A Systematic Workflow for the Design of Robust Batch Processes

Batch processing is an important operation mode in the chemical industry, commonly employed in the production of added-value products such as pharmaceuticals, flavor and fragrance chemicals, and pesticides [1]. The design of such processes is facilitated by using dynamic models, which aim to capture the performance of the unit operations used in the production of such products. In general, all the models involve parameters that i) are obtained experimentally and ii) distinguish processes one another, even if the same type of unit operation model is used to describe them. As an example, kinetic parameters to describe the rate of a reaction, or phase equilibrium coefficients for use in separation models, are obtained by conducting appropriate experiments and fitting the data to postulated models, taking into account measurement errors. The reliability of those numerical values will greatly impact the accuracy of the process simulations and therefore the capability of the final process design. For this reason, considering both the workflow followed in the determination of process parameters and their use in simulation and optimization routines is of great importance to develop robust designs for batch processes.

This contribution illustrates a systematic workflow leading to the optimal design of a batch process. The approach combines experimental procedures, parameter estimation, process modeling and simulation, global sensitivity analysis and process optimization . The synthesis of nopol, which involve reaction, filtration and distillation unit operations, is used as a case study to illustrate the application of the proposed workflow. Nopol is an oxygenated compound produced by the liquid-phase reaction between β-pinene and formaldehyde [2]. It can be produced via heterogeneous catalysis using a SnMCM-41 material [3]. The SnMCM-41 catalyzed process is an environmentally-friendly alternative to homogeneous synthesis procedures for nopol production [4], since in the heterogeneous process, catalyst can be reused and the utilization of toxic chemicals is greatly diminished.

As a first step, an experimental study of nopol chemical reaction rate was carried out in order to determine kinetic and adsorption parameters [5], [6]. Furthermore, a process flowsheet comprising the aforementioned stages and including a recycle stream was defined, and mathematical models consisting of both ordinary differential equations and algebraic equations [7]–[9], were developed for each unit operation. These mathematical models were implemented in a Matlab® process simulation. Next, a Monte-Carlo-based global sensitivity analysis [10], [11] was employed to rank the effect of the process variables and parameters on the model outputs. The sensitivity analysis identifies the model parameters that may require more attention at the experimental level, and prioritizes the process variables and thus reduce the complexity of the process optimization, by focusing attention of the variables that have the highest impact on the performance metrics, i.e. the net present value (NPV) and the break-even cost.

References

[1] A. Cybulski, J. A. Moulijn, M. M. Sharma, and R. A. Sheldon, Fine Chemicals Manufacture. Elsevier B.V., 2001.

[2] E. Kirk and D. F. Othmer, Encyclopedia of Chemical Technology, 4th Ed. New York: Wiley - Interscience, 1998.

[3] A. L. Villa, E. Alarcón, and C. Montes de Correa, “Synthesis of nopol over MCM-41 catalysts.” Chem. Commun. (Camb)., no. 22, pp. 2654–5, 2002.

[4] J. P. Bain, “Nopol . I. The Reaction of p-Pinene with Formaldehyde,” J. Am. Chem. Soc., vol. 68, pp. 638–641, 1946.

[5] A. L. Villa, L. F. Correa, and E. Alarcón, “Kinetics of the nopol synthesis by the Prins reaction over tin impregnated MCM-41 catalyst,” Chem. Eng. J., vol. 215–216, pp. 500–507, 2013.

[6] D. Casas-Orozco, E. Alarcón, and A.L. Villa, “Kinetic study of the nopol synthesis by the Prins reaction over tin impregnated MCM-41 catalyst with ethyl acetate as solvent,” Fuel, vol. 149, pp. 130–137, 2015.

[7] S. Fogler, Elements of Chemical Reaction Engineering, Fourth Ed. New York: Prentice Hall, 2006.

[8] S. Tarleton and R. Wakeman, Solid-Liquid Separation: Equipment Selection and Design, 1st Ed. Oxford: Elsevier Inc., 2007.

[9] I. M. Mujtaba and S. Macchietto, “Holdup issues in batch distillation-binary mixtures,” Chem. Eng. Sci., vol. 53, no. 14, pp. 2519–2530, 1998.

[10] I. M. Sobol, “Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates,” Math. Comput. Simul., vol. 55, no. 1–3, pp. 271–280, 2001.

[11] A. Saltelli, M. Ratto, S. Tarantola, and F. Campolongo, “Sensitivity analysis for chemical models,” Chem. Rev., vol. 105, no. 7, pp. 2811–2827, 2005.