Simulated Moving Bed Reactor for the Production of Triacetin. Part II: Model-Based Design and Optimization

The potential of a simulated moving bed reactor (SMBR) for the synthesis of triacetin is studied in the present work. Triacetin is produced by esterification of acetic acid using glycerol over the cation exchange resin catalyst Amberlyst-15 in three reaction steps. The experimental validation has already been presented in Part I of this article. As the next step, a simulation-based optimization method developed in our previous work for a single reaction (Reddy et al., Ind. Eng. Chem. Res., 2014, 53 (41), 15824–15835) is extended here for the multiple reaction system. To reduce the total annualized cost and energy consumption in the process, the study is mainly focused on maximizing productivity (PRD) and minimization of desorbent consumption (DC). We investigated the effects of different design parameters such as number of beds in each section, total catalyst/adsorbent loading, and the sectional flow rates on PRD and DC. A pareto plot based on the results of all the possible designs is generated to give maximum PRD and minimum DC. All the designs achieved the performance constraints, such as the desired conversion of glycerol, concentrations of monoacetin, diacetin, triacetin and water in both outlet streams, i.e., raffinate and extract. Furthermore, the SMBR process is compared with reactive distillation (RD) reported earlier. The energy consumption and catalyst requirements for producing 1 tonne/day of triacetin are nearly the same in both options. Reactive distillation needs heat duty of 30 kW and 160 kg of catalyst, while SMBR requires heat duty of 31.64 kW and 166.81 kg of catalyst for the given capacity. Despite the comparable energy demand and catalyst inventory of the two processes, SMBR has the potential to offer very high triacetin purity, addressing a key bottleneck associated with conventional process alternatives.