(101d) Techno-Economic Optimization of Heat-Integrated Multi-Functional System of Catalytic Reactors: The Case of Water-Gas Shift and COS Hydrolysis Reactions

1. Introduction

The design of the solid carbonaceous fuel-to-methanol production process needs to be carried out so as to reduce the production cost of methanol. However, the syngas derived from solid carbonaceous fuels contains multiple types of contaminants. Further, the relative concentrations of H₂, CO, and CO₂ present in the syngas are not ideal for obtaining maximum methanol yield from the syngas. Thus, a network of multiple processes is required for the clean-up and conditioning of the syngas. Due to the presence of multitude of sub-processes, syngas clean-up and conditioning unit results in an increase in the cost of production of methanol. The overall economy of the methanol production process can be improved by applications of principles of process intensification, as demonstrated by Stankiewicz [1], to individual parts of the process so as to reduce the capital cost and operating expenses. In this study, we demonstrate the intensification of catalytic systems for water-gas shift and COS hydrolysis reactions by treating it as a techno-economic optimization problem.

In a commercial plant, the water-gas shift reaction is carried out using H₂S-containing syngas with Co-Mo-based sour shift catalysts in a fixed-bed reactor. The COS hydrolysis reaction, on the other hand, is carried out over metal oxides such as γ-Al₂O₃ in a separate fixed-bed reactor [2]. However, it is noteworthy that the γ-Al₂O₃ support used in the Co-Mo/γ-Al₂O₃ has the capability to catalyse the COS hydrolysis reaction as well [3]. Further, both water-gas shift and COS hydrolysis are exothermic reactions requiring the same co-reactant, i.e., steam (H₂O). Hence, the possibility of carrying out both the reactions in a single multifunctional reactor rather than a system of multiple catalytic reactors needs to be assessed. Further, the overall catalytic process requires multiple heating and cooling arrangements leading to high capital and operating costs. The requirement of steam for the two reactions further increases the operating costs. To reduce the adverse impact of the catalytic processes on the overall economy of the methanol production process, optimization of the catalytic system needs to be carried out with an objective of minimizing the operating and capital costs.

2. Methodology

A one-dimensional heterogeneous model is developed for an adiabatic fixed-bed reactor. The model is adapted to simulate a single fixed-bed reactor loaded with Co-Mo/Al₂O₃ catalyst in which both the water-gas shift and COS hydrolysis reactions take place. The kinetics of the two reactions are adopted from the literature [2,4]. For optimization, the 1-dimensional reactor model is coupled with the elitist population-based non-dominated sorting genetic algorithm (NSGA-II). CO conversion of 40% is found to be required to attain the stoichiometric number of 2, which is considered optimal for methanol production. The minimization of deviation of the CO conversion (X_CO) from the required value (X_CO,req) is considered as the first objective of optimization. The maximization of the conversion of COS (X_COS) is considered as the second objective of the optimization problem. To address the problem of techno-economic optimization, minimization of the product of the capital costs and operating costs (CAPEX*OPEX) is considered as the third objective of the optimization. The capital costs are calculated as the sum of the cost of the fixed-bed reactor vessel including the catalyst and the cost of the heat exchangers required for a minimum energy target configuration of the heat-exchanger network [5]. Further, the operating costs considered include the cost of process and utility steam, cooling water, degradation of the guard bed placed downstream, and the cost incurred due to pressure drop. The cost of the utilities is calculated for the area-targeted heat exchanger network. The three-objective optimization problem is solved and the resulting Pareto fronts are analyzed. Based on the results, a heat-integrated reactor system capable of carrying out both water-gas shift and COS hydrolysis reactions with low capital and operating costs is proposed.

3. Results and discussion

Fig. 1 (a) shows the Pareto-optimal front obtained after solving the three-objective optimization problem. It is observed that at a constant value of deviation from the required conversion of CO (x-axis), the value of CAPEX*OPEX (contour) increases with an increase in the conversion of COS (y-axis). It is further observed that the maximum COS conversion possible in the reactor decreases as CO conversion becomes close to the required value. This can be explained based on the variation of the conversions of CO and COS with the feed temperature (Fig. 1 (b)). It can be noted that high temperatures correspond to lower deviation from the required CO conversion but also result in low COS conversions. The water-gas shift reaction with an appreciable conversion is favored at relatively higher temperatures (>280°C). At these temperatures however, the equilibrium limitations on the COS hydrolysis reaction set in and the reversible reaction dominates. This is because water-gas shift and COS hydrolysis are both reversible exothermic reactions and the activation energy for COS hydrolysis on the Co-Mo/γ-Al₂O₃ is comparatively lower than that for the water-gas shift reaction. Thus, increasing the temperature increases the conversion of CO, as the reaction operates in the kinetic regime, while increasing the temperature reduces the conversion of COS, as the reaction operates in equilibrium-limited regime.

Fig. 1 (c) shows the variation of the COS and CO conversions with the steam to CO ratio. It is observed that at a constant deviation from the conversion of CO (x-axis), the conversion of COS increases with an increase in the steam to CO ratio. This trend can also be explained on the basis of thermodynamic limitations on the COS conversion at the reactor operating temperatures. Here, higher steam to CO ratios increase the partial pressure of reactant (steam) thereby increasing the rate of forward COS hydrolysis reaction. However, comparison of Figs. 1 (a) and (c) shows that the chromosomes associated with higher steam to CO ratio are also associated with high values of CAPEX*OPEX. This is because process steam contributes significantly (60-85%) to the operating costs of the multi-functional catalytic system. Fig. 1 (d) shows the variation of the contribution of the cost of steam to the total operating costs of the reactor with the conversion of COS. Higher conversion of COS is found to be linked with higher steam costs. Cost incurred due to pressure drop, ZnO (guard bed) degradation, and cooling water are found to have a relatively lower contribution (0-25% each) to the operating costs. Fig. 1 (e) shows the variation of CAPEX and the contribution of heat exchangers to the CAPEX with the conversion of COS. It is observed that high conversion of COS (>95%) is linked to high CAPEX. On the other hand, this effect is not observed for relatively lower values of conversions of COS (85-90%). The heat exchangers (HE) contribute the highest to the CAPEX (40-70%) while reactor fabrication and catalyst costs have relatively lower individual contribution to the CAPEX (5-35%). In contrast to the heat exchanger costs, the combined reactor and catalyst costs are found to decrease with an increase in the COS conversion.

Fig. 1 (f) shows the final reactor design chosen based on the CAPEX and OPEX trends on the Pareto-optimal front and corresponds to 0.1% deviation from the required CO conversion while providing a COS conversion of 86.5%. For this design, the pinch method suggests no requirement of the hot utility since the heat surplus generated by the water-gas shift reaction is sufficient to heat the reactants. Thus, the techno-economic optimization performed in the present study provides a multi-functional reactor design for water-gas shift and COS hydrolysis reactions with lower number of equipment than the conventional process. It further filters the designs that lead to lower capital and operating costs while considering the hot and cold composite curves. Thus, by combining a detailed reactor model, economic analyses and principles of heat integration, a methodology is presented for obtaining the most optimal design for a system of exothermic reactors with multiple heating and cooling arrangements. The full paper will provide an in-depth discussion on other optimization runs that were carried out with the conventional reactor system, the comparison of the optimum designs of the conventional system with the multi-functional reactor, as well as the impact of other decision variables on the reactor performance and costs.

4. Conclusions

A single multi-functional fixed-bed reactor capable of catalyzing both water-gas shift and COS hydrolysis reactions is optimized while considering the capital and operating costs. The optimized multi-functional reactor operates with lesser number of equipment than the conventional system and with lower capital and operating costs. Overall, a methodology of techno-economic optimization of a heat-integrated system of exothermic catalytic reactor systems is presented.

References

[1] A.I. Stankiewicz, J.A. Moulijn, Chem. Eng. Prog. (2000) 22–34.

[2] S. Tong, I.G. Dalla Lana, K.T. Chuang, Can. J. Chem. Eng. 73 (1995) 220–227.

[3] Z.M. George, 271 (1974) 261–271.

[4] S.S. Hla, G.J. Duffy, L.D. Morpeth, A. Cousins, D.G. Roberts, J.H. Edwards, Int. J. Hydrogen Energy. 36 (2011) 6638–6645.

[5] B. Linnhoff, E. Hindmarsh, Chem. Eng. Sci. 38 (1983) 745–763.