Novel Catalysts and Reaction Pathways to Complex Nitrile Molecules

In the chemical and pharmaceutical industries, acetonitrile (CH₃CN) is a well-known solvent and it is used in drug synthesis, extractive distillation, and high-performance liquid chromatography, among others. However, during recent years, the supply of acetonitrile has been under pressure because it is only produced as a byproduct from the production of acrylonitrile. There is therefore a desire to develop novel and sustainable production methods, which can produce acetonitrile in a selective way. This is investigated at Haldor Topsøe A/S, which has identified a Co-Sn/Al₂O₃ catalyst for the purpose.

The reaction has ammonia and methanol as reactants and this is interesting for two reasons. First, the two substances are some of the main products in the sustainable Power-to-X technologies. Second, the reaction takes methanol – a C₁ molecule – and produces a C2 product. This has the potential to give a cheap process.

The reaction is not very well-studied and a lot of effort is thus needed if the process is to be commercialized. The work presented in this thesis has explored some of the aspects and features about the reaction parameters and catalysts. For this purpose, a lab-scale fixed-bed reactor was used. In the first part, the reaction conditions, when using a 5 wt% Co-10 wt% Sn/(θ+γ)-Al₂O₃ catalyst, were optimized. The results showed that the highest yield of acetonitrile could be obtained when the reaction was performed at 550 °C and had a gas hourly space velocity of 7.5 Nml h^-1 mg_cat^-1. When stoichiometric amounts of methanol and ammonia were used, the yield of acetonitrile was around 50 %, which could be maintained for 50 hours. However, the N-based yield reached over 80 % when the volumetric ratio of methanol and ammonia was approximately 4.

The reaction produced byproducts such as carbon monoxide, carbon dioxide, and hydrogen but significant amounts of hydrogen cyanide, methylamine, dimethyl ether, and longer hydrocarbons were produced under certain conditions. This lead to an investigation of some of the mechanistic aspects of the reaction. Here, it was suggested that the reaction could happen in three main steps, starting with the amination of methanol to form methylamine, which is then dehydrogenated to hydrogen cyanide. The last step involved the cyanation of another methanol molecule. For the three steps to happen, the catalysts needed to have a balance between two main catalytic sites – acid sites and metal sites. The acid sites were important in the amination and cyanation steps and this was confirmed by testing different carriers with different amounts of acid sites. When acidic carriers (Al₂O₃ and SiO₂-doped Al₂O₃) were used, a significant amount of methylamine was produced but the non-acidic carriers (SiO₂ and α-Al₂O₃) did not convert methanol. Moreover, the cyanation step did not happen over non-acidic carriers and this resulted in the N-based yield of hydrogen cyanide reaching over 70 % for the Co-Sn/SiO₂ catalysts. It could thus be concluded that the acid functionality was essential in order to carry out the C-C bond formation.

The role of the metals were also investigated and for those, it was shown that both cobalt and tin are needed if a selective catalyst is to be synthesized. When impregnating only cobalt on the carrier, extensive decomposition of methanol and coking happened, giving a limited amount of acetonitrile. Addition of only tin did not cause coking. However for this catalyst the main product was hydrogen cyanide. Conversely, a combination of the two metals resulted in a selective catalyst and only a small amount of coking. The reason for this increase in selectivity was that the metals formed the alloy CoSn. The optimal molar ratio of cobalt and tin in the catalyst was also 1. This was confirmed with powder X-ray diffractometry, temperature-programmed reduction with hydrogen and electron microscopy. Since the pure carriers did not produce hydrogen cyanide, it was concluded that the main role of the metals was to perform the dehydrogenation of methylamine. Additionally, since the catalyst with only tin impregnated failed to produce any significant amount of acetonitrile, it was seen that a combination of acidity and cobalt was needed to form the C-C bond in the cyanation step.

Because it was necessary to have a bifunctional catalyst, the dispersion/distribution of the acid and metal sites was important. To improve this, different strategies were used. One strategy was to employ different synthesis procedures and this study indicated that incipient wetness impregnation gave a more selective catalyst than using the deposition-precipitation, co-precipitation, sol-gel synthesis, and Pechini techniques. With this technique, it was also easier to control the elemental composition of the final catalyst.

The calcination procedure was also studied. In that series of experiments, it was concluded that a calcination temperature of 400 °C when calcining in static air gave the most selective catalyst. However, some results indicated that calcining at 600 °C resulted in a more stable catalyst. In particular, the stability was the main concern when treating the catalysts in H₂/Ar instead of calcining in static air. When doing that, the initial yield was high but a relatively fast decline in the acetonitrile yield was seen.