Iridium complexes with chiral N,P ligands have established themselves as efficient catalysts for the asymmetric hydrogenation of olefins, with largely complementary scope to Rh and Ru diphosphane complexes. In contrast to Rh and Ru catalysts, they do not require a coordinating polar group next to the C=C bond. Initial experiments with cationic phosphanyloxazoline (phox) complexes ([Ir(1)(cod)]X ) (cod= cyclooctadiene) showed that these catalysts are highly active in the hydrogenation of unfunctionalized triand even tetrasubstituted olefins. In this respect, they resemble Crabtree)s catalyst, [(Cy3P)(pyridine)Ir(cod)]PF6 (Cy= cyclohexyl), which provided the stimulus for the development of these catalysts. In these studies, we also found that the choice of solvent and anion is crucial as only in weakly coordinating solvents like dichloromethane or toluene with a virtually non-coordinating anion such as BArF (tetrakis[bis3,5-(trifluoromethyl)phenyl]borate) could high turnover numbers (> 5000) be obtained. Although high enantioselectivities were obtained in the hydrogenation of certain trisubstituted aryl alkenes such as (E)-methylstilbene, the application range of Ir-phox catalysts proved to be limited. However, subsequent work has led to new classes of N,P ligands, which have broadened the scope of Ir-catalyzed hydrogenation considerably. 7] Among the many structures we investigated, oxazolinephosphinites such as 3 and certain imidazoline analogues proved to be particularly efficient, giving high enantiomeric excesses with a wide range of unfunctionalized as well as certain functionalized olefins. With the intention of mimicking the coordination sphere of the Crabtree catalysts more closely, we also examined a series of pyridineand quinoline-derived ligands 4 and 5. As the results were quite encouraging, we decided to extend our studies to bicyclic analogues of type 6 because we thought that the more rigid conformation imposed by the additional ring could result in even higher enantioselectivities. Here we report the syntheses of a series of pyridyl–phosphinites 6 and their evaluation as ligands for Ir-catalyzed asymmetric hydrogenation. As shown in the schemes below, ligands of this type are readily accessible from simple, commercially available starting materials via the corresponding pyridyl alcohols. By changing the substituents at the pyridine ring and the P atom, or altering the size of the carbocyclic ring, the steric and electronic properties of these ligands and the coordination geometry can be optimized for a specific substrate. Ligands with unsubstituted backbones (6, R=H) were synthesized from commercially available precursors 12–14 via pyridyl alcohols 7–9 (Scheme 1). Oxidation to the corre-

Iridium Catalysts with Bicyclic Pyridine–Phosphinite Ligands: Asymmetric Hydrogenation of Olefins and Furan Derivatives

Molecular design, chemical synthesis, and biological action of enediynes

Photochemical Perfluoroalkylation with Pyridine N-Oxides: Mechanistic Insights and Performance on a Kilogram Scale.

Synthesis of 2,2-Bipyridines: Versatile Building Blocks for Sexy Architectures and Functional Nanomaterials

ion from either the RCH2 or the R CH2 group. However, some selective cases have been reported. The products 294 and 296 can be converted to imidazoles and oxazoles, respectively. Reaction of the oximes 297 with dimethyl carbonate in the presence of potassium carbonate under heating in an autoclave afforded N-methyloxazolones 299 in moderate yields (Scheme 93). The primary O-methoxycarbonylation of the initial oximes 297 has been established as leading to the target products. The intermediates 298 are further subjected to Nmethylation and thus give rise to the oxyenamines A. Subsequently, these undergo [3,3]-rearrangement with ultimate ring closure. Reaction is facile only for α-CH2-ketone-derived oximes. Tautomerization leading to A proceeds exclusively through abstraction of the more mobile proton. Recently, a similar transformation was reported for the O-perfluorobenzoyl oxime 300 (Scheme 94). In this case the reaction proceeds under milder conditions than those for dimethyl carbonate. An older multistep approach to the mentioned transformation consists of oxime O-acylation, followed by Nmethylation with a Meerwein salt and subsequent proton abstraction (Scheme 95). This method is the predecessor for the direct α-oxygenation methodology discussed in the next section. Acid-catalyzed conversion of conjugated cyclohexenone oximes into aniline derivatives has been known since the end of the 19th century. This reaction is often referred as “Semmler−Wolff aromatization”. Typical conditions consist of treating the oximes with acetic anhydride in the presence of a strong acid such as hydrogen chloride. The reaction was used in the syntheses of substituted anilines, quinolines, and other heteroaromatics as well as in the total syntheses of miltirone, sanguinarine, and chelerythrine. It also proved successful for more complex targets such as pseudopteroxazole, penitrem D, and HKI 0231B. Ketene and 1-ethoxyvinyl acetate were found to be useful reagents for the transformation (Scheme 96, eq 1). Other mild reaction conditions involve treatment of 303 with acetyl chloride in toluene at 80 °C. One of the mechanisms proposed includes a N,O-bis(acetylation) of the starting oximes, leading to the oxy-enamines A and/or B. Although alternative ways involving dienimine 307 or azirine 308 are also possible, acetic acid elimination from A or B affords the anilides 304. Another way for generation of the initial enoxime is the in situ enolization of the monooximes 305, which allows synthesis of the acetylated m-aminophenols 306 (eq 2). A substituent shift can be observed if the initial cyclohexenone oxime possesses quaternary carbon atoms. In certain cases a Beckmann rearrangement is a side process in the rearrangement. However, judicious choice of reaction conditions may allow selective transformations. Due to the acidic medium there is a high probability for formation of cationic species that can be trapped onto aromatics. In contrast, sometimes basic media may be preferable for aromatization, for example, if the cyclohexenone oxime ring possesses electron-withdrawing substituents. Scheme 94 Scheme 95 Chemical Reviews Review dx.doi.org/10.1021/cr400196x | Chem. Rev. 2014, 114, 5426−5476 5456 4.6. Direct α-Oxygenation of Carbonyl Compounds Reaction of carbonyl compounds with secondary amines is a classical method for enamine synthesis. The enamines formed can be involved in [3,3]-rearrangements with subsequent hydrolysis, affording functionalized carbonyl compound. Thus, it seems somewhat amazing that only recently has the reaction of aldehydes and ketones 309 with N-alkyl-O-acylhydroxylamines been applied for reliable introduction of the α-hydroxyl moiety (Scheme 97, Table 1). The bulky reagent 310 (X = C, R = t-Bu, R = Ph) reacts selectively with aldehydes but is, however, unactive toward ketones. The less sterically demanding derivative 310 (X = C, R = Me, R = Ph) reacts with cyclic ketones at room temperature and with acyclic and aromatic ones under mild heating (50 °C). This synthesis methodology tolerates different functionalities, such as esters, acetals, or phenols. Unsymmetrical ketones give selective rise to CH2-group oxygenation in the presence of a CH3 group. Note that methyl ketones (acetone, acetophenone) fail to react. A similar reaction is applicable for synthesis of carbonates (X = C, R = Me, R = OR′) and carbamate derivatives (X = C, R = Me, R = NR′2) 311. Use of N-methyl-O-tosylhydroxylamine in the presence of methanesulfonic acid converts aldehydes to the corresponding 2-tosyloxy derivatives 311 (X = SO, R = Me, R = n-Tol). In such a manner the functionalization of methyl ketones 309 (R = H) is possible but with only moderate regioselectivity (functionalization of the secondary site/primary site ≈ 2.6− 4.2:1). More electron-accepting substituents at the sulfur, such as p-nitrophenyl, lead to Beckmann-like rearrangements. The applicability of asymmetric reagents such as 310 was also studied. Substituents on both the nitrogen and the oxygen atoms, reaction temperature, solvent, and counteranion were all found to have a dramatic effect on both the yield and the asymmetric induction. After thorough screening the best Scheme 96

V. Boekelheide

Papers

Rearrangements of N-Oxides. A Novel Synthesis of Pyridyl Carbinols and Aldehydes