Mechanistic insights into the hydrocyanation reaction

The hydrocyanation of an alkene is a catalytic carbon-carbon bond formation reaction and the obtained nitriles can be converted into a variety of valuable products. The investigation of this reaction has mainly focused on the DuPont adiponitrile (AdN) process. This process is so far the only example of a large scale industrial application of an alkene hydrocyanation. Adiponitrile is produced from butadiene in 3 steps: the Ni-catalyzed hydrocyanation of butadiene leads to a mixture of 2-methyl-3-butenenitrile (2M3BN) and 3-pentenenitrile (3PN), obtained in varying ratio (typically 2:3) depending on the ligand employed. In a second step, the branched 2M3BN is isomerized to the desired linear 3PN in the presence of similar Ni-catalysts. The last step is the hydrocyanation of 3PN to AdN. The catalyst performance in this process still needs to be improved in terms of activity, especially for the hydrocyanation of 3PN, and, in selectivity for both hydrocyanation steps. Many investigations are also focusing on the hydrocyanation of vinylarenes. Several ligands have been applied in this reaction and their influence on activity, regio- and enantioselectivity has been considered. An overview on the hydrocyanation of alkenes is given in Chapter 1. The chemistry behind this reaction is discussed prevalently from a mechanistic point of view. The reactivity of different classes of substrates is underlined. Mainly examples of the Ni-catalyzed hydrocyanation are reported, along with a brief overview on catalysis based on other metals. In Chapter 2, a new route for the synthesis of the triptycene-based diphosphine ligand Tript(PPh2)2 is described, giving the desired compound in good yield. The corresponding Pt(II)- and Ni(0)-complexes are characterized. In butadiene hydrocyanation the [Ni(cod)(Tript(PPh2)2)] pre-catalyst leads to unprecedented high selectivities for the linear product 3PN, combining concurrently high activity for both, hydrocyanation and isomerization reaction. The double activity of the catalyst enables to reduce the synthesis of 3PN to a one-step procedure consisting of a hydrocyanation followed by an isomerization reaction. This new catalyst could be the key towards process intensification in the future. Chapter 3 describes for the first time an in situ FT-IR spectroscopic study of the isomerization of 2M3BN towards 3PN. The spectra were analyzed comprehensively to obtain conversion profiles from the different band dynamics. Each band was transformed to its second derivative to enhance peak resolution. Calculated spectra of the substrate and the products support the peak assignment. An average conversion profile was calculated from different bands of the substrate and the product, applying a "quasimultivariate analysis" technique to correlate different band dynamics. This approach was validated using advanced chemometrics. Furthermore, these profiles obtained by IR spectroscopic analysis of the formation of 3PN and the consumption of 2M3BN showed a zero order kinetic. The application of new tetraphenol-based diphosphite ligands (TP) in the hydrocyanation reaction is described in Chapter 4. Very high activities were observed in the hydrocyanation of 3-pentenenitrile. Surprisingly, these systems are neither active in the hydrocyanation of butadiene nor do they show any isomerization of 2M3BN. This peculiar behavior of the [Ni(TP)] catalysts was investigated by means of NMR and IR spectroscopy, considering the formation of s-alkyl and p-allyl intermediates. The s-alkyl species formation seems to be prevalent with the TP ligands, while the formation of p-allyl species is disfavored. Since the hydrocyanation of 3PN proceeds via s-alkyl intermediates and the first two steps of the DuPont process via the p-allyl species, these results provide an explanation for the observed catalytic activity. Moreover, the coordination of ZnCl2 to the [Ni(2M3BN)(TP2)] complex was studied by IR spectroscopy. The comparison with a binaphthol-based diphosphite (BIPPP) ligand, often applied in hydrocyanation reactions, is also presented in relation to the coordination and catalytic activity. Chapter 5 reports on the hydrocyanation of styrene. According to present knowledge, this reaction leads predominantly to the branched product 2-phenylpropionitrile (98%). A dramatic inversion of the regioselectivity upon addition of a Lewis acid is observed. Up to 83% of the linear product 3-phenylpropionitrile was obtained applying phosphite ligands in the presence of AlCl3. The mechanism of the Ni-catalyzed reaction and the influence of additional Lewis acids have been elucidated by means of deuterium labeling experiments, NMR studies, and DFT calculations. It was concluded that the selectivity towards the linear product 3-phenylpropionitrile in the presence of AlCl3 is due to the higher stability of the intermediate ¿3-benzyl complex. The selective stabilization of this intermediate in the presence of the Lewis acid leads to the formation of a "steady state" for the ¿3-benzyl intermediate and indirectly promotes the formation of the linear product 3-phenylpropionitrile via the s-alkyl intermediate. Furthermore, the influence of different Lewis acids, such as CuCN, could be predicted via DFT calculations. Chapter 6 deals with the hydrocyanation of simple monoalkenes. So far, this reaction did not attract much attention, due to the lower conversion generally obtained as compared to the hydrocyanation of 1,3-dienes and vinylarenes. Yet, this reaction leads to aliphatic nitriles, which are potentially valuable intermediates for both bulk and finechemical industry. The role of the Lewis acid in the mechanism of the reaction is still not completely clear; in particular its exact role in the increase of the reactivity and regioselectivity towards linear nitriles. A conversion of 89% in the hydrocyanation of 1-octene is reported, applying a binaphthol-based diphosphite (BIPPP) as ligand and AlCl3 as Lewis acid. The competition between ß-(H,D)-elimination and hydrocyanation in the reaction mechanism has been investigated by deuterium labeling experiments. Furthermore, preliminary DFT calculations have been performed to study the deactivation of the Ni-catalyst by formation of dicyano species.