Rhodium-nitrene

Padwa has shown that rhodium-catalyzed oxidation of indolyl carbamate 67 employing either Phl(OAc)2 or Phl=0 follows a path similar to that of the D-aUal carbamate (Scheme 17.26) [95]. In principle, indole attack of the putative rhodium-nitrene generates zwitterion 68, which is trapped subsequently by an exogenous nucleophile. Spiro-oxazolidinone products (for example, 69) are isolated as single diastereomers in yields ranging from 50 to 85%. As an intriguing aside, Padwa has found that certain carbamates react with Phl=0 in the absence of any metal catalyst to furnish oxazoHdinone products. This result may have implications for the mechanism of the rhodium-catalyzed process, although it should be noted that control experiments by Espino and Du Bois confirm the essential role of the metal catalyst for C-H amination [57]. 

Analysis of the reaction KIE using monodeuterated sulfamate 106 reinforces the proposed rhodium-nitrene model (Eq. 9). The ratio of H/D products 107/108 is a direct measure of the KIE and has been found to equal 1.5 0.2 by integration of the C NMR spectrum. An equivalent value is obtained for nitrene C-H insertion upon photolysis of a carbamoyl azide, and is taken as evidence for a nonlinear transition state. As noted previously, Muller s use of NsN=lPh with partially deuterated adamantane gives... 

Exposure of the (cycloalkenyl)methyl carbamates 304 to iodosylbenzene in the presence or absence of Rh2(OAc)4 gives the tricyclic aziridines 305 (Scheme 88) (02OL2137). Reactions of 305 with nucleophiles, facilitated with tosic acid or lithium perchlorate, proceed with cleavage of the C-N edge bond and afford the tf 7 z-spirooxazolidinones 306. Intramolecular aziridination of the indolyl carbamate 307 with DAIB, on the other hand, requires Rh(II)-catalysis and leads directly to the acetoxy-substituted YDi-spirooxazolidinone 308 (Scheme 88) (02OL2137). When iodosylbenzene is used instead of DAIB and alcohols are available in the reaction medium, alkoxy-substituted syn-spirooxazolidinones 309 are obtained. Whereas the conversion of 304 to 305 appears to proceed by direct cyclization of intermediate iminoiodanes, the production of 308 from 307 was attributed to the intervention of a rhodium nitrene, which collapses to 308 through zwitterionic intermediates (02OL2137). 

This chapter vhll focus only on the most recent aspect of rhodium nitrene chemistry to perform C—H amination of alkanes, which has not been covered in previous reviews. The literature has been surveyed from 2001 to September 2008. 

By analogy with the rhodium carbene intermediate proposed in the C H insertion reaction with diazo compounds, C— H amination is believed to proceed via a rhodium nitrene species, although such an intermediate has never been characterized. However, as chiral dimeric rhodium complexes lead to the formation of enantioen-riched amination products, it suggests that the metal center is closely associated with the reactive nitrogen during the C—H insertion step. Both a rhodium nitrene or rhodium phenyliminoiodinane species may be involved (Figure 5.2). 

This is in agreement with a transition state in which the C—H bond is at least partially broken. The current transition state hypothesis for the C—H insertion with a rhodium nitrene species involves an asynchronous-concerted pathway, in which the C—H—N angle is smaller than 180° (Figure 5.3). 

The coordination of the N-tosyloxycarbamate to the rhodium catalyst is required for the deprotonation to occur, as no reaction is observed between potassium carbonate and the N-tosyloxycarbamate alone. The formation of the rhodium nitrene may proceed through the formation of a sulfonyloxy metal species which rapidly liberates the tosylate group, or the deprotonation may be concerted with the departure of the leaving group. It is postulated that the formation of the rhodium nitrene is the rate-determining step, since the rate of C H insertion is the same for a deuterated and undeuterated substrate (Eqs. (5.28) and (5.29)). 

Furthermore, the C—H insertion with N-tosyloxycarbamates appears also to involve a singlet rhodium nitrene species, as the reaction is sterospecific, and no ring opening product is observed with a radical clock substrate (Scheme 5.11). 

In the presence of a chiral catalyst such as rhodium(II) (5)-/V l,8-naphthanoyl-tert-leucinate dimer, Troc-amino indane was produced with 56% yield and 2.57 1 enantiomeric ratio. In contrast to other methods, no hypervalent iodine reagent (typically used stoichiometrically or in excess and forming iodobenzene as by-product) is required for oxidation of the amine component. However, a slight excess of the aromatic alkane component (5 equiv) must be used to achieve good conversions. The reactivity of rhodium nitrenes generated from 2,2,2-trichloroethyl-/V-tosyloxycarbamate with aliphatic alkanes is similar to the one observed with metal nitrenes obtained from the oxidation of sulfamate with hypervalent iodine reagent. Troc-protected amino cyclohexane and cyclooctane were obtained, respectively, in 73 and 62% yields when 2 equiv of alkanes was used, whereas yields up to 85% were observed with 5 equiv (eq 3). 

Parker KA, Chang W. Regioselectivity of rhodium nitrene insertion. Syntheses of protected ycals of L-damiosamine, D-saccharosamine, and L-ristosamine. Org Lett. 2005 7 1785-1788. 

---