Comparative study of liquid split mechanisms for 3D-printed laboratory dividing wall columns

Abstract Dividing wall columns (DWCs) are a well-established example of intensified distillation technology, offering significant reductions in both energy consumption and capital investment. These benefits, however, can only be fully realized when the prefractionator is operated under optimal conditions. This operating point is known as the preferred split. The column profile in the prefractionator, and thus the preferred split, is influenced by the choice of the vapor and liquid splits. While the liquid split can typically be manipulated during column operation, this is not the case for the vapor split. The latter is determined by the initial column design and, in contrast to the liquid split, does not represent a dynamic degree of freedom that can be adjusted within control strategies. Consequently, a precise liquid split is essential for maintaining energy-efficient operation of a dividing wall column. At the industrial scale, the required splitting technology is well established and is often implemented outside the column. In contrast, the operational limits of standard laboratory-scale liquid dividers are rarely reported in the literature. However, as dividing wall columns at the miniplant scale are commonly used for process development, this lack of information is of significant relevance. For the future implementation in a 3D-printed laboratory-scale dividing wall column, existing designs of lab-scale liquid dividers were adapted, and an additively manufactured test rig for liquid splitting was developed. Within this setup, various liquid-splitting concepts and actuation mechanisms were investigated and compared using water and ethanol as test fluids. In total, three different approaches were examined, employing pneumatic rotary, pneumatic linear, and electromagnetic actuation. The effects of the supplied liquid flow rate, switching frequency, and solvent type were systematically analyzed, and operational boundaries were identified. The observed effects of flow rate, switching frequency, and solvent properties arise mainly from the underlying liquid-splitting mechanisms rather than from the use of 3D-printed materials. Consequently, the findings are applicable to a wide range of laboratory-scale liquid dividers, independent of the employed manufacturing technology. In addition, challenges arising from interactions between 3D-printed surfaces and solvents were examined. The resulting insights were used to determine the most suitable liquid-splitting approach. Among the investigated concepts, pneumatic rotary actuation exhibited superior performance, as its liquid-splitting behavior proved largely independent of switching frequency, flow rate, and operating fluid. Overall, this study helps bridge the existing knowledge gap regarding the operation of liquid dividers in laboratory-scale distillation equipment.