.
Angewandte
Communications
cally deactivated aryl halides tend to react more slowly than
their activated counterparts, they are especially challenging
coupling partners.[20] To overcome similar challenges with
other sensitive 2-heterocyclic boronic acids, we introduced
the slow-release cross-coupling strategy.[17] Specifically, in the
presence of mild bases and water as a cosolvent, air-stable
MIDA boronates undergo in situ hydrolysis to liberate the
corresponding boronic acids at a rate that is slower than
catalyst turnover.[17] Analogous to utilizing a syringe pump,
such conditions strongly favor cross-coupling over boronic
acid decomposition.[17,21]
Presumably as a result of the extreme lability of the 2-
pyridyl–boron bond, even under these slow-release condi-
tions, the cross-coupling of 2-pyridyl MIDA boronate
remained challenging. As shown in Scheme 1c, modified
reaction conditions employing isopropyl alcohol instead of
water as a cosolvent and Cu(OAc)2 as a substoichiometric
additive were somewhat effective with activated, electron-
deficient aryl chlorides such as 3a. However, when we
attempted to cross-couple 1a with more challenging deacti-
vated aryl chlorides, such as 3b, very little of the desired cross-
coupling product 4b was observed.
provided no notable advantage (entries 5–7). In contrast,
addition of the trivalent ligand diethanolamine (DEA)
resulted in the intriguing formation of a royal-blue reaction
mixture and the formation of 4b in a substantially increased
yield of 70% (entry 8).
To enable further optimization of these reaction condi-
tions, we sought to understand the mechanistic underpinnings
of this DEA-promoted increase in efficiency. Deng and co-
workers have shown that the cross-coupling of 2-pyridyl
boronic esters promoted by copper(I) salts likely involves an
À
À
initial C B to C Cu transmetalation to produce an inter-
mediate 2-pyridyl copper species which, in turn, undergoes
transmetalation with palladium(II).[14a] Starting with this
general mechanistic framework, we considered two possible
pathways for DEA to promote the transformation of 1a into
a the putative 2-pyridyl copper intermediate 6 (Scheme 2). In
pathway 1, DEA reacts with the conformationally rigid 1a in
An extensive survey of palladium/ligand combinations,[22]
copper salts,[14,16,17,23] bases, solvents, temperatures, and reac-
tion times resulted in reaction conditions that were somewhat
more effective, but the yield of 4b remained modest (Table 1,
entry 1). Driven by our then working hypothesis that the role
of IPA in these reactions was to promote initial transligation
of 1a to the corresponding 2-pyridyl isopropyl boronic ester,
we investigated a range of different alcohols as additives.
However, less (entries 2 and 3) or more (entry 4) sterically
bulky alcohols were all inferior to IPA, and common diols also
Scheme 2. Two possible pathways for the DEA-promoted coupling of
1a.
a novel transligation reaction to form a conformationally
flexible and thereby more reactive DEA adduct 5,[1d] which in
turn transmetalates with Cu(OAc)2 to form 6. In pathway 2,
DEA alternatively reacts with Cu(OAc)2 to yield a Cu(DEA)n
complex[24,25] and KOAc. The released KOAc then reacts with
1a to generate a reactive 2-pyridyl boronate intermediate 7
(X = acetate[26] or other ion), which in turn transmetalates
with Cu(DEA)n to form 6.
Table 1: Cross-coupling of 2-pyridyl MIDA boronate 1a with deactivated
aryl chloride 3b.[a]
To determine whether pathway 1 was operative, we first
mixed DEA with 1a in the presence of K3PO4 in deuterated
DMF at 1008C and monitored the reaction by 1H NMR
spectroscopy. Seeming to support this mechanism, we
observed the slow transligation of 1a to 5 over the course of
four hours (see the Supporting Information) and succeeded in
isolating 5 as a crystalline solid (Scheme 3). However, when
we attempted to couple to 5 to 3b with or without syringe-
pump-mediated slow addition of 5 over the course of four
hours to mimic the rate of its in situ formation,[17] we observed
only very low yields of 4b (Scheme 3). Thus, pathway 1
Entry
ROH
Equiv
Yield [%][b]
1
2
3
4
IPA
3
3
3
3
49
36
39
43
MeOH
EtOH
tBuOH
5
6
7
1.5
1.5
1.5
51
35
31
8
1.5
70
DEA
[a] Reaction conditions: 1.0 equiv 3b (1.0 mmol), 1.5 equiv 1a, 5 mol%
XphosPdcycle, 50 mol% Cu(OAc)2, 5 equiv K3PO4, 0.125m in DMF.
[b] Determined by GC analysis. XPhosPdcycle=chloro(2-dicyclohexyl-
phosphino-2’,4’,6’-triisopropyl-1,1’-biphenyl)[2-(2-aminoethyl)phenyl]-
palladium(II) methyl tert-butyl ether adduct.
Scheme 3.
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 2667 –2672