Die in situ rutenium(III) chloriedkatalisatorsisteem vir alkeenmetatese in 'n nywerheidsomgewing

Abstract

In this study, the behaviour of an in situ-prepared RuCI3.xH20 catalytic system, where the active catalyst is generated using RuCI3.xH20 in EtOH, together with an alkyne, a ligand and an alkene substrate in the presence of a small H2 sparge, was investigated. The influence of different reagents on the catalytic system was investigated by looking at metathesis of linear alkenes and NMR studies. A few additives that have an influence on the metathesis catalyst system were identified and the effects of these additives determined. The published mechanism of the catalyst system was verified by the results reported here, and some additional findings about the reaction rate were found. The reaction occurs via a hydride-to-carbene mechanism. It was found that the formation of the hydride is a fast reaction, and the subsequent formation of the carbene is the rate-determining step. The carbene is the metathesis active species. The carbene is formed by the reaction of RuCI3.xH20 dissolved in a reducing solvent (EtOH), in the presence of an alkyne, a ligand and a linear alkene. The reaction with which the EtOH reduces RullI to RuII was not investigated. It was found that the best alkyne is a short-chain, terminal, unfunctionalised alkyne (1-hexyne). BOD was also confirmed to be a good alkyne for the reaction, although 1-hexyne afforded the highest metathesis of the alkynes investigated. The metathesis reaction is sensitive to the order of addition of the reagents. NMR studies showed that the phosphine ligand coordinates at room temperature to the Ru core, while the formation of the carbene only took place after heating of the reaction mixture to at least 50 °C. If the order of addition of the reagents is changed, side reactions take place that inhibits the formation of the carbene. A reaction with PIBu3 as ligand showed that the steric bulk of the ligand is important, since it afforded considerably less metathesis than the more bulky PIPr3. In addition to trialkylphosphines, a bicyclic phosphine ligand (EP) was found that afforded metathesis activity in the same order than PCy3. By comparing this ligand with PCy3, it was found that the bicyclic ligand afforded slower carbene formation and faster deactivation. The overall metathesis yield was still comparable. The optimum Ru:ligand ratio was different for the two ligands, with EP reaching an optimum ratio of 1:1.5, compared to the optimum of 1:4 for PCy3. The metathesis reaction was followed using NMR. This, together with results found when the time of Hz sparge was varied, showed that the carbene forms continuously during the reaction, but also deactivates continuously. The exact nature of the deactivation process was not determined during this study. The influence of the addition of hydrochloric acid to the reaction was also determined using NMR techniques. It was found that HCI protonates the phosphine as well as the phosphine oxide. Metathesis with other acids showed that acids have a effect on the metathesis reaction. The addition of strong sulphonic acids suppressed metathesis while promoting isomerisation and, in some cases, hydrogenation. The similar results from metathesis reactions with the addition of acetic acid and trifluoro acetic acid showed that the Bronsted acid strength is not the only factor that influences the reaction. The increased isomerisation using acid addition supports a hydride isomerisation mechanism. The use of chlorinated solvents, especially chloroform and chlorobenzene, for the ligand and alkyne increased metathesis. The best metathesis conversions were found using EtOH as solvent for the RuC13.xH20. No adverse effect due to ageing of the stock solutions was found, even after 3 months. The in situ catalyst system showed resistance to some oxygenates, but not to all of them. This shows that the presence of an 0-atom is not sufficient for catalyst deactivation. Further studies in this field will be beneficial. Metathesis of an industrial C7 alkene cut afforded less 6-dodecene than expected from results obtained using 1 -octene. The turnover frequencies were lower using the industrially derived cut than using 1-octene. The reactions using EP as ligand afforded significantly higher yields of 6-dodecene than reactions using PCy3 as ligand, but still less than the yields afforded by reactions using 1 -octene.