Prediction of infinite dilution activity coefficient of alcohol in ionic liquids using group contribution method

Abstract

Group contribution (GC) method is one of the predictive models used to estimate the thermodynamics properties of a compound. In this work, GC is applied to develop a predictive model to estimate the infinite dilution activity coefficient (IDAC) of ethanol and propanol in ionic liquids (ILs). Developing a GC model is easier compared to other models that require 3 or more parameters. GC method has rarely been used to estimate the chemical potential of alcohol solutes in ILs. Thus, a simple two parameters van't Hoff model using multiple linear regression (MLR) is applied and fitted by generalized reduced gradient (GRG) non-linear algorithm to calculate the IDAC values of alcohol solutes in ILs. The GC model predicted IDAC values that are based on the structure of interacting species is then, compared to the experimental values obtained from literature. The developed model is known to be in best performance if the differences between predicted and experimental IDAC values are at the least. The percentage of difference or deviation between the estimated and experimental IDAC values was determined using the absolute average relative deviation (AARD). 46 ILs with varying anions and cations are used in developing a predictive model at different temperature based on the alcohol solutes. An overall AARD value of 7.49% and 9.91% is obtained from ethanol and propanol solute, respectively, that was formed by the combination of training and validation set. The efficiency of the model with lesser than 10% deviation has been validated by the comparison between predicted and experimental IDAC values based on anion, cation headgroup, and alkyl chain length. Hydrogen bonding basicity of anion and an increase in the alkyl chain of cation headgroup contribute an important role in their interaction with ethanol and propanol as a notable reduction of IDAC values could be observed. The effect of IL's structural characteristics on IDAC is discernible, which aids in designing several ILs that have enhanced interaction with alcohol solutes.