QUANTITATIVE RELATIONSHIPS BETWEEN MOLECULAR STRUCTURE AND MELTING POINT OF SEVERAL ORGANIC COMPOUNDS

Abstract

Quantitative  Structure - Property  Relationships (QSPR) analysis of the Melting Point (m.p) of 349 organic compounds using molecular descriptors based on linear regression have been done. The experiment m.p. values were collected from the CRC Hand Book of Chemistry and Physic 81^th [1]. Molecular description of each compound was described in the form of the number of atoms and number of chemical bonding. This study was made in the form of normal melting point (in Kelvin unit) as the linear function of all molecular descriptor. The best equation model was determined by regression multilinear analysis and the result showed that the melting point is linearly correlated  with:   n(C_{sek alf}), n(C_{ter alf}), n(C_{sek sik}), n(C_{ter sik}), n(C_{sek arm}), n(C_{ter arm}), n(O), n(N), n(F), n(Cl), n(C-C _sik), n(C=C _arm), n(C₁=C₂), n(C₂=C₃), n(C-O), n(C-Br), n(C-I), n(N-H), n(Z), n(ortho), n(meta), n(para), (BM), n(C-H)²_, n(H x BM) and n(C_prim x BM). The QSPR equation is significant at 95 % level and has statistical parameters: n = 349, r = 0.943, SD = 24.336 and F_exp = 99.326,  respectively.