Knowles’ catalyst

The results obtained with the Knowles catalyst system have led to a number of useful tools that have helped with the development of other ligand families. 

Although the chemistry of Knowles catalyst system has been highlighted here, many other systems are now available. Some have distinct advantages, such as DuPhos and the ability to reduce enamides with two P-substituents with high enantioselectivity (Chapter 13 see also Scheme 2.15), whereas the monodentate ligands (Chapter 14) have distinct cost advantages. 

Asymmetric hydrogenation with Knowles catalyst, [Rh-DIPAMP-COD]BF4 has been used on a large scale to prepare L-dopa. The general method has been taken and applied to a wide variety of unnatural amino acids that are required by the pharmaceutical industry as starting materials for complex drug candidates. The scope and limitations of the catalytic method are well understood which means that the approach can be applied with confidence to a wide variety of substrates. 

Although it has some limitations, the Rh-DIPAMP system developed by Knowles has been well studied and this allows for a fair degree of certainty that it can be used with untried substrates. For this reason, it has been used to prepare a range of unnatural amino acid derivatives at scale [7, 11]. A model has been reported that allows others to predict whether Knowles catalyst can be used in a productive manner with their desired enamide. The prediction of the model is based on dividing the enamide substrate into four quadrants (Fig. 2) [7]. 

If we start at the top left-hand corner and work round anti-clockwise, the scope and limitations of the system can be explained. This quadrant has to be hydrogen. Any substitution results in a drop in the reaction rates and a significant lowering of enantioselectivity. Thus, / -branched a-amino acid derivatives cannot be accessed with Knowles catalyst. This limitation has been overcome by the use of other catalyst systems such as Burk s DuPHOS [12]. The need for hydrogen at this position... 

Many of the ligands that have been advocated for analogous reductions provide high enantioselectivity and some can provide good turnover numbers and frequencies [4, 5], Knowles catalyst often results in an ee of about 94-95% if the reaction mixture is monitored. However, crystallization of the N-acylamino acid product often results in enantioenrichment [10, 11]. In addition, hydrolysis of the amide to provide the amino acid itself also provides an opportunity for enantioenrichment (Fig. 3) [10]. 

Fig. 5 Mechanism of enamide reductions with Knowles catalyst.

The mechanism for the asymmetric hydrogenation of enamides by Knowles catalyst is well understood due to the work of Halpern (Fig. 5) [20], The intermediates were identified by spectroscopy. The surprising finding was that two catalytic cycles were possible. The one that contains the lower concentrations of intermediates gives rise to the major product isomer as the reaction rates are faster compared with the cycle that has more detectable intermediates. 

Although Knowles catalyst has been known for 30 years, there are still applications that are coming to light. 

The phosphorus analogs of enamide esters, the dimethyl phosphonates can be reduced by Knowles catalyst to provide the corresponding a-amido phosphonates (Fig. 8) [25]. 

The enol acetates of a-keto acids can be accessed by a variation of the azlactone synthesis. They can also be reduced by Knowles catalyst, but ee values are 87-88% and the substrate to catalysis ratio needed for reduction does not allow this to be an economic approach to a-hydroxy acids although the reaction times are reduced compared with the corresponding carboxylate esters [9, 29]. 

Knowles catalyst has been used to prepare a wide range of a-amino acids at scale. L-Dopa is still produced at scale using the original process developed by Monsanto. Other amino acids are produced at a smaller scale for a variety of pharmaceutical applications. 

Despite its age, Knowles catalyst will continue to be a work horse in the fine chemical industry and is still used as the reference system for amino acid synthesis. After 25 years, although ee values may be slightly lower than more modern catalysts, the turnover numbers and turnover frequencies continue to make it a stiff competitor for new catalysts systems to better. 

Perhaps the most famous example of the use of asymmetric hydrogenation at scale for the product of an unnatural amino acid is the Monsanto synthesis of L-dopa, a drug used for the treatment of Parkinson s disease (Scheme 9.19). " The methodology with the Knowles catalyst system has been extended to a number of other unnatural amino acids. ... 

HORNER - KNOWLES - KAGAN Asymmetric Hydrogenation Enantnselective hydrogenation of prochirai olefins with chiral Rh catalysts... 

William Knowles at the Monsanto Company discovered some years ago that u-amino acids can be prepared enantioselectively by hydrogenation of a Z enam-ido acid with a chiral hydrogenation catalyst. (S)-Phenylalanine, for instance, is prepared in 98.7% purity contaminated by only 1.3% of the (H) enantiomer when a chiral rhodium catalyst is used. For this discovery, Knowles shared the 2001 Nobel Prize in chemistry. 

A different approach to making chiral drugs is asymmetric synthesis. An optically inactive precursor is converted to the drug by a reaction that uses a special catalyst, usually an enzyme (Chapter 11). If all goes well, the product is a single enantiomer with the desired physiological effect In 2001, William S. Knowles, Ryogi Noyori, and K. Barry Sharpless won the Nobel Prize in chemistry for work in this area. 

The standard work of Evans [2] as well as a survey of the papers produced in the Journal of Labeled Compounds and Radiopharmaceuticals over the last 20 years shows that the main tritiation routes are as given in Tab. 13.1. One can immediately see that unlike most 14C-labeling routes they consist of one step and frequently involve a catalyst, which can be either homogeneous or heterogeneous. One should therefore be able to exploit the tremendous developments that have been made in catalysis in recent years to benefit tritiation procedures. Chirally catalyzed hydrogenation reactions (Knowles and Noyori were recently awarded the Nobel prize for chemistry for their work in this area, sharing it with Sharpless for his work on the equivalent oxidation reactions) immediately come to mind. Already optically active compounds such as tritiated 1-alanine, 1-tyrosine, 1-dopa, etc. have been prepared in this way. 

Following Wilkinson s discovery of [RhCl(PPh3)3] as an homogeneous hydrogenation catalyst for unhindered alkenes [14b, 35], and the development of methods to prepare chiral phosphines by Mislow [36] and Horner [37], Knowles [38] and Horner [15, 39] each showed that, with the use of optically active tertiary phosphines as ligands in complexes of rhodium, the enantioselective asymmetric hydrogenation of prochiral C=C double bonds is possible (Scheme 1.8). 

Knowles reported the hydrogenation of a-phenylacrylic acid and itaconic acid with 15% and 3% optical purity, respectively, by using [RhCl3(P )3] [P = (R)-(-)-methyl-n-propylphenylphosphine] as homogeneous catalyst [38]. Horner found that a-ethylstyrene and a-methoxystyrene can be hydrogenated to (S)-(+)-2-phe-nylbutane (7-8% optical purity) and (R)-(+)-l-methoxy-l-phenylethane (3-4% optical purity), respectively, by using the complex formed in situ from [Rh(l,5-hexadiene)Cl]2 and (S)-(-)-methyl- -propylphenylphosphine as catalyst [39]. 

From the seminal studies of Sabatier [43] and Adams [44] to the more recent studies of Knowles [45] and Noyori [46], catalytic hydrogenation has been regarded as a method of reduction. The results herein demonstrate the feasibility of transforming catalytic hydrogenation into a powerful and atom-economical method for reductive C-C bond formation. Given the profound socioeconomic impact of al-kene hydroformylation, the development of catalysts for the hydrogen-mediated... 

During the late 1960s, Homer et al. [13] and Knowles and Sabacky [14] independently found that a chiral monodentate tertiary phosphine, in the presence of a rhodium complex, could provide enantioselective induction for a hydrogenation, although the amount of induction was small [15-20]. The chiral phosphine ligand replaced the triphenylphosphine in a Wilkinson-type catalyst [10, 21, 22]. At about this time, it was also found that [Rh(COD)2]+ or [Rh(NBD)2]+ could be used as catalyst precursors, without the need to perform ligand exchange reactions [23]. 

It is interesting to note that a few rales of thumb and myths came out of these early studies. Many of these have been perpetuated for decades, and the myths are only just being put to rest. Knowles showed that only two phosphorus ligands were needed on the metal to achieve reduction, and not three as in Wilkinson s catalyst [10]. The success of DIPAMP and DIOP led to the belief... 

In 1968, Knowles et al. [1] and Horner et al. [2] independently reported the use of a chiral, enantiomerically enriched, monodentate phosphine ligand in the rhodium-catalyzed homogeneous hydrogenation of a prochiral alkene (Scheme 28.1). Although enantioselectivities were low, this demonstrated the transformation of Wilkinson s catalyst, Rh(PPh3)3Cl [3] into an enantioselective homogeneous hydrogenation catalyst [4]. 

---