The cost of dilute ethanol purification is a significant component of the overall cost for fuel grade ethanol production through fermentation or other biological routes. In this study, we consider bioethanol purification to the fuel grade in a closed system of photobioreactors (PBRs) that produce a dilute (0.5−5 wt %) ethanol solution, and where the water is recycled back to the reactors after separation. To reduce the energy consumption and hence the cost of dilute ethanol purification, a reverse osmosis (RO) membrane process is introduced as a potential pretreatment in order to purify ethanol−water mixtures to an intermediate concentration. The pretreated mixture is separated across the azeotropic point to the fuel grade by a hybrid distillation-membrane-pervaporation (D-PV) process. A superstructure of the overall separation process is formulated in the General Algebraic Modeling System (GAMS) environment as a mixed-integer nonlinear programming (MINLP) problem and optimized to minimize the total separation cost. The problem is posed with a constraint on the minimum ethanol recovery by the overall system. For dilute feeds, the steam used in the distillation reboiler dominates both the energy and the cost of the overall system. Installing the RO system reduces the steam usage at the expense of larger capital investment in membrane modules, which has been found to be optimal for dilute feeds below 3 wt % ethanol. The optimal number of membrane stages and the feed location of individual RO modules change at different feed concentrations and ethanol recoveries. The optimal process design results in a higher ethanol recovery by the main hybrid separation unit compared to the RO membrane process due to the high capital cost of an RO system to achieve a high recovery. Developing a cheaper RO membrane may further reduce the separation cost and may expand an optimal dilute feed concentration range.

No abstract

No abstract

A new reactive separation process concept termed solid-phase reactive separation system (SPRSS) is proposed which integrates the progressing batch reactor (PBR) and simulated moving bed (SMB) chromatography processes. The key idea of this new process is that both PBR and SMB employ similar principles of the movement of the liquid feed to imitate the countercurrent movement of solid and liquid phases. This concept may be applied to the reactive separation system that involves solid reactants. In this work, SPRSS is applied to the production of sugar from cellulosic materials which contain different amounts of cellulose and hemicellulose in which the kinetics of each hydrolysis reaction path are different. SPRSS enables the variation of these reaction parameters within the reactor system to hydrolyze different portions of the biomass, which allows more flexible acid catalyst concentration profiles. The overarching goal is to improve the total sugar yield and concentration while minimizing the sugar decomposition reaction and undesired product formation. An optimization strategy using a superstructure formulation is applied to find the optimal process design of SPRSS using a continuous moving-bed model. The optimal results show a potential improvement of the sugar yield as well as less byproduct formation by changing the acid concentration using the SPRSS configuration design. The effect of different kinetic parameters is investigated, which could significantly affect the performance of SPRSS. Several observations of the values of these kinetic parameters which could further enhance the advantages of SPRSS design are demonstrated.

No abstract