Thermodynamic model for product and process design in biorefineries

SELVA PEREDA; MARIANA GONZÁLEZ PRIETO; FRANCISCO A. SÁNCHEZ

York

Conferencia; 11th International Conference on Renewable Resources and Biorefineries; 2015

University of York

A thermodynamic model able to predict phase behavior of typical mixtures found in biomass processing technologies is required for the development of a biorefinery process simulator. Moreover, the design of particular fuel/biofuel blends also requires the support of a thermodynamic model to predict the properties of the final products. Second generation biorefineries include in their platform the processing of ligno-cellulosic biomass, in addition to oils and sugars, for the production of chemicals as well as materials and biofuels. In any of these starting raw materials, engineers have to deal with mixtures containing a wide variety of oxygenated organic compounds. The phase behavior of these mixtures is highly non-ideal due to the presence of association and solvation effects. In some cases, important size asymmetry between components also contributes to non-ideality. Moreover, in the case of gasification, knowledge of gas solubilities in aqueous broths under a wide range of conditions is required. Even though the processing of ligno-cellulosic material produces numerous oxygenated species, all of them belong to certain families of organic compounds (alcohols, carboxylic acids, esters, etc). Therefore, all these mixtures can be described by a reduced number of functional groups, for which a group contribution approach is a logical choice for thermodynamic modeling. On the other hand, pressure intensified technologies have a great potential in the context of biomass refining. It has already been proved that the Group Contribution with Association Equation of State (GCA-EoS) is able to predict the complex phase behavior of mixtures containing natural products and biofuels under a wide range of conditions. The group-contribution equation of state, originally proposed to predict gas solubilities in liquid solvents, has proved to have excellent predictive capacity to represent the phase behavior of mixtures containing natural products. The extension of the equation to multiple-associating and solvating solutions makes GCA-EoS an appropriate model to predict the phase behavior of fluid mixtures typical of biomass conversion. On the other hand, the prediction of properties of fuel/biofuel blends calls for the extension of the model to represent mixtures containing also linear, branched, cyclic and aromatic hydrocarbons. It is a big challenge for the model to predict, with a single set of parameters, the wide range of operating conditions found in processing, blending, transportation and storage of these fluid mixtures. The GCA-EoS had already been applied to predict the phase behavior of mixtures involving first generation biofuels (ethanol and biodiesel), both for process simulation and final product properties prediction. In this work a review of the GCA-EoS capabilities will be presented together with its extension to biofuels with good future potential like butanol and 2,5 dimethylfurane.