Modellering en sintese van alisikliese fosfienverbindings as ligande
Abstract
In this study alicyclic phosphine compounds were identified from the literature that has the potential to be used as new ligands for the Grubbs-type catalyst system. Molecular modelling was used to investigate the productive metathesis of 1-octene. Lastly the synthesis of new alicyclic phosphine ligands for the Grubbs-type catalyst system was attempted. The short literature overview illustrates the diversity of alicyclic phosphine ligands that have already been synthesised for transition metal complexes. Several methods to synthesise alicyclic phosphine compounds that have the potential to be used as ligands for transition metal complexes have already been reported.
The molecular modelling study shows that a deeper insight into the mechanism of the catalytic reaction can be achieved if no simplifications are applied to the dissociation step, activation step and catalytic cycles of Grubbs-type catalyst systems. It is clearly illustrated that precatalyst stability of the conformers does not necessarily predict the rate of activation. The catalyst with the lowest dissociation energy does not necessarily activate the easiest. The same is true for catalytic activity. It is illustrated that the small differences in precatalyst stability of the conformers cannot predict the large differences in catalytic activity. Although the rate of catalyst initiation does not differ by much between the isomers, there is a clear difference in the rate of activation and catalytic activity. It is clearly illustrated that the small energy differences in the rate of activation between the conformers does not predict the clear difference in catalytic activity. The Ph-cub (Ru(=CHPh)CI2(C6H5PC8H8)2 en Cy-cub (Ru(=CHPh)CI2(C6H11PC8H8)2 catalysts showed a larger affinity for the binding of 1-octene than Phobcat and Grubbs 1. This larger affinity might indicate that Ph-cub and Cy-cub has a better selectivity for the coordination of alkenes. The steric and electronic effects responsible for the higher energy during the formation of some of the metallocyclobutane rings is not easily explained by the molecular modelling results and has to be investigated further. An idea of the catalytic activity of the catalyst can be formed once the complete mechanistic cycle is investigated, but experimental data is needed to confirm the observed trends.
The Gibbs free energy (AG) corrections has to be determined to calculate the thermodynamic properties of the reaction, by doing so the rate limiting step of the reaction can be determined more accurately. The time consuming calculation method of vibration modus of the atoms in the molecule is necessary to determine the Gibbs free energy corrections. The phenylphosphahomocubane oxide could not be synthesised during any of the irradiation experiments through Pyrex or quartz. Only the cubane oxide's precursor could be isolated from the irradiation experiment through Pyrex. Corex glass was used for the original irradiation but the glass is not manufactured anymore. To synthesise the phenylphosphahomocubane oxide successfully it is necessary to find an irradiation setup that has the correct transmittance properties and have correctly sealed glass tubes. The molecular modelling investigation of the formation of phenylphosphahomocubane oxide out of its precursor resulted from the inability to synthesise the homocubane. The last step energy values of neither of the proposed photocyclisation mechanisms falls in the Corex (λ ≥ 300 nm) to before Pyrex (λ ≥ 280 nm)range of 95.37-102.18 kcal/mol. If it is assumed that phenylphosphahomocubane oxide forms via a diradical stepwise triplet excited state [2π + 2π] cycloaddition mechanism the energy values are overestimated. If the phenylphosphahomocubane oxide forms via a concerted [2π + 2π] cycloaddition mechanism the energy values are underestimated. The photocyclisation has to be investigated experimentally to determine which of the two proposed mechanisms is correct. The synthesis of dichlorocyclohexylphosphine oxide was investigated with an esterification reaction. The MS-spectra fragmentation pattern of the ester correlated with the literature. This was taken as proof for the purposes of the research that the dichlorocyclohexylphosphine oxide was synthesised. Dichlorocyclohexylphosphine oxide could not be reduced with phenylsilane. Cyclohexylphosphahomocubane oxide could not be synthesised successfully from dichlorocyclohexylphosphine oxide and cyclooctatetraene. To successfully synthesise the cyclohexylphosphahomocubane oxide it is necessary to successfully synthesise dichlorocyclohexylphosphine. A different synthesis method for dichlorocyclohexylphosphine has to be found or it should be acquired from a commercially available source. The positive molecular modelling results should serve as enough motivation to study the synthesis further.