KINETIC UNDERSTANDING OF THE SYNGAS-TO-DME REACTION SYSTEM AND ITS IMPLICATIONS TO PROCESS AND ECONOMICS

Abstract

In a single-step synthesis gas-to-dimethyl ether process, synthesis gas (or syngas, a mixture of H{sub 2} and CO) is converted into dimethyl ether (DME) in a single reactor. The three reactions involved in this process, methanol synthesis, methanol dehydration and water gas shift, form an interesting reaction network. The interplay among these three reactions results in excellent syngas conversion or reactor productivity. A fundamental understanding of this interplay helps to explain many experimental and simulation observations, to identify optimal reaction conditions, and to provide guidelines for process development. The higher syngas conversion or reactor productivity in the syngas-to-DME reaction system, compared to that in the syngas-to-methanol reaction system, is referred to as chemical synergy. This synergy exhibits a strong dependence on the composition of the reactor feed. To demonstrate the extent of this dependence, simulations with adjusted activity for each reaction were performed to reveal the relative rate of each reaction. The results show that the water gas shift reaction is the most rapid, being practically controlled by the equilibrium. Both methanol synthesis and methanol dehydration reactions are kinetically controlled. The kinetics of the dehydration reactions is greater than that of the methanol synthesis reaction in the CO-rich regime. However, the rates of these two reactions come closer as the H{sub 2} concentration in the reactor feed increases. The role of the dehydration reaction is to remove the equilibrium barrier for the methanol synthesis reaction. The role of the water gas shift reaction is more complex; it helps the kinetics of methanol dehydration by keeping the water concentration low, which in turn enhances methanol synthesis. It also readjusts the H{sub 2}:CO ratio in the reactor as the reactions proceed. In the CO-rich regime, the water gas shift reaction supplements the limiting reactant, H{sub 2}, by reacting water with CO. This enhances both the kinetics and thermodynamic driving force of the methanol synthesis reaction. In the H{sub 2}-rich regime, water gas shift consumes the limiting reactant, CO, which harms both the kinetics and thermodynamics of methanol synthesis. An understanding of these complex roles of the methanol dehydration and water gas shift reactions and of their dependence on the syngas composition explains why the synergy is high in the CO-rich regime, but decreases with increasing H{sub 2} or CO{sub 2} content in the reactor feed. The methanol equivalent productivity of the syngas-to-DME reactor is also a strong function of the reactor feed. A mathematical approach was developed to understand this dependence. The approach divides a power law type of rate equation into two terms, the kinetic term (the rate of the forward reaction) and the thermodynamics or driving force term (1- approach to equilibrium). The equations for the best feed composition for each term were derived. The approach was developed for the single reaction system, and then extended to the syngas-to-DME reaction system. The equations provide insights into why and how the methanol synthesis in the syngasto-DME system depends on the other two reactions. They can also be used to calculate the best feed composition for a given conversion. The analysis shows that for typical commercial syngas conversion, the optimal H{sub 2}:CO ratio for the LPDME{trademark} reactor is around 1-to-1, in good agreement with the results from the simulation. While the 1-to-1 feed provides a good foundation for some process configurations, it does not match the composition of natural gas-derived syngas, which typically has a H{sub 2}:CO ratio of 2:1 or greater. The process would also produce one CO{sub 2} molecule for every DME product, both a materials utilization and an environmental problem. However, recycling CO{sub 2} to the syngas generation unit can solve all of these problems. Integration schemes with different syngas generation technologies (dry reforming, steam methane reforming and partial oxidation) were developed. The feasibility of these schemes was illustrated by simulations using realistic kinetics, thermodynamics, and commercial conditions. Finally, this report discusses the implications of the kinetic understanding and the resulting process schemes to the process economics. It was recognized that, for the overall process, the cost saving in the synthesis loop due to the reaction synergy is counteracted by the cost addition due to CO{sub 2} formation and the resulting costly separation.