Angewandte
Communications
Chemie
ꢀ
of the B H bonds in 4a to react, and thus raised an
uncertainty with our hypothesis. Nevertheless, we expected
that the second borylation would be facilitated in our system
owing to its intramolecular nature (Scheme 2a).
To test our hypothesis, we initially examined the reaction
of ditriflate 3a with 4a (2 equiv) in the presence of a palladium
catalyst at 658C for 15 h (Table 1). The yield of borylated
Table 1: Optimization of the reaction conditions.[a]
Entry
Substrate
Ligand
Additive
NMR yield of 1 [%]
1
2
3a
3b
DPEPhos
CyJohnPhos
none
KI
>99 (76)[b]
97 (75)[b]
[a] Reaction conditions: 3a or 3b (0.50 mmol), 4a (1.0 mmol), Pd(OAc)2
(0.05 mmol), ligand (0.10 mmol), Et3N (2.5 mmol), additive
(0.25 mmol) in THF (2 mL) at 658C for 15 h. [b] Isolated yields are
shown.
product 1a was estimated by 1H NMR spectroscopy after
treatment with methanol and an aqueous solution of NH4Cl.
A brief screening of the ligand revealed that a bisphosphine
with a diphenyl ether backbone (i.e., DPEPhos) displayed the
highest activity, giving 1a in > 99% NMR yield (entry 1). The
product 1a could be isolated in 76% yield by column
chromatography. Unfortunately, however, reactions under
these optimized conditions using DPEPhos failed to promote
the borylation of dibromide 3b. Reoptimization of the
reaction conditions showed that using KI together with
CyJohnPhos,[9] markedly improved the yield of 1b, affording
product 1b in 97% NMR yield (entry 2, see also the
Supporting Information (SI) for details of optimization and
some discussion regarding the effect of KI).[6h]
The reaction was successfully applied to the synthesis of
a diverse array of cyclic diarylborinic acids (Scheme 3). In
addition to diarylborinic acids bearing simple alkyl, aryl and
alkoxy groups (1c–1h), those containing cyano, chloro, ester,
amide, carbamate, and fluoro groups (1i–1o) were all
compatible with the present catalytic conditions. This high-
lights the synthetic advantage of our protocol over previously
reported methods using organolithium reagents.[3] In partic-
ular, the tolerance of an aryl chloride moiety, as shown for 1k,
is notable when considering that the report that the borylation
of aryl chlorides occurs with 4a using a Pd/XPhos catalyst.[6h]
Our protocol also allowed the synthesis of p-extended
analogue 1p as well as diarylborinic acids containing nitrogen
Scheme 3. Pd-catalyzed synthesis of cyclic diarylborinic acids by annu-
lative two-fold borylation.[a] [a] Reaction conditions: 3 (0.50 mmol), 4a
(1.0 mmol), Pd(OAc)2 (0.050 mmol), ligand (0.10 mmol), Et3N
(2.5 mmol) in THF (2 mL) at 658C for 15 h. Ligand: DPEPhos for
ditriflates 3c–3h; CyJohnPhos for dibromides 3i–3o and 3s; XPhos for
3p and 3r; RuPhos for 3q. 1H-NMR yields are shown. Numbers in
parentheses are isolated yields. [b] KI was added. [c] After the reaction,
2-aminoethanol (4.0 equiv) was added. [d] Using 1.0 g of 3s.
(1q) and sulfur (1r) tethers. Notably, seven-membered diaryl-
borinic acid 1s was successfully synthesized.[10] Although we
routinely isolated the product after hydrolysis in the form of
borinic acid 1, chromatographic separation led to a consid-
erable loss of 1 in some cases (numbers in parentheses in
Scheme 3 refer to isolated yields). However, this issue could
be easily addressed by formation of an aminoalcohol adduct,
such as 1j,[8] which could be isolated by simple filtration and
used directly for the subsequent two-fold cross-coupling (see
below).
Our catalytic protocol can be readily applied to the gram
scale production of cyclic diarylborinic acids (Scheme 4, 3i!
1i). The obtained functionalized boracycle can then serve as
a 1,5-dianion equivalent in annulative cross-coupling with
dihalides under palladium catalysis, and thus allows access to
a diverse range of p-extended heteroarenes (Scheme 4).[3e]
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!