Comparative study of the different approaches to the synthesis of isatins with a chiral substituent at the nitrogen atom

Article abstract of DOI:10.1007/s10593-011-0654-y

Through a comparative study of three different routes to the synthesis of isatins with a chiral substituent on the nitrogen atom it has been shown that better results for preparing the isatins are achieved using the Sandmeyer (>99% ee, yield 50%) rather t

Full text of DOI:10.1007/s10593-011-0654-y

Chemistry of Heterocyclic Compounds, Vol. 46, No. 10, 2011
COMPARATIVE STUDY OF THE DIFFERENT APPROACHES
TO THE SYNTHESIS OF ISATINS WITH A CHIRAL
SUBSTITUENT AT THE NITROGEN ATOM
A. V. Kurkin1*, A. A. Bernovskaya1, and M. A. Yurovskaya1
Through a comparative study of three different routes to the synthesis of isatins with a chiral substituent
on the nitrogen atom it has been shown that better results for preparing the isatins are achieved using
the Sandmeyer (>99% ee, yield 50%) rather than the Stolle method (95% ee, yield 16%). The procedure
proposed by Gassman for preparing isatins was unsuitable for the compounds described.
Keywords: chiral N-arylanilines, isatins with a chiral substituent on the nitrogen atom, the Sandmeyer,
Stolle, and Gassman methods, enantioselectivity.
Isatin and its derivatives are widely used in the synthesis of a broad range of heterocyclic compounds
and they are also as starting material for preparing drugs. The interest in isatin derivatives is due to their broad
spectrum of biological activity They show antimicrobial and antiviral activity, affect the CNS, are involved in
metabolism, and stimulate plant growth [1].
The preparation of novel isatins, a study of their properties, and their use as building blocks for the
synthesis of other heterocyclic structures are interesting and timely questions. This especially concerns isatin
derivatives with a chiral substituent on the nitrogen atom, evidence for which is absent in the literature. The
importance of problem of the synthesis of chiral derivatives is due to the fact that, among substances synthesized
every year and undergoing biological testing, ever more attention is paid to enantiomerically pure (not racemic)
compounds and these substances are increasingly used as drugs [2].
Several basic routes to the synthesis of isatins with an alkyl or aryl substituent on the nitrogen atom can
be identified [3]. The first of these is the preparation of isatins developed by Sandmeyer. It is based on the
reaction of aromatic amines with chloral hydrate and hydroxylamine hydrochloride to give isonitro-
soacetanilides. In the presence of sulfuric acid they cyclize to the corresponding isatins. In the Sandmeyer
reaction it is possible to use anilines containing electron acceptor groups in the aromatic ring and heterocyclic
amines [1]. The reaction is simple in use, the starting materials are cheap, and the cyclization product yields
often exceed 75% [1] hence this route is widely used in practice. However, the yields depend on the structure of
the starting amine, e.g. in the cyclization of 5-methylaniline to the isatin the yield amounts to only 22% [3]. One
of the important alternatives to the Sandmeyer method is that for preparing isatins due to Stolle. In this, aniline
_______
*To whom correspondence should be addressed, e-mail: kurkin@direction.chem.msu.ru.
1M. V. Lomonosov State University, Moscow 119991, Russia.
__________________________________________________________________________________________
Translated from Khimiya Geterotsiklicheskikh Soedinenii, No. 10, pp. 1497-1504, October, 2010.
Original article submitted June 8, 2010.
1208
0009-3122/11/4610-1208©2011 Springer Science+Business Media, Inc.

derivatives react with excess oxalyl chloride to give intermediate N-aryloxanilic acid chlorides which cyclize to
isatins in the presence of Lewis acids [1]. It should be noted that Stolle reaction yields are typically in the range
30-60% and substrates containing electron acceptor groups do not take part. A third method for preparing both
unsubstituted and N-alkyl-substituted isatins is the Gassman reaction. From an aniline derivative it produces an
intermediate 3-methylthio-2-oxindole which is oxidized to the corresponding isatin, e.g. using NaIO4 in THF.
The cyclization product yields in the Gassman reaction range from 40 to 80% but, with the presence in the
substrate structure of a readily oxidizable group (e.g. hydroxyl or aldehyde), this method loses its attraction [3].
As follows from the above, the majority of the widely used practical methods for preparing isatin
derivatives from anilines can be successfully used for obtaining N-alkylated isatins. Hence anilines which
contain a chiral substituent on the nitrogen can be involved in different cyclizations to give isatins. In this work
we have carried out a comparative study of the efficiency and stereoselectivity of the three different routes for
preparing isatins with a chiral substituent on the nitrogen through the Sandmeyer, Stolle, and Gassman
approaches.
R1
R1
1.
Cl3CCH(OH)2,
(COCl)2,
NH
R
NH
R
.
NH2OH HCl, Na2SO4
Lewis acid
2.
H2SO4, 
Stolle
Sandmeyer
R1
O
O
N
R
1.
2.
SO2Cl2, MeSCH2CO2Et, Et3N, AcOH
CuO, CuCl2
Gassman
R1
NH
R
R = H, Alk; R1 = H, Hal, Alk, OAlk
As starting materials for the preparation of the isatins with a chiral substituent on the nitrogen atom we
selected optically active N-arylalanines. They were synthesized under stereospecific conditions using the
Mitsunobu reaction by alkylation of N-arylsulfonylanilines with the commercially available ethyl (S)-lactate.
We have previously reported the potential preparation of such optically active N-arylalanines in good yields and
with high enantioselectivity (>99% ee) as well as the cyclization of these chiral anilines to isatins using the
Sandmeyer method [4]. The enantiomeric purity of the isatins prepared was determined by HPLC using a chiral
stationary phase as 97-99% ee with a yield of 52-67%. Because varying the substituent in the benzene ring of
the aniline (H, Br, Me) did not affect the reaction course we decide to select two approaches to the synthesis of
the racemic and enantiomerically pure (R)-isomer of ethyl N-(4-phenylamino)propionate as model substrates.
We selected and developed conditions for the racemic substrate which were then used for its enantiomerically
pure analog in order to study the enantioselectivity of the reaction.
A similar strategy was also used for studying the applicability of the Stolle method for preparing isatins
with a chiral substituent on the nitrogen atom.
1209

In the Stolle synthesis of isatins it is known that N-aryloxanilic acid chlorides (prepared by condensation
of anilines with excess oxalyl chloride) cyclize to the corresponding isatins in the presence of different Lewis
acids, viz. AlCl3 [5], BF3·Et2O [6], or TiCl4 [7]. The reaction is often carried out by heating with carbon
disulfide, methylene chloride, tetrachloroethane, benzene, or chloroform used as solvent [1]. We wanted to
select that synthetic method for isatins with a chiral substituent at the nitrogen atom which would take place
under the mildest conditions and would not lead to the racemization of optically active substrates. We chose the
method reported in the literature in which TiCl4 was used as Lewis acid and methylene chloride at 12ºC used as
solvent [7]. However, in these conditions, we were unable to observe formation of the target isatin and separated
only the starting amine 1. Change in the conditions (amount of TiCl4, reaction time, temperature) led to
improved results. The highest yield of isatin (58%) was achieved by heating the reaction mixture at 30-35ºC for
6 h (Table 1).
O
(COCl)2, TiCl4
O
CH2Cl2
NH
N
O
Me
H
Me
H
EtO
EtO
O
(R)-2
(R)-1
Thus having optimized the Stolle method of synthesizing isatins for the racemic N-alkylated amine we
extended it to the enantiomerically pure analog. An important feature of our study is the question of the
retention of the optical purity of the isatin prepared with a chiral substituent on the nitrogen atom. We carried
out the experiments numbered 2, 4, and 5 to reveal how the reaction time and temperature influenced the isatin
enantiomer yield. It was found that an increase in temperature affected the retention of isatin optical purity to a
lesser extent than an increase in the reaction time to 6 days. The highest enantiomeric purity for the isatin (R)-2
was achieved in experiment 2 (24 h, ~ 20ºC) with 95% ee, i.e. partial racemization was observed even here. The
angles of rotation were measured for the isatins obtained by the three experimental routes (Table 2).
TABLE 1. Optimization of the Stolle Cyclization Conditions*
Reaction conditions
Т, °С Time, h
Yield of
Experiment
TiCl4, equiv.
isatine, %
1
2
3
4
5
4
4
6
4
4
12
16 [7]
—
14
32
55
58
20
24
20
24
20
144 (6 days)
6
30-35
_______
* The conditions were developed for the racemic compounds.
TABLE 2. Characteristics of the Isatins with a Chiral Substituent on the
Nitrogen Atom
[]D20 of the isatin,
Enantiomeric purity
of the chiral isatin, % ее
c = 1 g/100 ml
Experiment
Isatin yield, %
dichloromethane
2
4
95
73
+18
+16
16
38
1210

The Gassman synthesis is another variant of the cyclization of an N-alkylated amine to the isatin. In this
case there occurs the formation and subsequent oxidation of an intermediate 3-methylsulfanyl-2-oxindole [8]
whose synthesis can involve two methods depending on the electronic effects of the substituent in the aromatic
ring. In the case of electron acceptor groups the oxindole derivative is prepared from the intermediate
N-chloroaniline which reacts with ethyl methylthioaceate to form an azosulfonium salt [9]. For electron donor
substituents (which destabilize the N-chloro intermediate and hence lower the yields of the azosulfonium salt) a
second method is used involving reaction of a chlorosulfonium salt with the corresponding aniline [10].
The intermediate 3-methylsulfanyl-2-oxindole is then oxidized to the isatin by several methods, e.g. by
carrying out a reaction with N-chlorosuccinimide to yield the unstable 3-chloro-3-methylsulfanyl-2-oxoindoles
which are hydrolyzed to isatins in the presence of mercuric oxide and boron trifluoride etherate in aqueous THF
[8]. It is possible to obtain the isatins by oxidation of the oxindole derivatives with a mixture of copper(II) oxide
and chloride [7].
We have used two routes to the Gassman synthesis of isatins, viz. its modification in which the
chlorosulfonium salt is prepared from the corresponding sulfoxide and oxalyl chloride [11] and the classical
variant. In the first case we could not prepare the target compound and only the starting amine 1 was isolated
from the reaction mixture. In the second case the reaction did not go to completion and LCMS data showed that
the reaction mixture contained the acyclic intermediate compound 3 as well as the starting amine 1.
O
–
Cl
H2C
S
OEt
SO2Cl2
Et3N
+
MeSCH2CO2Et
+
N
Me
Me
NH
Me
EtO
EtO
H
H
O
O
(R)-1
O
OEt
SMe
[2,3]-sigmatropic
rearrangement
SMe
+
O
H
N
O
NH
Me
H
Me
H
EtO
O
EtO
4
(R)-3*
[α]D20 = +35 (с = 1, CH2Cl2)
_______
* We have proposed an (R)-configuration for compound 3 because the chiral center is not affected in the
reaction process. A separate study to establish the absolute configuration has not been carried out.
Increasing the reaction time and the amount of the reagents led only to lowering of the yields of the
acyclic intermediate compound 3 (Table 3)
TABLE 3. Optimization of the Conditions for Preparing the Acyclic
Compound 3*
SO2Cl2 (MeSCH2CO2Et)
Experiment
Reaction time, h
Yield of compound 3, %
and Et3N, equiv.
1
2
3
1.06 and 1.04
1.5 and 1.47
1.06 and 1.04
3.5
3.5
5.5
27
18
13
_______
* The conditions were developed for the racemic compounds.
1211

TABLE 4. Selection of Conditions for Compound 3 Cyclization*
Reaction conditions
Experiment
H+
Т, °С
Time, h
1
2
3
4
5
6
7
8
АсОН [7]
20
20
20
20
20
65
65
78
1
24
18
24
24
2
CF3COOH
2 mol/l HCl [11]
6 mol/l HCl
12 mol/l HCl
1 mol/l HCl [5]
6 mol/l HCl
4.5
2
12 mol/l HCl
_______
* The cyclization product 4 was not seen in any of the experiments.
We considered that the reason for the lack of success is an insufficient medium acidity. However, the use
of the stronger trifluoroacetic and hydrochloric acids at different concentrations (Table 4) also did not yield the
cyclization product 4.
From LCMS data we observed the starting amine 3 as well as its hydrolysis product in the reaction
mixture. It is possible that we were unable to prepare the cyclization product 4 because of steric hindrances. In
[9] the authors reported the preparation of stable o-aminophenylacetic acid derivatives, one example of which is
compound 5. This was separated and fully characterized but attempts to cyclize it to the corresponding
3-methylthio-2-oxindole were not undertaken by the authors.
OEt
O
SMe
NHAc
5
In the case of the enantiomerically pure amine (R)-1 as starting material in the Gassman synthesis we
also observed the formation of compound 3 which showed optical activity and whose angle of rotation was
measured.
Hence the Sandmeyer and Stolle methods of the three proposed by us as routes to isatins with a chiral
substituent on the nitrtogen atom led to the target isatins. The Gassman synthesis proved to be unsuited to our
model substrate. In the case of the optically active isatins prepared according to Sandmeyer we achieved better
results (>99% ee, yield >50%) than for those compounds using the Stolle method (95% ee with a yield of 16%).
EXPERIMENTAL
1
IR spectra were recorded on a Thermal Nicolet IR200 FT instrument and H and 13C NMR spectra on a
Bruker AMX-400 (400 ad 100 MHz respectively) using CDCl3 with TMS as internal standard. Chromato-mass
spectrometry was carried out using a Carlo Erba/Kratos Fractovap Series 4200 gas liquid chromatograph with a
Hewlett Packard Ultra-1 column (25 m×0.2 mm, phase layer thickness 0.33 m, helium gas carrier at 1 ml/min,
flow split 1:10, evaporator temperature 280ºC, and temperature gradient from 150 to 280ºC at 5ºC/min). IDT-700
mass spectrometric detector (Finnigan MAT), EI mode, ionization energy 70 eV, mass range m/z 45-400. The
1212

specific rotation was measured on a Jasco DIP-360 (589 nm) polarimeter. Monitoring of the reaction course and
the purity of the separated compounds were carried out by TLC on Silufol UV-254 plates. The determination of
enantiomeric purity was carried out by HPLC using a Chiralpak AD-RH stationary phase (4.6×150 mm, 5 m,
eluent H2O + CF3COOH/MeCN (50:50), 1.0 ml/min flow rate, ~ 20ºC, 250 nm).
Stolle Synthesis of the Isatins (Table 1).
Experiment No. 2. A solution of ethyl 3-phenylaminopropionate (0.5 g, 2.59 mmol) in dichloromethane
(3 ml) was added dropwise with stirring to a cooled (0ºC) solution of oxalyl chloride (0.35 ml, 4.19 mmol) in
dichloromethane (1 ml). The reaction mixture was stirred at room temperature for 2 h and the solvent was
evaporated in vacuo. The residue obtained was dissolved in dichloromethane (3 ml) and treated with TiCl4
(1 molar dichloromethane solution, 1.13 ml, 0.01 mol). The reaction mixture was stirred at room temperature for
24 h, poured into ice, extracted with dichloromethane (3×5 ml), dried over Na2SO4, and the solvent was
evaporated in vacuo. The residue was flash chromatographed using dichloromethane.
Experiment No. 3. Similar to experiment No. 2 but using 6 equivalents of TiCl4 in place of 4 equivalents.
Experiment No. 4. Similar to experiment No. 2 but with stirring of the reaction mixture for 144 h
(6 days).
Experiment No. 5. Similar to experiment No. 2 but with stirring of the reaction mixture for 6 h at
30-35ºC.
Ethyl 2-(Isatin-1-yl)propionate (2). Yield 14% (experiment No. 2), 32% (No. 3), 55% (No. 4), 58%
(No. 5); mp 69ºC (a mixture of petroleum ether and ethyl acetate). IR spectrum, , cm-1: 3467, 2993, 1739
1
(CO), 1608, 1468, 1367, 1309, 1246, 1113, 750, 476. H NMR spectrum, , ppm (J, Hz): 1.22 (3H, t, J = 7.2,
CH2CH3); 1.69 (3H, d, J = 7.5, CHCH3); 4.18-4.27 (2H, m, CH2CH3); 5.16 (1H, q, J = 7.5, CHCH3); 6.85 (1H,
br. d, J = 7.7, CH Ar); 7.15 (1H, ddd, J = 7.7, J = 7.7, J = 0.7, CH Ar); 7.57 (1H, ddd, J = 7.7, J = 7.7, J = 1.4,
CH Ar); 7.65 (1H, ddd, J = 7.7, J = 1.4, J = 0.7, CH Ar). 13C NMR spectrum, , ppm: 14.08 (CH3); 14.27 (CH3);
49.18 (CH); 62.16 (CH2); 111.46 (CH); 117.89 (C); 123.89 (CH), 125.64 (CH); 138.19 (CH); 149.45 (C);
157.69 (CO); 169.43 (CO); 182.70 (CO). Mass spectrum, m/z (Irel, %): 247 [M]+ (11), 174 [M+ - CO2Et] (8), 146
[M+ - MeCHCO2Et] (100), 128 (0.8), 117 (6), 91 (12), 77 (26), 51 (9). Found, %: C 63.20; H 5.43; N 5.81.
C13H13NO4. Calculated, %: C 63.15; H 5.30; N 5.66.
(R)-Ethyl 2-(Isatin-1-yl)propionate ((R)-2) (see Table 2).
Gassman Synthesis of Isatins (Table 3).
Experiment No. 1. Sulfuryl chloride (0.22 ml, 2.74 mmol) was added dropwise with stirring to a cooled
(-78ºC) solution of methyl methylthioacetate (0.35 ml, 2.74 mmol) in dichloromethane (26 ml) and stirred for 20
min. A solution of ethyl N-(4-phenylamino)propionate (0.5 g, 2.59 mmol) and 1,8-bis(dimethyl-
amino)naphthalene (0.58 g, 2.69 mmol) in dichloromethane (16 ml) was added, stirred for 3.5 h at -78ºC,
triethylamine (0.38 ml, 2.69 mmol) was added, and the product was stirred at -78ºC for a further 15 min and
then for 20 h at room temperature. Acetic acid (2.6 ml) was added and the product was stirred for 1 h at room
temperature. The reaction mixture was washed with water, saturated NaCl solution, dried over Na2SO4, and the
solvent was removed in vacuo. The residue was chromatographed on a wet silica gel column eluting with a
mixture of petroleum ether and ethyl acetate (10:1).
Experiment No. 2. Similar to experiment No.1 using 1.5 and 1.47 equivalents of SO2Cl2,
(MeSCH2CO2Et), and Et3N respectively instead of 1.06 and 1.04 equivalents.
Experiment No. 3. Similar to experiment No. 1 with the reaction mixture stirred for 5.5 h at -78ºC.
Ethyl 2-[2-(Ethoxycarbonylmethylsulfanylmethyl)phenylamino]propionate (3). Yield 27% (experiment
1
No. 1), 18% (No. 2), 13% (method B). H NMR spectrum, , ppm (J, Hz): 1.27 (6H, t, J = 7.1, 2CH2CH3); 1.48
(3H, d, J = 6.7, CHCH3); 2.07 (3H, s, SCH3); 4.08-4.30 (5H, m, 2CH2CH3, CHCH3); 4.42 (1H, s, CHSCH3); 6.58
(2H, d, J = 8.2, CH Ar); 7.28 (2H, d, J = 8.2, CH Ar). 13C NMR spectrum, , ppm: 14.13 (CH3); 14.18 (CH3); 14.86
(CH3); 14.86 (CH3); 18.90 (CH3); 51.88 (CH); 53.07 (CH); 61.20 (CH2); 61.43 (CH2); 113.25 (2C
1213

CH); 124.91 (C); 129.53 (2C, CH); 146.42 (C); 171.08 (CO); 174.41 (CO). Mass spectrum, m/z (Irel, %): 325
[M]+ (9), 278 (100), 252 (42), 224 (6), 206 (19), 178 (25), 164 (32), 133 (55), 120 (16), 89 (22), 45 (17). Found,
%: C 59.01; H 7.30; N 4.10. C16H23NO4S. Calculated, %: C 59.05; H 7.12; N 4.30.
(R)-Ethyl 2-[2-(Ethoxycarbonylmethylsulfanylmethyl)phenylamino]propionate (3). []D20 = +35 (c = 1,
1
dichloromethane), yield 30% (experiment No. 1). The H NMR and 13C NMR spectra are identical to those of
the racemic sample.
REFERENCES
1.
2.
3.
4.
5.
G. I. Zhungietu and M. A. Rekhter, Isatins and its Derivatives [in Russian], Shtiintsa, Kishinev (1977).
V. Farinam, J. T. Reeves, C. H. Senanayake, and J. J. Song, Chem. Rev., 106, 2734 (2006).
J. F. Silva, S. J. Garden, and A. C. Pinto, J. Braz. Chem. Soc., 12, 273 (2001).
A. V. Kurkin, A. A. Bernovskaya, and M. A. Yurovskaya, Tetrahedron Asymmetry, 20, 1500 (2009).
W. J. Welstead, Jr., H. W. Moran, H. F. Stauffer, L. B. Turnbull, and L. F. Sancilio, J. Med. Chem., 22,
1074 (1979).
6.
7.
8.
G. Loloiu and O. Maior, Rev. Roum. Chim., 42, 67 (1997).
Y. Fukuda, Y. Itoh, K. Nakatani, and S. Terashima, Tetrahedron, 50, 2793 (1994).
P. G. Gassman, J. Cue, and T. Y. Luh, J. Org. Chem., 42, 1344 (1977).
9.
10.
11.
P. G. Gassman and T. J. van Bergen, J. Am. Chem. Soc., 96, 5508 (1974).
P. G. Gassman, G. Gruetzmacher, and T. J. van Bergen, J. Am. Chem. Soc., 96, 5512 (1974).
S. W. Wright, L. D. McClure, and D. L. Hageman, Tetrahedron Lett., 37, 4631 (1996).
1214