Ponytails catalysts

The interest in catalyst recyclability has led to the development of biphasic catalysts for hydro-boration.22 Derivitization of Wilkinson s catalyst with fluorocarbon ponytails affords [Rh(P (CH2)2(CF2)5CF3 3)3Cl] which catalyzes FIBcat addition to norbornene in a mixture of C6FnCF3 and tetrahydrofuran (TF1F) or toluene (alternatively a nonsolvent system can be used with just the fluorocarbon and norbornene) to give exo-norborneol in 76% yield with a turnover number up to 8,500 (Scheme 4). Mono-, di- and trisubstituted alkenes can all be reacted under these conditions. The catalyst can be readily recycled over three runs with no loss of activity.23... 

The concept makes use of the complimentary strengths and weaknesses of the two unconventional media. While ionic liquids are known to be excellent solvents for many transition metal catalysts, the solubility of most transition metal complexes in scC02 is poor (if not modified with e. g. phosphine ligands with fluorous "ponytails" [64]). However, product isolation from scC02 is always very simple, while from an ionic catalyst solution it may become more and more complicated depending on the solubility of the product in the ionic liquid and on the product s boiling point. 

Figure 1.6 Like dissolves like perfluoroalkyl ponytails make phosphines more soluble in a fluorous solvent. These phosphines are suitable ligands for metal catalysts, and will therefore aid the solubility of these catalysts in fluorous solvents...

If nonvolatile liquids are to be used to avoid the problems associated with volatile organic solvents, then it is very desirable that there is some convenient way of recovering the reaction products from the liquid. This approach is used in the biphasic systems described in Chapters 2-5. In the fluorous biphase (Chapter 3), reagents and catalysts are fine-tuned by adding perfluoroalkyl chains, known as ponytails , to ensure that only those chemicals will mix with the fluorous layer. Purification is simply a matter of separating the two phases. Transition metal catalysts with fluorous ligands will remain in the fluorous phase, and the whole catalyst-solvent mixture may be reused for another batch of reactions, as shown schematically in Figure 1.20b. 

Figure 8.7 Fluorous and control ligands screened to gauge the effect of the fluorous ponytails on catalysis using Wilkinson s catalyst analogues...

For systems where the catalyst is required in the CO2 phase a modification of the hgand periphery to increase the solubihty in the supercritical medium is usually necessary. This has to be mostly done via introduction of perfluorinated tags ( ponytails ) which causes expensive and/or sometimes difficult synthetic operations. 

Following the publication of the first example of fluorous biphase catalysis by Horvath and Rabai in 1994 [1], the immediate focus was to develop catalysts that would exhibit very biased partition coefficients with respect to fluorous and organic solvents. Such liquids are normally immiscible at room temperature. This was done by attaching ponytails of the formula (CH2)m(CF2) -iCF3 (abbreviated (CH2)mRf )> including arrays emanating from silicon atoms [2]. Catalysis was then effected at elevated temperatures, where fluorous and organic solvents are commonly miscible, with prod-uct/catalysis separation at the low-temperature two-phase limit. 

As detailed elsewhere, the fluorous palladacycle acetates and hahdes 7 and 8 were synthesized [38,39]. These feature three Rfg ponytails, and were poorly soluble in common organic solvents at room temperature, and insoluble in DMF. However, they were very soluble in DMF at higher temperatures. All were effective catalyst precursors for Heck reactions (100-140 °C), and precipitated (as the halides) upon cooling. However, a number of control experiments established that 7 and 8 served as steady-state sources of colloidal palladium nanoparticles, formed anew with each cycle imtil the palladacycles were exhausted. These, or low-valent Pd(0) species derived therefrom, were the true catalysts. 

To our knowledge, there have been no previous attempts to develop a broad class of molecular catalysts that have temperature-dependent solubilities. When molecular catalysts are covalently bound to polymeric supports, they generally assume the solubihty properties of the host polymer. In the above fluorous catalysts, we Hke to think that a short segment of polymer is being grafted onto a molecular catalyst, hi other words, the ponytails can be viewed as pieces of Teflon , which impart more and more of the solubility characteristics of the polymer as they are lengthened. 

In conclusion, the new recycHng procedures described above offer virtually unhmited possibihties for optimization and catalyst engineering . The lengths and other structural features of the ponytails are easily varied. There are innumerable types of possible fluoropolymer supports, as well many additional classes of fluorous supports. Accordingly, a variety of further refinements and developments can be expected in the near future. 

In order to perform fluorous biphasic catalysis the (organometallic) catalyst needs to be solubilized in the fluorous phase by deploying fluorophilic ligands, analogous to the hydrophilic ligands used in aqueous biphasic catalysis. This is accomplished by incorporating so-called fluorous ponytails . 

In Chapter 7 we have already discussed the use of fluorous biphasic systems to facilitate recovery of catalysts that have been derivatized with fluorous ponytails . The relatively high costs of perfluoroalkane solvents coupled with their persistent properties pose serious limitations for their industrial application. Consequently, second generation methods have been directed towards the elimination of the need for perfluoro solvents by exploiting the temperature-dependent solubilities of fluorous catalysts in common organic solvents [42]. Thus, appropriately designed fluorous catalysts are soluble at elevated temperatures and essentially insoluble at lower temperatures, allowing for catalyst recovery by simple filtration. 

Alternatively, an insoluble fluorous support, such as fluorous silica [43], can be used to adsorb the fluorous catalyst. Recently, an eminently simple and effective method has been reported in which common commercial Teflon tape is used for this purpose [44]. This procedure was demonstrated with a rhodium-catalyzed hydrosilylation of a ketone (Fig. 9.27). A strip of Teflon tape was introduced into the reaction vessel and when the temperature was raised the rhodium complex, containing fluorous ponytails, dissolved. When the reaction was complete the temperature was reduced and the catalyst precipitated onto the Teflon tape which could be removed and recycled to the next batch. 

The possibility of asymmetric induction under the fluorous biphase conditions was first speculated upon by Horvath and Rabai [10], and this year has seen the first report of asymmetric catalysis in a fluorous biphase [69]. Two, C2 symmetric salen ligands (29a, b) with four C8Fi7 ponytails have been prepared (Scheme 5) and their Mn(II) complexes evaluated as chiral catalysts for the aerobic oxidation of alkenes under FBS-modified Mukaiyama conditions. Both complexes are active catalysts (isolated yields of epoxides up to 85%) under unusually low catalyst loadings (1.5% cf. the usual 12%). Although catalyst recovery and re-use was demonstrated, low enantioselectivities were observed in most cases. 

---