(569bk) Direct Conversion of CH4 and N2 to Ammonia and C2 Products over Metal Composite Oxides at Plasma Conditions

The ammonia synthesis from natural gas consumes a lot of energy and contributes considerable CO₂ emission. It’s better if methane directly delivers its hydrogen to nitrogen to product ammonia and C₂ hydrocarbon. To achieve this goal, nonthermal plasma generated by dielectric barrier discharge (DBD) is used to input high-quality energy to overcome the huge activation energy barrier.

Another key point in this study is catalysts. A series of oxide catalysts (SiO₂, Al₂O₃, Fe₂O₃, Mn₂O₃), Al₂O₃ supported metal oxide catalysts (Fe₂O₃/CoO/NiO/CuO/Ag₂O on Al₂O₃) prepared by the impregnation method, and metal composite oxide catalysts (Al-Fe, Al-Mn, Fe-Mn, Al-Fe-Mn) prepared by the coprecipitation method were test with plasma. All catalyst tests were conducted at the same gas flow rate and discharge power. The catalysts were made into uniformly sized particles.

The setup is shown in figure 1. According to table 1, if prioritize ammonia concentration, Fe₂O₃ performs best in oxide catalysts. All supported catalysts show a decrease compared to pure Al₂O₃, exhibiting consistent negative effects. It is worth mentioning that Mn₂O₃ caused a very low breakdown voltage, making it impossible to achieve the same discharge power as other groups. Fortunately, the combination of Al and Mn in composite oxide catalysts significantly increases ammonia concentration and demonstrates potential in carbon nitrogen coupling. The most important product among C₂ products, ethylene, reaches its peak with the combination of Fe and Mn.

To understand the synergistic effect of plasma and Al-Mn oxide catalyst, related work is still ongoing. It can be affirmed that Al-Mn oxide catalyst has a mesoporous structure, good crystallization, and significantly affects the plasma discharge. the reaction mechanism will be determined through spectral characterization. Correlations between operating parameters, catalyst structure, and plasma discharge state with products distribution and energy efficiency will be established using machine learning methods.