Development of a practical one-pot synthesis of indigo from indole

Article abstract of DOI:10.1246/bcsj.20100134

A novel and highly practical one-pot synthesis of indigo from indole via 3-position selective oxidation and dimerization of the indole framework was developed. Using 0.1 mol% of molybdenum complex and 2.2 equivalents of cumene hydroperoxide in tert-butyl alcohol, the reaction was complete in 7 h and pure indigo was obtained in 81% yield as a deep-blue solid just by filtration. The described one-step method renders the pure indigo readily available on a large scale using only inexpensive raw materials.

Full text of DOI:10.1246/bcsj.20100134

8
2
Bull. Chem. Soc. Jpn. Vol. 84, No. 1, 82–89 (2011)
© 2011 The Chemical Society of Japan
Development of a Practical One-Pot Synthesis of Indigo from Indole
1
1
1
2
Yoshihiro Yamamoto,* Yoshihisa Inoue, Usaji Takaki, and Hiroharu Suzuki
1Catalyst Science Laboratory, Mitsui Chemicals, Inc., 580-32 Nagaura, Sodegaura, Chiba 299-0265
2Department of Applied Chemistry, Graduate School of Science and Engineering, Faculty of Engineering,
Tokyo Institute of Technology, 2-12-1 O-okayama, Meguro-ku, Tokyo 152-8552
Received May 6, 2010; E-mail: Yoshihiro.Yamamoto@mitsui-chem.co.jp
A novel and highly practical one-pot synthesis of indigo from indole via 3-position selective oxidation and
dimerization of the indole framework was developed. Using 0.1 mol % of molybdenum complex and 2.2 equivalents of
cumene hydroperoxide in tert-butyl alcohol, the reaction was complete in 7 h and pure indigo was obtained in 81% yield
as a deep-blue solid just by filtration. The described one-step method renders the pure indigo readily available on a large
scale using only inexpensive raw materials.
Indigo is among the oldest dyes to be used for textile dyeing
and printing. Evidence of the use of indigo dyes dates back to
000 B.C. in Egypt. In many Asian countries, such as India,
still a big demand for improvement in terms of practicality.
If one could synthesize indigo (1) in a single step from
inexpensive raw materials, it could become a highly attractive
new industrial process. Toward this aim, we recently succeeded
in developing a one-pot synthesis of indigo (1) from indole (6)
through molybdenum complex catalyzed selective oxidation of
1
2
China, and Japan, indigo has been used as a dye for centuries. In
the United States, the primary use for indigo is as a dye for
cotton work clothes and blue jeans. Over one billion pairs of
jeans around the world are dyed blue with indigo. Indigo-
bearing plants can be found all over the world and a variety of
plants, including woad, have provided indigo throughout
history. The synthesis of indigo and its precursor was performed
using various strategies by von Baeyer, Heumann, and co-
workers so that by 1914 only 4% of the world production of
indigo had a plant origin. The manufacturing process developed
3
6 (Scheme 2). In the presence of 0.1 mol % of molybdenum
complex and 2.2 equivalents of peroxide, the reaction
completed in 7 h and pure indigo (1) was obtained in 81%
yield as a deep-blue solid just by filtration. In this article, we
report a full account of this work, including catalyst screening,
investigation of solvent effects and additive effects, optimiza-
tion of reaction conditions, and mechanistic investigation.
2
a
by Pfleger in 1901 is still in use throughout the world. In this
process, sodium phenylglycinate (4), which is synthesized by
the coupling of aniline (2) and chloroacetic acid (3), is treated
with an alkaline melt of sodium and potassium hydroxides
containing sodamide to produce indoxyl (5), which is then
oxidized in air to form indigo (1) (Scheme 1). In an alternative
O
H
N
cat.
1/2
N
H
N
H
process, indigo (1) is produced by the use of aniline (2),
O
2
ROOH
2 ROH
H O
formaldehyde, and HCN as starting materials.2b
indole (6)
1
+
2
In these processes however, multi-step organic reactions and
rather severe reaction conditions are required. Thus, there is
Scheme 2. One-pot synthesis of indigo (1) from indole (6).
O
ONa
O
NaOH
+
Cl
NH2
OH
N
H
2
3
4
NaOH
KOH
OH
O
H
N
NaNH2
[O]
1
/2
>200 °C
N
N
H
H
O
indoxyl (5)
indigo (1)
Scheme 1. Manufacturing process of indigo (1).
Published on the web January 1, 2011; doi:10.1246/bcsj.20100134

Y. Yamamoto et al.
Bull. Chem. Soc. Jpn. Vol. 84, No. 1 (2011)
Table 2. Solvent Effects
83
Table 1. Catalyst Screening
O
O
cat. (1 mol%)
CHP (x equiv)
H
Mo(CO)6 (1 mol%)
CHP (x equiv)
H
N
N
solvent
80 °C, 5 h
solvent
0 °C, 5 h
N
N
N
N
8
H
H
H
H
O
O
6
1
6
1
Entry
Catalyst
CHP (x)a) Yield/%b)
Entry
Solventa)
CHP(x)b)
Yield/%c)
1
2
3
4
5
6
7
8
9
0
®
Mo(CO)6
MoO2(acac)2
Mo­Nap
MoS2
10
5
5
5
5
5
5
5
5
5
5
4
46
23
35
14
19
17
10
7
1
2
3
4
5
6
7
8
9
benzene
benzene
toluene
toluene
propylbenzene
butylbenzene
isobutylbenzene
cumene
5
3
5
3
3
3
3
5
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
47
50
46
52
53
54
55
49
60
28
55
45
49
43
60
51
58
3
c)
MoO3
[H3(PW6Mo6O40)]¢nH2O
Ti(O-i-Pr)4
TiO(acac)2
RuCl2(PPh3)3
(C6H11OBO)3
cumene
1
8
13
10
11
¡-methylstyrene
tert-butylbenzene
p-cymene
mesitylene
anisole
heptane
hexadecane
isooctane
dimethoxyethane
dimethyl carbonate
butyl acetate
ethyl benzoate
£-butyrolactone
cyclohexanone
11
12
13
14
15
16
a) CHP 83 wt % in cumene solution. b) Isolated yield of 1.
c) Mo­Nap: Molybdenum naphthenate.
d)
d)
d)
Results and Discussion
17
18
19
20
21
22
23
Optimization of Reaction Conditions. Indole (6), which
used to be obtained from a fraction of coal tar, is now produced
on an industrial scale by the heterogeneous silver catalyzed
reaction of aniline (2) and ethylene glycol and is easily
available as a raw material. Several studies on the oxidation
of 6 have been reported, including photooxidation, anodic
11
34
41
9
4
14
5
6
7
a) Solvent/6 = 30 (weight ratio). b) CHP 83 wt % in cumene
solution. c) Isolated yield of 1. d) Reaction temperature was
0 °C.
oxidation, ozonization, and oxidations with manganese
dioxide, persulphates, hydrogen peroxide, and peracids.
8
9
10
11
9
The object of these reports, however, was to study the reactivity
of 6 and not to produce indigo (1). Even when 1 was detected,
it was obtained just as a very low-yield by-product, indicating
that an efficient direct process from 6 to 1 has not been
developed. Because organic hydroperoxide has not been
utilized for the oxidation of 6, we first examined various metal
complexes for the direct conversion of 6 to 1 in the presence of
cumene hydroperoxide (CHP). Group 1 (Na, K), 2 (Ba), 3 (Y,
La, Ce, Eu), 4 (Ti, Zr, Hf), 5 (V, Nb, Ta), 6 (Cr, Mo, W), 7
Using 1 mol % of Mo(CO) as a catalyst and CHP as an
6
oxidant, we then extensively examined solvent effects
(Table 2). In the preliminary studies using benzene (Entries 1
and 2), toluene (Entries 3 and 4), and cumene (Entries 8 and 9)
as a solvent, we found that reducing the amount of CHP from
5 equivalents to 3 equivalents (see also the next section)
significantly improved chemical yield of 1. Thus, further
studies were performed in the presence of 3 equivalents of
CHP. First of all, a variety of aromatic hydrocarbons was
examined (Entries 1­14). Apparently, more sterically congest-
ed solvents, such as isobutylbenzene (Entry 7, 55%), cumene
(Entry 9, 60%), and tert-butylbenzene (Entry 11, 55%), gave
better yield. Aliphatic hydrocarbons (Entries 15­17) also gave
comparative results to cumene when the reaction was per-
formed at higher temperature (90 °C). In contrast, polar
solvents (Entries 18­23) gave unsatisfactory results. The use
of solvents that have high coordination ability to a metal cation,
such as DMF, aniline, and pyridine, completely prevented the
reaction with the interesting exception of alcohol (Table 8).
Next, the influence of oxidant was investigated (Table 3). At
the beginning further details of the amount of CHP were
studied (Entries 1­7). As shown in Scheme 2, for the full
conversion of indole (6) to indigo (1), at least 2 equivalents of
CHP to 6 are required. One equivalent of CHP is consumed for
the oxidation at the 3-position of 6 to form 3-oxyindole (7)
(Mn), 8 (Fe, Ru, Os), 9 (Co), 10 (Ni, Pd), 11 (Cu, Ag), 12 (Zn),
1
3 (B, Al, Tl), 14 (Ge, Sn), 15 (As), and 16 (Se) metal
complexes were tested as catalysts. Among them, only
molybdenum (Entries 2­7), titanium (Entries 8 and 9),
ruthenium (Entry 10), and boron (Entry 11) complexes listed
in Table 1 afforded indigo (1) in yields that exceeded those
obtained in a control reaction (Entry 1). In particular, molyb-
denum complexes considerably enhanced the yield and
12
Mo(CO) gave the best (46%, Entry 2). As a general trend,
6
the oxidation state of molybdenum complexes did not influence
the catalyst activity to a large extent. It is likely that
molybdenum complexes of low oxidation state are oxidized
to that of high oxidation state in situ and the resulting
complexes act as active species. Because 1 is insoluble in
toluene, a deep-blue solid of 1 precipitated from the solution
with progress of the reaction. Then, chemically pure indigo (1)
was obtained just by filtration and washing with methanol.

8
4
Bull. Chem. Soc. Jpn. Vol. 84, No. 1 (2011)
One-Pot Synthesis of Indigo from Indole
Table 3. Screening of Oxidant
obtained yield of 1 remained at low level (20%). Taking into
account the fact that CHP is the raw material of phenol
production and is easily available on an industrial scale, CHP
was selected as an oxidant for further experiments.
Determining the best catalyst and oxidant allowed us to
perform precise optimization of reaction temperature, catalyst
amount, and reaction time. In the presence of 1 mol % of
Mo(CO)6 and 3 equivalents of CHP, the reaction was
performed within a temperature range of (60­110 °C) (Table 4,
Entries 1­6, Figure 2a). At temperature lower than 80 °C, both
reactivity and selectivity of 1 [(yield of 1)/(conversion of
O
Mo(CO) (1 mol%)
H
6
N
oxidant (x equiv)
solvent
N
N
H
80 °C, 5 h
H
O
6
1
Solvent Yield/%a)
Entry
Oxidant
x
b)
1
0.5 cumene
11
27
52
53
56
60
56
b)
2
1
2
cumene
cumene
b)
3
6
) © 100 (%)] were quite low, affording 1 in very low yield.
b)
4
2.2 cumene
2.5 cumene
b)
At 80 °C or higher, all of indole (6) was consumed in 5 h. Due
to the increase of side reactions however, selectivity of 1
decreased at temperature higher than 80 °C. Moreover, at
higher temperature, consumption of CHP considerably ex-
ceeded the ideal value of 2 equivalents (at 110 °C, full
conversion of 3 equivalents of CHP), suggesting undesired
decomposition of CHP. Therefore, 80 °C was the best temper-
ature for the reaction using 1 mol % of the catalyst. When
5
b)
6
3
5
cumene
cumene
b)
7
8c)
3
3
toluene
toluene
47
50
9d)
0
.1 mol % of the catalyst was used, however, higher temper-
ature (100 °C, Entry 13) was required to obtained comparable
results (59% yield). Under these conditions (0.1 mol % of
1
0e)
3
toluene
49
Mo(CO) at 100 °C), the reaction completed within 5 h (Entries
6
9
­14, Figure 2b). The fact that selectivity of 1 varied with
1f)
reaction time indicated that this reaction involves at least two
basic steps (Scheme 3).
Additive Effects. By the intensive optimization of reaction
conditions described above, we found two promising systems:
1
3
3
toluene
toluene
37
30
1
2g)
(
1) at 80 °C (Table 4, Entry 3) and (2) at 100 °C (Table 4,
1
3h)
4i)
3
3
cumene
octane
50
20
Entry 13). To further improve chemical yield of 1, we next
examined effects of additives. After screening of various types
of additives, such as aliphatic amines (inhibition), aromatic
amines, (inhibition or no effect), nitriles (no effect), sulfones
(no effect), sulfides (inhibition), phosphines (inhibition or no
effect), and dehydration agents (almost no effect), carboxylic
acids have been found to have positive effect on chemical yield
of 1. Results at 80 °C are summarized in Table 5. In the
presence of 10 mol % of carboxylic acid, yield of 1 was
improved from 60% (Table 4, Entry 3) to up to 69% (Entry 7).
Based on HPLC analysis, these additives accelerate the reaction
(faster conversion of 6, for example, without additive: 29%,
with benzoic acid: 45% after 1 h) and reduce the formation of
by-product 12 (ca. 6% to <2%). Moreover, since additive
effects of carboxylic acids appeared even when the reaction
was performed in the absence of catalyst, the carboxylic acid
may activate either hydroperoxide or indole (6). Using highly
1
H2O2
a) Isolated yield of 1. b) 83 wt % in cumene. c) 50 wt % in
aromatic hydrocarbon. d) 50 wt % in hydrocarbon. e) >90 wt %
purity. f) 67 wt % in toluene. g) 67 wt % in water. h) 65 wt % in
ethylbenzene. i) 30 wt % in water.
(
Figure 1) and the other one for the oxidative coupling of
two indole moieties to form a C­C double bond between each
-position. Thus, not surprisingly, the use of less than 2
equivalents of CHP gave significant decrease of yield (Entries
and 2). In general, higher amounts of CHP led to higher
reactivity. In the presence of a large excess of CHP however,
selectivity decreased, resulting in the formation of many by-
products such as 2-oxyindole (8). As a result, the yield of 1 had
a maximum value at 3 equivalents of CHP. Several other by-
products shown in Figure 1 were also detected in the reaction
mixture. For example, when 3 equivalents of CHP were used,
2
1
acidic carboxylic acids (pK < 3), however, decomposition of
a
CHP and formation of by-products increased, resulting in a
considerable decrease of yield. While adjustment of acidity by
reducing the amount of acid greatly improved yield (Entries 15
and 17), obtained yields did not exceed the result without an
additive (60% yield). Effects of carboxylic acids appear to be
related not to steric factors but to pK with a value of around 4
being optimum.
The amount of benzoic acid was then optimized for both 80
and 100 °C systems. As shown in Table 6 and Figure 3, yield
of 1 increased with increasing amount of benzoic acid to
30 mol % at 80 °C and to 10 mol % at 100 °C. In the presence of
7
% of 8, 5% of 12, 4% of 9 and 10, and <1% of 11, 13, and 14
were detected by HPLC analysis. Then, reactivity of several
other tertiary (Entries 8­12) and secondary (Entry 13) organic
hydroperoxides was investigated. Although these organic
hydroperoxides except tert-butyl hydroperoxide showed good
reactivities, they did not improve on the yield obtained using
CHP. As in Entry 12, the existence of water interfered with
catalyst activity, lowering the yield of 1. We also examined the
use of hydrogen peroxide as an oxidant. Although screening of
various solvents revealed that octane was the best solvent, the
a

Y. Yamamoto et al.
Bull. Chem. Soc. Jpn. Vol. 84, No. 1 (2011)
85
H
N
O
O
O
O
NH
N
N
H
H
N
H
3
2
-oxyindole (7)
indirubin (10)
1
3
N
H
N
H
O
O
N
H
-oxyindole (8)
HN
O
NH
O
N
H
O
isoindigo (11)
O
N
H
N
H
O
1
4
isatin (9)
O
N
NH
1
2
Figure 1. Structure of detected products in the reaction mixture.
Table 4. Effects of Reaction Temperature, Catalyst Loading,
and Reaction Time
1
00
8
0
O
Mo(CO) (y mol%)
CHP (3 equiv)
H
6
Conv. of 6
N
60
40
Selectivity of 1
Yield of 1
cumene
Temp., Time
N
N
H
H
O
2
0
6
1
0
Cat. Temp Time Conv. of 6a) Selectivity Yield of 1
c)
Entry
50 60 70 80 90 100 110 120
b)
(
y) /°C
/h
/%
of 1 /%
/%
Reaction temperature /°C
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
0.1
0.1
0.1 100
0.1 100
0.1 100
0.1 100
0.1 100
0.1 100
0.1 110
60
70
80
5
5
5
5
5
5
5
5
0.3
1
3
4
5
17
52
99
>99
>99
>99
44
82
10
50
97
6
44
60
55
50
46
44
60
0
39
57
58
59
59
56
1
23
60
55
50
46
19
50
0
19
55
58
59
59
56
(
a)
90
100
100
110
80
8
6
4
2
0
0
0
0
0
Conv. of 6
Selectivity of 1
Yield of 1
90
1
0
1
1
12
13
14
15
>99
>99
>99
>99
0
1
2
3
4
5
6
6
5
Reaction time /h
(
b)
a) Conversion yield of 6 was determined by HPLC analysis. b)
Selectivity of 1) = (yield of 1)/(conversion of 6) © 100 (%).
c) Isolated yield of 1.
Figure 2. (a) Effects of reaction temperature. (b) Effects of
reaction time.
(
excess benzoic acid, undesired decomposition of CHP was
accelerated.
We also found that addition of silanol had some positive
effects on yield of desired product 1 (Table 7, Entries 1­7). In
contrast to the additive effects of carboxylic acid, even in the
presence of excess amount of silanol, negative effects, such
as acceleration of CHP decomposition, was not observed
(Entry 6). In the presence of silanol, the formation of by-

8
6
Bull. Chem. Soc. Jpn. Vol. 84, No. 1 (2011)
One-Pot Synthesis of Indigo from Indole
O
Table 5. Additive Effects of Various Carboxylic Acids
ROH
Mo
O
R
Mo(CO) (1 mol%)
CHP (3 equiv)
6
O
H
16
N
carboxylic acid (10 mol%)
ROOH
cumene
80 °C, 5 h
N
H
N
N
H
H
O
O R
indole (6)
6
1
Mo
Entry
Acid
pKa
Yield/%a)
15
1
2
3
4
5
6
7
8
9
3,4-(O2N)2-C6H3-CO2H
2,4-Cl2-C6H3-CO2H
4-O2N-C6H4-CO2H
4-NC-C6H4-CO2H
2-Cl-C6H4-CO2H
3-Cl-C6H4-CO2H
4-Cl-C6H4-CO2H
4-Br-C6H4-CO2H
Ph-CO2H
3-HO-C6H4-CO2H
4-HO-C6H4-CO2H
3-MeO-C6H4-CO2H
4-MeO-C6H4-CO2H
Cl3C-CO2H
Cl3C-CO2H
Cl-CH2-CO2H
Cl-CH2-CO2H
Ph-CH2-CO2H
Me-CO2H
Et-CO2H
20
49
20
0
H O
R
O
Mo
3.44
3.55
2.94
3.83
3.99
N
H
17
H
62
68
69
68
67
65
63
65
63
3
58
15
59
66
62
63
63
62
29
59
50
57
OH
indoxyl (5)
O
N
H
N
H
4.2
cat.
1
0
4.08
4.58
4.09
4.47
0.65
0.65
2.86
2.86
4.31
4.76
4.87
1/2
O
N
11
12
13
H
O
ROOH
ROH
indigo (1)
+
H O
2
N
14
15
16
17
18
19
20
21
22
23
24
25
26
H
b)
3
-oxyindole (7)
Scheme 3. Proposed mechanism.
c)
product 12 was successfully prevented. Positive effects of
silanols may not be attributed to their pK value. For example,
addition of triphenylmethanol, whose pKa value (12.9) is
close to that of trimethylsilanol (12.7) and triphenylsilanol
11.1), had almost no effect on the reaction. Other silyl
compounds, such as silyl ether, silane, disilane, and silyl-
chloride, gave unsatisfactory results except diacetoxydiphenyl-
silane (Entry 8). In this case, diacetoxydiphenylsilane seemed
to partially decompose to acetic acid and silanol. Thus, we next
examined the combination of carboxylic acid and silanol. As
we expected, this combination exhibited a synergetic effect
Entries 9 and 10), achieving a yield of 74%. This suggests that
the additive effects of the carboxylic acid and of the silanol are
independent of each other.
Use of Alcohol as a Solvent. On the one hand, we found
that polar solvent prevented the formation of indigo (1) as
described above, but on the other, at least 2 equivalents of
alcohol and 1 equivalent of water form during the progress of
the reaction (Scheme 2). Considering the results of additive
effects of silanol, we studied the use of alcohol as a solvent.
Results using various alcohols are summarized in Table 8.
Although primary alcohols gave only unsatisfactory results
Entries 1­5), secondary alcohols (Entries 6­10), and tertiary
alcohols (Entries 11­13) gave better yields than cumene
Table 4, Entries 3 and 13). In the case of secondary alcohols
however, some amount of alcohol reacted with CHP to afford
the corresponding ketone, resulting in conversion of CHP
Me-(CH2)6-CO2H
Me3C-CO2H
C6H4-1,2-(CO2H)2
C6H4-1,3-(CO2H)2
C6H4-1,4-(CO2H)2
HO2C-(CH2)2-CO2H
a
2.98
3.46
3.51
(
a) Isolated yield of 1. b) 0.02 mol % of carboxylic acid was
used. c) 0.1 mol % of carboxylic acid was used.
Table 6. Optimization of the Amount of Carboxylic Acid
Mo(CO) (1 or 0.1 mol%)
(
6
O
CHP (3 equiv)
PhCO H (z mol%)
H
N
2
cumene
N
N
H
80 or 100 °C, 5 h
H
O
6
1
Ph-CO H
2
Entry
Systema)
Yield/%b)
(z)
1
2
3
4
5
6
7
8
9
80 °C system
80 °C system
80 °C system
80 °C system
80 °C system
80 °C system
100 °C system
100 °C system
100 °C system
100 °C system
100 °C system
5
10
20
30
40
50
1
64
67
70
71
70
69
61
64
67
64
57
(
(
5
10
30
50
(
2.4­3.0 equivalents) much higher than the ideal value of 2
1
0
equivalents. When tert-amyl alcohol was used, the isolated
yield of 1 exceeded 70% without an additive (Entry 12). Under
these reaction conditions, the formation of by-products 10­14
was effectively diminished.
11
a) 80 °C system: Mo(CO) (1 mol %) at 80 °C. 100 °C system:
Mo(CO)6 (0.1 mol %) at 100 °C. b) Isolated yield of 1.
6

Y. Yamamoto et al.
Bull. Chem. Soc. Jpn. Vol. 84, No. 1 (2011)
Table 8. Effects of Alcohol Solvents
87
8
7
6
5
0
0
0
0
O
Mo(CO) (y mol%)
CHP (3 equiv)
H
6
N
Solvent
Temp, 5 h
8
1
0°C system
00°C system
N
N
H
H
O
6
1
Entry Solvent
Cat. (y) Temp/°C Yield/%a)
1
2
3
4
5
6
7
8
9
ethyl alcohol
n-butyl alcohol
octyl alcohol
1
78 (reflux)
100
44
55
53
59
27
62
70
67
63
66
56
72
62
0
0.1
0.1
0.1
0.1
1
0.1
0.1
0.1
0.1
1
0
10
20
30
40
50
100
100
100
80
The amount of benzoic acid/mol%
benzyl alcohol
2-phenylethanol
isopropyl alcohol
2-butyl alcohol
2-pentyl alcohol
2-octyl alcohol
cyclohexyl alcohol
tert-butyl alcohol
tert-amyl alcohol
¡-cumyl alcohol
di(ethylene glycol)
Figure 3. The effects of the amount of benzoic acid.
98 (reflux)
100
Table 7. Additive Effects of Various Silyl Compounds
Mo(CO) (1 mol%)
CHP (3 equiv)
additive (z mol%)
100
100
80
100
100
80
6
O
H
10
N
1
1
cumene
80 °C, 5 h
12
13
14
0.1
0.1
1
N
N
H
H
O
6
1
Yield/%a)
a) Isolated yield of 1.
Entry
Additive
z
1
2
3
4
5
6
7
8
Me3SiOH
Et3SiOH
Ph3SiOH
Ph3SiOH
Ph3SiOH
Ph3SiOH
Ph3Si(OH)2
Ph3Si(OAc)2
Ph3SiOH + PhCO2H
Ph3SiOH + PhCO2H
10
10
10
30
50
100
10
10
10
10
63
66
64
66
66
65
63
67
74
70
Table 9. Additive Effects of Carboxylic Acids in tert-Amyl
Alcohol
Mo(CO) (0.1 mol%)
CHP (x equiv)
carboxylic acid (z mol%)
6
O
H
N
tert-amyl alcohol
100 °C, 5 h
N
N
H
H
O
b)
6
1
9
c)
1
0
CHP (x) Concentrationa) Yield/%b)
Entry Carboxylic acid
z
a) Isolated yield of 1. b) Mo(CO)6 (1 mol %), PhCO2H
1
2
3
4
5
6
7
8
PhCO2H
PhCO2H
10
10
®
10
5
10
20
30
3.0
3.0
3.0
3.0
2.2
2.2
2.2
2.2
30
30
30
30
15
15
15
15
78
78
72
78
76
77
76
75
(
(
30 mol %) at 80 °C. c) Mo(CO)6 (0.1 mol %), PhCO2H
10 mol %) at 100 °C.
c)
c)
MeCO2H
MeCO2H
MeCO2H
MeCO2H
We then investigated additive effects of carboxylic acid and
silanol using tert-amyl alcohol as a solvent (Table 9). Addition
of 10 mol % of benzoic acid improved chemical yield of 1 from
7
2% (Table 8, Entry 12) to 78% (Entry 1) while addition of
MeCO H
2
triphenylsilanol had no effect (Entries 2 and 3). It seems that
the functions of the alcohol and of the silanol are essentially
identical and that they stabilize molybdenum catalyst by
forming alkoxy or silanoxy complexes. When tert-amyl alcohol
was used, acetic acid gave comparable result to benzoic acid
a) Concentration: solvent/6 (weight ratio). b) Isolated yield of
. c) Ph3SiOH (10 mol %) was added.
1
thenate, which is readily available on an industrial scale, was
identical to that of Mo(CO)6 (Entry 3). The reaction also
proceeded with lower catalyst loading (Entry 5) and with less
solvent (Entries 6 and 7) in slightly lower yield. Thus, the best
result was obtained under conditions shown in Entries 2 and 3
(81%) (Table 10).
Mechanistic Investigations. From a mechanistic point of
view, this reaction would involve at least two basic steps
namely the oxidation of the 3-position of indole (6) and
dimerization of the indole moiety. To elucidate the reaction
mechanism, we tried to detect reaction intermediates by HPLC-
Mass spectrum measurement under inert gas. When the reaction
solution at early reaction stage within approximately 15 min
was analyzed, one HPLC peak with considerably high intensity
(Entry 4). In this system, the amount of CHP can be reduced to
2
1
.2 equivalents and the optimal amount of acetic acid was
0 mol % (Entry 6).
In terms of practicality, final optimization of the reaction
was performed. For this purpose, we used tert-butyl alcohol as
a solvent due to the limited availability of tert-amyl alcohol.
Although at 80 °C the use of tert-butyl alcohol gave unsat-
isfactory results (Table 8, Entry 11), under reflux conditions
(85.7­86.5 °C) chemical yield greatly improved to 73% even
with less catalyst (Table 10, Entry 1). Moreover, addition of
acetic acid had a positive effect of further improving yield
to 81% (Entry 2). In contrast to using cumene as a solvent
(Table 1, Entry 4), catalyst efficiency of molybdenum naph-

8
8
Bull. Chem. Soc. Jpn. Vol. 84, No. 1 (2011)
One-Pot Synthesis of Indigo from Indole
Table 10. Optimization of Reaction Conditions in Terms of
Practicality
of molybdenum complex and 2.2 equivalents of cumene
hydroperoxide in tert-butyl alcohol, the reaction was complete
in 7 h and pure indigo (1) was obtained in 81% yield as a deep-
blue solid just by filtration. This catalysis is noteworthy in pure
chemistry and applied chemistry as well.
Cat. (y mol%)
O
CHP (2.2 equiv)
H
N
MeCO H (z mol%)
2
tert-butyl alcohol
reflux, Time
N
N
Experimental
H
H
O
1H and 13C NMR
6
1
General Methods and Materials.
spectra were recorded on a JEOL EX-400 spectrometer,
Entry Cat.
y
z
Concentrationa) Time/h Yield/%b)
1
operating at 400 MHz for H NMR and 100.35 MHz for
c)
1
2
3
4
5
6
Mo(CO)6 0.1
Mo(CO)6 0.1
Mo­Nap 0.1
Mo­Nap 0.1
Mo­Nap 0.05 10
Mo­Nap 0.1
®
10
10
20
15
15
15
15
15
10
6
7
7
7
7
7
5
73
81
81
79
75
78
76
13
C NMR. The chemical shifts (ppm) were determined relative
c)
c)
c)
c)
d)
to Me Si. EI mass spectra were measured on a JEOL JMS-
4
DX300. HPLC analyses were carried out on a Shimadzu
system equipped with a UV­vis detector, and a TSK-GEL
packed column (15 cm © º4.6 mm) using n-hexane/chloro-
form/acetonitrile (65/30/10) as eluent for the quantitative
determination of by-product or an ODS-80TS packed column
10
10
7e) Mo­Nap 0.1
7.5
a) Concentration: solvent/6 (weight ratio). b) Isolated yield of
. c) Internal temperature was in a range of 85.7­86.5 °C. d)
(
15 cm © º4.6 mm) using acetonitrile/water (30/70) as eluent
1
for the quantitative determination of peroxides and indole.
Indole was purchased from Kanto Chemical Co., Inc.
Solvents were obtained from commercial suppliers and distilled
according to standard protocol to use. [H (PW Mo O )]¢nH O
Internal temperature was in a range of 86.8­89.0 °C. e) Internal
temperature was in a range of 87.4­90.3 °C.
3
6
6
40
2
showed the parent ion peak at 133 in the mass-spectrum. This
number is the molecular weight of 3-oxyindole (7). The HPLC-
Mass spectrum coincided with that of an authentic sample of
was purchased from Japan New Metals Co., Ltd. (C6H11OBO)3
1
6
was prepared according to a literature procedure. Molybde-
num naphthenate was purchased from Nihon Kagaku Sangyo
Co., Ltd. The other catalysts were purchased from Strem
Chemicals, Inc. H O (30 wt % in water) was purchased from
13
7
, proving the existence of 7 in the early reaction stage. It was
also found that 7 was easily oxidized to indigo (1) even by air
2
2
2
in a way similar to the conventional industrial process.
Wako Pure Chemical Industries, Ltd. Organic hydroperoxides
were purchased from NOF Corporation and were used without
further purification. Carboxylic acids were purchased from
Kanto Chemical Co., Inc. or Sigma-Aldrich Corporation, and
were used without further purification. Silyl compounds were
purchased from Azmax Co. and were used without further
purification. The other chemicals were purchased from com-
mercial sources and were used without further purification.
General Procedure for the Preparation of Indigo (1)
(Table 10, Entry 2). An 82 wt % solution of 1-methyl-1-
phenylethyl hydroperoxide (CHP) in cumene (34.6 g,
188.0 mmol of CHP) was added to a solution of indole (6)
(10.0 g, 85.4 mmol), acetic acid (0.51 g, 8.5 mmol), and
molybdenum hexacarbonyl (22.5 mg, 0.085 mmol) in tert-butyl
alcohol (150 g) at room temperature. This mixture was then
allowed to warm up to 86 °C (reflux) and stirred for 7 h. A
deep-blue solid gradually precipitated from the solution. The
reaction mixture was cooled to room temperature and filtered
through a 0.5 ¯m PTFE membrane filter (ADVANTEC MFS,
Inc.). The resulting solid was washed with methanol and then
dried under reduced pressure to afford indigo (1) (9.10 g, 81%)
as a deep-blue solid. The spectral data and analytical data of 1
Therefore, we believe that 7 is the reaction intermediate. The
proposed mechanism is shown in Scheme 3. The first step is
oxidation of the 3-position of indole (6) to 7. A molybdenum­
peroxy complex is postulated to be an active species of the first
step, which then reacts with 6 to form five-membered metal-
lacycle complex 17. Then, 7 forms through the dissociation of
1
7. This mechanism is quite similar to that of the catalytic
epoxidation of olefins by organic hydroperoxides proposed by
1
4
Mimoun et al. The high selectivity of the 3-position oxidation
is due to electronic reasons. It is well known that ³-electrons
of 6 are localized on the 3-position. One example of the
experimental evidence of this fact is the Friedel­Crafts reaction
of indole (6), in which an electrophilic substitution occurs
mainly at the 3-position of 6. In the present reaction, the
electrophilic oxygen atom bound to molybdenum atom of the
molybdenum­peroxy complex 16 should attack the electron
rich 3-position of 6 selectively. It has been reported that 7 is
oxidized to form indigo (1) by radical mechanism in basic
1
5
solution. In the present reaction, oxidative coupling of 7 to
form 1 is probably also a radical reaction, although it occurs in
neutral solution. Since the addition of the same amount of Mo
complex as initiate one after 2 h had almost no effect on the rate
of reaction and the yield of indigo, Mo complex most probably
promotes only the 1st step (6 to 7) and may not participate in
the 2nd step (7 to 1). Attempts to more thoroughly elucidate the
mechanism of this latter step are ongoing.
2
were agreement with those previously reported. The purity of
the solid was estimated to be greater than 98% on the basis of
1
7
the following procedure. The deep-blue solid of indigo (1)
was treated with conc. sulfuric acid to afford indigo-5,5¤-
sulfonic acid and then the aqueous solution of the resulting
mixture was titrated with potassium permanganate.
Conclusion
Large-Scale Procedure for the Production of Indigo (1).
A four neck flask having a capacity of 3 L and fitted with a
mechanical stirrer, a thermometer, a dropping funnel and a
cooling coil was charged with tert-butyl alcohol (1.0 kg),
We developed a novel and highly practical one-pot synthesis
of indigo from indole via 3-position selective oxidation of
indole and dimerization of the indole moiety. Using 0.1 mol %

Y. Yamamoto et al.
Bull. Chem. Soc. Jpn. Vol. 84, No. 1 (2011)
89
indole (6) (100.0 g, 853.6 mmol), molybdenum naphthenate
Developments in Selective Oxidation II, ed. by V. Cortés Corberán,
S. Vic Bellón, Elsevier, Amsterdam, 1994, pp. 615­622.
(Mo 5%, Nihon Kagaku Sangyo Co., Ltd.) (1.63 g, 0.85 mmol),
4
T. Honda, F. Matsuda, K. Kiyoura, K. Terada, EP Patent 69
and acetic acid (5.1 g, 84.9 mmol) sequentially. While this
mixture was being heated at 86 °C (reflux) on an oil bath and
stirred, an 82 wt % solution of CHP in cumene (348.5 g,
2
5
42, 1983.
5
61.
6
T. Yoshimura, T. Ohno, Photochem. Photobiol. 1988, 48,
C. J. Nielsen, R. Stotz, G. T. Cheek, R. F. Nelson, J.
1
878 mmol of CHP) dissolved in tert-butyl alcohol (0.5 kg) was
added dropwise thereto over a period of 1.5 h. Thereafter, the
reaction was continued for 7 h under the same conditions. The
reaction mixture was homogeneous at the start of the reaction,
but a deep-blue solid gradually precipitated out with the
progress of the reaction. The reaction mixture was cooled to
room temperature and filtered through 0.5 ¯m PTFE type
membrane filter (ADVANTEC MFS, Inc.). The resulting solid
was washed with methanol and then dried at 50 °C under
reduced pressure to deep-blue solid product. Elemental analysis
and IR spectroscopic analysis revealed that this product was
indigo (1). The molar yield of the isolated indigo as based on
the charged indole was 79.8%.
Electroanal. Chem. 1978, 90, 127.
7
8
9
0
B. Witkop, Justus Liebigs Ann. Chem. 1944, 556, 103.
B. Hughes, H. Suschitzky, J. Chem. Soc. 1965, 875.
C. E. Dalgliesh, W. Kelly, J. Org. Soc. 1958, 3726.
a) A. K. Sheinkman, H. A. Klyuev, L. A. Rbibenko, E. X.
1
Dank, Khim. Geterotsikl. Soedin. 1978, 11, 1490. b) T. Kaneko, M.
Matsuo, Y. Iitaka, Heterocycles 1979, 12, 471.
11
a) S. Sakamura, Y. Obata, Bull. Agric. Chem. Soc. Jpn.
1
1
956, 20, 80. b) B. Witkop, H. Fiedlerl, Justus Liebigs Ann. Chem.
947, 558, 91.
12 Using cumene as a solvent instead of toluene (Table 2),
further catalyst screening was performed. Although we found that
Mo(OBz)2]2, [Mo(OAc)2]2, Ta(OBu)5, and 1% Ti/SiO2 also
promoted the reaction, they did not exceed the result of Mo(CO)6.
B. Capon, F. C. Kwok, J. Am. Chem. Soc. 1989, 111, 5346.
[
References
1
3
1
For a recent review, see: a) E. S. B. Ferreira, A. N. Hulme,
14 a) H. Mimoun, Angew. Chem., Int. Ed. Engl. 1982, 21, 734.
b) P. Chaumette, H. Mimoun, L. Saussine, J. Fischer, A. Mitschler,
J. Organomet. Chem. 1983, 250, 291.
15 G. A. Russell, G. Kaupp, J. Am. Chem. Soc. 1969, 91,
3851.
H. McNab, A. Quye, Chem. Soc. Rev. 2004, 33, 329. b) E.
Steingruber, in Ullmann’s Encyclopedia of Industrial Chemistry,
ed. by B. Elvers, S. Hawkins, M. Ravenscroft, G. Schulz, VCH,
Weinheim, Germany, 1989, Vol. A14, p. 149.
2
a) Degussa, DRP No. 137955, 1901; J. Pfleger, Fortschritte
16 G. L. O’Connor, H. R. Nace, J. Am. Chem. Soc. 1955, 77,
1578.
17 Japanese Industrial Standard (JIS) K8092-1972.
der Theerfarben Fabrikation, Part 6, p. 560. b) J. G. Kern, H.
Stenerson, Chem. Eng. News 1946, 24, 3164.
3
Y. Inoue, Y. Yamamoto, H. Suzuki, U. Takaki, in New