M.-Y. Chang et al. / Tetrahedron Letters 55 (2014) 2876–2878
2877
Table 2
OR
OR
3
MeO
MeO
Synthesis of skeleton 4a
OR
OR
MeO
MeO
Me
R
N
H
NH
2
1
R
1
10% Pd/C, N H
2 4 (aq)
4
4
Me
toluene, reflux, 25 h
N
1
NO
2
H
OR
1
O
OH
MeO
MeO
MeO
4
1
R
R
1
1
Entry
1, R =, R1 =,
4, Yieldb (%)
NO
NO
NO
2
2
2
1
2
3
4
5
6
1a, Methyl, H
1b, Isopropyl, H
1c, n-Butyl, H
1d, Cyclopropyl, H
1e, n-Octyl, H
1f, Methyl, methyl
4a, 84
4b, 80
4c, 81
4d, 76
4e, 83
4f, 70
5a, R = H
5b, R = Me
2
1
1
Scheme 2. Retrosynthesis of oxygenated 2-methylindolines 4.
a
The reactions were run on a 1.0 mmol scale with 1, N2H4 (80% in H2O), Pd/C
ratio of toluene and N2H4(aq) (10 mL) is an optimal combination for
elevating the yields of 4a under boiling conditions for 25 h (entry
4). With the results in hand, one-pot facile preparation of substi-
(10%, 50 mg) in refluxing temperature.
b
The isolated 4 were >95% pure as determined by 1H NMR analysis.
tuted indolines 4 was examined. From our previous reports,11
a
facile three-step route was employed to create skeleton 1, starting
with 2 via (1) O-allylation of 2 with allyl bromide (Table 2, entries
1–5) or trans-crotyl bromide (entry 6), (2) Claisen rearrangement
of O-allyl-1-nitrobenzene and (3) O-alkylation of the resulting 2-
allyl-3-nitrophenol 5a or 5b with alkyl bromides (R = methyl, iso-
propyl, n-butyl, cyclopentyl, n-octyl).
When O-3-substituted 2-allyl-nitrobenzenes 1a–f (Table 2, en-
tries 1–6) were treated under optimal reaction conditions, we ob-
served that indolines 4a–f provided 70–84% yields.12 Changing the
R substituent, a dioxygenated indoline was isolated via the one-pot
reduction/hydroamination protocol. Following the procedure, syn-
thesis of bis-indoline was examined next. As shown in Scheme 3,
four bis-3-O-linked 2-allyl-nitrobenzenes 7a–d with different
lengths of carbon linear chain (propyl, butyl, hexyl and octyl) re-
acted with the one-pot condition to afford a skeleton of bis-indo-
lines 7a–d at 70–78% yields. Adjusting the reaction temperature
to rt, the skeleton of o-allylanilines 8a–d (58–65%) was isolated
in major products along with trace amounts of o-allylanilino-indo-
lines 9a–d (8–14%), as shown in Scheme 4. From the phenomenon,
we found that the reaction temperature could control product
distribution. To deserve to be mentioned, the color of the isolated
skeletons 8 and 9 (pale yellow) was easily converted into a brown-
ish color at room temperature within 1 h under air atmosphere.
This color change shows that quinine derivatives should be formed.
Attempts to construct other oxygenated indoline skeletons are
examined next. Skeleton 10 was subjected to the one-pot reduc-
tion/hydroamination, but 2-allylanilines 11 was isolated as the sole
compound with good yields (80–84%) and no cyclized skeleton 12
was observed (Scheme 5). Increasing the time (25 ? 100 h) in the
reaction of 10a with a one-pot protocol showed no 12a was
yielded. To explore the feasibility, scope, and limitations of this ap-
proach, reaction of 13 or 15 (prepared from ring-closing metathesis
of 4a or cross metathesis of 1a) provided bis-anilines 14 or 16 at
65% or 58% yields (Scheme 6). A skeleton of indoline was still not
found. The structure of 15 was determined by single-crystal
X-ray crystallography.13
According to the above results, we envision that the position of
the methoxy group on a benzene ring is an important substituent
controlling the formation of indoline via the hydroamination
ring-closing process. As shown in Table 2, the C4 electron-donating
methoxy group of skeleton 1 could promote the C1 amino group
(from nitro-reduction) to easily cyclize with the o-allyl group.
When the position of the C4 methoxy group was moved to a C3 po-
sition, the indoline ring was not formed (for 10d). The plausible
reason should be C4-substituent forces the o-allyl moiety closer
to the C1 amino group via the required aminopalladation proce-
dure. The resulting phenomenon with the accelerated cyclization
is similar to Thorpe-Ingold effect.14 On the other hand, the o-allyl
group was a key arm in the formation of an indoline skeleton, as
shown in Scheme 5. For the o-cinnamyl side chain (for 10a–c),
the C1 amino group was not cyclized with the olefinic motif of
the styryl group since it possessed the steric hindrance which im-
peded its coordination to palladium center. The similar results
have been found for the ineffective cyclization of skeleton 13 and
Table 1
Synthesis of compound 4aa
OMe
OMe
MeO
MeO
Pd/C, N H
2
4 (aq)
Me
reflux
N
H
NO
2
4a
1a
Entry
Solvent (mL), N2H4 (mL), time (h)
4a, Yieldb (%)
1
2
3
4
5
6
7
8
9
Toluene (10), 1, 10
Toluene (10), 5, 10
Toluene (10), 5, 25
Toluene (5), 5, 25
Toluene (5), 5, 50
Toluene (1), 10, 25
DME (5), 5, 25
1,4-Dioxane (5), 5, 25
THF (10), 5, 25
EtOH (5), 5, 25
42
70
74
84
78
70
66
60
44
56
65c
76d
MeO
O
O
OMe
10% Pd/C, N H
MeO
4 (aq)
O
O
OMe
(
)
( )
n
n
10
11
12
2
Toluene (5), 5, 25
Toluene (5), 5, 25
toluene, reflux, 25 h
N
H
N
H
O N
2
NO
2
n = 1, 2, 4, 6
Me Me
a
The reactions were run on a 1.0 mmol scale with 1a, N2H4 (80% in H2O), Pd/C
(10%, 50 mg) in refluxing temperature.
6a, n = 1; 6b, n = 2
6c, n = 4; 6d, n = 6
7a (72%); 7b (70%)
7c (75%); 7d (78%)
b
The isolated 4a were >95% pure as determined by 1H NMR analysis.
c
Pd/C (10%, 10 mg) was added.
d
Pd/C (10%, 100 mg) was added.
Scheme 3. Synthesis of skeleton 7.