isomerization. Increasing the H 2 /CO ratio while maintaining a low CO partial
pressure increases the initial TOF substantially along with the aldehyde l:b ratio.
But all the monometallic catalysts deactivate before they can complete the 1,000
turnovers (100% conversion of alkene). This is tied into Rh-induced phosphine
orthometalations and P–Ph or P–benzyl bond cleavage reactions that lead to
monometallic catalyst deactivation [4]. Bisbi is especially susceptible to deactivation under these conditions, barely making it past 50% conversion of the alkene.
The mainly alkylated et,ph-P4 ligand does not seem to suffer from Rh-induced
phosphine fragmentations under these conditions, but does tend to lose a rhodium
center and “fragment” losing the bimetallic cooperativity.
The Xantphos-based monometallic hydroformylation catalyst also deactivates
before converting all the alkene to aldehyde under reduced CO pressures but
behaves differently in that both the initial TOF and aldehyde l:b regioselectivity
decreases with higher H 2 /CO ratios while keeping the CO partial pressure low. We
believe the reason for this is that the Xantphos ligand can coordinate to the Rh
center via the central oxygen atom, but this is a weaker interaction than the Rh–P
bonds. At low CO partial pressures, the Xantphos favors the κ
3 -mode using the
phosphines and central oxygen atom leading to less reactive and selective catalysts.
Higher CO pressures favor dissociation of the Xantphos oxygen leading to a more
active catalyst with a folded Xantphos configuration that is more sterically directing
relative to the flatter Xantphos structure when the oxygen atom is coordinated to the
rhodium [47, 48].
5 Catalyst Binding Site Considerations
The high product aldehyde regioselectivity observed for our bimetallic catalyst,
either dicationic or monocationic, is a result of the relatively rigid dinuclear
structure of 11r/11r*/15r and the proper arrangement of steric effects on the et,
ph-P4 ligand/catalyst. When an alkene coordinates to a typical monometallic
square-planar hydroformylation catalyst, the other ligands will bend away to form
a trigonal bipyramid or square pyramid, which is the least congested coordination
geometry (Fig. 15). This geometric reorganization results from electronic orbital
rehybridization on the metal center and causes the steric directing groups on the
phosphine ligands to swing away from the incoming alkene substrate. This, in turn,
reduces the steric effectiveness of the phosphine for orienting the alkene to insert
properly into the Rh–H bond to give the desired linear alkyl intermediate species.
The bimetallic catalysts, however, cannot distort this way on alkene coordination because the Rh–Rh bond and bridging ligand(s) prevents any significant
movement of the ligand environment away from the alkene. Minimizing the
geometric reorganization about the rhodium maximizes the steric effect of the et,
ph-P4 ligand, directing the alkene insertion into the M–H bond to form a linear
alkyl group, which goes on to form the linear aldehyde product.
Bimetallic Homogeneous Hydroformylation
25
pressure increases the initial TOF substantially along with the aldehyde l:b ratio.
But all the monometallic catalysts deactivate before they can complete the 1,000
turnovers (100% conversion of alkene). This is tied into Rh-induced phosphine
orthometalations and P–Ph or P–benzyl bond cleavage reactions that lead to
monometallic catalyst deactivation [4]. Bisbi is especially susceptible to deactivation under these conditions, barely making it past 50% conversion of the alkene.
The mainly alkylated et,ph-P4 ligand does not seem to suffer from Rh-induced
phosphine fragmentations under these conditions, but does tend to lose a rhodium
center and “fragment” losing the bimetallic cooperativity.
The Xantphos-based monometallic hydroformylation catalyst also deactivates
before converting all the alkene to aldehyde under reduced CO pressures but
behaves differently in that both the initial TOF and aldehyde l:b regioselectivity
decreases with higher H 2 /CO ratios while keeping the CO partial pressure low. We
believe the reason for this is that the Xantphos ligand can coordinate to the Rh
center via the central oxygen atom, but this is a weaker interaction than the Rh–P
bonds. At low CO partial pressures, the Xantphos favors the κ
3 -mode using the
phosphines and central oxygen atom leading to less reactive and selective catalysts.
Higher CO pressures favor dissociation of the Xantphos oxygen leading to a more
active catalyst with a folded Xantphos configuration that is more sterically directing
relative to the flatter Xantphos structure when the oxygen atom is coordinated to the
rhodium [47, 48].
5 Catalyst Binding Site Considerations
The high product aldehyde regioselectivity observed for our bimetallic catalyst,
either dicationic or monocationic, is a result of the relatively rigid dinuclear
structure of 11r/11r*/15r and the proper arrangement of steric effects on the et,
ph-P4 ligand/catalyst. When an alkene coordinates to a typical monometallic
square-planar hydroformylation catalyst, the other ligands will bend away to form
a trigonal bipyramid or square pyramid, which is the least congested coordination
geometry (Fig. 15). This geometric reorganization results from electronic orbital
rehybridization on the metal center and causes the steric directing groups on the
phosphine ligands to swing away from the incoming alkene substrate. This, in turn,
reduces the steric effectiveness of the phosphine for orienting the alkene to insert
properly into the Rh–H bond to give the desired linear alkyl intermediate species.
The bimetallic catalysts, however, cannot distort this way on alkene coordination because the Rh–Rh bond and bridging ligand(s) prevents any significant
movement of the ligand environment away from the alkene. Minimizing the
geometric reorganization about the rhodium maximizes the steric effect of the et,
ph-P4 ligand, directing the alkene insertion into the M–H bond to form a linear
alkyl group, which goes on to form the linear aldehyde product.
Bimetallic Homogeneous Hydroformylation
25