The synthesis of some 3-acylindoles revisited

Article abstract of DOI:10.1002/jhet.5570440540

(Chemical Equation Presented) A study probing the scope of acylation of indoles with dicarboxylic acids in acetic anhydride has been performed, resulting in products incorporating 3-acylindole- or 1-acylindole motifs depending on the choice of the acid reactant. Synthetically useful results were only obtained from reactions involving malonic acid or Meldrum's acid. Correlations to previous studies have also been made and discussed.

Full text of DOI:10.1002/jhet.5570440540

Sep-Oct 2007
1213
The Synthesis of Some 3-Acylindoles Revisited
Vedran Hasimbegovic, Johnny Slätt, Jan Bergman* and Tomasz Janosik
Unit for Organic Chemistry, Department of Biosciences and Nutrition, Karolinska Institute, Novum Research
Park, SE-141 57 Huddinge, Sweden
Received March 19, 2007
O
SCN
O
O
NH
N
H
N
H
A study probing the scope of acylation of indoles with dicarboxylic acids in acetic anhydride has been
performed, resulting in products incorporating 3-acylindole- or 1-acylindole motifs depending on the
choice of the acid reactant. Synthetically useful results were only obtained from reactions involving
malonic acid or Meldrum’s acid. Correlations to previous studies have also been made and discussed.
J. Heterocyclic Chem., 44, 1213 (2007).
INTRODUCTION
RESULTS AND DISCUSSION
Acylation reactions are highly useful tools for
functionalization of indoles, and have consequently been
studied in considerable detail over the years. The common
practical routes to 3-acylindoles and related compounds
involve for instance treatment of indoles with acyl
chlorides in the presence of pyridine [1], exposure of
zincated indoles to acyl chlorides [2], Friedel–Crafts type
acylation of indoles with acyl chlorides in the presence of
Lewis acids [3,4], or Friedel–Crafts reactions of indoles
possessing electron withdrawing N-substituents [5]. In
addition, dimeric 3-acylindoles connected by long alkyl
chains have been prepared by exposure of indole to
bisimidazoline derivatives in acetic anhydride [6]. A
recent contribution to this field features a new useful
technique for facile preparation of 3-(cyanoacetyl)indole
derivatives by acylation employing cyanoacetic acid in
acetic anhydride [7]. The resulting 3-(cyanoacetyl)-
indoles, for instance the parent molecule 1, are very useful
starting materials for construction of other related indole
containing systems [8], whereas the similarly prepared
phosphonate 2 has been employed in a stereoselective
synthesis of the alkaloid murrayacarine [9]. Further
synthetic applications of 3-(cyanoacetyl)indole 1 have
emerged recently, leading to preparation of cytotoxic
analogues of the marine natural product meridianin D [10]
via the previously reported enamine product originating
from treatment of 1 with DMFDMA [8].
It is well established that heating of indole with
chloroacetyl chloride in toluene gives 1-(chloroacetyl)-
indole 3 [11], which has previously also been obtained
from the reaction of indole with chloroacetic anhydride in
refluxing dioxane [12]. Therefore, we were rather
surprised to learn that a very recent paper claims prepar-
ation of 3-(chloroacetyl)indole 4 by heating indole with
chloroacetyl chloride in refluxing dioxane [13]. However,
the reported melting point of the resulting product (mp.
105 °C [13]) does not match that of the well-known
compound 4 (mp. 233–234 °C [14], 230–232 °C [1]), but
is quite close to that previously recorded for 1-(chloro-
acetyl)indole 3 (mp 116–117 °C [12], 115 °C [11], 120–
121 °C [15]). The structure of 3 as a 1-substituted
derivative of indole has also been rigorously proven a
long time ago by dehydrogenation of 1-(chloroacetyl)-
indole with DDQ [12]. Not unexpectedly, a repetition of
the experiment described by Elnagdi [13] gave a product
which was easily identified as the known 1-(chloro-
1
acetyl)indole 3. This assignment was supported by H
NMR data in DMSO-d6, which displayed a characteristic
doublet of a doublet at ꢀ 6.81 corresponding to the indole
H-3, coupling to a doublet at ꢀ 7.87 (indole H-2), as well
as the absence of an acidic hydrogen resonance at low
field. The 13C NMR spectrum in combination with a
DEPT experiment revealed the presence of six doublets,
giving further indication that substitution had occurred at
the indole nitrogen atom. In addition, the IR absorption
observed for the carbonyl functionality (ꢁ = 1698 cm-1) is
typical for an N-acylated indole. It was therefore clear that
Elnagdi and co-workers have in fact obtained 1-
(chloroacetyl)indole 3 instead of the claimed compound 4.
Moreover, the 1H NMR data for 3 recorded in CDCl3 were
in good agreement with those attributed to 4 in Elnagdi’s
O
R
N
H
N
(O)
= C
= P
1 R
2 R
t)
(OE
2

1214
V. Hasimbegovic, J. Slätt, J. Bergman and T. Janosik
Vol 44
paper [13]. Ottoni and co-workers have published an
interesting procedure (unduly criticized by Elnagdi)
involving pre-mixing of indole in dichloromethane with a
Lewis acid (e.g. SnCl4), followed by addition of the
appropriate acid halide [3]. We have prepared 3-
(chloroacetyl)indole 4 in good yield using this method,
which is in fact competitive with the most commonly used
literature procedure [1]. The advantage with the Brazilian
method is the higher yield, and the drawback is a
somewhat more laborious procedure.
point 287 °C [16], which is considerably higher than that
reported by Elnagdi (152 °C) [13]. In order to shed some
light on these discrepancies, the reaction involving two
equivalents of indole with malonic acid in acetic
anhydride was performed essentially as described by
Elnagdi, giving a rather complex mixture of products.
Repeated crystallization of the crude product mixture or
purification using column chromatography gave a low
yield of a material which was unsymmetrical, as it
exhibited two sets of indolic resonances in the NMR
1
The incorrect assignment by Elnagdi discussed above
has serious consequences for some other reactions
embodied in their paper. For example, a substitution
experiment involving their material and ammonium
thiocyanate does of course not lead to the purported
product 2-(1H-indol-3-yl)-2-oxoethyl thiocyanate 5. This
was easy to verify by treatment of an authentic sample of
3-(chloroacetyl)indole 4 prepared according to the well-
established literature procedure [1] with ammonium
thiocyanate in refluxing acetonitrile, which gave the new
compound 5 in good yield. This material displayed data
fully consistent with the expected structure, including an
indole NH resonace at ꢀ 12.21 in the 1H NMR recorded in
DMSO-d6, a ketone carbonyl signal at ꢀ 186.1 in the 13C
NMR spectrum, as well as a typical ketone absorption at ꢁ
= 1616 cm-1 in the IR spectrum. In addition, the
conversion of the claimed 4 with cyanide into 3-
(cyanoacetyl)indole 1 described by Elnagdi [13] is hard to
understand unless some mysterious rearrangement leading
to transfer of the acyl group from the indole nitrogen to C-
3 was in operation, which is an unlikely scenario under
the reported reaction conditions.
spectrum. Moreover, the H NMR data featured only one
acidic hydrogen signal, whereas the 13C NMR and IR
spectra suggested the presence of two different carbonyl
functionalities, namely one ketonic and one amidic. Based
on these data, the product was assigned the previously
described unsymmetrical bisindole structure 7 [17]. There
were no indications of keto–enol tautomerism in any of
the NMR spectra. It is also important to note that Elnagdi
used malonic acid and indole in the molar ratio 1:1. When
repeated, this experiment also failed to give the product 6,
leading instead to isolation of 3-acetylindole as the major
component after careful column chromatography of the
complex crude product mixture. In addition, small
amounts of the diindole derivative 8 could be obtained.
Interestingly enough, all 13C NMR resonances recorded
for the indole derivative 8 find rather good matches in the
data given by Elnagdi for the purported compound 6,
suggesting that he might perhaps have isolated an impure
sample of 8. Furthermore, we found that heating of indole
with Meldrum’s acid (2,2-dimethyl-1,3-dioxane-2,5-
dione) in acetic anhydride gave the product 8 in 22%
yield, proving a more convenient route despite the low
yield. It should also be mentioned that there were no
indications of tautomerism during NMR studies of 8 in
DMSO-d6 solution. In this context some interesting
experiments reported by Majima and Shigematsu should
be mentioned. After reaction of indole with diethyl
malonate, they isolated in minimal amounts a product that
was not further studied. We have not repeated this
experiment, but we consider it likely that the Japanese
workers had compound 8 in their hands. In a similar
experiment with diethyl succinate they were able to
isolate the 1,1'-connected isomer of 9 [18].
O
R
N
O
N
H
Cl
4 R = Cl
5 R = SCN
3
A deeper scrutiny of some other of the experiments
detailed by Elnagdi [13] revealed additional
inconsistencies. It was stated that treatment of indole with
a reagent generated from malonic acid in acetic anhydride
gives the symmetrical product 1,3-di(1H-indol-3-
yl)propane-1,3-dione 6 as a mixture of the tautomers
6a/6b in a good yield. However, there are no ketone
resonances in the listed 13C NMR data, while an IR
absorption at ꢁ = 1701 cm-1 raised instead our suspicion
that the actual product may instead incorporate at least
one indole unit connected via its nitrogen atom. It should
also be noted that compound 6 had already been reported
in 1922 by Sanna from the reaction of the indole Grignard
reagent with malonyl chloride, and displayed the melting
Finally, the target 6 could be obtained in a low yield by
exposure of indole to freshly distilled malonyl chloride in
dioxane. The NMR spectra in DMSO-d6 clearly revealed
the co-existence of the tautomers 6a/6b in the ratio 1:0.33.
However, neither of the forms 6a/6b matched the data
given by Elnagdi. For example, we found that the ketone
carbon atom of the tautomer 6a resonates at ꢀ 189.4, just
as might be expected for this type of structure, whereas
the two most downfield signals reported by Elnagdi
appear at ꢀ 165.75 and 162.34, suggesting amide carbonyl
signals. These facts, combined with the other

Sep-Oct 2007
Comumication to the Editor
1215
discrepancies discussed above raise serious doubts
concerning the identity of Elnagdi’s product.
have questioned previous results which have been verified
and used [21-24] by numerous successors over the years
is particularly misleading and ignorant, as they have
clearly misinterpreted the spectral evidence which
contradicts their conclusions and supports the previous
findings. Readers interested in the preparation, reactivity,
and applications of 3-acylindole derivatives are instead
advised to consult the leading references cited in this
work.
O
HO
O
O
NH
NH
N
H
N
H
6b
6a
EXPERIMENTAL
O
O
O
O
N
N
General remarks. NMR spectra were recorded at
1
300.1 MHz for H and 75.5 MHz for 13C, respectively,
N
using the residual solvent signal as reference, unless
otherwise stated. Coupling constants are given in Hz. The
IR spectra were acquired using a FT-IR instrument.
Melting points were determined on a capillary melting
point or a hot stage apparatus. All reagents were
commercially available and used as received. All solvents
were purified by distillation or were of analytical grade.
Chromatographic separations were performed on silica
gel (35–70 μm particle size).
N
H
8
7
The situation becomes slightly different when longer
chain dicarboxylic acids are involved in similar reactions,
as the longer distance between the carbonyl units decrease
the reactivity of the mixed anhydride. This may be
illustrated by performing an acylation of indole with
succinic acid in acetic anhydride (1:1 molar ratio) as
described by Elnagdi, which does not give the known
dione 9 [19,20], again giving rise to a complex mixture
from which the well-known product 3-acetylindole could
be isolated. The other products were present only in
minute quantities and were not isolated or characterized.
Interestingly enough, the data provided by Elnagdi for the
1-(Chloroacetyl)indole [2-Chloro-1-(1H-indol-1-yl)-
ethanone] (3). This material was prepared by heating
choroacetyl chloride with indole in 1,4-dioxane as
described by Elnagdi [13]. The resulting pinkish
crystalline solid was identified as 3, mp 115–117 °C
(lit.[12] 116–117 °C, lit.[11] 115 °C, lit.[15] 120–121 °C);
ir (neat): 1698 (CO) cmꢀ1; 1H nmr (DMSO-d6): 8.33 (dd, J
= 8.1, 0.6 Hz, 1H), 7.87 (d, J = 3.8 Hz, 1H), 7.66–7.63 (m,
1H), 7.39–7.27 (m, 2H), 6.81 (dd, J = 3.8, 0.6 Hz, 1H), 5.12
purported dione
9 disagree with those published
previously, in particular concerning the melting point
(233 °C) [13], which does not match with the reported
value (189 °C) [19,20]. Also in this case, the correctness
of the assignment of Elnagdi’s material must be
questioned. In addition, there were no NMR data
provided, thus precluding further comparison.
1
(s, 2H); H nmr (CDCl3, 500.2 MHz): 8.44 (d, J = 7.8 Hz,
1H), 7.58 (d, J = 7.8 Hz, 1H), 7.43 (d, J = 4.1 Hz, 1H),
7.41–7.38 (m, 1H), 7.33–7.30 (m, 1H), 6.72 (d, J = 4.1 Hz,
1H), 4.58 (s, 2H); 13C nmr (DMSO-d6): 165.4 (s), 135.0 (s),
130.1 (s), 126.2 (d), 124.9 (d), 123.9 (d), 121.0 (d), 115.7
(d), 109.1 (d), 43.5 (t); MS (ESI+) m/z 194 [M + H]+.
3-(Chloroacetyl)indole [2-Chloro-1-(1H-indol-3-yl)-
ethanone] (4). The procedure described here is based on a
communication by Ottoni et al. [3]. Indole (5.85 g, 50
mmol) was dissolved in CH2Cl2 (80 mL) whereupon
SnCl4 (13.0 g, 50 mmol) was added with stirring at 0–5
°C. After completed addition, chloroacetyl chloride (5.65
g, 50 mmol) was added dropwise to the blue mixture.
Addition of nitromethane (30 mL) as a co-solvent was
necessary. This mixture was left at room temperature for
12 h, whereupon water (80 mL) and conc. HCl (25 mL)
were added. The solid formed was collected after 2 h at
room temperature and washed carefully with 2-propanol
which gave 4.65 g (48%) of 3-(chloroacetyl)indole.
Evaporation of the organic phase followed by trituration
with ethanol gave an additional quantity (3.05 g, 31%) of
4 after rather laborious purification. The total yield of 4
was 79%. All data are in excellent agreement with those
published previously [1,14].
H
N
O
O
N
H
9
In conclusion, it appears clear from the discussion
above that many of the results presented in the Elnagdi
paper are seriously flawed, as particularly illustrated by
one of the key experiments claiming preparation of
3-(chloroacetyl)indole by heating indole with chloroacetyl
chloride in dioxane, a reaction which in our hands
gave the expected well-known isomeric product
1-(chloroacetyl)indole. The fact that Elnagdi et al. also

1216
V. Hasimbegovic, J. Slätt, J. Bergman and T. Janosik
Vol 44
2-(1H-Indol-3-yl)-2-oxoethyl thiocyanate (5). A mixture
of 3-(chloroacetyl)indole [1] (4) (1.94 g, 10 mmol),
ammonium thiocyanate (1.52 g, 20 mmol) and a catalytic
amount of tetrabutylammonium iodide (18 mg) in
anhydrous acetonitrile (50 mL) was heated at reflux for 3
h, and was thereafter allowed to cool to room temperature.
Removal of the solvent in vacuo gave a solid residue,
which was triturated with water (50 mL) giving a beige
precipitate. This material was collected by filtration,
dried, and crystallized from ethanol/acetonitrile to provide
the title compound 5 (1.11 g) as tan crystals. A second
crop (0.30 g) could be collected from the mother liquor.
The total yield of 5 was 1.31 g (61%). Tan crystals, mp
218–219 °C (decomp.); ir (neat): 3250 (NH), 2158 (SCN),
Alternatively, the reaction was performed using indole
(23.4 g, 0.2 mol) and malonic acid (10.4 g, 0.1 mol) in
acetic anhydride (100 mL). Workup as above, followed
by suspension of the crude complex product mixture in
2-propanol give a solid material, which was finally
subjected to repeated crystallizations from acetonitrile/
DMF providing small amounts of 7 (1.5 g, 5%) as a white
crystalline solid, mp 246–248 °C (lit.[17] 237–238 °C); ir
1
(neat): 3213 (NH), 1697 (CO), 1617 (CO) cm-1; H nmr
(DMSO-d6): 12.15 (s, 1H), 8.53 (d, J = 3.2 Hz, 1H), 8.38
(d, J = 8.0 Hz, 1H), 8.14 (d, J = 7.1 Hz, 1H), 7.88 (d, J =
3.8 Hz, 1H), 7.65–7.62 (m, 1H), 7.52 (d, J = 7.3 Hz, 1H),
7.35–7.20 (m, 4H), 6.75 (d, J = 3.7 Hz, 1H), 4.77 (s, 2H);
13C nmr (DMSO-d6): 187.6 (s), 167.4 (s), 136.7 (s), 135.6
(d), 135.0 (s), 130.4 (s), 127.3 (d), 125.3 (s), 124.7 (d),
123.7 (d), 123.2 (d), 122.1 (d), 121.1 (s), 120.9 (d), 115.9
(d), 115.9 (d), 112.3 (d), 108.5 (d), 47.9 (t); MS (ESI+)
m/z 303 [M + H]+.
1
1616 (CO) cm-1; H nmr (DMSO-d6): 12.21 (s, 1H), 8.49
(s, 1H), 8.16–8.13 (m, 1H), 7.53–7.50 (m, 1H), 7.27–7.23
(m, 2H), 4.85 (s, 2H); 13C nmr (DMSO-d6): 186.1 (s),
136.6 (s), 135.7 (d), 125.1 (s), 123.3 (d), 122.3 (d), 121.0
(d), 113.9 (s), 113.2 (s), 112.4 (d), 40.9 (t); MS (ESI–) m/z
215 [M – H]–. Anal. Calcd. for C11H8N2OS: C, 61.09; H,
3.73; N, 12.95. Found: C, 61.18; H, 3.80; N, 12.86.
Treatment of indole with malonic acid in acetic
anhydride. A suspension of malonic acid (3.12 g, 30
mmol) in acetic anhydride (30 mL) was heated to 85
°C for 10 min. The resulting solution was then treated
with indole (3.52 g, 30 mmol), heated at reflux for 30
min, allowed to cool to room temperature, and poured
over ice/water (250 g). The obtained solid was
collected by filtration, washed with several portions of
water, dried, and separated by column chromatography
(EtOAc/n-heptane 1:2) to yield 1,3-di(1H-indol-1-
yl)propane-1,3-dione (8) (93 mg, 1%), which gave
data identical with those reported for the material
prepared by exposure of indole to Meldrum’s acid in
acetic anhydride (see below), followed by 3-
acetylindole (1.06 g, 22%). This procedure gave a
plethora of products, most of which were present in
trace quantities, thus only the two components
indicated above were isolated and characterized. Data
for 3-acetylindole: 191.5–192 °C (lit.[2] 187–189 °C;
lit.[14] 191–193 °C), ir (neat): 3118 (NH), 1610 (CO)
1,3-Di(1H-indol-3-yl)propane-1,3-dione (6a/6b).
Freshly distilled malonyl chloride (1.41 g, 10 mmol) was
added to indole (2.34 g, 20 mmol) in dioxane (20 mL). The
resulting solution was kept at 60 °C for 10 min., allowed to
cool, and poured into water. The resulting semi-solid was
treated with 2-propanol which gave the title compound in a
low yield (350 mg). However, slow concentration of the
mother liquor gave a more substantial quantity (650 mg),
totalling the yield to 1.0 g (30%). White solid with a pinkish
tinge, mp 290–292 °C (lit.[16] 287 °C); ir (neat): 3294
1
(NH), 1597 (CO) cm-1; H nmr (DMSO-d6, 500.2 MHz)
(only resonances for 6a are listed): 12.01 (s, 2H), 8.42 (d, J
= 3.2 Hz, 2H), 8.16 (d, J = 7.8 Hz, 2H), 7.49–7.47 (m, 2H),
7.16–7.23 (m, 4H), 4.46 (s, 2H); 13C nmr (DMSO-d6) (only
resonances for 6a are listed): 189.4 (s), 136.7 (s), 135.4 (d),
125.5 (s), 123.0 (d), 121.9 (d), 121.3 (d), 116.8 (s), 112.2
(d), 52.3 (t); MS (ESI+) m/z 303 [M + H]+.
1
1
The nmr resonances (in particular for H) of the enol
cm-1; H nmr (DMSO-d6): 11.9 (br s, 1H), 8.30 (s, 1H),
tautomer 6b overlap much with those of 6a, but the
8.20–8.17 (m, 1H), 7.48–7.45 (m, 1H), 7.21–7.16 (m,
2H), 2.45 (s, 3H); 13C nmr (DMSO-d6): 192.6 (s),
136.7 (s), 134.3 (d), 125.3 (s), 122.7 (d), 121.6 (d),
121.3 (d), 116.8 (s), 112.1 (d), 27.3 (q).
1
diagnostic signals for the alkene unit at 6.88 (s) in the H
nmr and 92.1 (d) in the 13C nmr could be discerned
clearly.
1-(1H-Indol-1-yl)-3-(1H-indol-3-yl)propane-1,3-dione
(7). This reaction was performed according to Elnagdi,
but using the starting materials indole and malonic acid in
a molar ratio 2:1. A suspension of malonic acid (1.04 g,
10 mmol) in acetic anhydride (10 mL) was heated to 85
°C for 10 min. The resulting solution was then treated
with indole (2.34 g, 20 mmol), heated at reflux for 30
min, allowed to cool to room temperature, and poured
over ice/water (200 g). The obtained solid was collected
by filtration, washed with several portions of water, dried,
and purified by column chromatography (EtOAc/n-
heptane 1:2) to yield 1-(1H-indol-1-yl)-3-(1H-indol-3-
yl)propane-1,3-dione (7) (154 mg, 5%) which displayed
data as below.
1,3-Di(1H-indol-1-yl)propane-1,3-dione (8). Indole
(1.10 g, 9.4 mmol) was added to a mixture of Meldrum’s
acid (2,2-dimethyl-1,3-dioxane-2,5-dione) (0.65 g, 4.5
mmol) in acetic anhydride (20 mL) at 70 °C. The resulting
mixture was heated at 70 °C for 2 h, and was thereafter
allowed to cool and concentrated to a semisolid.
Trituration of this material with Et2O afforded 8 (0.30 g,
22%) as a yellowish solid, mp 192.5–193.5 °C; ir (neat):
1
1695, 1677 (CO) cm-1; H nmr (DMSO-d6): 8.35 (d, J =
7.8 Hz, 2H), 7.97 (d, J = 3.8 Hz, 2H), 7.67–7.64 (m, 2H),
7.39–7.28 (m, 4H), 6.81 (d, J = 3.8 Hz, 2H), 5.03 (s, 2H);
13C nmr (DMSO-d6): 166.0 (s), 135.1 (s), 130.5 (s), 127.2
(d), 124.9 (d), 124.0 (d), 121.0 (d), 115.9 (d), 109.0 (d),
44.8 (t); MS (ESI+) m/z 303 [M + H]+.Anal. Calcd. for

Sep-Oct 2007
Comumication to the Editor
1217
[8] Slätt, J.; Janosik, T.; Wahlström, N.; Bergman, J. J.
Heterocycl. Chem. 2005, 42, 141.
[9] Johnson, A.-L.; Slätt, J.; Janosik, T.; Bergman, J.
Heterocycles 2006, 68, 2165.
[10] Radwan, M. A. A.; El-Sherbiny, M. Bioorg. Med. Chem.
2007, 15, 1206.
[11] Mutschler, E.; Winkler, W. Arch. Pharm. (Weinheim) 1978,
311, 248.
C19H14N2O2: C, 75.48; H, 4.67; N, 9.27. Found: C, 75.61;
H, 4.60; N, 9.20.
Treatment of indole with succinic acid in acetic
anhydride. A solution of succinic acid (1.18 g, 0.01 mol)
in acetic anhydride (10 mL) was heated to 85 °C for 10
min. The resulting solution was then treated with indole
(1.17 g, 0.01 mol), heated at reflux for 30 min, allowed to
cool to the room temperature, and poured over ice/water
(150 g). The obtained solid was collected by filtration,
washed with several portions of water, dried, and purified
by column chromatography (EtOAc/n-heptane 1:2) to
yield 3-acetylindole (0.67 g, 42%) which displayed data
as above.
[12] Bergman, J.; Carlsson, R.; Misztal, S. Acta Chem. Scand.
1976, B30, 853.
[13] Abdel-Motaleb, R. M.; Makhloof, A.-M. A.-S.; Ibrahim, H.
M.; Elnagdi, M. H. J. Heterocycl. Chem. 2007, 44, 109.
[14] Anthony, W. C. J. Org. Chem. 1960, 25, 2049.
[15] Martin, P. Helv. Chim. Acta. 1984, 67, 1647.
[16] Sanna, G. Gazz. Chim. Ital. 1922, 52, 170.
[17] Vinograd, L. K.; Sorokina, N. P.; Turchin, K. F.; Dubinskii,
R. A.; Suvorov, N. N. Zh. Org. Khim. 1980, 16, 2594.
[18] Majima, R.; Shigematsu, T. Chem. Ber. 1924, 57, 1449.
[19] Banerji, A.; Bandyopadhyay, D.; Basak, B. Heterocycles
2004, 63, 2371.
REFERENCES
[1] Bergman, J.; Bäckvall, J.-E.; Lindström, J.-O. Tetrahedron
1973, 29, 971.
[2] Bergman, J.; Venemalm, L. Tetrahedron 1990, 46, 6061.
[3] Ottoni, O.; Neder, A. d. V. F.; Dias, A. K. B.; Cruz, R. P. A.;
Aquino, L. B. Org. Lett. 2001, 3, 1005.
[20] Banerji, A.; Bandyopadhyay, D.; Basak, S.; Sur, K. R.; Paul,
J. N.; Banerji, J.; Chatterjee, A. Tetrahedron Lett. 2005, 46, 7033.
[21] Moody, C. J.; Ward, J. G. J. Chem. Soc., Perkin Trans. 1
1984, 2903.
[22] Aubé, J.; Wang, Y.; Hammond, M.; Tanol, M.; Takusagawa,
F.; Velde, D. V. J. Am. Chem. Soc. 1990, 112, 4879.
[23] Dong, W.-P.; Nam, G.; Fan, J.; Kirk, K. L. J. Org. Chem.
2003, 68, 2798.
[4] Wynne, J. H.; Lloyd, C. T.; Jensen, S. D.; Boson, S.; Stalick,
W. M. Synthesis 2004, 2277.
[5] Ketcha, D. M.; Gribble, G. W. J. Org. Chem. 1985, 50, 5451.
[6] Bergman, J.; Koch, D. Heterocycl. Commun. 1996, 2, 305.
[7] Slätt, J.; Romero, I.; Bergman, J. Synthesis 2004, 2760.
[24] Johnson, A.-L.; Bergman, J. Tetrahedron 2006, 62, 10815.