Industrial & Engineering Chemistry Research, Vol.48, No.11, 5360-5369, 2009
Characterization and Dynamic Optimization of Membrane-Assisted Crystallization of Adipic Acid
The concept of membrane-assisted crystallization using reverse osmosis (MaC-RO) shows a high potential in decreasing energy conversion as compared to evaporative crystallization. A setup has been constructed for MaC-RO, using two separate loops for membrane separation and crystallization. In this study we focus on the characterization of the membrane performance in solvent removal from an aqueous adipic acid solution under reverse osmosis conditions. A semiempirical model was developed for the flux as a function of temperature, concentration, and pressure difference across the membrane. The presence of adipic acid strongly reduces the flux at the same feed temperature and pressure, as compared to the flux from a pure water feed. This effect could be attributed. to competitive adsorption on the membrane surface and, to a lesser extent, concentration polarization. For a batch MaC-RO process a dynamic optimization was conducted by changing the operational policy (membrane feed pressure, flow rate, and temperature). The energy conversion is minimized while maintaining a high, prior optimized product yield, defined as the volume-based mean size. The so-determined optimal operational policy results in a six times lower energy conversion as compared to evaporative crystallization with a final mean size that closely corresponds to the maximum mean size that can be achieved in a crystallizer with ideal control of supersaturation. This study demonstrates the potential of the concept of MaC-RO in terms of control over product quality and reduction of energy conversion.