Metal-catalyzed asymmetric hydroformylation : towards the understanding of stereoselection processes

The metal-catalyzed hydroformylation reaction is one of the most important homogeneously catalyzed reactions. The asymmetric hydroformylation is an elegant an atom-efficient way to produce chiral aldehydes, which are valuable intermediates in fine chemistry. Much is still unknown about the factors determining the enantioselectivity in this reaction. In order to develop efficient catalysts for this reaction, more knowledge about the stereoselection mechanism is required. In this thesis, the asymmetric hydroformylation reaction was studied from four different points of view. An overview of the history of the reaction and the most important recent developments were described in chapter 1. The selective hydroformylation of 1,1-disubstituted alkenes has been described in chapter 2. For this class of substrates, the linear hydroformylation product contains a stereogenic center. High enantioselectivity, together with high regioselectivity towards the linear product were obtained applying sugar-based diphosphite ligands in the rhodium-catalyzed hydroformylation of methyl methacrylate (MMA) and ??- methylstyrene. In case of MMA, an ee of 71% was reached, which is the highest ee ever obtained for this substrate. The remarkable selectivity of these catalytic systems was further investigated by deuterioformylation experiments using MMA as a substrate. Analysis of the reaction mixture by 2H NMR spectroscopy and GC-MS analysis showed that both at T = 100 and 60°C, coordination of the alkene to the rhodium center as well as alkyl formation are highly reversible. Analysis by 1H NMR spectroscopy showed the presence of hydrogen atoms in the aldehyde functionality, indicating that a source of H2 or HD was present in the gas phase. This was confirmed by MS analysis of the residual gas after a deuterioformylation experiment. Monitoring the gas composition during a deuterioformylation experiment as a function of time enabled for the first time, to the best of our knowledge, to estimate the rate of ??-hydrogen elimination versus the rate of hydroformylation. It was shown that ??-hydrogen elimination is at least 29 times faster than hydroformylation in the system studied. We feel that this new method bears great potential to study the performance of catalysts, especially in the isomerizing hydroformylation of internal alkenes. In chapter 3, the coordination behavior of three selected chiral bidentate phosphorus ligands towards the rhodium center in trigonal bipyramidal hydrido-carbonyl complexes was investigated by high pressure in-situ NMR and FT-IR spectroscopy. All three ligands were found to coordinate to the metal center in one mode exclusively. This is assumed to be a prerequisite for efficient chiral induction during catalysis. (R,R)-Ph-BPE and (S)- Binapine were shown to coordinate in equatorial-axial fashion. Moreover, (S)-Binapine was found to form dimeric species of the type [Rh(CO)2(P^P)]2 under hydroformylation conditions. (S,S)-Kelliphite coordinates to the rhodium center in equatorial-equatorial fashion. All three ligands showed fluxional behavior on the NMR timescale. Slow exchange limits were reached at low temperature. For the [RhH(CO)2((S,S)-Kelliphite)] complex, two ee-coordinated conformations were detected in HP FT-IR spectroscopy. This was confirmed by computational methods. In chapter 4, the synthesis and application in the rhodium-catalyzed hydroformylation of styrene of a series of new ligands based on benzo[b]thiophene was described. Phosphine-phosphonite as well as monodentate and bidentate phosphine ligands were considered. The influence of the ligand structure on the activity and selectivity of the corresponding catalysts was investigated by varying the substituents of both the phosphine and the phosphonite moiety. The molecular structures of a P-stereogenic, borane-protected monodentate phosphine as well as a bidentate diphosphine ligand were investigated by X-ray crystal structure determination. HP NMR spectroscopy showed that the latter ligand coordinates in ea fashion to the rhodium center in a tbpy hydrido- carbonyl complex. In chapter 5, the platinum-catalyzed hydroformylation was considered. Asymmetric hydroformylation using the classical platinum/tin catalyst [PtCl(SnCl3)((R,R)-XantBino)] resulted in 77% ee for vinyl acetate and 80% ee for allyl acetate. Hydroformylation of 4- methylstyrene resulted in 12.4% ee with a b/l ratio of 78:22 and a chemoselectivity of 79%. The platinum/tin system based on a diphosphine ligand with a triptycene backbone showed low activity in the hydroformylation of 1-octene and 2-octene. 31P NMR studies together with X-ray crystal structure determination revealed the formation of a pincer- type [Pt(PCP)(SnCl3)] complex. Besides the classical platinum/tin catalysts, also in-situ generated cationic [PtMe2(P^P)]/BR4 systems were applied in the hydroformylation reaction. The in-situ generated catalyst based on [PtMe2(Xantphos)] showed activity in the hydroformylation of 1-octene. On the other hand, no activity towards hydroformylation or isomerization was observed for 2-octene. The diphosphonite complex [PtMe2((R,R)-XantBino)]/B(C6F5)3 did not show any activity for styrene. For 1- octene on the other hand, some hydroformylation activity was observed. In none of the cases, hydrogenation by-products were detected, in clear contrast to the classical platinum/tin-systems.