COMMUNICATION
rising of the highest occupied molecular orbital (HOMO) of
the e-center of the original dienal (C6), powered by an
amino-catalyst. Consequently, regiospecifity is successfully
governed by optimal orbital overlap of the HOMO of the
catalyst-bound trienamine with the lowest unoccupied mo-
lecular orbital (LUMO) of the dienophile (C3’).[10] More-
over, it is suggested that secondary orbital interactions and
catalyst-induced face-shielding should provide the desired
diastereo- and enantiocontrol of the reaction. Noteworthy,
the obtained a-quaternary azlactones can be easily depro-
tected in both acidic and basic media to afford the parent
amino acids.[11]
A preliminary screening indicated that the presence of an
aromatic ring-substituent on the azlactone was vital for the
course of the reaction. Results from previous work also indi-
cated that the rate of the trienamine formation was signifi-
cantly enhanced by performing the reaction in polar aprotic
media co-catalyzing with o-fluorobenzoic acid (OFBA).[6]
Summarizing these factors, the initial reaction was set up by
reacting 2,4-hexadienal (1a) with 2-naphthyl-substituted ole-
finic azlactone 2a in the presence of 20 mol% catalyst 3a[o,p]
and OFBA. Gratifyingly, the desired cycloadduct 4a was
formed in good stereoselectivity, albeit with moderate con-
version (Table 1, entry 1). To determine the enantiomeric
ing a weak acid, base, or no additive at all, the conversion
of reactants improved (Table 1, entries 3–5). Full conversion
could be achieved by using no additive and a twofold in-
crease of substrate concentration (Table 1, entry 6). Lower-
ing the catalyst loading to 10 mol% was also possible; how-
ever, to achieve complete consumption of 2a within a simi-
lar time frame, heating the reaction media at 408C was re-
quired (Table 1, entry 8). Under the optimal reaction condi-
tions, product 4a was obtained in 70% yield (isolated as a
diastereomeric mixture), 5:1 d.r., and 98% ee.
The scope of the reaction is summarized in Table 2. The
use of different homologation reagents had a significant
effect on the yield and the resulting diastereomeric ratio of
the isolated products (compare entry 1, A/B). The shorter
reaction time of the Ramirez olefination (method B)[12] com-
pared with that of the Wittig reaction (method A) is be-
lieved to significantly reduce the degree of partial decom-
postition of the intermediary aldehyde taking place during
prolonged stirring, apparently affecting the major diastereo-
isomer more. A variety of aromatic azlactones 2 was evalu-
ated as dienophiles in the developed Diels–Alder reaction.
Both electron-donating (-OMe, -Me) and electron-with-
drawing aryl substituents (-F, -Cl, -Br, -CO2Me, -NO2) were
tolerated, providing the desired cycloaddition products with
Table 2. Scope of the reaction.
Table 1. Initial screening results.
Entry[a]
Additive
Conv. [%][b]
d.r.[c]
ee [%][d]
Entry[a] R (2)
Yield [%][b]
A (4) B (5)
d.r.[c]
A
ee [%][d]
1
2
3
4
OFBA
ONBA
BzOH
none
NaOAc
none
65
<15
75
75
75
>95
85
>95 (70)
5:1
–
98
–
B
A
B
5:1
5:1
5:1
5:1
5:1
5:1
98
98
98
98
98
98
1
2
2-naphthyl (2a)
Ph (2b)
70
82
90
66
60
83
51
68
67
84
71
51
84
56
63
63
5:1 7:1 98 98[e]
8:1
98[e]
5
3
4
4-Me-Ph (2c)
4-OMe-Ph (2d)
2-Br-Ph (2e)
3-pyridyl (2 f)
2-thiophenyl (2g)
51
47
55
7:1 7:1 98 99[e]
6:1 7:1 98 99[e]
4:1 7:1 98 99[e]
6[e]
7[e,f]
8[e,f,g]
none
none
5[f]
6[g]
7
5:1
8:1
99[e]
99[e]
[a] Reactions performed with 2a (0.1 mmol), 1a (0.15 mmol),
3
(0.02 mmol), and the given additive (0.02 mmol, 20 mol%) in CHCl3
(0.5 mL) at room temperature for 20 h. In situ homologation carried out
with PPh3=CHCO2Et (0.15 mmol), room temperature, 18 h. [b] Deter-
8
3,4-OCH2O-Ph (2h) 53
4-CO2Me-Ph (2i)
4-F-Ph (2j)
4-Cl-Ph (2k)
3-Cl-Ph (2l)
4-NO2-Ph (2m)
2-NO2-Ph (2n)
1-heptyne (2o)
5:1 7:1 98 99
9[f]
10[f]
11
12[f]
13
14[f]
15[h]
8:1
7:1
6:1
8:1
6:1
6:1
5:1
96
98
99
98
96
99
97
1
mined by H NMR spectroscopy. Yield of isolated product as a diastereo-
meric mixture is given in parenthesis. [c] Determined by 1H NMR spec-
troscopy. [d] Determined by CSP-HPLC analysis. [e] Using 0.25 mL
CHCl3 as solvent. [f] 10 mol% of 3 was used. [g] Reaction performed at
408C for 18 h.
[a] Reaction conditions: i) 2 (0.1 mmol), 1 (0.15 mmol), 3 (0.01 mmol) in
CHCl3 (0.25 mL), 408C, 18 h. ii) Method A: PPh3=CHCO2Et
(0.15 mmol), room temperature, 18 h. Method B: PPh3 (0.6 mmol), CBr4
(0.3 mmol), À108C, 15 min. [b] Yield of isolated product (as a diastereo-
isomeric mixture). [c] Determined by 1H NMR spectroscopy. [d] Deter-
mined by CSP-HPLC analysis. [e] Determined after opening of the azlac-
tone (see Table 3). [f] Reaction at room temperature. [g] Using ent-3 as
the catalyst. [h] Addition of OFBA (0.01 mmol) as co-catalyst was neces-
sary for full conversion.
excess and improve the product stability, the aldehyde cyclo-
adduct was in situ homologated by using a stabilized phos-
phonium ylide. Increasing the acid strength of the additive
by using o-nitrobenzoic acid (ONBA) diminished the reac-
tivity completely (Table 1, entry 2). Contrary to the inhibito-
ry effect of stronger acids, it was discovered that by employ-
Chem. Eur. J. 2011, 17, 9032 – 9036
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9033