Biginelli reactions catalyzed by hydrazine type organocatalyst

Article abstract of DOI:10.1016/j.tetlet.2008.03.080

Pyrazolidine dihydrochloride can be used in the acceleration of Biginelli reactions between urea, ethyl acetoacetate and various aldehydes to provide DHPMs in good to excellent yields.

Full text of DOI:10.1016/j.tetlet.2008.03.080

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 3238–3241
Biginelli reactions catalyzed by hydrazine type organocatalyst
*
Ichiro Suzuki , Yukari Iwata, Kei Takeda
Department of Synthetic Organic Chemistry, Graduate School of Biomedical Sciences, Hirosahima University,
Kasumi 1-2-3, Minami-ku, Hiroshima 734-8553, Japan
Received 27 February 2008; revised 15 March 2008; accepted 18 March 2008
Available online 20 March 2008
Abstract
Pyrazolidine dihydrochloride can be used in the acceleration of Biginelli reactions between urea, ethyl acetoacetate and various
aldehydes to provide DHPMs in good to excellent yields.
Ó 2008 Elsevier Ltd. All rights reserved.
2
Biginelli-type condensation reaction, which is one of the
multicomponent reactions, is a very useful reaction for pre-
paring 3,4-dihydropyrimidin-2(1H)-ones. Traditionally,
Biginelli reactions were carried out under strongly acidic
conditions with heating, and the yields were only low to
peptide Y (NPY) antagonists. Thus, in the past ten years,
Biginelli reactions have attracted much attention, and
many improved methods, including enantioselective ver-
sions, have been exploited. In these advanced methods,
Lewis or Brønsted acid was mainly used as a catalyst under
milder conditions with much better results compared to the
1
moderate (see Scheme 1).
3
Recently, several 3,4-dihydropyrimidin-2(1H)-ones
results obtained by employing traditional conditions.
(
DHPMs) have been revealed to have interesting biological
Recently, secondary amine-based organocatalysts that pro-
mote the reactions via enamine formation have been exten-
sively studied; however, for Biginelli reactions, only a few
and pharmaceutical properties, including antiviral, antitu-
mour, antibacterial, antiinflammatory and antihyperten-
sive properties, and these compounds have emerged as
integral backbones of several calcium channel blockers,
antihypertensives, a -adrenergic antagonists and neuro-
4
examples using (S)-proline or chiral secondary amines as
5
organocatalysts have been reported. In these examples,
refluxing in EtOH and a neat condition were employed,
but their efficiencies were not sufficiently high and only
racemic products were obtained. In the course of our study
to develop efficient nucleophilic chiral organocatalysts, we
1
a
O
O
acid
R
H3C
O
OEt
catalyst
EtO2C
NH
became interested in hydrazines because hydrazines are
+
6a–c
more nucleophilic due to the a-heteroatom effect
and
H C
N
H
O
RCHO
3
H2N
NH2
NH2
6d
their salts with acids are more acidic than related amines.
Some hydrazides have been reported to catalyze Diels-
Alder reactions and 1,3-dipolar cycloaddition reactions
R
7
O
H
efficiently, but the number of examples is not so large.
OH
O
EtO C
H
N
2
NH
In this Letter, we report that pyrazolidine dihydrochloride
1b is a more effective organocatalyst than the widely used
modified pyrolidine derivatives for Biginelli reactions.
First, we prepared hydrazines 1a–c as shown in Scheme
+
H3C
OEt
H3C
N
H
O
R
HO
Scheme 1. Biginelli reaction.
2
and screened them for their catalytic abilities in Biginelli
8
*
reactions. Hydrazine salts 1a–c were prepared as dihydro-
Corresponding author. Tel.: +81 082 257 5321; fax: +81 082 257 5184.
E-mail address: isuzuki@hiroshima-u.ac.jp (I. Suzuki).
chlorides. We also tried to prepare 1,2-dibenzyhydrazine,
0
040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.03.080

I. Suzuki et al. / Tetrahedron Letters 49 (2008) 3238–3241
3239
Bn
difference. In general, basicity and nucleophilicity of typical
amines can be estimated roughly based on the electronic
effects of substituents and steric congestion around a
nitrogen atom. On the other hand, in hydrazines, lone-
pair—lone-pair repulsion leads to an electronic preference
for a nearly 90° dihedral angle between the lone-pair orbi-
tal axes, and the pyramidality at nitrogen changes with ring
(
i)
(iii)
BocHN NHBoc
BocN NBoc
Bn
BnHN NHBn 2HCl
1a
(ii)
n
n
(iii)
n=1: 1b;
HN NH
HCl
BocN NBoc
n=2: 1c
2
Scheme 2. Reagents and conditions: (i) NaH (2.4 equiv), BnBr (2.2 equiv)
in DMF, 0 °C to rt, 75%; (ii) NaH (2.4 equiv), BrCH (CH CH Br
1.2 equiv), 0 °C to rt, 86% for n = 1 and 80% for n = 2; (iii) HCl (5 equiv)
in 1,4-dioxane, rt, 94% for 1a, 90% for 1b and 95% for 1c.
)
n
2
10
2
2
size and with NN twist angle. Due to these electronic and
(
structural features, it is difficult to approximate the reactiv-
ity of catalysts 1a–c in a manner similar to the method used
for typical amines. Yadav et al. reported that (S)-proline
could catalyze Biginelli reactions under solvent-free condi-
pyrazolidine and 1,2-piperazine in acid-free forms; how-
ever, these acid-free hydrazines were unstable and suffered
from air-oxidation under neutral to basic conditions. The
5
a
tions at rt; however, we could not reproduce their results
and only a trace amount of DHPM 2a was obtained with
recovery of p-anisaldehyde and ethyl acetoacetate at rt
9
reactions were carried out using p-anisaldehyde, urea and
ethyl acetoacetate in the presence of 5 mol % of catalysts
(
entry 6). On the other hand, when (S)-proline-catalyzed
5
b
1
a–c, 10 mol % of dibenzylamine hydrochloride, (S)-pro-
reaction was carried out in MeOH, cyclic product 3
line and HCl in methanol at room temperature. The results
are summarized in Table 1.
and adduct 4 (E/Z mixture) were obtained in 7% and 5%
yields, respectively, with no DHPM 2a.
1
1
When dihydrochloride 1a was used as a catalyst, the
reaction occurred and DHPM 2a was obtained quantita-
tively after 24 h, whereas dibenzylamine hydrochloride
did not show any remarkable catalytic activity (entries 1
and 2). These results encouraged us to test the catalytic
activities of other hydrazine dihydrochlorides 1b and 1c,
and we found pyrazolidine dihydrochloride 1b showed
superior catalytic activity to that of 1a and 1c. The reaction
proceeded very smoothly in the presence of 5 mol % of
catalyst 1b to provide DHPM 2a quantitatively in 3.5 h,
whereas catalyst 1c showed only a low level of catalytic
activity (entries 4 and 5). Although we cannot clearly
explain this difference in catalytic activity at present, we
do not think that there is a simple interpretation for this
OCH3
EtO2C
H3C
HO
CO2Et
CH3
HN
NH
H3CO
O
O
3
4
In the case of using HCl, which is a typical catalyst for
Biginelli reaction, DHPM 2a was obtained in only 18%
yield at room temperature (entry 8). In Biginelli reaction,
a Brønsted acid is essential to facilitate N-acyliminium
ion formation, which is a rate-determining step. Catalyst
1
b can also work as a Brønsted acid catalyst, but its acidity
should be lower than that of HCl. Despite the lower acid-
ity, catalyst 1b showed a considerably higher level of cata-
lytic activity than did HCl. These results suggest that
catalyst 1b worked not only as an acid catalyst but also
as a nucleophilic catalyst to form a reactive enehydrazine
intermediate from ethyl acetoacetate. While (S)-proline
also affected the reaction in a manner similar to catalyst
1b, (S)-proline catalyzed the reactions less efficiently than
did catalyst 1b, because the nucleophilicity of catalyst 1b
is enhanced by the a-heteroatom effect. Next we screened
the reaction solvents for their efficiency by using three alde-
hydes, p-anisaldehyde, p-nitrobenzaldehyde and hexanal.
The results are shown in Table 2.
Table 1
Hydrazine-catalyzed Biginelli reaction using p-anisaldehyde, ethyl aceto-
a
acetate and urea
O
OMe
EtO2C
Me
1
a-1c
H2N
NH2
O
O
(5 mol%)
+
HN
NH
Me
OEt
p-anisaldehyde
2a
O
Entry
Catalyst
Solvent
MeOH
Time (h)
Yield (%)
b
c
1
2
3
4
5
6
7
8
Bn
1a
1a
1b
1c
2
NHꢀHCl
24
24
<5
For p-anisaldehyde, the reactions proceeded to give
excellent yields in MeOH, EtOH, i-PrOH, CH CN, DMSO
MeOH
MeOH
MeOH
MeOH
98
30
3
c
3.5
3.5
3.5
3.5
3.5
3.5
and THF; however, we encountered some difficulties when
p-nitrobenzaldehyde and hexanal were employed. When p-
nitrobenzaldehyde was used, the reaction was very slow
and provided DHPM 2b in only 49% yield with the recov-
ery of p-nitrobenzaldehyde even after 24 h in MeOH. For
other solvents, i-PrOH and DMSO gave somewhat better
98
<5
<5
c
b
d
c
(S)-Proline
(S)-Proline
None
b
MeOH
MeOH
0
18
b
c
HCl
a
The reactions were carried out using aldehydes (1.0 mmol), ethyl
acetoacetate (1.5 mmol) and urea (1.5 mmol) in MeOH (1 mL) with
–10 mol % of catalysts at rt.
results than did EtOH, CH CN and THF. On the other
3
5
hand, in the reactions of hexanal, only decomposed prod-
ucts were formed with no DHPM 2c in MeOH and EtOH,
though hexanal disappeared within 1 h. Although we could
not specify the decomposition pathway, an enehydrazine
b
1
0 mol % of catalysts were loaded.
p-Anisaldehyde was recovered and any other byproducts were not
formed.
c
d
Solvent-free conditions were employed; see Ref. 5a.

3240
I. Suzuki et al. / Tetrahedron Letters 49 (2008) 3238–3241
Table 2
in Table 3. Considering the relatively slow reaction rate in
i-PrOH, we used 10 mol % of catalyst 1b.
a
Solvent-effects on pyrazolidineꢀ2HCl-catalyzed Biginelli reactions
O
p-Anisaldehyde and p-nitrobenzaldehyde afforded
DHPM 2a and 2b in 97% and 90% yields, respectively
EtO2C
Me
1
b
H2N
NH2
R
O
O
(5 mol%)
solvent
(entries 1 and 2). For p-nitrobenzaldehyde, we also
+
HN
NH
Me
OEt
employed reflux conditions; however, the yield was lowered
RCHO
2a - 2c
O
to 23%, probably due to the decomposition of catalyst 1b
1
3
Solvent
R, Yield/% (reaction time)
(2a) p-NO (2b)
(
entry 2). When the reaction of hexanal was carried out
p-MeOC
6
H
4
2
C
6
H
4
n-C
5
H11 (2c)
in the presence of 10 mol % of catalyst 1b, interestingly,
only decomposed products were obtained with no DHPM
2c as in the reactions carried out in EtOH or MeOH,
though DHPM 2c was obtained in 79% yield in the pres-
ence of 5 mol % of catalyst 1b in i-PrOH (entry 3 in Tables
3 and 2). In the reaction of hexanal, the amount of catalyst
b
b
b
b
b
b
MeOH
EtOH
i-PrOH
98 (4 h)
93 (4 h)
78 (4 h) , 95 (24 h)
86 (4 h) , 92 (24 h)
61 (4 h) , 94 (24 h)
49 (24 h)
51 (24 h)
68 (24 h)
51 (24 h)
67 (24 h)
61 (24 h)
dec (1 h)
dec (1 h)
79 (24 h)
41 (24 h)
62 (24 h)
45 (18 h)
b
c
c
c
c
b
CH
3
CN
b
DMSO
THF
a
b
79 (4 h) , 89 (24 h)
1b seemed to be important and to affect the reaction course.
The reactions were carried out using aldehydes (1.0 mmol), ethyl ace-
On the other hand, cyclohexanecarbaldehyde and pivalal-
dehyde afforded DHPMs 2d and 2e in 93% and 95% yields,
respectively (entries 4 and 5). For cyclohexanecarbalde-
hyde, its steric hindrance would prevent enehydrazine
and/or acetal formation to facilitate the reactions. Other
aromatic aldehydes also gave DHPMs 2f–i in excellent
yields (entries 6–9). Unexpectedly, m-nitrobenzaldehyde
did not react to give DHPM 2j, and diisopropyl acetal of
m-nitrobenzaldehyde was formed in 39% yield. For m-
nitrobenzaldehyde, DMSO was better solvents to provide
DHPM 2j in 88% yield (entry 10). Although we cannot
clearly explain this solvent effect, low solubility of an
intermediary N-acyliminium ion and related N-acyl species
to i-PrOH may affect the reaction efficiency. o-Nitrobenz-
aldehyde gave DHPM 2k in 46% yield along with the
toacetate (1.5 mmol) and urea (1.5 mmol) with 5 mol % of catalyst 1b at rt.
b
An aldehyde was recovered.
Decomposed products were accompanied.
c
formation from hexanal and catalyst 1b may be involved in
1
2
the decomposition pathway. Among the solvents tested,
i-PrOH gave a better result, providing DHPM 2c in 79%
yield, than the results obtained using CH CN, THF and
3
DMSO. For an N-acyliminium ion formation step in alco-
holic solvents, an acetal and an a-alkoxy-N-acylamine can
also be formed, and these are in equilibrium as shown in
Scheme 3. p-Anisaldehyde is considered to form a relatively
stable N-acyliminium ion by resonance stabilization, and
consequently Biginelli reaction could proceed smoothly.
On the other hand, an N-acyliminium ions from p-nitro-
benzaldehyde and hexanal are unstable due to the lack of
resonance stabilization and are prone to suffer from nucleo-
philic attack by the solvent to give unreactive a-alkoxy-N-
acylamines. When i-PrOH was used as a solvent, its steric
hindrance should suppress the formation of the acetal
and/or the a-alkoxy-N-acylamine, leading to improvement
in the N-acyliminium ion formation step. While less nucleo-
Table 3
a
Biginelli reactions using 1b as a catalyst in i-PrOH and DMSO
X
EtO2C
Me
H2N
NH2
1b
R
O
O
(X = O, S)
(10 mol%)
+
HN
NH
2a-2n
Me
OEt
i
-PrOH
R
CHO
or
X
DMSO
philic solvents, such as CH CN, THF and DMSO, are
3
also more appropriate, in CH CN and THF, urea is hardly
Entry
R
X
Time (h)
Product
Yield (%)
3
soluble, and catalyst 1b could work less efficiently as an
acid catalyst in DMSO, resulting in slower reaction rates
than that in i-PrOH.
Based on the results of solvent screening, we then exam-
ined Biginelli reactions using various aldehydes in i-PrOH
1
2
3
4
5
6
7
8
9
p-MeOC
p-NO C H
4
6
H
4
O
O
O
O
O
O
O
O
O
O
O
O
O
S
8
18
2
8
8
4
4
3
5
10
24
24
24
48
2a
2b
2c
2d
2e
2f
97
90, 23
dec
93
95
92
95
93
97
0, 88
<46, 91
70, dec
62, dec
91
b
2
6
c
n-C
5
H
11
Cyclohexyl
tert-Butyl
p-ClC
p-BrC
m-MeOC
o-MeOC
m-NO
o-NO
CH@CH
6
H
4
(
for some cases in DMSO), and the results are summarized
6
H
4
2g
2h
2i
6
H
4
6
H
4
O
d
H+
10
11
2
C
6
H
4
2j
H
R
RCHO + Urea
DHPM
d
N
NH2
C
6
H
2k
2l
2
4
R'OH
d
1
1
2
3
C
6
H
5
H
d
2-furyl
p-MeOC H
6 4
2m
2n
H+
H+
14
a
The reactions were carried out using an aldehyde (1.0 mmol), ethyl
O
acetoacetate (1.5 mmol), urea (1.5 mmol) and 1b (10 mol %) in i-PrOH or
DMSO (1 mL) at rt.
b
Urea, H+
H
N
NH2
OR'
RCH(OR')2
The reaction was carried out at reflux for 24 h.
Hexanal was consumed completely in 2 h.
The reactions were carried out in DMSO at rt for 24 h.
R
c
0
d
Scheme 3. Equilibrium of an N-acyliminium ion in R OH.

I. Suzuki et al. / Tetrahedron Letters 49 (2008) 3238–3241
3241
considerable amounts of inseparable impurities, and in this
case also DMSO was an optimal solvent to give DHPM 2k
in 91% yield (entry 11). Biginelli reactions using cinnamal-
dehyde and acid-sensitive furfural were also catalyzed by
catalyst 1b, providing DHPMs 2l and 2m in 70% and
T. N. B. Synlett 2001, 863; (m) Kumar, K. A.; Kasthuraiah, M.;
Reddy, C. S.; Reddy, C. D. Tetrahedron Lett. 2001, 42, 7873; (n) Lu,
J.; Ma, H. Synlett 2000, 63; (o) Ma, Y.; Qian, C.; Wang, L.; Yang, M.
J. Org. Chem. 2000, 65, 3864; (p) Lu, J.; Bai, Y.; Wang, Z.; Yang, B.;
Ma, H. Tetrahedron Lett. 2000, 41, 9075; (q) Ranu, B. C.; Hajra, A.;
Jana, U. J. Org. Chem. 2000, 65, 6270; (r) Hu, E. H.; Sidler, D. R.;
Dolling, U.-H. J. Org. Chem. 1998, 63, 3454.
6
2% yields, respectively (entries 12 and 13). We examined
4
5
6
. (a) List, B. Chem. Commun. 2006, 819; (b) List, B. Acc. Chem. Res.
the same reactions in DMSO, but only decomposed prod-
ucts were obtained. It is noteworthy that relatively low-
reactive thiourea could be used as a component of Biginelli
reactions and 2n was obtained in 91% yield, whereas
prolonged reaction time (48 h) was needed (entry 14).
In conclusion, we demonstrated that pyrazolidine di-
hydrochloride 1b worked as a better catalyst for the
Biginelli reaction than do secondary amines to provide
DHPMs in good to excellent yields under mild conditions.
Development of new hydrazine-type organocatalysts for
enantioselective Biginelli reactions is now underway.
2
004, 37, 548; (c) Notz, W.; Tanaka, F.; Barbas, C. F., III. Acc. Chem.
Res. 2004, 37, 580.
. (a) Mabry, J.; Ganem, B. Tetrahedron Lett. 2006, 47, 55; (b) Yadav, J.
S.; Kumar, S. P.; Kondaji, G.; Rao, R. S.; Nagaiah, K. Chem. Lett.
2004, 33, 1168.
. (a) Grekov, A. P.; Veselov, V. Y. Russ. Chem. Rev. 1978, 16, 87; (b)
Fina, N. J.; Edwards, J. O. Int. J. Chem. Kinet. 1973, 5, 1; (c) Fleming,
I. In Frontier Orbitals and Organic Chemical Reactions; Wiley-
Interscience: Chichester, 1976; (d) Hinman, R. L. J. Org. Chem. 1958,
23, 1587.
7. (a) Lemay, M.; Ogilvie, W. W. J. Org. Chem. 2006, 71, 4663; (b)
Lemay, M.; Ogilvie, W. W. Org. Lett. 2005, 7, 4141; (c) Cavill, J. L.;
Peters, J.-U.; Tomkinson, N. C. O. Chem. Commun. 2003, 728.
8
. 1a–c are known compounds; see ref. for 1b and 1c: Ahn, J.-H.; Shin,
M.-S.; Jun, M.-A.; Jung, S.-H.; Kang, S.-K.; Kim, K.-R.; Rhee, S.-D.;
Kang, N.-S.; Kim, S.-Y.; Sohn, S.-K.; Kim, S.-G.; Jin, M.-S.; Lee, J.-
O.; Cheon, H.-G.; Kim, S.-S. Bioorg. Med. Chem. Lett. 2007, 17,
2622.
References and notes
1
. (a) Biginelli, P. Ber. Deutsche. Chem. Gessel. 1891, 24, 2962; (b)
Kappe, C. O. Acc. Chem. Res. 2000, 33, 879; (c) Kappe, C. O.;
Stadler, A. In Org. Reaction; Paquette, L. A., Ed.; Wiley-Interscience:
New York, 2004; Vol. 63, p 1 and references cited herein.
9. Engel, P. S.; Wu, W.-X. J. Am. Chem. Soc. 1989, 111, 1830.
10. (a) Song, J.-W.; Li, H.-J.; Choi, Y.-S.; Yoon, C.-J. J. Phys. Chem. A
2006, 110, 2065; (b) Nelsen, S. F.; Rumack, D. T.; Sieck, L. Y.; Meot-
Ner (Mautner), M. J. Am. Chem. Soc. 1988, 110, 6303; (c) Nelsen, S.
F.; Peacock, V.; Weisman, G. R. J. Am. Chem. Soc. 1976, 98, 5269.
11. We attempted to isolate 3; however, compound 3 was partially
2
3
. Kappe, C. O. Eur. J. Med. Chem. 2000, 35, 1043.
. (a) Polshettiwar, V.; Varma, R. S. Tetrahedron Lett. 2007, 48, 7343;
(b) Suzuki, I.; Suzumura, Y.; Takeda, K. Tetrahedron Lett. 2006, 47,
7861; (c) Huang, Y.; Yang, F.; Zhu, C. J. Am. Chem. Soc. 2005, 127,
16386; (d) Su, W.; Li, J.; Zheng, Z.; Shen, Y. Tetrahedron Lett. 2005,
46, 6037; (e) Sun, Q.; Wang, Y.; Ge, Z.; Cheng, T.; Li, R. Synthesis
2004, 1047; (f) Mu n˜ oz-Mu n˜ iz, O.; Juaristi, E. Arkivoc 2003, xi, 16; (g)
dehydrated to give DHPM 2a during purification using a SiO column
2
chromatography.
12. When hexanal was treated with 5 mol % of catalyst 1b in methanol at
rt, hexanal disappeared within 1 h, and considerable amount of
inseparable decomposed products were formed along with hexanal
dimethylacetal.
Maiti, G.; Kundu, P.; Guin, C. Tetrahedron Lett. 2003, 44, 2757; (h)
Bose, D. S.; Fatima, L.; Mereyala, H. B. J. Org. Chem. 2003, 68, 587;
(i) Paraskar, A. S.; Dewkar, G. K.; Sudalai, A. Tetrahedron Lett.
2
003, 44, 3305; (j) Reddy, C. V.; Mahesh, M.; Raju, P. V. K.; Babu, T.
13. When a solution of catalyst 1b in i-PrOH was refluxed, the solution
immediately became yellow and the decomposition of catalyst 1b was
observed on TLC.
R.; Reddy, V. V. N. Tetrahedron Lett. 2002, 43, 2657; (k) Lu, J.; Bai,
Y. Synthesis 2002, 466; (l) Ramalinga, K.; Vijayalakshmi, P.; Kaimal,