Diastereoselection substrate control

Until the end of the 1980s it was believed that the high reactivity and flexibility of acyclic radicals prevent stereoselective reactions. This opinion changed in 1991 when the review of Porter, Giese, and Curran appeared [1], In the middle of the 1990s, it became obvious that in most cases acyclic radicals follow the same rules of stereoselectivity as non-radicals [2]. This chapter describes diastereoselective, substrate-controlled reactions of acyclic radicals. The chemistry of cyclic radicals, the influence of chiral auxiliaries and of Lewis acids as well as enantioselective radical reactions are reviewed in Chapters 4.2-4.5. Actually, radicals are suitable intermediates for an understanding of stereoselectivity because (a) their conformation can be determined by ESR spectroscopy, and (b) the transition states of synthetically relevant radical reactions are very early on the reaction coordinate. The present ehapter makes use of these features. 

Brown s discovery of the hydroboration reaction in the 1950s opened new avenues for the selective functionalization of olefins. The recognition that acyclic olefins can have well-defined conformational biases allowed the development of diastereoselective, substrate-controlled processes and the advent of the principles of acyclic stereocontrol. Given the central role the hydroboration of olefins has played, it is hardly surprising that the earliest examples in this field involve such transformations. For the organic chemist, diastereoselective and enantioselective hydrometalation reactions have rapidly become an indispensable tool equally useful in simple olefin functionalizations and in the stereoselective construction of highly complex molecules. 

The examples from the preceding discussion catalog the development of the field, which has largely led to increasing sophistication in the implementation of dipolar cycloadditions in target-specific molecule synthesis. The development of diastereoselective substrate-controlled methods as a general synthesis of chiral building blocks by use of nitrile-oxide dipolar cycloaddition reactions has only recently been heralded by the work of Kanemasa and Carreira. 

Boland applied this methodology to Garner s aldehyde, and found the addition to be substrate-controlled rather than reagent-controlled (Scheme 9.13b) [68]. Viny-lepoxide 15 could thus also be obtained with high diastereoselectivity with achiral 9-MeO-9-BBN. 

Diastereoselective hydroformylation can be achieved in special cases through passive substrate control in which conformational preferences are transferred in the corresponding selectivity-determining hydrometalation step [4-6]. A recent example is the highly diastereoselective hydroformylation of a kainic acid derivative (Scheme 17) [64], The selective formation of the major diastereomer has been explained via a reactive substrate conformation in which allylic 1,2-strain has been minimized. In this situation the czs-positioned methylene carbonylmethoxy group controls the catalyst attack to occur from the si face exclusively. 

Scheme 17 Diastereoselective hydroformylation of a kainic acid derivative relying on passive substrate control...

Alternatively, substrate control of diastereoselectivity can rely on attractive catalyst substrate interactions. This requires in general special functional groups which allow for a directed hydroformylation, which is summarized in Sect. 6 (vide infra). 

As described hitherto, diastereoselectivity is controlled by the stereogenic center present in the starting material (intramolecular chiral induction). If these chiral substrates are hydrogenated with a chiral catalyst, which exerts chiral induction intermolecularly, then the hydrogenation stereoselectivity will be controlled both by the substrate (substrate-controlled) and by the chiral catalyst (catalyst-controlled). On occasion, this will amplify the stereoselectivity, or suppress the selectivity, and is termed double stereo-differentiation or double asymmetric induction [68]. If the directions of substrate-control and catalyst-control are the same this is a matched pair, but if the directions of the two types of control are opposite then it is a mismatched pair. 

Figure 32.22 shows the diastereoselective hydrogenation of (R)-/ -keto sulfoxides with Meo-BIPHEP-Ru catalysts [72]. The R chiral center of the substrate matches with the S catalyst, giving the S,R alcohols in >99 1 selectivity, whereas reactions with the R catalyst affords a 6 94 to 10 90 mixture of the S,R and R,R diastereomeric alcohols. The diastereoselection is controlled mainly by the configuration of the catalyst. 

Substrate control This refers to the addition of an achiral enolate (or allyl metal reagent) to a chiral aldehyde (generally bearing a chiral center at the a-position). In this case, diastereoselectivity is determined by transition state preference according to Cram-Felkin-Ahn considerations.2... 

It is also possible to carry out a substrate-controlled reaction with aldehydes in an asymmetric way by starting with an acetylene bearing an optically active ester group, as shown in Eq. 9.8 [22]. The titanium—acetylene complexes derived from silyl propiolates having a camphor-derived auxiliary react with aldehydes with excellent diastereoselectivity. The reaction thus offers a convenient entry to optically active Baylis—Hillman-type allyl alcohols bearing a substituent (3 to the acrylate group, which have hitherto proved difficult to prepare by the Baylis—Hillman reaction itself. 

The substrate-controlled diastereoselective addition of lithiated alkoxyallenes to chiral nitrones such as 123, 125 and 126 (Scheme 8.32) furnish allenylhydroxyl-amines as unstable products, which immediately cydize to give enantiopure mono-orbicyclic 1,2-oxazines (Eqs 8.25 and 8.26) [72, 76]. Starting with (R)-glyceraldehyde-derived nitrone 123, cydization products 124 were formed with excellent syn selectivity in tetrahydrofuran as solvent, whereas precomplexation of nitrone 123 with... 

Access to the corresponding enantiopure hydroxy esters 133 and 134 of smaller fragments 2 with R =Me employed a highly stereoselective (ds>95%) Evans aldol reaction of allenic aldehydes 113 and rac-114 with boron enolate 124 followed by silylation to arrive at the y-trimethylsilyloxy allene substrates 125 and 126, respectively, for the crucial oxymercuration/methoxycarbonylation process (Scheme 19). Again, this operation provided the desired tetrahydrofurans 127 and 128 with excellent diastereoselectivity (dr=95 5). Chemoselective hydrolytic cleavage of the chiral auxiliary, chemoselective carboxylic acid reduction, and subsequent diastereoselective chelation-controlled enoate reduction (133 dr of crude product=80 20, 134 dr of crude product=84 16) eventually provided the pure stereoisomers 133 and 134 after preparative HPLC. 

Fig. 7 Diastereoselective hydrogenation based on catalyst and substrate control...

The simplest case of substrate-controlled diastereoselection is the incorporation of the controlling stereocenter and the prostereogenic center into a cyclohexane or cyclopentane ring. In the classical example of nucleophilic attack on a conformationally anchored cyclohexanone, axial and equatorial attack are possible, leading to diastereomers 1 and 2, respectively. 

O-Benzyllactaldehyde dimethylhydrazone 230 allows a substrate control in the addition reaction of organomagnesium halides, leading almost exclusively to the 5yn-isomer 231 (equation 155) . The resulting hydrazide can be reduced on Raney Ni to the corresponding iyw-aminoalcohol 232. The stereoselective Grignard addition to a similar A-formyl hydrazone 233 proceeds with 92% diastereoselectivity (equation 156). The silylation of the amide nitrogen by TMSCl provides the pure iyw-adduct . 

Hupe, E. Calaza, M. I. Knochel, P. Substrate-controlled highly diastereoselective synthesis of primary and secondary diorganozinc reagents by a hydroboration/B-Zn exchange sequence. Chem. Eur. J. 2003, 9, 2789-2796. 

The stereochemical result is no longer characterized solely by the fact that the newly formed stereocenters have a uniform configuration relative to each other. This was the only type of stereocontrol possible in the reference reaction 9-BBN + 1-methylcyclohexene (Figure 3.25). In the hydroborations of the cited chiral alkenes with 9-BBN, an additional question arises. What is the relationship between the new stereocenters and the stereocenter(s) already present in the alkene When a uniform relationship between the old and the new stereocenters arises, a type of diastereoselectivity not mentioned previously is present. It is called induced or relative diastereoselectivity. It is based on the fact that the substituents on the stereocenter(s) of the chiral alkene hinder one face of the chiral alkene more than the other. This is an example of what is called substrate control of stereoselectivity. Accordingly, in the hydroborations/oxidations of Figures 3.26 and 3.27, 9-BBN does not add to the top and the bottom sides of the alkenes with the same reaction rate. The transition states of the two modes of addition are not equivalent with respect to energy. The reason for this inequality is that the associated transition states are diastereotopic. They thus have different energies—just diastereomers. 

Fig. 3.26. cis-Selective hydration of a chiral, racemic, trisubstituted alkene with induced diastereoselectivity A corresponds to the extent of the substrate control of diastereoselectivity. 

Conversely, the addition of enantiomerically pure chiral dialkylboranes to enantiomerically pure chiral alkenes can also take place in such a way that substrate control and reagent control of diastereoselectivity act in the same direction. Then we have a matched pair. It reacts faster than the corresponding mismatched pair and with especially high diastereoselectivity. This approach to stereoselective synthesis is also referred to as double stereodifferentiation. 

---