Angewandte
Chemie
sponding metal salts and a chiral spiro bisoxazoline ligand
Increasing the amount of NaBArF to 2 equiv relative to
[10]
(Sa,S,S)-4a with NaBArF
as an additive. As shown in
[Pd(PhCN)2Cl2] shortened the reaction time to 1.5 h and
markedly improved both the yield and the enantioselectivity
(entry 12). By contrast, the yield dropped dramatically to less
than 10% in the absence of NaBArF. The role of NaBArF
remains unclear; however, the bulky and noncoordinating
BArF anion of the resulting palladium catalyst may increase its
Lewis acidity and stability, which is helpful for getting higher
Table 1, chiral catalysts derived from copper, iron, nickel,
À
cobalt, gold, iridium, and ruthenium produced the desired O
H insertion product with low enantioselectivity (entries 1–7).
By contrast, when PdCl2 was used as the catalyst precursor,
promising enantioselectivity (85% ee) was observed,
although the yield was still low (entry 8). The nature of the
palladium precursor strongly affected both the yield and the
enantioselectivity of the reaction; [Pd(CH3CN)2Cl2] and
[Pd(PhCN)2Cl2], which have nitrile ligands, provided higher
yields and enantioselectivities (entries 9 and 10).
À
reactivity and enantioselectivity. Because an O H insertion
reaction with water was detected as a side reaction, molecular
sieves were introduced to absorb the water in the reaction
system and this change further increased the yield of desired
product to 83% and the enantioselectivity to 98% ee
(entry 13). The reaction could be performed at a catalyst
loading of 1 mol% without compromising enantioselectivity;
however, the reaction rate and yield decreased (entry 14).
Ligands (Sa,S,S)-4b, (Sa,S,S)-4c and (Sa)-4d gave the same
level of enantioselectivity (entries 15–17) as that of ligand
(Sa,S,S)-4a, thus indicating that the substituents at the oxazo-
line rings have a negligible influence on the chiral induction of
the catalyst. A bisoxazoline ligand with a binaphthyl back-
bone, (Sa,S,S)-5, also gave good yield and enantioselectivity
(entry 18). However, a bisoxazoline ligand with a pyridine
spacer, (S,S)-6, gave very low enantioselectivity (entry 19).
The optical purity of product 3aa remains unchanged even
after stirring under the standard reaction conditions for an
additional 20 h, a result that highlights the advantage of the
current neutral and mild reaction conditions in the synthesis
of a-aryl-a-aryloxyacetates.
Furthermore, [Pd(CH3CN)2Cl2] and [Pd(PhCN)2Cl2] were
stable under the reaction conditions, whereas palladium black
formed when PdCl2 was used. A Pd0 precursor, [Pd(dba)2],
gave a faster reaction but no chiral induction (entry 11).
À
Table 1: Palladium-catalyzed asymmetric O H insertion of methyl a-
diazo-a-phenylacetate into phenol.[a]
Under the optimal reaction conditions, various phenols
and a-aryl-a-diazoacetates were evaluated as substrates
(Table 2). Impressively, all the reactions exhibited good to
high yields (68–89%) and excellent enantioselectivity (96–
99% ee; entries 1–27). Although the highly sterically hin-
dered reactants phenol 2k and diazoester 1o reduced the
reaction rate (entries 11 and 27), the steric and electronic
properties of the phenol and diazoester substituents had
a negligible effect on the enantioselectivity, thus indicating
that the spiro palladium catalyst is highly tolerant of various
substrate structures. The broad substrate scope observed
suggests that this reaction could prove a powerful method for
the preparation of optically active a-aryl-a-aryloxyacetates.
Moreover, a gram-scale synthesis of the insertion product 3aa
was performed with excellent yield and enantioselectivity
(entry 28). This experiment further highlighted the potential
Entry [M]
Ligand
t [h]
Yield [%][b] ee [%][c]
1[d]
2
CuCl
FeCl2
(Sa,S,S)-4a
3
71
10
59
22
2
(Sa,S,S)-4a 24 (8)[e] 15
(Sa,S,S)-4a 24 (8)[e] 11
(Sa,S,S)-4a 24 (8)[e] 29
(Sa,S,S)-4a 24 (8)[e] 12
(Sa,S,S)-4a 24 (8)[e] 25
3
NiCl2
4
CoBr2
5
AuCl
22
3
6
[Ir(COD)Cl]2
7
[RuCl2(benzene)]2 (Sa,S,S)-4a 15
10
23
51
47
54
69
83
66
66
72
72
74
74
36
85
87
92
rac
96
98
98
97
98
95
90
19
8
PdCl2
(Sa,S,S)-4a 18
(Sa,S,S)-4a 18
(Sa,S,S)-4a 11
(Sa,S,S)-4a
(Sa,S,S)-4a
(Sa,S,S)-4a
(Sa,S,S)-4a
(Sa,S,S)-4b
(Sa,S,S)-4c
(Sa)-4d
9
[Pd(CH3CN)2Cl2]
[Pd(PhCN)2Cl2]
[Pd(dba)2]
10
11
12[f]
3.5
1.5
2
[Pd(PhCN)2Cl2]
13[f,g] [Pd(PhCN)2Cl2]
14[g,h] [Pd(PhCN)2Cl2]
15[f,g] [Pd(PhCN)2Cl2]
16[f,g] [Pd(PhCN)2Cl2]
17[f,g] [Pd(PhCN)2Cl2]
18[f,g] [Pd(PhCN)2Cl2]
19[f,g] [Pd(PhCN)2Cl2]
À
of this palladium-catalyzed asymmetric O H insertion reac-
tion.
5
1.5
1.5
5
2
2
In addition to a-aryl-a-diazoacetates, ethyl and benzyl a-
diazopropionates (1p and 1q) also reacted with excellent
enantioselectivity (96% ee and 94% ee, respectively), but the
yields were low (Scheme 2).
(Sa,S,S)-5
(S,S)-6
À
In addition to phenols, other O H donors, including n-
butanol and water, were also tested (Scheme 3). The reactions
[a] Reaction conditions: 1a/2a/[M]/ligand/
NaBArF =0.3:0.45:0.015:0.018:0.018 (mmol), in 4 mL solvent at 408C.
[b] Yield of isolated product. [c] Determined by supercritical fluid
chromatography (SFC) on a Chiralcel OD-H column. [d] The data is from
Ref. [6d]. [e] The diazoester 1a was not completely consumed after 24 h
at 408C so the reaction was heated at 608C for additional 8 h. [f] With
12 mol% NaBArF. [g] With 300 mg 5 ꢀ molecular sieves as additive.
[h] With 1 mol% catalyst and 2.4 mol% NaBArF. COD=1,5-cycloocta-
diene, dba=dibenzylideneacetone.
of n-butanol and water proceeded smoothly to afford the
corresponding O H insertion products with moderate yields
À
and enantioselectivity.
À
The O H insertion products 3 could easily be transformed
into various important chiral compounds (Scheme 4). For
instance, the hydrolysis of methyl 2-(naphthalen-2-yloxy)-2-
Angew. Chem. Int. Ed. 2014, 53, 2978 –2981
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim