ACS Catalysis
Research Article
borylation of allylic alcohols is also an attractive tool in allylic
borylation,9 the problem of limited substrate scope is still
unresolved. On the other hand, the use of heterogeneous
catalysis based on supported metal nanoparticles (NPs) can
provide a reliable solution thanks to their high stability,
reusability, and ease with which they can be separated from the
products. These features make it possible to synthesize
valuable chemicals on a practical scale in a green manner.
Although Morken and co-workers reported that Pd/C showed
activity for the conversion of an allyl chloride to an allyl
boronate,6a the development of a novel catalytic system that
shows high activity and reusability, as well as a wide substrate
scope is still largely unexplored. Meanwhile, in the last decade,
supported gold nanoparticles have emerged as efficient
heterogeneous catalysts for the synthesis of organoboron
compounds,10 which stimulated us to devise gold-based
catalytic system for C−X bond borylation. Herein, we describe
the catalytic borylation of sp3 C−O bonds over supported Au
catalysts. While the Au-catalyzed borylation of allyl, benzyl,
and propargyl esters required no aid of additives, a series of the
corresponding allyl, allenyl, and benzyl boronates were
obtained in high yields. A detailed investigation on the
reaction mechanism elucidated that the formation of C−B
bonds over Lewis acidic Au NPs proceeded in the SN1′ manner
involving the formation of a cationic intermediate (Scheme
1c), thereby leading to a high reaction efficiency and broad
substrate scope. Furthermore, supported Au catalysts showed
excellent reusability, scalability, and environmentally friendly
nature, which should be key characteristics for the practical
synthesis of organoboron compounds.
reported to be effective catalytic components under homoge-
neous conditions, their metal-NP variants were ineffective
(entry 2). A survey to identify the optimal solvent revealed that
1,4-dioxane was the solvent of choice for the present Au NP-
catalyzed borylation, while THF, toluene, acetonitrile, and 1,2-
dichloroethane were unsuitable (entries 3−5). The supported
Au catalyst showed high activity, and the reaction, even at 30
°C, gave 3a in 83% yield (entry 6). In addition to an acetoxy
t
(−OAc) group, carbonate (−OCO2Me and −OCO2 Bu) and
phosphate (−OPO(OEt)2) groups also served as suitable
leaving groups to give 3a in comparable yields (entry 7). In
contrast, 3a was not obtained at all in the reaction of cinnamyl
chloride (entry 8). The high catalytic activity of Au/TiO2 was
further well reflected in the reaction with a low catalyst loading
(substrate/Au = 5000), to provide 3a in satisfactory yield.
Notably, the turnover number of 3150 was the highest value
ever reported in the catalytic borylation of sp3 C−X bonds
(entry 9, data for comparison are shown in the Supporting
The scope of allylic substrates in the Au NP-catalyzed C−O
bond borylation was investigated (Table 2). The reactions of
primary allyl acetates provided the corresponding allyl
boronates in good to high yields, and boryl groups were
delivered predominantly at terminal positions (entries 2−5).
Although the reaction of the simplest allylic acetate required at
least 1 h to furnish 3b in moderate yield (entry 1), the
installation of substituents on the alkene moiety shortened the
time required for full conversion of substrate and increased
product yields (entries 2 and 3). The present heterogeneous
Au-catalyzed reaction can deliver a boryl group in terpene-
based scaffolds to provide 3e in excellent yields (entries 4 and
5). The E/Z configuration of the starting substrates (1e and
1f) did not affect the reaction efficiency, and the E/Z
configurations at alkene moieties were not transferred to the
corresponding allyl boronates. The borylation of secondary
and tertiary allyl acetates also proceeded efficiently to give the
corresponding allyl boronates in excellent yields (entries 6−
15). In the reaction of secondary allyl acetates bearing a
terminal alkene moiety (1g−1j), C−B bonds were formed
exclusively at the γ-position (entries 6−9). On the other hand,
the introduction of a methyl group at the γ-position of
secondary allyl acetate led to the formation of a mixture of
regioisomers (3k-α and 3k-γ) (entry 10). In contrast, a
substrate containing a phenyl substituent at the γ-position of
allyl acetate underwent C−B bond formation selectively at the
α-position to afford secondary allyl boronate 3l (entry 11). A
methyl substituent at the β-position was also tolerated in the
Au NP-catalyzed reaction to give the corresponding boronate
(entry 12). Such regiochemistry in the present borylation
enables us to envisage that C−O borylation over Au NPs
proceeds through reaction mechanism that is different from an
SN2′ mechanism, which is often seen in C−X bond borylations
under Cu catalysis.5 The terminal chloro group remained
intact during the catalytic borylation of 1p (entry 15).
2. RESULTS AND DISCUSSION
Initially, we examined the catalytic activity of a series of metal
NPs supported on TiO2 for the borylation of cinnamyl acetate
(1a) with bis(pinacolato)diboron (2) (Table 1). The reaction
in the presence of supported Au catalyst (2 mol %) at 60 °C
completed within only 15 min and delivered a boryl group
exclusively at the α-position, to give cinnamyl boronate (3a) in
94% yield (entry 1). In contrast, a supported Pt NP catalyst
showed no activity. Although Ni, Cu, and Pd have been
a
Table 1. Borylation of 1a Over Supported Catalysts
b
entry
variation from the standard condition
yield of 3a (%)
1
2
3
4
5
6
7
none
94 (93)
0
11
19
0
Ni, Cu, Pd, or Pt instead of Au
THF instead of 1,4-dioxane
toluene instead of 1,4-dioxane
1,2-DCE or CH3CN instead of 1,4-dioxane
at 30 °C for 1 h
83
t
−OCO2Me, −OCO2 Bu, or −OPO(OEt)2 as LG 72−83
Borylation of cinnamyl acetate (1a) with various diboron
compounds (2′−2‴) other than bis(pinacolato)diboron (2)
was investigated (Scheme 2). Although a higher reaction
temperature was required, the reaction of bis-
(neopentylglycolato)diboron (2′) and bis(hexyleneglycolato)-
diboron (2″) proceeded to furnish the corresponding cinnamyl
boronates (3a′ and 3a″) in 63 and 89% yields, respectively. In
contrast, tetrahydroxydiboron (2‴) was not suitable as a boron
source for the Au-catalyzed borylation.
instead of −OAc, at 30 °C for 1 h
8
9
−Cl as LG instead of −OAc
0.02 mol % of Au/TiO2 at 100 °C for 3 h
0
63
(TON = 3150)
a
Reaction condition: 1a (0.30 mmol), 2 (0.60 mmol), Au/TiO2 (2
b
mol %),1,4-dioxane (1.0 mL), 60 °C, 15 min under Ar. Determined
by H NMR with mesitylene as an internal standard. Isolated yield is
1
shown in parentheses.
759
ACS Catal. 2021, 11, 758−766