Pentene Catalyst deactivation

Sandelin, F., Salmi, T., and Murzin, D. (2006) Dynamic modelling of catalyst deactivation in fixed bed reactors skeletal isomerization of 1-pentene on ferrierite. Ind. Eng. Chem. Res., 45, 558-566. 

On the contrary, when vanadyl pyrophosphate is used to catalyze the transformation of butenes and pentenes, the catalyst deactivates easily and the performance is very poor (11). This occurs because the acidity of the catalyst is responsible for side reactions when olefins are directly used as the feedstock. In this case, in fact, due to its high nucleophilicity, the olefin may easily interact with different types of sites, other than those able to transform it directly to the final anhydrides. Therefore, the surface acidity of the vanadyl pyrophosphate, which seems to be a necessary property for the transformation of n-butane, is a negative feature when the corresponding olefin is the reactant. 

The same neopentylidene-alkoxo complexes react with cis-2-pentene to give the two initial metathesis products (4,4-dimethyl-2-pentene and 5,5-dimethyl-3-hexene) and catalyze the metathesis of cis-2-pentene to 2-butenes and 3-hexenes. Furthermore, propylene and ethylene appear in the reaction medium as the catalyst deactivates. These latter olefins are formed by rearrangement of the ethylidene and propylidene intermediates, providing the mechanism for the metathesis chain termination step ... 

Table 5.3-1 summarizes the most relevant results of this early study. Although the reactants show only limited solubility in the catalyst phase, the rates of hydrogenation in [BMIM][SbFis] are almost five times faster than for the comparable reaction in acetone. However, the reaction was found to be much slower using a hexafluo-rophosphate ionic liquid. This effect was attributed to the better solubility of pentene in the hexafluoroantimonate ionic liquid. The very poor yield in [BMIM](BF4], however, was due to a high amount of residual Cl ions in the ionic liquid leading to catalyst deactivation. At that time the preparation of this tetrafluoroborate ionic liquid in a chloride-free quality was obviously a problem. 

A few experiments were perfonned in which a partly deactivated CuO/Si02 catalyst was treated in an Oz/He mixture at 250°C (6h) for regeneration. In all cases this treatment was sufficient to regain the initial activity. A technical process for the production of 4-methyl-4-penten-2,3-dione from 3-hydroxy-4-methyl-4-penten-2-one consisting of alternate reaction and subsequent regeneration cycles seems therefore feasible. 

Alkenes (ethylene, propylene) do not react the hindered 4,4-dimethyl-2-pentene also does not react. Surprisingly, cyclohexene does not undergo this reaction. The yields are also low when both R and R are alkyl groups. The most serious competing reaction is polymerization of the alkene with deactivation of the catalyst. 

The turnover rates (Table 3) are lower than those obtained with 1 in the metathesis of 2-pentene or ethyl oleate, suggesting a reversible coordination of the sulfur compound to the metallocarbene leading to a partial deactivation of the catalyst. Nevertheless, the reaction is highly selective (only the expected metathesis products are detected) and the conversion of 4 can reach a high value when an excess of the co-reactant olefin is used. 

The probable role of tricarbonyl complexes, e.g., Fe(CO)3(alkene)2, in these isomerization cycles has been demonstrated by preparing these species photolytically from Fe(CO)5 [66,79]. Both Fe(CO)3(alkene)2 (and Ru(CO)3(alkene)2) proved active thermal catalysts for the isomerization of 1-pentene at 293 K. However, a competing dehydrogenation process which leads to the formation of stable and catalytically inactive (at 298 K) M(CO)3(Tl -l,3-pentadiene) complexes serves to deactivate these... 

---