Angewandte
Communications
Chemie
Taylor research group established that the boron “ate”
complex 5 (generated in a catalytic amount from 4) is more
reactive than the parent free alcohol. We deliberated that,
should the chelating hydroxy groups in the “ate” complex 5
become nucleophilic enough to rapidly trap intermediate 3,
propargyl carbonate 7 (Table 1). Guided by our reaction
design (Scheme 1b), we identified conditions that delivered
the desired product 9 in excellent yield and enantioselectivity
(Table 1, entry 1). This reaction occurs under extremely mild
conditions (with Et N as the base, at ꢀ208C) and requires
3
[
9]
then merging
of the boron-based alcohol-activation
only 5 mol% each of the copper complex and the borinic acid
catalyst B1. Some key factors affecting the performance of
this transformation are listed in Table 1. First, if the copper
catalyst was omitted, the reaction did not proceed, and both
starting materials were recovered (entry 2); if the borinic acid
catalyst was absent (entry 3), no detectable amount of 9 was
formed, although (ꢁ)-7 fully decomposed (to unidentifiable
compounds). These observations underscore the essential
roles of each catalyst. Besides B1, borinic acid B2 (entry 4) or
boronic acid B3 (entry 5) also promoted this reaction, but
furnished 9 in lower yields with lower enantioselectivity. Of
all the ligands screened, (S,S)-L1 gave the best results in terms
of reaction efficiency and selectivity (entry 6; for the results,
see Figure S1 in the Supporting Information). The identity of
copper salt is critical, as the use of CuI dramatically reduced
the product yield (entry 7). Solitary hydroxy groups are inert
under these reaction conditions, as iPrOH could be employed
as a non-interfering solvent (entry 8) and no product arising
from further propargylation of 9 was detected in any case.
Halving of the catalyst loadings had minimal impact on the
stereoselectivity, but slightly diminished the yield of 9
(entry 9). This transformation is homogeneous and readily
scalable (entry 10). Finally, this reaction reached completion
within 2 h when performed at 258C (entry 11), although 9 was
formed with slightly lower enantioselectivity.
approach and the copper-mediated catalytic cycle might
alleviate the low reactivity issue intrinsic to aliphatic hydroxy
groups and enable the efficient propargylation of alcohols 4 to
give compounds 6 (Scheme 1b). We were aware, however,
that the success of this reaction design hinged upon the rapid
reaction between two species of low concentrations (3 and 5),
if both copper and boron-based reagents were to be used in
catalytic amounts, and the compatibility of the two catalytic
cycles.
In this study, we established that the Cu/B dual catalytic
strategy depicted in Scheme 1b is feasible, and that only
a stoichiometric amount of the alcohol substrate is required
for a high-yielding process. This reaction, which employs
readily available reagents and catalysts, occurs under very
mild conditions, proceeds with high site- and stereoselectivity,
and is compatible with a broad range of substrates. Impor-
tantly, this method is also amenable to the desymmetrization
of various meso 1,2-diols to generate products that contain
three stereogenic centers and one terminal alkyne group in
a single step.
We commenced our study by investigating the model
reaction between ethylene glycol (8, 1.2 equiv) and a racemic
[
a]
Table 1: Validation of the Cu/B dual catalysis in a model system.
After validating the Cu/B dual catalytic reaction in the
model system, we proceeded to investigate the scope of this
transformation. Various propargyl carbonates participated in
this reaction (Scheme 2a). For example, the substituent at the
propargylic position could be an aryl group with different
electronic (products 10a,b) and steric properties (product
1
1
0e). This substituent could also be heterocyclic (products
0i–l) or aliphatic (product 10m) in nature. Moreover,
Entry Deviation from standard conditions
ee [%] Yield [%]
1
2
3
4
5
6
7
8
9
0
1
none
no [Cu(CH CN) ]BF
no borinic acid
B2 instead of B1
B3 instead of B1
ligands other than L1
CuI instead of [Cu(CH CN) ]BF
4
iPrOH instead of THF as the solvent
2.5 mol% “Cu”, 5 mol% L1, 2.5 mol% B1 94
94
–
94
0
functional groups including halogen atoms (products
10c,d,f), carbamates (product 10g), and nitriles (product
1
with respect to the nucleophile as well (products 11a–g and
12, Scheme 2b). For example, an alkyl halide was tolerated
(product 11c). Besides 1,2-diols, 1,3-diols were also compe-
tent reaction partners (products 11d,e). The use of (S)-1,2,4-
butanetriol as a substrate gave some interesting results
3
4
b]
4
[
–
65
80
<5
30
64
0h) were accommodated. The reaction displayed good scope
see Figure S1
80
87
5
85
78
98
86
3
4
[
c]
1
1
7 mmol scale
258C, 2 h
95
87
(product 11 f/f’). First, of the two primary hydroxy groups
present in this substrate, the C1-OH group was selectively
functionalized, presumably because 1,2-diol moieties chelate
with the borinic acid catalyst more efficiently than do 1,3-
[
a] Unless otherwise noted, reactions were performed on a 0.2 mmol
1
scale. Yields were determined by H NMR spectroscopy with 2-
methylnaphthalene as an internal standard; ee values were determined
by HPLC analysis. See the Supporting Information for the absolute
stereochemical assignment of 9. [b] When 8 was used as the solvent, 9
was formed in 22% yield with 50% ee. [c] Yield of the isolated product.
[
8,10]
diols.
Second, the use of a pair of antipodal ligands led to
the formation of two complementary diastereoisomers with
different selectivity, thus suggesting there is a match/mis-
[11]
match relationship between the catalysts and substrates.
We also applied this method to the direct propargylation
of biologically relevant polyols. We found that the expecto-
rant drug guaifenesin could be propargylated by our method
in high yield to give 11g. As a highlight of the power of this
method, it installed a terminal propargyl group on the polyol
2
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!