Sulfoxantphos

Figure 7.8. Unmodified xantphos (a), sulfoxantphos (b) and guanidinium-modified xantphos ligand (c) as used in the biphasic, Rh-catalysed hydroformylation of 1-octene...

The use of the sulfoxantphos ligand (compound (b) in Figure 7.8) in the biphasic hydroformylation of 1-octene with [BMIM][PF6] has been studied by Dupont and coworkers [58], The ligand allowed recycling of the catalyst solution up to four times with no loss in activity or selectivity. Flighly regioselective hydroformylation (n/iso = 13) was reported for a Rh/phosphine-ratio of 4 (100°C, 15 bar syngas pressure). 

These systems have been successfully applied as catalyst media for the biphasic, Rh-catalysed hydroformylation of 1-octene [11], [62]. The catalyst activities obtained with these systems were in all cases equal or even higher than those with the commonly used [BMIM][PF6]. Table 7.4 compares some typical results using Rh-sulfoxantphos in alkylsulfate and sulfonate ionic liquids with analogous experiments in a hexafluorophosphate system. 

Reaction conditions aUcenerCOrEL = 1 1 1, T = 100 °C, 0.2 wt% Rh metal loading, silica 100 p = 10 bar for guanidinium and norbos-Cs, p = 5 bar for sulfoxantphos ionic liquid used for impregnation [BMIM][PF6] a) molar ligand to metal ratio b) ratio of ionic liquid volume to support pore volume c) mol aldehyde per mol rhodium per hour d) ratio between linear and branched aldehyde e) support loaded with ionic liquid only. 

To estimate costs for the liquid-liquid biphasic hydroformylation using ionic liquids, a process was designed for the production of 100,000 tons per year of nonanal. The use of ionic liquids in hydroformylation catalysis is a fairly new technology and exact kinetic data are scarce, thus the TOFs reported for the Rh-sulfoxantphos system [80] have been used to determine catalyst inventory and reactor dimensions. In a similar way the plant design for the SILP process for a production capacity of 100,000 tons per year of butanal has been derived based on preliminary literature results [68]. The process flow sheets for both process variations are shown in Figures 7.12 and 7.13. 

Recently, van Leeuwen and co-workers provided support for the similarity between the active catalytic species in ionic liquids and in organic solvents by spectroscopic investigations [80]. These authors compared the complex formation of [Rh(acac)(CO)2] in the presence of 4 equivalents of sulfoxantphos ligand dissolved in [BMIM][PF6] and... 

The first fixed-bed application of a supported ionic liquid-phase catalyst was hydroformylation of propylene, with the reactants concentrated in the gas phase (265). The catalyst was a rhodium-sulfoxantphos complex in two ionic liquids on a silica support. The supported ionic liquid phase catalysts were conveniently prepared by impregnation of a silica gel with Rh(acac)(CO) and ligands in a mixture of methanol and ionic liquids, [BMIMJPFg and [BMIM][h-C8Hi70S03], under an argon atmosphere. 

C4C1im][PF6] Sulfoxantphos 1-Olefins 7-47 (1.7-61) TOF stable for at least four cycles low selectivity for 1-octene (4.8), high selectivity for 1-decene (61) and 1-octadecene (49). [52]... 

C mJtPFs] Sulfoxantphos 1-Octene 382-411 (6-7) High pressure IR and NMR studies. [53]... 

C4C,im][PF6] [C4C,im][C8H17S04] Continuous fixed-bed gas-phase hydroformylation with non-modified and partially dehydroxynated silica gel propene as substrate sulfoxantphos as ligand with dehydroxynated support, activity and selectivity remained stable for > 60 hours. TOF 1.5-40.8 l b 1.7-23.7 [66] [67]... 

A well established way in multiphasic catalysis to improve the substrate solubility in the catalyst phase is the addition of appropriate co-solvents. This approach has been studied for the hydroformylation of 1-octene with a [Rh(CO)2(acac)]-(sulfoxantphos) catalyst system in [BMIM][PFe] by Dupont et al. [183]. They found that the selectivity to n-nonanal was best in pure [BMIM][PF6] compared to the same ionic liquid with e.g. added co-solvent toluene. They reported catalyst... 

The use of catalyst systems containing the bidentate phosphine ligand sulfoxantphos proved particular interesting, as excellent n/iso ratios of 23 (i.e. linear product selectivities up to 96%) were attained with these systems (entries 12 to 14). 

Remarkably, at a low ligand/rhodium ratio of 2.5 (entries 9 and 10), the selectivity of the Rh-sulfoxantphos system was significantly lower and comparable to the best selectivity obtained with the monodentate phosphines. This results was in contrast to analogous experiments performed in ionic liquid-liquid biphasic mode where high selectivities were obtained [81,98], and suggested some ligand loss due to reactions between the ligand and the solid surface. 

Importantly, it was also realized in these initial studies that the catalysts deactivated in prolonged use with simultaneous decrease in catalytic activity and selectivity, independent of the type of ionic liquid, ionic liquid loading a, and the ligand/rhodium ratio of the system. However, additional studies demonstrated a suitable thermal pre-treatment of the silica support to be necessary to obtain stable SILP catalysts. This technique was applied to prepare Rh-sulfoxantphos catalysts with [BMIMjfn-CsHiyOSOs] and similar systems without ionic liquid. Both systems were compared in 60 h continuous propene hydroformylation reactions using similar conditions as previously reported (Table 5.6-4, entries 11 and 12, Fig. 5.6-8) [93]. 

Fig. 5. 6-8 Continuous hydroformylation of propene with siiica Rh-sulfoxantphos/[BMiM][n-C8Hi70S03] cataiysts (iigand/rhodium = 10) having ionic liquid loadings of a = 0 ( , ) and a = 0.1 (o. ). Activity (closed symbols) and selectivity (n/iso ratio, open symbols) [93].

In the reaction using the Rh-sulfoxantphos catalyst without ionic liquid, the activity and selectivity decreased sharply after 5-10 h on stream, while the SILP system (a = 0.1) reached its maximum activity only after 30 h and maintained this level stable up to 60 h (i.e. TON > 2400) along with a high selectivity corresponding to a n/iso ratio of 21-23. It can also be deduced that the apparently negative ionic liquid effect on the catalyst activity measmed in the short time reactions can be explained by a delayed formation of the catalytically active species. 

---