COMMUNICATION
menol 2a, 5 mol% of the chiral phosphoric acid 1,[13] and
1.2 equiv of the allyl ester 3[14] in toluene at room tempera-
ture for 1 h. As shown in Table 1, entry 1, the reaction pro-
ceeded smoothly to afford the 4H-chromene 4a in a good
(Table 1, entry 5). When a methoxy group was introduced in
the 7-position, a significant reduction in enantioselectivity
was observed. As expected, the enantiomeric excess in-
creased with decreasing reaction temperature, reaching 87%
ee at À208C (Table 1, entry 6). However, further reduction
of the reaction temperature to À358C led to a disappointing
stereochemical outcome (Table 1, entry 7).[17]
Table 1. Initial screening of the reaction conditions.[a]
The effect of the chromenol core substituent prompted us
to further screen oxygenated substituents at the 6-position
of chromenol 2 (Table 1, entries 8–13). Subsequent manipu-
lation of the oxygenated substituents had a strong impact on
the stereochemical outcome. Protection of the 6-hydroxy
moiety with either a methoxymethyl (MOM) group (2 f) or
a tert-butoxycarbonyl (Boc) group (2g) led to a considerable
decrease in the enantioselectivity (Table 1, entries 8 and 9).
Whereas substitution with benzyl ether (2h) at the 6-posi-
tion slightly compromised the enantioselectivity (Table 1,
entry 10), ethyl ether (2i) enhanced the selectivity to 90%
ee (Table 1, entry 11). Further improvement in the enantio-
selectivity was achieved with the introduction of sterically
hindered substituents, such as isopropyl ether (2j) and tert-
butyl ether (2k). Enantioselectivities as high as 93% ee
were obtained for these derivatives (Table 1, entries 12 and
13). Considering the ease of removal of the protective group
of the 6-hydroxy moiety, the tert-butyl ether group was em-
ployed for subsequent investigations.
Having identified the optimal protective group for chro-
menol 2 as tert-butyl ether and the optimized reaction condi-
tions, we further investigated the enantioselective reduction
of a range of chromenols 2.[18] Table 2 summarizes the ex-
periments carried out to probe the substrate scope of the
present chiral anion catalysis reaction. In all cases, the re-
duction products 4 were obtained in good to excellent chem-
ical yields, irrespective of the electronic and steric properties
of the aromatic substituents (Ar1 and Ar2). Investigation of
the substituent effect of the Ar1 moiety indicated that Ar1
substituents with either electron-withdrawing or -donating
groups (Table 2, entries 1 and 2) led to a slight reduction in
the enantioselectivity. In addition, the enantioselectivities
were slightly dependent on the Ar2 moiety (Table 2, en-
tries 3–6), however, the highest enantioselectivity was ob-
served in the reaction of 4p bearing a p-methoxyphenyl
group (Table 2, entry 5). Introduction of an ortho substitu-
ent in the Ar2 moiety, however, led to a considerable reduc-
tion in the enantioselectivity with quantitative formation of
the 4H-chromene derivative 4r (Table 2, entry 7). A slight
reduction in the enantioselectivity was also observed in the
reaction of chromenol 2s with electron-donating groups in
both the Ar1 and Ar2 moieties (Table 2, entry 8). However,
in this case, the stereochemical outcome could be improved
by lowering the reaction temperature to À358C (Table 2,
entry 9).[17]
Entry
2
4
Yield [%][b]
ee [%])[c]
1
2
3
4
2a: X=6-Me
2b: X=H
2c: X=6-Br
2d: X=6-MeO
2e: X=7-MeO
2d
4a
4b
4c
4d
4e
4d
4d
4 f
4g
4h
4i
91
90
87
89
99
98
97
83
>99
98
58
43
47
72
19
87
85
41
57
84
90
93
93
5
6[d]
7[e]
8[d]
9[d]
10[d]
11[d]
12[d]
13[d]
2d
2 f: X=6-MOMO
2g: X=6-BocO
2h: X=6-BnO
2i: X=6-EtO
2j: X=6-iPrO
2k: X=6-tBuO
>99
95
91
4j
4k
[a] Unless otherwise noted, all reactions were carried out with
0.005 mmol of (R)-1 (5 mol%), 0.1 mmol of 2, and 0.12 mmol of 3
(1.2 equiv) in toluene (0.5 mL) at room temperature for 1 h. [b] Isolated
yield of 4. [c] The enantiomeric excess of 4 was determined by chiral sta-
tionary phase HPLC analysis. The absolute configuration at the 4-posi-
tion of 4 was assigned to be (R) by analogy. See Table 2, entry 3 and the
Supporting Information for details. [d] At À208C for 24 h. [e] With
0.01 mmol of (R)-1 (10 mol%) at À358C for 24 h.
yield with moderate enantioselectivity. The exclusive forma-
tion of the 4H-chromene regio-isomer is of particular inter-
est. Neither the 2H-chromene nor ring-opened by-products
were observed. The regio-selectivity observed is in contrast
to that found for typical nucleophilic additions to 1-benzo-
pyrylium ions bearing a substituent at the 4-position.[2a,15]
The formation of 2H-chromene derivatives, in most
cases,[2b–d,16] although it markedly depends on the substituent
pattern, is well-documented.[15] Encouraged by this prelimi-
nary result, we investigated the effect of substituents on the
chromenol core (Table 1, entries 2–4). The substituents in-
troduced to the chromenol core had a marked effect on the
enantioselectivity. Whereas removal of the methyl substitu-
ent from the 6-position, giving 2b, or introduction of a bro-
mide, giving 2c, led to a slight decrease in the enantioselec-
tivity (Table 1, entries 2 and 3), the reaction of chromenol
2d with a methoxy substituent resulted in a noticeable in-
crease in the stereoselectivity (Table 1, entry 4). It is note-
worthy that the positional effect on enantioselectivity of the
substituent introduced to the chromenol core was significant
To shed light on the reaction mechanism, we confirmed
the formation of 1-benzopyrylium ion A by UV/Vis spectro-
scopic analysis (the UV/Vis absorbance of 1-benzopyrylium
species has been reported to appear at 400–600 nm).[20]
Indeed, a pale yellow solution of chromenol 2k in toluene
Chem. Eur. J. 2013, 19, 13658 – 13662
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
13659