition and molecular descriptors was investigated. Also, the multiple linear-regression method based on QSPR methodology was applied to predict the Log P of saturated mono-carboxylic acids C1-C22. On the other hand, the relation [ Log P = - 0.426 ( Platt ) + 0.190 ( V/ A°3 ) - 0.155 ( Max.P.A/ A°2 ) - 1.914 ( X ) - 1.576 ; N = 22, R2 = 0.995 , F = 917.005, DW=1.391] was generated for selected mono-carboxylic acids. The results of study indicated that the Platt, Randic, Volume and Maximum-Projection-Area descriptors have an important role in predicting the octanol/water coefficient partition of saturated monocarboxylic acids (C1- C22).