While tertiary boranes undergo efficient protodeboronation with carboxylic acids, tertiary boronic esters do not. Instead, we have discovered that CsF with 1.1 equiv of H2O (on tertiary diarylalkyl boronic esters) or TBAF·3H2O (on tertiary aryldialkyl boronic esters) effect highly efficient protodeboronation of tertiary boronic esters with essentially complete retention of configuration. Furthermore, substituting D2O for H2O provides ready access to deuterium-labeled enantioenriched tertiary alkanes. The methodology has been applied to a short synthesis of the sesquiterpene, (S)-turmerone.

The addition of an aryllithium reagent to a secondary boronic ester leads to an intermediate boron-ate complex that behaves as a chiral nucleophile, reacting with a broad range of electrophiles with inversion of stereochemistry. Depending on the electrophile, the C-B bond can be converted into C-I, C-Br, C-Cl, C-N, C-O, and C-C, all with very high levels of stereocontrol. This discovery now adds a new, readily available, configurationally stable, chiral organometallic-type reagent to the arsenal of methods for use in asymmetric organic synthesis.

Lithiation of carbamates followed by borylation provides a powerful method for the homologation of boron reagents. However, when applied to hindered systems (secondary carbamates with tBu-boronic esters) for the construction of two quaternary centers, this methodology fails. Instead, using mixed boranes (tBuBMe2), the synthesis of adjacent quaternary stereogenic centers with full stereocontrol was successful. The process can be repeated two or three times in one pot leading to carbon chains bearing multiple contiguous quaternary stereogenic centers. The boranes were converted into tertiary alcohols or C-tertiary amines using chloramine. The origin of the high selectivity for alkyl over Me group migration was determined computationally.

The lithiation/borylation-protodeboronation of a homoallyl carbamate was applied to the synthesis of (+)-sertraline and (+)-indatraline. Due to the presence of the alkene, significant modifications of the methodology were required (use of 12-crown-4, TMSCl, H(2)O), or a solvent switch to CHCl(3), to achieve high yields and high selectivities.

Tertiary alkylamines are common motifs in many natural products and pharmaceuticals but access to them in enantiomerically enriched form can sometimes present a major challenge. A standard route to these compounds involves the addition of nucleophiles, for example, organometallic reagents [1] or cyanide [2] to ketimines, but other indirect methods have also been reported.[3] Whilst many of these methods have been successful in delivering high levels of stereocontrol, the level of selectivity is highly substratedependent.We recently reported a conceptually new method for the synthesis of tertiary alcohols that routinely delivered more than 98 % ee over a broad range of substrates (Scheme 1).[4] In this process, lithiation of secondary carbamates 1 followed by treatment with boronic esters and subsequent addition of MgBr 2 /MeOH gave the tertiary boronic esters 2, which were finally oxidized to tertiary alcohols in high ee. We reasoned that isolation of the intermediate tertiary boronic esters and subsequent amination could provide a new route to C-tertiary alkylamines in high ee. Whilst amination of primary and even secondary boronic esters had been reported, [5][6][7] the much more challenging tertiary boronic esters had not. We therefore embarked on this study and now report that C-tertiary alkylamines can indeed be obtained in more than 98 % ee using this methodology.Of the reported amination transformations, Mattesons direct conversion of a potassium trifluoroborate salt into an amine [7] turned out to be the most efficient and reliable, [8] and after some modification of the reaction conditions [9] this protocol was ultimately successful. Thus, a broad range of tertiary boronic esters were first converted into the corresponding trifluoroborates 3 and then treated with SiCl 4 and an alkyl azide (Table 1). Using this protocol with benzyl azide 4 a, the tertiary trifluoroborate 3 a was converted into the tertiary benzylamine 5 aa in 94 % yield and 99 % ee (entry 1). The methodology could be extended to other substituted benzylic and alkyl azides (entries 2, 3). In terms of the scope of the tertiary trifluoroborate, hindered alkyl groups (iPr, cHex; entries 4-6) and even diarylalkyl trifluoroborates (entries 7, 8) could all be employed, leading to C-tertiary alkyl amines with very high ee. The latter two examples are noteworthy as the diarylalkyl boron intermediates are especially prone to homolysis and radical recombination but no erosion in ee was observed.Some limitations of the methodology however, were uncovered. For example, the homoallylic trifluoroborate salt 6 only gave the tertiary amine 7 in 14 % yield together with ketone 8 in 75 % yield (Scheme 2, see Supporting Information for a mechanistic discussion on the origin of 8).The para-methoxy analogue of the trifluoroborate 3 f was also ineffective. Since both the meta-methoxy substrate 3 f and the para-methoxybenzylic secondary trifluoroborate salt [10] worked well, it showed that this limitation was specific to the tertiary, electron-rich diaryl trifl...