화학공학소재연구정보센터
Energy, Vol.72, 274-290, 2014
Predicting the physical-chemical properties of biodiesel fuels assessing the molecular structure with the SAFT-gamma group contribution approach
The properties of biodiesel as fuel are strongly defined by the molecular structure of its constituent species (saturated, unsaturated and hydroxylated fatty acid alkyl esters). Improving fuel properties and energetic patterns in biodiesel implies optimising its fatty ester structures and compositions. Biodiesel fuels derived from different sources can have significantly varying fatty acid profiles and properties, which claims for the theoretical prediction of the thermophysical and phase equilibria properties of biodiesel compounds and its mixtures. In this work the SAFT-gamma (Statistical Associating Fluid Theory by group contribution) is applied for predicting these properties in biodiesel fuels by adequately representing the physical behaviour and stereochemistry of biodiesel molecules. A realistic representation of the molecular structure of long chain methyl esters and esters that contains hydroxyl groups, which are the typical biodiesel fuel constituents, is obtained. We implemented a simplex simulated annealing algorithm as global optimisation method to determine the SAFT-gamma parameters for the groups that fully represent the biodiesel compounds, which were fitting to experimental data available for analogous chemical families like secondary alkanols and short chain esters. These parameters are theoretically justifiable following physically meaningful trends and can be generalised to others fatty acid methyl esters in the same homologous series. The group of like and unlike parameters obtained were used to represent the thermophysical properties of several commercial biodiesel fuels. The theory provides a very good description of the liquid vapour equilibria behaviour of the chemical families used to estimate the set of parameters. With the proposed model, any potential biodiesel fuel from any feedstock can be represented and modelled. (C) 2014 Elsevier Ltd. All rights reserved.