(235h) Automated Synthesis of Hybrid Distillation Sequences Considering Mixed Azeotropes

For the construction of the next-generation chemical industry that provides useful products at a sufficient rate even under the various constraints imposed by carbon neutrality, a process systems design that minimizes the energy requirements is needed more than ever. Currently, the chemical industry consumes approximately 10% of the world's total energy (derived from fossil fuels)¹. In particular, due to its generally large heating and cooling utility requirements, distillation alone accounts for approximately 40% of the energy consumption across the chemical industry². Designing distillation sequences with effective strategies is a significant challenge in this area, particularly as the number of sequences increases factorially as components increase³. It is even more so when the chemical industry has to transition its feedstocks from fossil-based to renewable sources, such as biomass, CO₂, or recycled materials, as the separation task is diversified. Although significant progresses have been made in optimizing distillation sequences⁴, few studies have focused on the design of distillation systems that involve mixed azeotropes. In chemical processes where azeotropes are prevalent, the optimal design of such hybrid separation sequences could result in a more cost-effective solution than before. In this context, a mathematical programming-based systematic approach can help determine optimal solutions.

Our framework incorporates an extractive distillation method based on the theory of traditional distillation sequences⁵, giving the distillation sequences an ability to handle azeotropes that was not possible with traditional distillation sequences in the past. The design process is handled in three steps. Firstly, the superstructure of the distillation sequences (in this study, only sharp separated columns are considered) is generated by the automation server of the process simulator⁶. We propose an algorithm that generates a complex superstructure containing both normal and extractive distillation columns, using pre-specified heat duties and stages for the extractive distillation columns. Secondly, a mixed-integer linear program is automatically built using the computational results of the process simulator, with the objective of minimizing the total annual cost. Finally, the program solves the distillation sequence for the first best, second best, etc., and performs a sensitivity analysis on the variability of these optimal solutions with utility prices as variables. The framework⁷ is highly flexible, as the number of optimization variables can be dynamically varied according to the optimization problem. Therefore, it is possible to perform separation sequences synthesis for any number or even 10-component mixtures.

We demonstrated the effectiveness of our framework through a case study involving the production of dipropyl carbonate (DPrC) from CO₂ as a feedstock. The process involved a one-stage reactor where propanol and CO₂ reacted to produce DPrC and water. Dehydration was facilitated by adding 2-cyanopyridine (2-CP) to improve conversion⁸, which reacted with water to form 2-picolinamide (2-PA). The effluent from the reactor was separated, and the remaining stream was sent to a two-stage reactor for the regeneration of 2-PA. The effluent from the second-stage reactor contained regenerated 2-CP, water, a small amount of 2-PA, and cyclopentanone. Distillation sequence synthesis was performed for this stream, which contained a set of cyclopentanone and water azeotropes separated using 2-CP as an extractant. Our proposed framework identified optimal separation sequences that reduced total annual costs by 14.7% compared to the conventional distillation sequence, which prefers to separate the azeotropes at the end.

References

1. Kätelhön, A., Meys, R., Deutz, S., Suh, S. & Bardow, A. Climate change mitigation potential of carbon capture and utilization in the chemical industry. Proceedings of the National Academy of Sciences of the United States of America 166, 11187–11194 (2019).
2. Kiss, A. A. & Smith, R. Rethinking energy use in distillation processes for a more sustainable chemical industry. Energy 203, 117788 (2020).
3. Thompson, R. W. & King, C. J. Systematic synthesis of separation schemes. AIChE J. 18, 941–948 (1972).
4. Caballero, J. A. & Grossmann, I. E. Design of distillation sequences: from conventional to fully thermally coupled distillation systems. Computers and Chemical Engineering 28, 2307–2329 (2004).
5. Nadgir, V. M. & Liu, Y. A. Studies in chemical process design and synthesis: Part V: A simple heuristic method for systematic synthesis of initial sequences for multicomponent separations. AIChE Journal 29, 926–934 (1983).
6. Na, J. et al. General technoeconomic analysis for electrochemical coproduction coupling carbon dioxide reduction with organic oxidation. Nature Communications 10, (2019).
7. Wang, A., Guzman, U. A., Ohno, H. & Fukushima, Y. An automatic distillation sequences synthesis framework based on depth-first pre-order traversal algorithm. (2023). (In preparation)
8. Honda, M. et al. Organic carbonate synthesis from CO2 and alcohol over CeO2 with 2-cyanopyridine: Scope and mechanistic studies. Journal of Catalysis 318, 95–107 (2014).