COMMUNICATION
Table 1. N-Alkylation of indole with hexanol.[a]
Entry Alcohol/ Shvo PTSA Solvent
Conv.[b] Yield[b]
indole
[%]
[%]
[%]
[%]
1
2
3
4
1.2
1.2
1.1
1.1
1.1
1.1
1.1
1.1
0.5
0.5
0.2
0.2
0.2
0.2
0.2
0.2
–
toluene
toluene
toluene
toluene
toluene
toluene
heptane
85
99
99
99
95
93
99
77
45
69
85
81
79
82
72
2
0.1
0.025
0.001
0.025
0.025
0.025
5
6[c]
7
8
t-amyl alcohol 90
[a] Reaction conditions: indole (1 mmol), solvent (0.5 mL), 24 h. [b] Con-
version and yield are determined by GC analysis with hexadecane as the
internal standard; conversion and yield are based on the indole. [c] 18 h.
Scheme 3. Possible mechanism for the N-alkylation of indoles with alco-
hols.
is increased further by adding p-toluenesulfonic acid
(PTSA; Table 1, entries 2–5). However, by using 2 mol% of
PTSA the selectivity of the reaction dropped considerably.
To our delight, lowering the amount of PTSA to
0.025 mol% (Table 1, entry 4) in the presence of only
0.2 mol% catalyst still resulted in quantitative conversion
and an excellent product yield of 85%. Further reduction of
the amount of PTSA (Table 1, entry 5) as well as a reduc-
tion of the reaction time resulted in a decrease of the con-
version. Variation of the solvent (Table 1, entries 7 and 8)
revealed that heptane can be used too, whereas tert-amyl al-
cohol reduced the reactivity of the system.
the optimized conditions (Table 1, entry 4). In fact, N-hex-
AHCTUNGTREGyNNUN lindole is formed in 72% yield. When the experiment was
repeated without catalyst, no product was formed, but the
enamine 3 was obtained as the major product. This shows
the necessity of the catalyst in the final isomerization step.
The groups of Bꢀckvall[22] and Casey[23] demonstrated that
transfer-hydrogenation processes in the presence of 2 pro-
ceed through a monohydride mechanism. This implies that
the deuterium removed from the electron-deficient
a
carbon of the butanol should be transferred to the electron
deficient C-2-position of indole. The isomerization step
should lead to an H:D ratio of 50/50 in the C-2-position. In
agreement with our mechanistic proposal, the resulting N-
butylindole contained 40% deuterium in the C-2-position
(Scheme 4).[24] The discrepancy of 10%, as well as 15% deu-
In addition to the desired transformation, small amounts
of C-3-alkylation and dialkylation of indole, as well as the
formation of hexyl hexanoate are observed as side reactions
(<5%).
Considering the previous work of Grigg and co-workers,[6]
who performed their reactions in the presence of KOH
(20 mol%), our selective N-alkylation is rather remarkable.
The difference in pH seems to be the main reason for the al-
tered regioselectivity. Apparently, the typical “borrowing hy-
drogen”[7] mechanism, in which hexanol is oxidized by the
catalyst to give hexanal, which is attacked by indole, cannot
be applied here. Indeed, direct reaction of hexanal with
indole gave a mixture of N-, C-, and disubstituted indoles,
but only a small amount of the N-substituted product.
Careful analysis of the side products revealed the possibil-
ity of another reaction mechanism. In addition to the N-al-
kylated product the formation of around 3% N-hexyl-2,3-di-
hydroindole is observed under the optimized conditions.
Evidently, this product requires hydrogenation of the indole
ring under the reaction conditions. As shown in Scheme 3,
we propose an initial transfer hydrogenation from the alco-
hol to the indole to give the corresponding aldehyde and in-
doline. Subsequent condensation and final isomerization
leads to the N-alkylated product. To the best of our knowl-
edge, transfer hydrogenations of indoles to indolines have
not been reported before. To support this novel mechanism
we performed the reaction of indoline and hexanal under
Scheme 4. Deuterium incorporation (%) during the alkylation reaction.
terium incorporation at the C-3-position can be explained
by H/D exchange reactions between water and the Shvo
monomer 2b (Scheme 2). In addition, hydrogen incorpora-
tion into the formerly completely deuterated butyl chain
was observed. The increase in reactivity by the addition of
PTSA can also be explained by the proposed mechanism.
Here, we presume that protonation at the C-3-position of
indole speeds up the indoline formation. Yet the question
remains open as to whether the 2,3-double bond is hydro-
genated directly or if subsequent formation of a 1,2-double
bond occurs.
Next, we investigated the scope and limitation of our cat-
alytic system. For this purpose, we studied the reaction of
indole and 5-methoxyindole with different alcohols
Chem. Eur. J. 2010, 16, 3590 – 3593
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3591