(176e) Synthesis and Design of Energy Efficient Distillation

Batch and continuous distillation and crystallization have been the workhorses for separations in the petroleum, chemical, pharmaceutical, and other industries for many years, and this is unlikely to change. These unit operations, as well as others, will remain the primary means of separation in many industries for many years to come. However, distillation and crystallization consume significant amounts of energy even when compared to the US GNP. While some believe that distillation is a mature technology and that there is little to be gained from research in separations like distillation and crystallization, we disagree with this viewpoint for two reasons. First, with the recent significant increase in global energy demands and every indication that demand will remain high, it is important to consider ways of designing or retrofitting separations so they are energy efficient. Second, the approach taken in this work is a direct outgrowth of recent results that shed new light on residue curves and separation boundaries. It is unlikely that we would have uncovered the proposed characterization of energy efficient separations without our initial results.

Lucia and Taylor (2005) have recently presented a geometric methodology for finding exact boundaries in separation processes like azeotropic distillation, crystallization, and reactive distillation and show that for ternary mixtures all separation boundaries are given by locally longest residue curves that run from a given unstable node to all reachable stable nodes. For four-component mixtures, boundaries are local maxima in surface areas while for five or more components boundaries correspond to local maxima in volumes. This geometric theory has led to an efficient feasible path optimization algorithm for computing exact separation boundaries for batch or continuous separations. Moreover, rigorous proof and a number of challenging numerical examples have been used to validate the theory.

In this work, we conjecture that local maxima in line integrals, surface areas, and volumes (i.e., the longest paths) represent the most difficult and therefore the least energy efficient separations in a given separation region. Conversely, local minima in line integrals, surface areas, and volumes (i.e., the shortest paths) correspond to the easiest and most energy efficient separations. We quantify these conjectures and show that locally longest and shortest line integrals for residue curves do, in fact, correspond to the most and least difficult separations respectively. These facts provide definitive guidance for the synthesis, design, and retrofitting of energy efficient finite separators. Because we are interested in finite designs (i.e., finite stages and finite internal flows), distillation lines are used in place of residue curves. First, distillation lines at infinite reflux and infinite reboil are used to show that the minimum number of stages (or height of packing), Nmin, required to achieve a desired separation and therefore the easiest separation occurs in the neighborhood of the shortest distillation lines. Second, we show that the separations with the smallest minimum reflux, and thus the lowest energy demands, again appear in the neighborhood of the shortest distillation path. This approach allows us to easily screen alternative designs with regard to energy use and to find the most energy efficient separations for a variety of situations (i.e., fixed number of stages, fixed safety margins, batch operation, continuous operation, and so on). It also allows us to retrofit existing columns with a fixed number of stages to improve energy efficiency. In addition, our results are general and apply to crystallization, reactive separations, non-equilibrium models, and other separations. This work includes theoretical development as well as numerical experimentation and validation. Numerical examples and geometric illustrations are used to elucidate key concepts and to demonstrate the practical value of the proposed approach in finding energy efficient separator designs.