Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article


Doctor of Philosophy


Chemical and Biochemical Engineering


Ray, Ajay K.


Chromatographic separation is a promising alternative for separation and purification of sugars in industry. Simulated moving bed (SMB) technique has been proven as an efficient chromatographic separation method due to its enhanced productivity and purity, reduced solvent consumption, convenient operating control, and improved separation performance for some systems with low resolution and selectivity. The sequential simulated moving bed (SSMB) is a modification of the conventional SMB process, which currently has some applications for sugar separation due to its low solvent consumption. This work mainly investigates the design strategy of the innovative SSMB process and explore its advantages and disadvantages over the SMB process based on the xylo-oligosaccharides (XOSs) and fructose-glucose systems. SSMB separation of XOSs, a functional food additive in the form of a oligomeric saccharide, was firstly conducted. DOWEX MONOSPHERETM 99/310 resin ionized with K+, which has better selectivity compared with Ca2+ and Na+ was used as the stationary phase.Breakthrough experiments showed that XOSs and the two major industrial impurities, xylose and arabinose, all exhibit linear isotherms. Transport-dispersive (TD) model parameters were determined by pulse experiments carried out at various flowrates. Finally, both the averaged and individual parameters of XOSs and XOS2-XOS7 were obtained. Lab-scale SSMB experiments and the corresponding simulations were carried out to validate the acquired TD model parameters and adsorption isotherms. After that, in order to investigate the optimal operating conditions of this process, the multi-objective optimizations were carried out for three cases with various objectives and constraints. It was found that, for a given SSMB unit, there exist a pareto curve for simultaneous maximization of purity and unit throughput. The flowrate ratios (m values), however, exhibit some trends that are different from those of conventional SMB and cannot be explained by the direct use of Triangle Theory with averaged m values. According to the literature, the fructose-glucose system is representative and have linear isotherms over a wide concentration range, which makes it an excellent example system to conduct some basic analysis and performance prediction. Therefore, the multi-objective optimization of SMB and SSMB processes was conducted and compared based on the fructose-glucose system. The results show that the solvent consumption of SSMB is always less than that of SMB unit.

Summary for Lay Audience

Simulated moving bed (SMB) chromatography, a well-developed continuous liquid chromatographic method, is widely used in petro-chemical industry, food industry, and pharmaceutical and biotechnological field. A typical 4-zone SMB unit consists of several packed chromatographic columns which are series-connected and form a closed loop. The mechanism of this separation process is the counter-current movement between the liquid and solid phases, which is achieved by periodically and simultaneously switching the inlet and outlet ports in the direction of the fluid flow. The sequential simulated moving bed (SSMB) is a modification of SMB process, which divides one switch in SMB into three steps with different flow patterns. Actually, SSMB technique is a preferential choice for some large-scale separation processes in industry due to its low solvent consumption. However, there is on one reported this SSMB process and explored its application in industry. Therefore, a comprehensive study of this SSMB technique was mainly conducted in this thesis. Xylo-oligosaccharides (XOS), produced by hydrolysis of xylan-rich hemicelluloses, is a kind of oligosaccharides with prebiotic effect. Due to the nature of the enzyme hydrolysis reaction of XOS synthesis, the industrial produced XOS syrup generally contains 70% XOS, 22% unreacted xylose, and 8% by-product arabinose. To our knowledge, there is only one study on XOS purification by using SMB which does not contain the detailed modeling and optimization works. In this work, XOS was purified by using an economic SSMB process, and the optimal operating conditions were selected by using multi-objective optimization corresponding to different industrial requirements. The multi-objective optimization was conducted based on XOS and fructose-glucose systems by using both SMB and SSMB processes in this work. The purpose of this method is to search several groups of equally good solutions (called a pareto set) for a certain problem. A pareto set means in which when we go from any one point to another, at least one objective function becomes better and at least one worsens. Therefore, any one solution in a pareto set is optimal and acceptable.