.
Angewandte
Communications
We started our investigation with readily available a-
chloro N-benzoyl hydrazone 2a as the 1,2-diazo-1,3-diene
precursor, to probe the feasibility of the aza-Diels Alder
reaction between 1,3-dimethly indole 1a and the in situ
formed azoalkene. With Na2CO3 as the base, the uncatalyzed
inverse-electron-demand aza-Diels–Alder reaction pro-
ceeded smoothly, affording the [2,3]-fused indoline/tetrahy-
dropyridazine cycloadduct 3aa in 75% yield, with exclusive
regioselectivity and excellent diastereoselectivity (Table 1,
When CuI/Binap was employed as the catalyst, the reaction
was finished in less than 12 h affording the desired adduct 3aa
in 91% yield with high diastereoselectivity and 55% ee
(Table 1, entry 2). However, further screening of several
Binap-type axially bisphosphine ligands failed to improve the
enantioselectivity (see the Supporting Information for
details). A chiral CuI/Box (L2) complex was also tested, and
only racemic product was obtained with low conversion
(entry 3). The chiral CuI complex generated from the (S)-
Phox ligand promoted the annulation process smoothly,
leading to the desired adduct in good yield, albeit with very
low enantioselectivity (entry 4). To our delight, by switching
the backbone of the chiral P,N-ligand ligand from benzene to
ferrocene, both the conversion and the enantioselectivity of
this cycloaddition were improved significantly (entry 5).
Remarkably, (S,Sp)-tBu-Phosferrox[16] (L7) bearing bulky
tert-butyl group on the oxazoline ring delivered the best
results in terms of the yields and stereoselectivity (entry 8).
When N-unsubstituted 3-methyl indole was employed, much
lower conversion and moderate enantioselectivity were
observed, probably owing to reduced nucleophilicity at C3
and reduced electrophilicity at C2 of 3-methyl indole
(entry 9). A study of this cycloaddition with CuI/(S)-L7 in
various solvents identified MeCN as a suitable alternative to
CH2Cl2, whereas toluene and THF had a detrimental effect on
this transformation (entries 8 and 10–12). Reducing the
reaction temperature proved beneficial to the control of
enantioselectivity, and 97% ee with full conversion were
achieved at À208C (entry 14).
Table 1: Optimization of the reaction conditions.[a]
Entry Ligand Solvent T [8C] Time [h] Yield [%][c] ee [%][d]
1[b]
2
3
4
5
6
7
8
–
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
toluene
CH3CN
THF
0
0
0
0
0
0
0
0
0
24
12
12
12
12
12
8
8
24
8
75
91
30
83
92
71
91
94
35
29
79
13
95
95
–
55
0
L1
L2
L3
L4
L5
L6
L7
L7
L7
L7
L7
L7
L7
3
79
52
87
95
66
5
92
7
92
97
9[e]
10
11
12
13
14
0
0
0
8
8
8
18
Under the optimized reaction conditions, we tested
various substituted indoles to examine the generality of this
annulation, and the results are shown in Table 2. In general,
a broad range of differently substituted indoles, bearing
electron-neutral (Table 1, entry 1) or electron-deficient
(entries 2–4) on the indole framework, reacted with N-
benzoyl hydrazone 1a smoothly to form a series of [2,3]-
fused indoline molecules in high yields (85–95%) with high
diastereoselectivities (> 20:1 d.r.) and excellent enantioselec-
tivities (93–98% ee). Remarkably, this method was also
compatible with the sterically hindered 4-bromo- (1b), 4-
methyl- (1e), and 7-methyl-substituted (1h) indoles in terms
of enantioselectivity and reactivity (entries 2, 5, and 8).
Various C3-substituted indoles have also been employed as
the reaction partners. It was found that both simple alkyl
groups and functionalized alkyl groups containing ether,
amide, and ester moieties were tolerated at the C3-position of
the indoles, affording the desired cycloadducts in good yields
with high stereoselectivities (entries 10–16). Notably, indoles
bearing a sterically hindered C3-substituted group, such as
iso-propyl and phenyl, also worked well in this transformation
(entries 11 and 13). No annulation occurred when indoles
without C3-substitution were employed under the optimized
reaction conditions. Subsequent studies revealed that N-allyl-
and N-benzyl-substituted indoles 1q and 1r are also viable
substrates in this catalytic system, affording the correspond-
ing heterocycles in good yields and with excellent enantiose-
lectivity (entries 17 and 18). Further exploration revealed that
1,2,3-trimethyl indole 1s was also a suitable substrate,
providing the cycloadduct 3sa, which bears two contiguous
CH2Cl2
CH2Cl2
RT
À20
[a] All reactions were carried out with 1a (0.3 mmol) and 2a (0.5 mmol)
in solvent (4.0 mL). [b] Without catalyst. [c] Yield of isolated product.
[d] ee value was determined by HPLC analysis; >20:1 d.r. was estab-
lished by 1H NMR analysis of the crude product. [e] N-unsubstituted 3-
methylindole was used.
entry 1). Thus, an efficient catalyst for this challenging
transformation needs to 1) control the annulation process
between heterodienes and dienophiles in stereoselective
manner, and 2) exhibit higher reactivity than the uncatalyzed
background reaction.
We envisioned that the reactivity of the in situ formed
azoalkene could be enhanced by coordination with a chiral
transition-metal complex, and therefore realize this annula-
tion with stereoselective control. Hence, different transition-
metal sources and some readily available chiral ligands were
examined, and the representative results are shown in Table 1.
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!