Communications
Table 1: Suzuki–Miyaura coupling of 2-naphthol (1a) with different aryl
boroxines 2.[a]
2-Naphtholates 1 with different metal counterions were
treated with phenylboroxine (2a) in the presence of nickel
catalysts to prove our concept. Although the magnesium
naphtholate gave no desired product, we found that the
lithium, potassium, and zinc salts afforded a trace amount of
product 3aa, and the sodium salt showed the highest
reactivity (see Table S1 in the Supporting Information).
Notably, the mixed nonpolar solvent system of THF and o-
xylene is a key requirement to facilitate this transformation.
Although the observation of the desired product 3aa was
exciting to us, initially further investigations gave no progress.
After numerous trials in which the different parameters were
varied, no improvement was seen in the reaction efficiency.
Pleasingly, the addition of the mild Lewis acid BEt3
(triethylborane) gave a surprising result. Both the conversion
and the yield of the isolated product significantly increased in
the presence of the appropriate amount of BEt3, despite being
accompanied by a trace amount of reduced and/or ethylated
by-products. Trimethoxyborane also promoted the reaction,
albeit with a slightly lower efficiency (see Table S1 in the
Supporting Information). The combination of [Ni(cod)2]
Entry
Ar (2)
3
Yield
[%]
1
2
3
82
90
80
4
80
(cod = cycloocta-1,5-diene)
and
tricyclohexylphosphine
(PCy3) showed the best efficiency, while the prepared [NiCl2-
(PCy3)2] plus additional PCy3 gave a lower catalytic activity.
Other catalysts completely failed in this transformation (see
Table S3 in the Supporting Information). Different phenyl-
boronic reagents were tested and phenylboroxine (2a) turned
out to be the best (see Table S4 in the Supporting Informa-
tion).
Further investigations demonstrated the broad substrate
scope of this transformation. Various aryl boroxines could be
applied in the reaction with good efficiency (Table 1).
Electron-donating groups, such as OMe and NMe2, promoted
the reaction (Table 1, entries 4 and 5). In contrast, aryl
boroxines that contained electron-withdrawing groups mainly
underwent protodeborylation and gave relatively poor con-
version into the desired product (Table 1, entry 6). Steric bulk
on the aryl group did not hamper the reaction and 2-
tolylboroxine also gave the desired product in an excellent
5
81
43
65
6[b]
7
[a] Reaction conditions: 0.4 mmol of 1a, 1.0 equiv of aryl boroxine 2,
10 mol% of [Ni(cod)2], 40 mol% of PCy3, 1.0 equiv of NaH, 1.5 equiv of
BEt3 in the mixture of 0.7 mL of o-xylene and 0.2 mL of THF at 1108C for
48 h. [b] 0.2 mmol of 1a and 3.0 equiv of 5,5-dimethyl-2-(4-trifluorome-
thylphenyl)-1,3,2-dioxaborinane was used.
À
À
yield (Table 1, entry 3). Notably the C OMe and C F bonds
(Table 1, entries 5 and 7), which can be easily cleaved by
nickel catalysis, were tolerated well, and therefore could
provide the opportunity for orthogonal cross-coupling reac-
tions with different functionalities.[10,11]
that the nickel species played a vital role in the activation and
À
cleavage of the C O bond of the phenolates. Importantly the
significant effect of the BEt3 is due to its Lewis acidity. Several
different hypotheses were proposed, as illustrated in
Scheme 2. As a Lewis acid, BEt3 might directly react with
À
From the aspect of the phenols (Table 2), 1-naphthol gave
the desired product (3bb), although in a relatively lower
yield, under the slightly modified reaction conditions
(Table 2, entry 2). Phenanthren-9-ol (1c) also showed high
reactivity (Table 2, entry 3). Notably the selective cleavage of
the phenolate to form a borate and thus activate the C O
bond (Scheme 2A). Another possibility is a second Lewis
acid/Lewis base interaction between the in situ generated
borate and BEt3, which can be defined as a Lewis acid assisted
Lewis acid effect (LA/LA) to promote the reactivities of both
[12]
À
À
À
À
the C OH bond over the C OR bond was observed; for
example, the tert-butoxy group on the naphthyl ring (1e) was
tolerated (Table 2, entry 5). Notably, an ester group (1 d) is
also stable under the reaction conditions (Table 2, entry 4).
Most importantly, the selective cleavage of simple phenol
derivatives could also take place, albeit in a low yield (Table 2,
entry 6), thus indicating the potential applicability of this
method to the functionalization of simple phenol derivatives.
Mechanistically, the absence of the desired product when
the reaction was carried out without a nickel catalyst implied
C O and C B bonds (Scheme 2B).
Experimentally, both BEt3 and the phenylboronic reagent
could indeed react with sodium naphtholate to form borate, as
shown by 11B NMR spectroscopy (Scheme 3, spectra (1)–(4)).
However, the 11B NMR signal of BEt3 in the presence of the
phenyl boronate ester and sodium naphtholate moved only
slightly upfield, thus implying that the formation of the borate
from BEt3 and naphtholate (Scheme 2A) did not occur in the
presence of the phenyl boronate ester (Scheme 3, spectrum
(5)). The observed reactivity in the absence of BEt3 also
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7097 –7100