Use of the multifunctional autocorrelation method to estimate molar volumes of alkanes and oxygenated compounds. Comparison between components of autocorrelation vectors and topological indices
Driss Zakarya, Fathallah Tiyal, M. Chastrette · Journal of Physical Organic Chemistry · 1993
Abstract The Multifunctional Autocorrelation Method (MAM) derived from the autocorrelation method of Moreau et al. was used to calculate molar volumes for 300 linear or branched alkanes from C6 to C11 and 104 oxygenated compounds including 40 alcohols, 30 ethers and 34 ketones. Components Ci of a MAM connectivity vector were compared with several topological descriptors. MAM descriptors were generally more efficient than topological descriptors but good regressions were obtained for alkanes only when more than two components Ci were used. An excellent regression was obtained with four components (C0 to C4): Models obtained using descriptors based on connectivity were good enough for many practical purposes. However, to derive models adapted to estimate molar volumes for alkanes, an autocorrelation vector V based on Van der Waals volumes (as given by Bondi) was considered. The best equation obtained again includes three components: To extend the use of MAM to compounds containing heteroatoms, a set of alcohols, ketones and ethers, were each functional group represents a different interaction type, was considered. Chemical families were first studied individually, then the total set was considered. The components of an autocorrelation vector V based on Van der Waals volumes were used as molecular descriptors. Oxygenated compounds are correctly described by only one component V0. The correlation coefficient is 0·993 for all the chemical families studied, and the standard error of the estimate is low (s = 2·56). Comparison of standard errors of the estimate for Wiener's indices (s = 17·38) and autocorrelation components (s = 3·95), for all chemical families, shows that components of MAM are more efficient as molecular descriptors. The quality of regression equations was not significantly changed when 104 oxygenated compounds and 141 alkanes were studied in a global set.