Directed electrophilic C-H borylation using Lewis-Base stabilised boranes

Abstract

Aryl boranes are widely utilised in a range of applications, including as emissive materials, as active pharmaceuticals and, most commonly, as intermediates in synthetic transformations, with the most notable being the Suzuki-Miyaura cross-coupling reaction. Consequently, there has been considerable interest in the development of efficient methodologies to synthesise aryl boranes. C-H borylation is an atom-economical method to produce aryl boranes by functionalising the (hetero)arene directly, with the use of iridium and cobalt catalysts in these reactions particularly well established. An alternative that avoids expensive / toxic metal catalysts is the use of boron electrophiles, which react in C-H borylation processes via electrophilic aromatic substitution (SₑAr) mechanisms. The application of directing groups in C-H borylation processes enables the regioselective functionalisation of C-H positions that are otherwise unreactive. Directed electrophilic C-H borylation of (hetero)arenes has been demonstrated primarily with trihaloboranes, in contrast the use of Lewis-base stabilised, four-coordinate boranes as electrophiles in such reactions is underdeveloped. Directed electrophilic C-H borylation processes that utilise an E-H unit (E = C, N or O) as the directing group are also underexplored. The results in this thesis seek to address each of these points.

The synthesis of pharmaceutically relevant benzoxaboroles via the directed electrophilic C-H borylation of benzyl alcohols was explored using NHC-boranes (Chapter 2). Activation of an NHC-BH₃ complex (to install a good leaving group), followed by addition of the benzyl alcohol generates a three-coordinate borenium cation, with the boron substituents dependent on the stoichiometry of the benzyl alcohol. However, these in situ generated borenium cations were demonstrated to be insufficiently electrophilic to enable the ortho C-H borylation of benzyl alcohols to form benzoxaboroles. The use of Lewis bases other than NHCs in LB-BH3 complexes was also ineffective due to the decomplexation of the Lewis base from the borane.

The use of pyrazabole ([H₂B(μ-C₃N₂H₃)]2), a readily synthesised diboron compound, as a transient directing group was established and applied to the ‘borylation directed borylation’ of indoles (Chapter 3). Pyrazabole has a B···B separation of ca. 3 Å, so is well suited for the regioselective N/C7-diborylation of indoles/indolines and can be converted to a ditopic electrophile using bis-(trifluoromethane sulfonyl) amine (HNTf₂). Pyrazabole electrophiles were shown to transform N-H indole into an N/C7-diborylated indoline species, with mechanistic studies demonstrating that this proceeds via a hydroboration / protodeboronation / C-H borylation pathway. The pyrazabole unit can be removed at the end of the reaction, generating useful 7-Bpin N-H indolines in a one-pot, metal-free process.

Whilst an effective activator of pyrazabole, HNTf₂ is expensive and must be handled inside a glovebox. To improve the utility of pyrazabole as a traceless directing group, the use of iodine as an activator was explored (Chapter 4). Though the synthesis of 7-Bpin indolines was demonstrated, the yields were lower than the comparative reactions with HNTf₂. Analysis of by-products revealed that the degradation of the pyrazabole unit occurs as a competitive side reaction, decreasing the amount of C7-H borylation. This is believed to be the result of the more coordinating iodide group (relative to [NTf₂]-).

In the hope of achieving the C3/C4-diborylation of indoles without C2-C3 hydroboration, synthetic routes to pyrazabole electrophiles that lack B-H bonds were investigated (Chapter 5). The synthesis and activation of di-halo and tetra-halo pyrazaboles was explored using either HNTf₂ or TMSNTf₂, respectively. The reaction of tetra-chloro or -bromo pyrazabole with TMSNTf₂ was shown to generate a mono-electrophilic pyrazabole derivative, but access to the di-electrophile species in large amounts was challenging. The reactivity of the mono-electrophile was explored, and whilst the C3-H borylation of N-Me indole was possible, C3/C4-diborylation was not achieved via this approach (though some evidence suggested minor C2/C3-diborylation was achieved).