Youngmi Kim1, Rick Hendrickson1, Nathan S. Mosier2, and Michael R. Ladisch1. (1) Laboratory of Renewable Resources Engineering (LORRE), Purdue University, Potter Engineering Center, 500 Central Drive, West Lafayette, IN 47907-2022, (2) Agricultural and Biological Engineering & Laboratory of Renewable Resources Engineering (LORRE), Purdue University, Potter Engineering Center, 500 Central Drive, West Lafayette, IN 47907-2022
Starch based adsorbents have found utility in the fuel ethanol industry to remove water from fuel ethanol in an energy efficient manner. Corn grits packed in a fixed bed absorber are used in a thermal swing mode to remove water from sub-azeotropic ethanol to yield a fuel-grade alcohol product. The heat of adsorption is stored in the bed, and combined with a counter-flow of CO2 (a co-product of the ethanol fermentation), this heat provides the energy required to regenerate the bed. A multi-bed, staged operation is used to obtain a continuous stream of fuel-grade product ethanol. A challenge in modeling this system has been the paucity of equilibrium data, which is obtained at representative conditions. A dynamic method has been developed that enables equilibrium data to be obtained in a flow-through system. Equilibrium adsorption of water on starch based adsorbents has been determined for sub-azeotropic ethanol that contains 90-95% (v/v) ethanol. Linear isotherm is valid for this range of feed ethanol. The adsorbent is packed in a basket inserted in a long column. Flow through the basket is managed to minimize dispersion, and enable the extent of adsorption to be measured using the change in composition between inlet and outlet hydrous ethanol streams as well as regenerant condensate composition. Equilibria data, a model, and the method are presented in this paper. These data are used to model the dynamic operation of a fixed bed system where the feedstock for corn ethanol is also the separations media.