Tom G. Driver

Recent advances in transition metal-catalyzed N-atom transfer reactions of azides

Transition metal-catalyzed N-atom transfer reactions of azides provide efficient ways to construct new carbon-nitrogen and sulfur-nitrogen bonds. These reactions are inherently green: no additive besides catalyst is needed to form the nitrenoid reactive intermediate, and the by-product of the reaction is environmentally benign N(2) gas. As such, azides can be useful precursors for transition metal-catalyzed N-atom transfer to sulfides, olefins and C-H bonds. These methods offer competitive selectivities and comparable substrate scope as alternative processes to generate metal nitrenoids.

Dirhodium(II)-Catalyzed Intramolecular C-H Amination of Aryl Azides

The development of new transition-metal-catalyzed methods for selective functionalization of carbon–hydrogen bonds continues to be an active area of research. Whereas many transition-metal complexes exhibit activity, rhodium(II) dimers are well established to react with a-diazo compounds or sulfonyliminoiodinanes 6] to access metal carbenoids or nitrenoids, which can functionalize proximal aliphatic C H bonds to form new C C bonds or C N bonds in a stereoselective manner. Transition-metal-mediated formation of new carbon–nitrogen bonds from vinyl or aryl C H bonds, however, is much less common. Azides can be employed in the amination of aromatic or vinyl C H bonds. Thermolysis or photolysis of azides produces nitrenes, 9] which react with proximal C H bonds to form N-heterocycles. Nitrenes, however, are highly reactive and can decompose into a variety of byproducts, including amines, azobenzenes, or tars. Whereas metalmediated nitrogen atom transfer reactions from azides are well-known to attenuate this extreme nitrene reactivity, dirhodium(II) carboxylates have been rarely employed to catalyze these processes despite their proven utility in other related atom transfer reactions. Since azides are readily available, 17] their use in new transition-metal-mediated methods that create new C N bonds is highly appealing. We recently discovered that indoles could be generated from azidoacrylates 2 through exposure to catalytic amounts of rhodium(II) perfluorobutyrate. Whereas this reaction exhibited a broad substrate scope, it required an a-azidomethylacetate, which restricted product formation to 2-indolecarboxylate esters. Achievement of indole synthesis from aryl azides 3 through rhodium-catalyzed vinyl C H bond amination would address this limitation as a broader range of aryl azides are readily available from commercial starting materials in two mild, functional group tolerant steps: palladium-catalyzed Suzuki cross-coupling of 2-bromoanilines and subsequent diazo transfer would produce 3. The combination of this potential method with our earlier one would enable the rhodium-catalyzed synthesis of indoles by the formation of either the aryl C N bond (from azidoacrylates) or the creation of the vinyl C N bond (from aryl azides; Scheme 1). As the thermal variant of this reaction, the Sundberg indole synthesis, requires heating of the poten-

Rh(2)(II)-catalyzed synthesis of carbazoles from biaryl azides.

An array of carbazoles (23 examples) can be synthesized from substituted biaryl azides at 60 degrees C using substoichiometric quantities of Rh(2)(O(2)CC(3)F(7))(4) or Rh(2)(O(2)CC(7)H(15))(4).

Rh2(II)-Catalyzed Intramolecular Aliphatic C–H Bond Amination Reactions Using Aryl Azides as the N-Atom Source

Rhodium(II) dicarboxylate complexes were discovered to catalyze the intramolecular amination of unactivated primary, secondary, or tertiary aliphatic C-H bonds using aryl azides as the N-atom precursor. While a strong electron-withdrawing group on the nitrogen atom is typically required to achieve this reaction, we found that both electron-rich and electron-poor aryl azides are efficient sources for the metal nitrene reactive intermediate.

Iron(II) Bromide-Catalyzed Synthesis of Benzimidazoles from Aryl Azides

The identity of the ortho-substituent of an aryl azide influences its reactivity toward transition metals. Substitution of a vinyl group with an imine disables rhodium(II)-mediated C-H amination and triggers a Lewis acid mechanism catalyzed by iron(II) bromide to facilitate benzimidazole formation.

Intramolecular Fe(II)-Catalyzed N–O or N–N Bond Formation from Aryl Azides

Iron(II) bromide catalyzes the transformation of aryl and vinyl azides with ketone or methyl oxime substituents into 2,1-benzisoxazoles, indazoles, or pyrazoles through the formation of an N-O or N-N bond. This transformation tolerates a variety of different functional groups to facilitate access to a range of benzisoxazoles or indazoles. The unreactivity of the Z-methyloxime indicates that N-heterocycle formation occurs through a nucleophilic attack of the ketone or oxime onto an activated planar iron azide complex.

Rhodium-Catalyzed Synthesis of 2,3-Disubstituted Indoles from β,β-Disubstituted Stryryl Azides

Transition metal-catalyzed migratorial processes that form new carbon–carbon bonds can enable the formation of complex products from readily accessible, simple starting materials. Controlling the selectivity of the migration step is critical to the success of these transformations.[1] Sequential reaction processes that involve metal nitrenes are rare despite their electrophilicity,[2] which enables reaction with carbon–hydrogen bonds or olefins.[3–6] Our mechanistic study of rhodium(II)-catalyzed carbazole formation from biaryl azides which suggested that C–N bond formation preceded C–H bond cleavage through a 4π-electron–5-atom electrocyclization.[7] Consequently, we anticipated that substrates lacking functionalizable C–H bonds might participate in a migratorial process where a new C–C bond is formed in addition to the C–N bond. In support of this hypothesis, rhodium octanoate catalyzed the conversion of β,β-diphenylstryryl azide 1 to 2,3-diphenylindole 3 (Scheme 1).[8] This result, however, does not indicate whether this process can be rendered selective for styryl azides 4 that contain two different β-substituents to form 2,3-disubstituted indoles. Because these N-heterocycles are important pharmaceutical scaffolds,[9] new methods, which streamline their synthesis, remain an ongoing goal.[10,11] Herein, we report our initial studies that resulted in the development of a general method to form 2,3-disubstituted indoles—as single regioisomers—from readily available β,β-disubstituted stryryl azides. Scheme 1 Potential for selective 2,3-disubstituted indole formation. The effect of transition metal complexes on the desired migration was investigated using a mixture of the E- and Z-isomer of β,β-disubstituted aryl azide 8 (Table 1). This azide is readily accessible in two steps from commercially available 2-nitrobenzaldehyde.[12] Examination of a range of dirhodium(II) complexes revealed that selective formation of 9 was obtained using with [Rh2(O2CC3F7)4],[8,13] [Rh2(O2CC7H15)4], or [Rh2(esp)2][14] (Table 1, entries 1–7).[15] Importantly, both the E- and Z-isomer of 8 were converted to indole 9 revealing that the selectivity of the reaction did not depend on the stereochemistry of the starting material. Other rhodium carboxylate complexes provided attenuated selectivities or reduced yields. Other transition metal complexes, such as [(cod)Ir(OMe)2],[16] [Co(tpp)],[17] RuCl3,[18] or copper salts,[19] known to decompose azides or π-Lewis acids,[20] did not promote indole formation (Table 1, entries 8–13). Consequently, the reaction conditions were further optimized using rhodium hexaflourobutyrate, and incomplete conversions were observed when either the catalyst loading or the reaction temperature was lowered (<5 mol%; <70°C). The optimal solvent was found to be either toluene or dichloroethane. Purification proved to be facile: analytically pure indole was obtained by filtering the reaction mixture through a pipette of alumina. Table 1 Development of optimal conditions for indole formation. Using these optimized conditions, the scope and limitations of the rhodium(II)-catalyzed formation of 2,3-disubstituted indoles from β,β-disubstituted stryryl azides was examined (Table 2). In every example, only aryl group migration was observed even if the electronic nature of the aryl azide moiety was modulated. High yields were observed with electron-donating substituents such as methoxide (Table 2, entries 1 and 2). Electron-withdrawing groups also did not lower the reaction yield or migration selectivity (Table 2, entries 3–8). Among these, azides bearing potentially reactive bromides, esters, or sulfones were competent substrates in our process. The reaction was also not sensitive to the steric nature around the azide: nearly quantitative yield of 12a was observed with 11a, which contained two ortho-substituents. Purification of every 2,3-disubstituted indole by simple filtration through alumina further underscores the synthetic utility of our reaction. Table 2 Scope of Rh2II-catalyzed migratorial reactions. The nature of the migrating group on the aryl azide was subsequently investigated (Table 3). For these substrates, only aryl group migration was observed. While rhodium perfluorobutyrate was a competent catalyst, [Rh2(esp)2] provided the highest yields of the reaction. Only indole 15a was observed when the tether was shortened (Table 3, entry 1). Appending the electron-withdrawing trifluoromethyl group or the electron-donating methoxy group to the migrating arene did not change the outcome of the reaction (Table 3, entries 2 and 3). In both cases, only aryl group migration was observed. High yields and selective formation of indole 15d was obtained when an oxygen atom was incorporated into the tether. The reaction was not limited to ring expansion: despite changing the electronic nature of the migrating aryl group, only indoles 15e and 15 f were formed from azides 14 e and 14 f. Table 3 Scope of Rh2II-catalyzed 2,3-disubstituted indole formation. The effect of ring size on the reaction efficiency was further examined using styryl azides 17. For this series of substrates, rhodium octanoate proved to be the most reliable catalyst. While ring-expanded products were formed from 4-, 5-, and 6-membered substrates, poor conversion was observed for 7-membered 17d (Table 3, entries 7–10). Varying the electronic nature of the aryl azide did not attenuate the yield of the reaction (Table 3, entries 11–13). Oxygen atoms were tolerated in the tether without lowering the yield of the ring expansion (Table 3, entry 14). While many mechanisms are possible to explain the reaction outcome, our data suggests that the migration occurs once an intermediate (21) is generated with positive charge on the benzylic carbon. We propose that this intermediate is formed by the mechanism outlined in Scheme 2. Coordination of the rhodium carboxylate complex to the azide produces either α-19 or γ-19.[21] Extrusion of N2 from 19 forms rhodium nitrene 20,[22] which participates in a 4π-electron–5-atom electrocyclization to establish the carbon–nitrogen bond in 21.[7] Aryl migration forms the more stable tertiary iminium ion 22, which tautermizes to produce 9. Alternatively, the ortho-double bond could assist in N2 extrusion to form the intermediate 23, or this intermediate could be formed from [2+1] cycloaddition of the pendant double bond with the electrophilic metallonitrene 20. While 23 is strained,[23] its intermediacy would account for the enhanced reactivity of azides with unsaturated ortho-substituents. Scheme 2 Potential mechanisms for indole formation. We performed several experiments to test the validity of our mechanism. To examine whether N2 was lost before C–N bond formation, we performed an intermolecular competition experiment between azides 11 a and 8 (Scheme 3). Our previous Hammett correlation study indicated that N2 extrusion occurred faster with electron-rich aryl azides.[7] Acceleration of metallonitrene formation was attributed to the ability of the electron-donating group to assist in N2 loss (24 to 25). In contrast, if N2 loss occurred simultaneously with C–N bond formation, we anticipated that 8 would react faster because the azide moiety was more electrophilic than in 11a. To test these assertions, a 1:1 mixture of styryl azides 11 a and 8 were exposed to reaction conditions. Despite the increased steric pressure around the azide, the more electron-rich substrate reacted faster to produce indole 12a as the major product to support our proposed electrocyclization mechanism. Scheme 3 Intermolecular competition experiment. If the migration mechanism involved the formation of a partial positive charge on the α-carbon, we anticipated that electron-rich aryl groups would migrate preferentially. To test this hypothesis, a series of styryl azides, which systematically varied the identity of the para-substituent R, were exposed to reaction conditions (Figure 1). Examination of the product ratios using the Hammett equation revealed that the best linear correlation was obtained with σpara values to give a ρ value of −1.49. The greater propensity of the more electron-rich aryl group to participate in the 1,2-shift was interpreted to suggest that the migration occurs through phenonium ion reactive intermediate 30,[24,25] where the more stable ion leads to the major product. Figure 1 Correlation of product ratios with the Hammett equation. y= −1.49x+0.17; R2 =0.98. In conclusion, we have demonstrated that rhodium carboxylate complexes catalyze cascade reactions of β,β-disubstituted styryl azides to selectively produce 2,3-disubstituted indoles. Our data suggests that the selectivity of the migratorial process is controlled by the formation of a phenonium ion. Future experiments will be aimed at clarifying the mechanism of this reaction as well as determining if the benzylic cation can be intercepted with additional nucleophiles to produce complex, functionalized N-heterocycles from simple, readily available styryl azides.

Rh2(II)-Catalyzed Nitro-Group Migration Reactions: Selective Synthesis of 3-Nitroindoles from β-Nitro Styryl Azides

Rhodium carboxylate complexes (1 mol %) catalyze the migration of electron-withdrawing groups to selectively produce 3-substituted indoles from β-substituted styryl azides. The relative order of migratorial aptitude for this transformation is ester ≪ amide < H < sulfonyl < benzoyl ≪ nitro.

Iron(II) Bromide-Catalyzed Intramolecular C–H Bond Amination–[1,2]-shift Tandem Reactions of Aryl Azides

Iron(II) bromide catalyzes the transformation of ortho-substituted aryl azides into 2,3-disubstituted indoles through a tandem ethereal C-H bond amination [1,2]-shift reaction. The preference for the 1,2-shift component of the tandem reaction was established to be Me < 1° < 2° < Ph.

Examination of the Mechanism of Rh2(II)-Catalyzed Carbazole Formation Using Intramolecular Competition Experiments

The use of a rhodium(II) carboxylate catalyst enables the mild and stereoselective formation of carbazoles from biaryl azides. Intramolecular competition experiments of triaryl azides suggested the source of the selectivity. A primary intramolecular kinetic isotope effect was not observed, and correlation of the product ratios with Hammett sigma(+) values produced a plot with two intersecting lines with opposite rho values. These data suggest that electronic donation by the biaryl pi-system accelerates the formation of rhodium nitrenoid and that C-N bond formation occurs through a 4pi-electron-5-atom electrocyclization.