Whilst attempts to crystallize compounds 1 and 2 were
unsuccessful, their formulation as borenium cations was
supported by X-ray diffraction on a related compound
derived from tetrachlorocatecholato (Cl4-Cat) [Cl4-CatB-
(NEt3)][AlCl4] (3; Figure 1). Crystallization of 3 is facilitated
sufficiently electrophilic for electrophilic aromatic substitu-
tion of N,N-dimethylaniline, which releases the base, Et3N,
necessary for proton scavenging. Electrophilic borylation of
N,N-dimethylaniline occurs exclusively in the para position
due to synergic electronic (meta deactivating) and steric
effects (ortho deactivating). Borenium cation borylation
therefore provides complimentary selectivity to iridium-
catalyzed borylation, which for mono-substituted arenes
gives predominantly meta-borylated products.[4]
This complementary behavior is further emphasized by
the electrophilic borylation of 3-bromo-N,N-dimethylaniline
that proceeds under electronic control to provide only the
1,3,4- borylated isomer, in high yield. In contrast, iridium-
catalyzed borylation of 1,3-disubstituted arenes gives the
1,3,5-trisubstituted isomer.[15] Whilst the borylation of 3-
bromo-N,N-dimethylaniline was extremely slow with 1 (only
ca. 50% after seven days), borylation proceeded to comple-
tion within 32 h using 3. This reactivity disparity can be
attributed to the greater electrophilicity of 3 relative to 1,
engendered by the inferior electron-donor ability of the
tetrachlorocatecholato ligand.
Figure 1. ORTEP drawing of compound 3 (thermal ellipsoids at 50%
probability and hydrogen atoms omitted for clarity). Selected bond
ꢀ
ꢀ
ꢀ
lengths [ꢀ]: B1 N1 1.505(3), B1 O1 1.364(3), B1 O2 1.370(3).
The wider borylation substrate scope using 1 was sub-
sequently investigated. At 208C in CH2Cl2 N-methylindole
was readily borylated exclusively at the 3-position, consistent
with the electrophilic aromatic substitution of indoles.[39]
However, attempts to isolate the catechol boronate ester
were hampered by protodeboronation to N-methylindole and
B-hydroxycatecholborane. This susceptibility to protodeboro-
nation can be attributed to the low steric environment and the
electron deficiency of boron in catechol boronate esters,
which allows coordination of protic species (e.g., H2O) to
boron, the first step in protodeboronation in weakly acidic
media.[40] The bulkier and more electron-donating diol,
pinacol, forms analogous pinacol boronate esters that are
significantly more resistant to protodeboronation in the
presence of H2O. Therefore, in situ transesterification of
catechol for pinacol was performed, eliminating the undesir-
able protodeboronation and enabling product isolation in
high yield. The use of pinacolato-ligated borenium cations
would conceptually enable the single-step synthesis of pinacol
boronate esters. This approach was initially complicated by
the instability of PinBCl,[41] and then precluded by the failure
of [PinB(amine)]+, synthesized from PinBH (amine = N,N-
dimethylaniline or 2,6-lutidine), to borylate N,N-dimethylani-
line and N-methylpyrrole.[31] This is attributable to the
reduced electrophilicity of boron on replacing catechol for
the more electron-donating pinacol.
by intermolecular CCl···H2CN hydrogen bonding.[35] The
boron center in 3 is trigonal planar (angles sum to 3608),
with the closest B···ClAlCl3 contact long at 3.444 ꢀ, consistent
ꢀ
with an ionic species. The short B N bond (1.505(3) ꢀ) is
comparable to the borenium cations [(aryl)2B(DMAP)]+
ꢀ
(e.g., B N 1.500(6) ꢀ; DMAP = 4-dimethylaminopyri-
dine).[36] The B O bonds in 3 (1.364(3) ꢀ and 1.370(3) ꢀ)
ꢀ
are short, shorter than in CatBCl (both 1.381(2) ꢀ) and
comparable
with
[CatB(PtBu3)]+
(1.369(6) ꢀ
and
1.373(5) ꢀ), indicative of significant O!B p donation in
3.[30,37] Despite this p donation from the catecholato moiety
compounds 1–3 are still highly electrophilic, presumably a
direct consequence of their cationic character.
For arene borylation studies 1 and 3 were chosen as the
boron electrophiles due to the neutral adducts, (e.g., CatBCl-
(Et3N)) not undergoing complicating disproportionation
reactions in contrast to 2.[37] Initial confirmation that 1 is
viable for arene borylation came on addition of 1 equivalent
of the activated arene N,N-dimethylaniline in CD2Cl2 at 208C.
This rapidly and quantitatively (by NMR spectroscopy) led to
regioselective arene borylation in the para position of N,N-
dimethylaniline, simultaneously producing 1 equivalent of
[Et3NH][AlCl4]. The sequestering of the proton by-product
from electrophilic aromatic substitution by Et3N prevents
formation of Brønsted superacidic H[AlCl4] which would
otherwise lead to competitive protodeboronation of the
boronate ester,[18] as observed in the analogous electrophilic
arene silylation.[38] In the previously reported intermolecular
arene borylation using BCl3/AlCl3 mixtures, activated alumi-
nium metal was required to consume the H[AlCl4] by-product
and prevent protodeboronation.[18,19] Therefore high-yielding
arene borylation by electrophilic aromatic substitution
requires both a strongly electrophilic boron source and a
proton scavenger that does not deactivate the electrophile.
These requirements are fulfilled by 1 where, despite ligation
by two p-donating alkoxy groups, the boron center is
A broad range of electron-rich N-heterocycles[42] were
amenable to borylation using this methodology (Table 1),
which proceeded effectively quantitatively (by in situ 1H and
11B NMR spectroscopy). Borylation is highly regioselective
and subsequent transesterification to the synthetically desir-
able pinacol derivatives is facile allowing for isolation of
pinacol boronate esters in high yield. Longer reaction times
were required for borylation of TIPS-protected indole,
compared to the alkyl-protected analogue, due to the
weaker electron-donating ability of trialkylsilyl groups, rela-
+
tive to alkyl groups (as indicated by sp values of ꢀ0.09 and
ꢀ0.31 for Me3Si and Me, respectively).[43] Borylation of
Angew. Chem. Int. Ed. 2011, 50, 2102 –2106
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2103