Quantitative Structure–Property Relationship Prediction of Liquid Heat Capacity at 298.15 K for Organic Compounds

Abstract

Novel QSPR models were developed and evaluated for the prediction of heat capacity of liquids at 298.15 K with only three descriptors. Two linear and nonlinear models were produced using genetic function approximation (GFA) and adaptive neurofuzzy inference system (ANFIS) methods based on a data set of 706 compounds with a wide variety of functional groups. The results showed that both GFA and ANFIS methods could model the relationship between the liquid heat capacity of organic compounds and their structures with high accuracy. The predictive quality of the QSPR models were tested for an external test set, where the squared correlation coefficients of prediction for the GFA and ANFIS methods were 0.970 and 0.973, respectively.

Prediction of the heat capacity for compounds based on the conjugate gradient and support vector machine methods

“…The S1K is one of the Kier α‐modified shape descriptors, representing paths of order 1, which encodes information about the count of atoms and relative cyclicity of molecules. This descriptor and therefore heat capacity increase with the increasing size of molecules and decrease with the cyclicity of molecules . C‐024 indicates the number of the R‐CH‐R group, which belongs to atomic center segments.…”

“…Meanwhile, in our present model, the AD of the ACO-CG and ACO-SVM models is validated, whereas the AD was not validated in Mostafa's work. Aboozar Khajeh et al [37] has developed two linear and nonlinear models using genetic function approximation and adaptive neurofuzzy inference system methods for predicting the liquid heat capacity at 298.15 K of organic compounds. Compared with our present model, the two models developed by the Aboozar Khajeh group have great predictive quality.…”

Calculation of Abraham model L-descriptor and standard molar enthalpies of vaporization and sublimation for C9 - C26 mono-alkyl alkanes and polymethyl alkanes

“…QSPR expressions have been derived for a large range of physicochemical and thermodynamic properties, including vapor pressure, 12 flash point temperatures, 13,14 Gibbs energies of solvation and Ostwald solubility coefficients, 15,16 liquid viscosity, 17 enthalpies of solvation, 18,19 and liquid and gas molar heat capacities. 20,21 Group contribution methods belong to a class of empirical property prediction methods that base calculations upon the functional groups or "molecular building blocks" contained within the chemical compound. The molecule is broken down into individual building blocks and the physicochemical and/or thermodynamic property is then estimated as:…”

Prediction of Thermophysical Properties for Binary Mixtures of Common Ionic Liquids with Water or Alcohol at Several Temperatures and Atmospheric Pressure by Means of Artificial Neural Network

“…The computational prediction methods including GFA, ANN, ANFIS and RBF were not only applied to predict some desired properties of ILs but they were also used for predicting other important properties. − For example, by using the artificial neural network (ANN) technique, Mohanty estimated the vapor liquid solubility of binary mixtures with an average absolute deviation of 3% for liquid phase and less than 0.02% for vapor phase mole fractions. Nguyen et al .…”