In this work, we show that the continuous flow stirred tank reactor (CFSTR) equivalence principle, developed by Feinberg and Ellison, can be used to obtain practical upper bounds on reaction selectivity for any chemistry of interest. The CFSTR equivalence principle allows one to explore the attainable reaction region by decomposing any arbitrary, steady‐state reactor‐mixer‐separator system with total reaction volume V > 0 into a new system comprising R+1 CFSTRs (where scriptR is the number of linearly independent chemical reactions) with the same total reaction volume and a perfect separator system. This work further refines the allowable selectivities by incorporating capacity constraints into the CFSTR equivalence principle to prevent arbitrarily large recycle streams between the CFSTRs and the separators and infinitesimally small CFSTR conversions. These constraints provide practical upper bounds on reaction selectivities of chemistries completely independent of reactor design. We present the methodology and the results for a selection of realistic chemistries. © 2017 American Institute of Chemical Engineers AIChE J, 64: 926–939, 2018

The Continuous Flow Stirred Tank Reactor (CFSTR) Equivalence Principle, developed by Feinberg and Ellison, proves that any and every reaction/mixing/separation process is equivalent to a process comprising at most +1 CFSTRs and a perfect mixer−separator, where is the number of linearly independent chemical reactions. Frumkin and Doherty showed that the CFSTR Equivalence Principle can be used together with global optimization to find the maximum selectivity of a chemistry independent of process design. These selectivity targets are useful in the context of process intensification because they represent ultimate selectivity improvements that can be achieved by combining multiple unit operations into a single device. In this work, the model is reformulated as a mixed-integer nonlinear program to solve this nonlinear and nonconvex optimization problem. We implement a more robust, deterministic global optimization using a spatial branch-and-bound algorithm (BARON) to investigate the selectivity limits for production on acrolein, a chemistry that has 13 components and 17 reactions. We find maximum selectivities that are lower than the stoichiometric selectivity limit and can be used as a target for process intensification. Article pubs.acs.org/IECR

No abstract

No abstract

No abstract