Kinetic modeling and optimization of the methanol and dimethyl ether synthesis

Introduction
An increasingly relevant alternative to deal with the fluctuating nature of solar and wind power plants is to spend the excess power generation in water electrolysis, and further convert the produced hydrogen to secondary energy carriers or added-value chemicals [1]. In this desired route, methanol and dimethyl ether (DME) are key intermediates.
The catalyst mixture of Cu/ZnO/ZrO2 (CZZ), for methanol synthesis from syngas, and ferrierite (FER), for methanol dehydration, is highly active for both CO-rich and CO2-rich feeds [2]. In order to optimize and scale-up the CZZ/FER system, accurate kinetic models are necessary, which, to the best of our knowledge, are not available in literature yet. In this contribution, a kinetic model for the methanol synthesis [3] and the methanol dehydration to DME is developed, validated experimentally, and used for the optimization of the catalyst bed composition.

Methodology
The CZZ was prepared by a continuous co-precipitation method at KIT [4]. Experiments were performed in a plug flow reactor filled with 4.2 mL of a catalyst mixture between CZZ and a commercial ferrierite (i.e.H-FER 20). ... mehrThe operating conditions include variations in pressure (31 – 55 bar), temperature (200 – 250 °C), space velocity (0.79 – 3.77 s-1), feed concentration, and catalyst bed distribution (CZZ/FER from 32/68 until 97/3% v/v).
The reaction rate equations of the methanol synthesis [3] were developed by reducing the complexity of a recent published microkinetic model [5]. In order to extend the model with the methanol dehydration, the associative mechanism is assumed as dominant. 80% of the experiments were used to estimate the kinetic parameters of the extended model, while the remaining 20% was used for validation.

Results and Discussion
Although the kinetic model of the methanol synthesis simulated the experiments with
high accuracy [3], this discussion will be focused on the direct DME synthesis. The
extended model simulated the experiments adequately, as most deviations to
experimental values were below 10%. The experimental trends were also correctly
predicted by the model.
According to the experimental results, an optimal DME productivity should occur
around CZZ/FER = 90/10% v/v for these operating conditions. The optimum of
CZZ/FER = 91.5/8.5% v/v was numerically obtained for the feed of H2/CO/CO2/N2 =
45/2/18/35% v/v, a result which was confirmed experimentally in further
measurements. Optimal catalyst distribution was then calculated for different operating
conditions to see their influence in the optimal value. While feed concentration has little
effect in the optimum, lower space velocities and lower temperatures tend to further
increase the optimal amount of CZZ.

References
1. Artz et al. Chem. Rev. 118 (2018), 2, 434-504.
2. Wild et al., RSC Advances 11 (2021), 2556-2564.
3. Campos et al., Ind. Eng. Chem. Res. 60, 15074-15086 (2021).
4. Polierer et al., Catalysts 10, 816 (2020).
5. Campos et al., Reac. Chem. Eng. 6, 868-887 (2021).