Total-recycling partial-discard strategy for improved performance of simulated moving-bed chromatography

Effect of radial distribution of injected flow on simulated moving bed performance

In the modeling of a simulated moving bed, several assumptions are considered, the key assumption is there are no radial concentration gradients based on perfect mixing. However, it is difficult to achieve perfect mixing because the injected flowrate of the bed is periodically changed in the process. In this study, the performance of the simulated moving bed process was analyzed when the injected flow such as the feed or desorbent stream was unevenly distributed. To this end, the distribution function of the injected flow was calculated and applied to the model. Two types of distribution functions were obtained using the experimental results of a previous study, and the simulation results were compared with classical modeling assuming perfect mixing. In the base case simulation, the purity was similar in all cases, the productivity was higher more than 5% in the even distribution case compared to the most uneven distribution case. The effect of distribution was analyzed through sensitivity analysis by changing the overall flow rate, switching time, bed length, and flow rate of sections 2 and 3. As a result, regardless of the distribution applied, the trends of the performance parameters were the same. However, the more uneven the distribution, the greater the difference in productivity, recovery, and desorbent consumption compared to the even distribution case. It was confirmed that the design that distributes the injected flow more evenly has a better performance.