Predicting thermochemical parameters of oxygen-containing heterocycles using simple QSPR models - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Molecular Simulation Année : 2006

Predicting thermochemical parameters of oxygen-containing heterocycles using simple QSPR models

Joachim Clauss
  • Fonction : Auteur
  • PersonId : 878494
Marc Meunier
  • Fonction : Auteur
  • PersonId : 878495

Résumé

Quantitative structure-property relationships for the prediction of standard enthalpies and entropies of formation as well as standard molar heat capacities for small oxygen heterocyclic compounds were developed, using 1D, 2D and 3D descriptors and experimental or computed thermochemical data. To develop the models, the data set was split into test and training sets using D-optimal experimental design to generate a diverse training set. Internal (R2cross-validated = 0.898 – 0.998) and external (R2cross-validated = 0.847 - 0.996) validation showed the models to be both stable and highly predictive. Enthalpies of formation were best described by elctrotopological, atomic composition and molecular refractivity descriptors, while Kier and Hall χ and κ descriptors as well as the number of rotatable bonds appear frequently in models describing the entropy of formation of these compounds. Heat Capacity models often feature the molecular area descriptor as well as well as the Kier and Hall 0χ descriptor and the number of methyl groups present in the molecule.

Mots clés

Fichier principal
Vignette du fichier
PEER_stage2_10.1080%2F08927020500474300.pdf (640.8 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00514974 , version 1 (04-09-2010)

Identifiants

Citer

Nico Adams, Joachim Clauss, Marc Meunier, Ulrich Schubert. Predicting thermochemical parameters of oxygen-containing heterocycles using simple QSPR models. Molecular Simulation, 2006, 32 (02), pp.125-134. ⟨10.1080/08927020500474300⟩. ⟨hal-00514974⟩

Collections

PEER
133 Consultations
169 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More