Catalysis in water: Synthesis of β-amino amides by Sc(III) promoted condensation of silylketene pyridylthioacetal and imines

Article abstract of DOI:10.1016/j.jorganchem.2007.10.013

The scandium (III) catalysed condensation of silylketene thioacetals and pyridylthioacetals with imines in pure water was studied. The reaction of the phenylthioacetal derivatives brings always to β-aminoesters; the reaction of pyridylthioacetals leads to

Full text of DOI:10.1016/j.jorganchem.2007.10.013

Available online at www.sciencedirect.com
Journal of Organometallic Chemistry 692 (2007) 5795–5798
www.elsevier.com/locate/jorganchem
Note
Catalysis in water: Synthesis of b-amino amides by Sc(III)
promoted condensation of silylketene pyridylthioacetal and imines
Cinzia Biaggi, Maurizio Benaglia , Alessandra Puglisi *
*
Dipartimento di Chimica Organica e Industriale, Centro di Eccellenza CISI, Universit a` degli Studi di Milano, Via Golgi 19, 20133 Milano, Italy
Received 19 September 2007; accepted 9 October 2007
Available online 16 October 2007
Abstract
The scandium (III) catalysed condensation of silylketene thioacetals and pyridylthioacetals with imines in pure water was studied. The
reaction of the phenylthioacetal derivatives brings always to b-aminoesters; the reaction of pyridylthioacetals leads to the stereoselective
formation of b-lactams in organic solvents, while in the presence of a large amount of water affords b-amino amides in good yields. That
represents a new, easy, one-pot catalytic synthesis in water of an interesting class of compounds, direct precursors of 1,3-diamino
hydroxy compounds. The possibility of recovering and recycling the catalytic systems was also briefly investigated.
ꢀ
2007 Elsevier B.V. All rights reserved.
Keywords: Scandium triflate; Catalysis; Reaction in water; Synthetic method
1
. Introduction
Therefore, the identification of catalytic procedures that
can promote stereoselective reactions in pure water is a true
challenge [5]. We wish to report here the results of our stud-
ies in the scandium (III) catalysed reaction of silyl ketene
phenylthioacetals and silyl ketene pyridylthioacetals with
imines in water, that was shown to lead to different com-
pounds depending on the substrate and the reaction med-
ium employed. The study is a further demonstration how
water may play a decisive role in determining the output
of stereoselective organic reactions.
The development of new catalytic systems able to effi-
ciently promote stereoselective reactions in water repre-
sents a fundamental research activity towards greener
synthetic methodologies, alternative to those employing
the traditional organic solvents [1]. However, despite the
several advantages offered by the use of water as a solvent
in terms of cost, safety and environmental impact, only a
relatively limited number of organic reactions can be effec-
tively carried out in this solvent [2]. One of the major rea-
sons responsible for this situation is the solubility problem
that has been tackled in different ways, including the use of
organic cosolvents [3]. It must be noted, however, that a
reaction medium with THF or ethanol as predominant
component mixed to water, or a system where a large
excess of organic reagents is employed in water loses most
of its appeal and can be better described as a wet organic
system rather than a real aqueous medium [4].
2. Results and discussion
The reaction between silylketene thioacetal (SKTA) (1),
readily obtained as a single (E)-isomer from 2-pyridylthio
propanoate [6], and the N-4-methoxyphenyl imine of ethyl
glyoxylate 2 carried out in the presence of Sc(OTf) [7] was
3
first studied (Scheme 1). The reaction was run for 18 h at
room temperature in different solvent systems in the pres-
ence of 10% mol cat of scandium (III) species.
The condensation between 2.0 mol equiv. of 1 with
imine 2 run in DCM afforded a mixture of trans b-lactam
*
Corresponding authors. Tel.: +39 02 50314168; fax: +39 02 50314159
3
t and of its cis diastereoisomer 3c, in a 50:50 ratio and
(
A. Puglisi).
1
E-mail address: alessandra.puglisi@unimi.it (A. Puglisi).
80% yield as determined by 300-MHz H NMR analysis
0
022-328X/$ - see front matter ꢀ 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2007.10.013

5796
C. Biaggi et al. / Journal of Organometallic Chemistry 692 (2007) 5795–5798
O
COOEt
EtOOC
Sc(OTf) 10 mol%
3
OTBS
SPy
1
PMP
PyS
N
Me
+
N
Me H
PMP
2
water
PMP
H H
N
H H
O
COOEt
PMP
Me
COOEt Me
+
PMP
COOEt
PMP
N
N
N
H
H
Py = 2-pyridyl
O
O
Me
TBS = t-BuMe Si
3c
3t
4
2
PMP = 4-MeOPh
Scheme 1. Sc(III)-catalysed condensation of 1 and 2.
Table 1
the imine 2. The second step in the synthesis of b-lactams
is the nucleophilic attack of the nitrogen on the pyridyl-
thio-group. In water, however, a competition might occur
between the nucleophilic intramolecular attack of the sec-
ondary amine to afford the b-lactam 3 and the nucleophilic
attack of p-methoxyaniline, derived from the hydrolysis of
imine 2, which would give the amide 4.
Sc(III)-catalysed condensation of 1 and 2
a
Entry
1:2
Solvent Catalyst Reaction 4 yield 3 + 4 yield
b
b
(equiv.)
time (h)
(%)
(%)
1
2
3
4
5
6
7
8
2.0:1.0
2.0:1.0
2.0:1.0
2.0:1.0
1.0:2.0
2.0:1.0
2.0:1.0
1.0:2.0
1.0:2.0
DCM
THF
Neat
Water
Water
Water
Water
Water
Water
Sc(OTf)
Sc(OTf)
Sc(OTf)
Sc(OTf)
Sc(OTf)
ScTDS
ScTDS
ScTDS
ScTDS
3
3
3
3
3
20
20
20
20
72
20
72
40
48
–
–
–
20
67
23
69
45
50
80
61
71
33
84
33
83
74
77
To test this hypothesis, 2.0 mol equiv. of 2 were reacted
with 1 for 72 h at room temperature in water and in this
case amide 4 was obtained with 67% yield [12] (entry 5).
In an attempt to increase the yield of the amide forma-
tion, scandium tris(dodecylsulfate) (ScTDS) was used as a
catalyst. ScTDS was first used by Prof. Kobayashi’s group
as a catalyst in which the Lewis acidic properties were com-
bined with surfactant properties [13]. The catalyst has
shown to be specially effective in water due to the presence
of long lipophilic chains that form hydrophobic reaction
fields, like micelles, in which there is a high concentration
of the active scandium and the reaction takes place. Unfor-
tunately in our hands the use of ScTDS did not give any
improvement. The reaction between 2.0 mol equiv. of 1
and 2 carried out in the presence of 10% mol ScTDS for
20 h at room temperature afforded the amide 4 in 23% yield
(entry 6). The reaction between 2.0 mol equiv. of 1 and 2
carried out in the presence of 10% mol ScTDS afforded
the amide 4 in 45% yield after 40 h and 69% yield after
72 h (entries 7 and 8). These values are comparable with
c
9
a
General: all reactions carried out using 10 mol% Sc(III).
Isolated yields after flash chromatography.
b
c
2
mol% Sc(III) was used.
of the crude reaction mixture and confirmed after chro-
matographic purification [8] (Table 1, entry 1).
When THF was used as a solvent after 18 h complete
conversion of the imine was observed, and a mixture of
trans b-lactam 3t and of its cis diastereoisomer 3c, in a
5
0:50 ratio and 61% isolated yield was gained (Table 1,
entry 2).
When 1 was reacted with imine 2 for 20 h at room tem-
perature in the absence of solvent a 50:50 mixture of trans-
and cis-b-lactams 3t and 3c, in 71% combined yield was
obtained (entry 3). In addition to the products, the spec-
trum of the crude reaction mixture showed the presence
of unreacted SKTA 1 and of the b-aminothioesters precur-
sors of the azetidinones. No traces of the imine could be
detected indicating that the reaction proceeded in high con-
version [9].
However, the condensation of pyridylthioacetal 1 with 2
in water [10] lead to a different compoud as major reaction
product, that was identified to be b-amino-amide 4 [11],
obtained as a 65/35 mixture of isomers in 20% yield (entry
those obtained using Sc(OTf) as a catalyst [14].
3
To further increase the efficiency of the method the cat-
alyst loading was lowered down to 2 mol% of ScTDS; also
in this case the reaction between 1 and 2.0 mol equiv. of 2
worked well and afforded amide 4 in 50% yield along with
26% of b-lactams mixture after 48 h reaction time (entry 9).
The unexpected outcome of the reaction might be deter-
mined by the combination of the ability of water to interact
with the substrates together with a good leaving group as –
SPy on the starting material. To prove this hypothesis the
reaction between imine 2 and the thioacetal 5, which does
not bear a good leaving group (SPh) was run in the pres-
4
). trans- and cis-b-lactams 3t and 3c were obtained as side-
products in 13% yield and 50:50 ratio.
We believe that the formation of the amide 4 arises from
the peculiar reaction conditions used in the condensation
between 1 and 2 in water. The first step is the formation
of the b-amino pyridilthioester from the thioacetal 1 and
ence of 10% mol Sc(OTf) using water as a solvent. The
3
only isolated product was the b-amino thioester 6 in 47%
yield, confirming that both water as a solvent and the 2-
pyridylthio group on the starting material are necessary

C. Biaggi et al. / Journal of Organometallic Chemistry 692 (2007) 5795–5798
5797
O
COOEt
EtOOC
OTBS
SPh
Sc(OTf) 10 mol%
3
PMP
Me
Me
+
PhS
N
H
N
water
PMP
Me
5
2
6
Scheme 2. Synthesis of b-aminothioester 6.
Ph
O
Ph
OTBS
SPy
1
Sc(III) 10 mol%
solvent
PMP
+
N
PyS
N
PMP
H
Me
7
water
H H
H H
Ph
O
Ph
Me
Ph Me
+
PMP
PMP
N
N
N
N
H
Me H
O
PMP O
PMP
9
Py = 2-pyridyl
8c
8t
TBS = t-BuMe Si
2
PMP = 4-MeOPh
Scheme 3. Sc(III)-catalysed condensation of 1 and 7.
for the formation of amino-amide to occur [15] (see Scheme
).
The scope of the reaction was extended to a unactivated
imine, derived from benzaldehyde and 4-methoxy aniline
2, ScTDS proved to be a better catalyst than Sc(OTf) to
3
2
promote the condensation with pyridylthioacetal 1.
Recovery and recycling of the catalyst was also studied.
First, the recyling of the aqueous phase containing ScTDS
after separation of the products with an organic solvent
was attempted [18]. For example, after the reaction of thi-
oacetal 1 with imine 7, the crude mixture was extracted
with an organic solvent and purified by flash chromatogra-
phy to give amide 9 in 42% isolated yield. To the aqueous
phase new fresh reagents were added and the reaction was
run again but no product formation was observed after
72 h reaction time. Probably due to its lipophilic character,
ScTDS is soluble in most of the organic solvents and is
extracted from the aqueous medium along with the prod-
ucts in the organic phase [19].
It was, then, decided to formally ‘‘recycle’’ the catalyst
in situ by adding fresh reagents to the reaction vessel with-
out isolating the products to mimic a continuous flow pro-
cess [20]. Thus, at the end of a first reaction of SKTA 1
with imine 7 in the presence of 10 mol% ScTDS, the same
amounts of reagents were added and a second reaction
allowed to proceed for 48 h. After three further iterations
of this procedure (for a total five reaction cycles) amide 9
was isolated in 31% yield along with 10% b-lactams 8c
and 8t [21].
(
Scheme 3).
The condensation between 2.0 mol equiv. of pyr-
idylthioacetal 1 and the imine 7, promoted by 10 mol%
of Sc(OTf) gave no reaction after 48 h (Table 2, entry
3
1
). When ScTDS was used as a catalyst 26% of amide 9
[
16] in a 70: 30 diastereoisomeric ratio along with 21% of
a mixture of cis and trans b-lactams 8c and 8t in a 50:50
ratio was obtained after only 18 h reaction time (entry 2).
Using 2.0 mol equiv. of 7 gave better results: the condensa-
tion promoted by Sc(OTf) afforded amide 9 although in
3
low yield (13% after 72 h, entry 3), while the reaction pro-
moted by ScTDS gave amide 9 in 53% yield after 60 h reac-
tion time (entry 4). It was possible to run the reaction for a
shorter reaction time: ScTDS promoted the formation of
amide 9 in 42% yield after 32 h (entry 5). Finally, using
2
mol% of ScTDS amide 9 was obtained in 13% yield along
with b-lactams 8c + 8t in 9% yield after 32 h reaction time
[
17] (entry 6).
As shown by data collected in Table 2, in the case of
imine 7, a non-activated and less water-soluble imine than
Finally, the b-amino amide easily prepared by this meth-
odology represents a useful starting point for further syn-
thetic transformations. For example, amide 4 may be
reduced to the corresponding 1,3-diamino-alcohol 10
Table 2
Sc(III)-catalysed condensation of 1 and 7
a
Entry
1:2
Catalyst
Reaction
time (h)
9 yield
(%)
8 + 9 yield
(%)
b
b
(equiv.)
(
Scheme 4) [22,23].
1
2
3
4
5
2.0:1.0
2.0:1.0
1.0:2.0
1.0:2.0
1.0:2.0
1.0:2.0
Sc(OTf)
ScTDS
Sc(OTf)
ScTDS
ScTDS
ScTDS
3
48
18
72
60
32
32
–
–
26
13
53
42
13
47
17
65
51
23
3
OH
PMP
H
Me
O
COOEt
LiAlH4
PMP
PMP
PMP
N
H
N
Me H
c
N
N
THF, rt, 3h
6
H
a
General: all reactions carried out using 10 mol% Sc(III) in water.
Isolated yields after flash chromatography.
mol% Sc(III) was used.
4
10
b
c
2
Scheme 4. Functionalisation of amide 4.

5
798
C. Biaggi et al. / Journal of Organometallic Chemistry 692 (2007) 5795–5798
[10] General procedure for the synthesis of amides: In a 10-mL round-
The reaction of 4 with 2.0 mol equiv. LiAlH in THF at
4
bottom flask Sc(III) (0.03 mmol), water (2 mL), imine (0.6 mmol) and
SKTA (0.3 mmol) were added in this order and the resulting mixture
was allowed to stir at room temperature for 72 h. The crude mixture
room temperature for 3 h, afforded the expected product 10
in a non-optimised 30% yield. Compound 10 represents a
highly functionalised building block, easily prepared in a
only 2 step-procedure.
was extracted with dichloromethane (3 · 3 mL), dried over Na
filtered and concentrated. The resulting oil was then charged on top of
column and purified by flash chromatography with 8:2
2 4
SO ,
a
a
hexane:AcOEt mixture followed by 7:3 hexane:AcOEt mixture as
eluant.
3
. Conclusions
[
11] Amide 4: IR (DCM): m 3322, 2934, 1732, 1662, 1512, 1244, 1181,
In conclusion, a novel synthesis of b-amino-amides was
À1
1
1
034 cm
Me of ethyl group and Me-C3); 3.00 (m, 1H, H-C3); 3.79, 3.81 (2 s,
3H each, OMe); 4.20 (q, 2H, J 4.0 Hz CH of ethyl group); 4.30 (d,
H, J 2.6 Hz, H-C4); 6.80–6.88 (m, 6H, aromatic protons); 7.39 (m,
. H NMR (CDCl ) (major isomer): d 1.27–1.40 (m, 6H,
3
developed by employing a Sc(III) catalysed condensation
between pyridylthioacetal and imines in water as a reaction
solvent. Amides were obtained in fair to good yields (20–
2
1
2
5
1
1
3
H, aromatic protons). C NMR (CDCl
5.6, 61.3, 61.5, 114.0, 114.5, 117.6, 121.6, 131.0, 139.6, 154.3, 156.3,
71.4. Elem. Anal. Calc. for C21 : C, 65.27; H, 6.78; N, 7.25.
3
): d 13.4, 14.1, 43.7, 55.4,
6
7%) along with mixture of diastereoisomeric b-lactams.
Experiments were run in order to elucidate the reaction
mechanism and it was concluded that water plays a decisive
role in the amide formation, in association with the ability
of the pyridylthio group to act as a good leaving group.
Preliminary experiments to study further manipulations
of the products and recycle of the catalyst were also
performed.
26 2 5
H N O
Found: C, 65.31; H, 6.76; N, 7.23%.
[12] The reaction afforded always product 4 with the same diastereiso-
meric ratio (65/35) that was shown to be independent from the
experimental conditions employed.
[
[
13] S. Kobayashi, T. Wakabayashi, Tetrahedron Lett. 39 (1998) 5389.
14] For recent examples of reactions catalysed by ScTDS, see: (a) S.
Azoulay, K. Manabe, S. Kobayashi, Org. Lett. 7 (2005) 4593;
(
(
b) M. Boudou, C. Ogawa, S. Kobayashi, Adv. Synth. Catal. 348
2006) 2585.
Acknowledgements
[
15] The use of 1 or 2 mol/eq of free 4-methoxy aniline in the condensation
between silyl ketene acetal 1 (2 mol/eq) and imine 2 (1 mol/eq) did not
improve the reaction yield and did not alter the diastereoisomeric
ratio of product 4.
This work was supported by MIUR (Rome) within the
national project ‘‘Nuovi metodi catalitici stereoselettivi e
sintesi stereoselettiva di molecole funzionali’’. We thank
Prof. Cozzi for valuable discussion.
[16] Amide 9 m.p. = 133–136 ꢁC; IR (DCM): m 3054, 2986, 1676, 1510,
À1
1
1
1
265, 740 cm . H NMR (CDCl
3
) (major isomer): d 1.38 (d, 3H, J
0.3 Hz Me-C3); 2.85 (m, 1H, H-C3); 3.80 (s, 3H, OMe); 4.50 (d, 1H,
J 7.7 Hz, H-C4); 6.70–6.85 (m, 4H, aromatic protons); 7.11–7.27 (m,
References
13
4
H, aromatic protons); 7.32–7.37 (m, 5H, aromatic protons).
C
NMR (CDCl ): d 16.1, 48.6, 55.4, 55.6, 62.1, 114.0, 114.5, 114.8,
3
[
[
1] P.T. Anastas, J.C. Warner, Green Chemistry: Theory and Practice,
Oxford University Press, New York, 1998, p. 30.
2] (a) P.A. Grieco (Ed.), Organic Synthesis in Water, Blackie Academic
122.5, 126.6, 127.4, 127.5, 128.8, 130.0, 141.2, 142.3, 151.8, 156.7,
172.5. Elem. Anal. Calc. for C24H26N O : C, 73.82; H, 6.71; N, 7.17.
Found: C, 73.79; H, 6.72; N, 7.19%.
2
3
&
(
Professional, London, 1998;
b) S. Ribe, P. Wipf, Chem. Commun. (2001) 299.
3] (a) U.M. Lindstr o¨ m, Chem. Rev. 102 (2002) 2751;
b) C.-J. Li, Chem. Rev. 105 (2005) 3095.
4] For a very interesting discussion of enantioselective organocatalysis
‘in water’’ or ‘‘in the presence of water’’, see: (a) A.P. Brogan, T.J.
Dickerson, K.D. Janda, Angew. Chem., Int. Ed. 45 (2006) 8100;
b) Y. Hayashi, Angew. Chem., Int. Ed. 45 (2006) 8103.
[17] The reaction afforded always product 9 with the same diastereiso-
meric ratio (70/30) that was shown to be independent from the
experimental conditions employed.
[18] Recovery and recycling of scandium tris(dodecyl sulfate) have not
been clearly accomplished: K. Manabe, Y. Mori, T. Wakabayashi, S.
Nagayama, S. Kobayashi, J. Am. Chem. Soc. 122 (2000) 7202;
For an example of recycle of supported scandium catalyst, see: S.
Nagayama, S. Kobayashi, Angew. Chem., Int. Ed. 39 (2000) 567.
[19] In the present experimental conditions in our hands, ScTDS was
almost quantitatevely extracted from water with many organic
[
[
(
‘
(
[
5] For a recent contribution from our group, see: S. Guizzetti, M.
Benaglia, L. Raimondi, G. Celentano, Org. Lett. 9 (2007) 1247;
See also: M. Benaglia, M. Cinquini, F. Cozzi, G. Celentano, Org.
Biomol. Chem. 2 (2004) 3401.
solvents such as THF, Et O, DCM, AcOEt, or hexanes/AcOEt
2
mixtures.
[
6] For the synthesis of these silylketene 2-pyridylthioacetals and their
use in a non-catalytic synthesis of b-lactams, see: (a) R. Annunziata,
M. Cinquini, F. Cozzi, V. Molteni, O. Schupp, Tetrahedron 52
[20] Review: (a) G. Jas, A. Kirschning, Chem. Eur. J. 9 (2003) 5708;
For examples of organocatalysed reactions carried out under the
continuous flow mode, see: (b) K. Ishiara, A. Hasegawa, H.
Yamamoto, Synlett (2002) 1296;
(
(
1996) 2573;
b) R. Annunziata, M. Cinquini, F. Cozzi, V. Molteni, O. Schupp, J.
(c) M. Br u¨ njes, G. Sourkouni-Argirusi, A. Kirschning, Adv. Synth.
Catal. 345 (2003) 635.
Org. Chem. 61 (1996) 8293;
c) The one-pot formation of the azetidinone ring takes advantage of
(
[21] Procedure for catalyst recycling: SKTA 1 was reacted with imine 7 as
described in note [10]. After 48 h 0.085 g of SKTA 1 and 0.124 g of
imine 7 were added to the aqueous phase and a second reaction
allowed to proceed for 48 h. After three further iterations of this
procedure (for a total five reaction cycles) amide 9 was isolated after
silica gel chromatography in 31% yield along with 10% b-lactams 8c
and 8t as described above.
[22] R. Annunziata, M. Benaglia, M. Caporale, L. Raimondi, Tetrahe-
dron: Asymmetry 24 (2002) 2727.
[23] R. Annunziata, M. Benaglia, M. Cinquini, F. Cozzi, L. Raimondi,
Tetrahedron Lett. 39 (1998) 3333.
the excellent leaving group nature of the 2-pyridylthio residue.
7] Reviews: (a) S. Kobayashi, Synlett (1994) 689, and references cited
therein;
[
(b) S. Kobayashi, H. Ishitani, Chem. Rev. 99 (1999) 1069, and
references cited therein.
8] trans:cis Configurations were assigned on the basis of the H-C3/H-C4
coupling constant values (Jtrans = ca. 2.0–2.7 Hz; Jcis = ca. 5.0–
[
[
6.0 Hz).
9] For a solvent-free, one-pot synthesis of b-lactams, see: A. Puglisi, M.
Benaglia, F. Cozzi, Eur. J. Org. Chem. (2007) 2865.