H-Y zeolite is a versatile heterogeneous catalyst for the synthesis of β-nitroamines

Article abstract of DOI:10.1016/j.apcata.2010.08.054

Heterogeneous aza-Henry reactions of aldimes with nitromethane were examined with zeolites as catalysts. Among the zeolite used, H-Y zeolite showed satisfactory catalytic activity for various aldimines, resulting in corresponding β-nitroamines in good yields. H-Y zeolite also showed the excellent reusability until the 3rd usage without loss of its activity; further, the calcination of deactivated H-Y zeolite allowed it to rejuvenate its catalytic activity. Solid-state ¹³C MAS NMR revealed the predominant activation of an aldimine into the H-Y zeolite.

Full text of DOI:10.1016/j.apcata.2010.08.054

Applied Catalysis A: General 388 (2010) 211–215
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
H–Y zeolite is a versatile heterogeneous catalyst for the synthesis of
␤-nitroamines
Kenichi Komura∗, Yuki Taninaka, Yoshifumi Ohtaki, Yoshihiro Sugi
Department of Materials Science and Technology, Faculty of Engineering, Gifu University, Yanagido 1-1, 501-1193 Gifu, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 15 April 2010
Received in revised form 20 August 2010
Accepted 25 August 2010
Available online 27 September 2010
Heterogeneous aza-Henry reactions of aldimes with nitromethane were examined with zeolites as cat-
alysts. Among the zeolite used, H–Y zeolite showed satisfactory catalytic activity for various aldimines,
resulting in corresponding ␤-nitroamines in good yields. H–Y zeolite also showed the excellent reusability
until the 3rd usage without loss of its activity; further, the calcination of deactivated H–Y zeolite allowed
it to rejuvenate its catalytic activity. Solid-state 13C MAS NMR revealed the predominant activation of an
aldimine into the H–Y zeolite.
Keywords:
H–Y zeolite
© 2010 Elsevier B.V. All rights reserved.
Aza-Henry reaction
Heterogeneous catalysis
␤-Nitroamine
1. Introduction
2. Experimental
Aza-Henry reaction, a nucleophilic addition of nitroalkanes to
imines, is an important reaction in organic synthesis [1]. The
reaction is because it directly offers the valuable chemical inter-
mediates, ␤-nitroamines; these can be feasibly transformed to
vicinal diamines by the reduction [2–6], or to ␣-amino acids by
Nef reactions [7,8]. Typically, the reaction has been carried out
by homogeneous catalysts such as organic weak bases or Lewis
acids [9–13]. However, the usage of heavy metal salts gives some
drawbacks, such as a tedious work-up, separation of catalyst, and
waste production. To overcome these issues, researchers have paid
much attention to the development of heterogeneous catalysts
for environmental benign organic syntheses. Especially, zeolites,
a microporous aluminosilicate with ordered pores, have been used
for catalysts in fine chemical syntheses [14], because they have acid
sites in their frameworks and are reusable by some reactivation
such as calcination under air flow. Recently, we have found that
Y zeolite works as an efficient catalyst for the hydroamination of
methyl acrylate to aniline [15–17]. This encouraged us to develop
some other fine chemical syntheses over zeolite as a versatile het-
erogeneous catalyst.
All of reagents are commercially available and were used with-
out any purification. Na–Y and H–Y zeolite (SiO2/Al2O3 (SAR) = 5.6,
HSZ-320NAA and HSZ-320 HOA) and H-MOR (SAR = 25, HSZ-
650HOA) were purchased from Tosoh Corporation, Tokyo, Japan.
H-BEA (SAR = 25, CP 814B-25, Zeolyst International, Valley Forge,
PA, USA). Prior to use, all zeolite samples were calcined by the fol-
lowing calcination program: room temperature to 550 ◦C (heating
rate 1 ◦C/min), 550 ◦C for 7 h under air flow. Sc(OTf)3 and the ion-
exchanged resin Nafion® (NR50, 0.8 mmol H+/g) were purchased
from Tokyo Kasei and Aldrich, respectively.
The acidity of zeolites was determined by ammonia
temperature-programmed desorption (NH3-TPD) using TPD-
66 (Bel Japan). The specific surface area and the pore volume were
determined by the BET method using nitrogen adsorption analysis
(BELSORP 28SA; Bel Japan) at 77 K. Estimation of yields and chemi-
cal structure of product were performed by NMR measurements in
CDCl3, and solid-state 13C magic angle scanning (MAS) NMR was
also carried out with an ECA-500 (JEOL).
2.1. Aza-Henry reaction
Here, we wish to report the heterogeneous aza-Henry reaction
of aldimines with nitromethane over zeolite catalysts. H–Y zeolite
found to be a suitable catalyst for this reaction with good yields.
To a test tube containing a magnetic stir bar and H–Y zeolite
(50 mg) was added a solution of aldimine (1 mmol) in nitromethane
(1.5 ml). The resulting mixture was then heated at 60 ◦C for 9 h.
After cooling to ambient temperature, the mixture was filtrated to
remove the catalyst and the residue was washed with chloroform
(2 ml × 3). The obtained filtrate was concentrated by evapora-
∗
Corresponding author.
E-mail address: kkomura@gifu-u.ac.jp (K. Komura).
0926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.08.054

212
K. Komura et al. / Applied Catalysis A: General 388 (2010) 211–215
Scheme 1.
Table 1
Aza-Henry reaction of 2a and nitromethane.a
Run
Catalystb
Surface area (m2/g)c
Acid amount (mmol/g)d
Conversion of imine (%)e
Yield (%)e
2a
2b
1
2
3
4
5
6
7
8
9
10
None
–
–
–
700
660
725
613
220
170
720
–
–
<1
96
13
99
48
33
77
7
<1
75
13
82
46
31
72
7
0
21
0
17
2
2
5
0
2
f
Sc(OTf)3
Nafion®g
0.80
0.54
0.01
n.d.h
0.60
0.99
1.15
–
H–Y (5.6)
H–Y (80)
H–Y (320)
H-BEA (25)
H-MOR (30)
H-ZSM-5 (40)
Na–Y (5.5)
8
26
6
26
0
Reactions were performed by using imine (0.5 mmol) and zeolite (50 mg) in nitromethene (1.5 ml, 28 mmol) at 50 ◦C for 6 h.
a
b
c
d
e
f
Number in parentheses is SiO2/Al2O3 ratio.
Measured by N2 adsorption using BET method.
Measured by NH3-TPD. Values are estimated from h-temperature peak.
Estimated by 1H NMR.
10 mol% of catalyst was used.
g
h
Amount of installed H+ was 10 mol%.
Not detectable.
tion and dried in vacuo. Resulting crude mixture was resolved in
H–Y zeolite catalyst was calcined at 550 ◦C under air flow and
used.
CDCl3 and measured by 1H NMR to determine the yield: compar-
ing with methyne proton (CH) of product, aryl protons (Ph–H) of
substrate and ␤-nitrostyrene. In the reaction screenings given in
Tables 1 and 2, the reaction was performed at 90 ◦C for 6 h under
the same amounts of substrates and catalyst as above.
2.3. Solid-state 13C MAS NMR measurement of imine occluded HY
zeolite
Samples of imine and CH3NO2 occluded HY zeolite were pre-
pared as follows: the prescribed amounts of imine or CH3NO2 were
resolved in dry toluene, and to the resulting clear solution was
added HY zeolite. After stirring for 2 h at room temperature, toluene
was removed by reduced pressure. The obtained zeolite powder
was further exposed under vacuum for 6 h at room temperature,
and then each sample was put into a solid-state NMR sample folder
(4 mm zirconia rotor) immediately. 13C MAS NMR measurement
was done at 5 kHz spinning.
2.2. Reuse and recycle experiment
The reaction was performed using the same supporting con-
ditions as mentioned above. After the reaction was completed,
the resulting mixture was filtrated and washed with chloro-
form. Collected H–Y zeolite catalyst was dried in vacuo and put
into a fresh test tube containing a magnetic stirring bar again.
To this was added the solution of imine 1a in nitromethane to
demonstrate the reaction. After the 4th time running, collected
3. Results and discussion
Table 2
We examined the zeolite catalyzed aza-Henry reaction using an
N-p-anisyl-p-tolualdimine (1a) in nitromethane as a model reac-
tion, as shown in Scheme 1. Aldimine 1a was easily prepared
from p-tolualdehyde and p-anisidine, and the reaction 1a with
nitromethane gives a 2-(p-anisyl)amino-2-p-tolyl-1-nitroethane
(2a) as a main product; however, the ␤-nitrostyrene derivative (2b)
was observed as a minor product.
Table 1 summarizes the results of the aza-Henry reaction. The
reaction was performed at 50 ◦C for 6 h in the presence of catalyst.
The reaction did not occur in the absence of catalyst (run 1). By com-
paring with conventional homogeneous catalysts, we carried out
the reaction by using 10 mol% of Sc(OTf)3 as a catalyst [18]. The high
conversion of 1a was observed and the yield of 2a was 75% accom-
panying with 21% of 2b. On the other hand, the ion-exchanged resin
as a versatile heterogeneous acid catalyst, Nafion catalyzed the
reaction of 1a with nitromethane to afford 2a with low yield (run
Effects of solvent in aza-Henry reaction over H–Y zeolite.a
Run
Solvent
Conversion of imine (%)
Yield (%)
2a
2b
1
2
3
4
5
6
7
8
9
10
MeOH
EtOH
75
48
36
99
40
43
0
65
42
29
82
32
29
–
–
–
84
10
6
7
17
8
14
–
–
–
16
2-Propanol
CH3NO2
Toluene
Chloroform
THF
EtOAc
0
DMF
0
>99
CH3NO2/H2Ob
a
Reactions were performed by using imine (0.5 mmol), zeolite (50 mg) and
nitromethane (2.5 mmol) in solvent (1.5 ml) at 50 ◦C for 6 h.
b
0.5 mmol of H2O was added.

K. Komura et al. / Applied Catalysis A: General 388 (2010) 211–215
213
Table 3
Aza-Henry reaction of aldimines in nitromethane over H–Y zeolite.a
Run
R
Product
Yield of ␤-nitroamineb
1
81 (10)
2
90 (9)
3
75 (25)
4
79 (17)
5
44 (51)
6
47 (17)
44 (15)
7
a
All reactions were performed by using imine (0.5 mmol) and nitromethane (1.5 ml, 27.9 mmol) in the presence of H–Y (SiO2/Al2O3= 5.6, 50 mg) at 60 ◦C for 9 h.
b
Estimated by 1H NMR, and the number of parenthesis indicates the yield of the corresponding ␤-nitrostryrene derivative.
3). However, H–Y (SiO2/Al2O3 (SAR) = 5.6) and H-Beta (SAR = 25)
is highest to give 2a in 82% yield. This suggests that pore size and
structure of zeolite affect the catalysis, and that H–Y zeolite is the
most appropriate catalyst for the aza-Henry reaction. However, we
found that the yield of 2a was decreased with increasing in SAR of
H–Y zeolite (runs 5 and 6), suggesting that acidity of the zeolite is
also essential for catalysis. Na–Y zeolite (SAR = 5.5) gave 2a in 26%
yield (run 6), although it is effective catalyst for the hydroamina-
tion [15]. These results show that both pore structure and acidity of
zeolite are essential, and that H–Y zeolite with low SAR is a suitable
catalyst to achieve the aza-Henry reaction.
zeolites with 3-dimensional 12-membered ring (MR) showed high
catalytic activity: H–Y zeolite gave the product 2a in 82% yield
with quantitative conversion and H-Beta zeolite was in 72% yield,
respectively (runs 4 and 7). It can be found that H–Y zeolite showed
the catalytic activity comparable with the homogeneous catalyst,
Sc(OTf)3. The higher yield of H–Y zeolite than of H-Beta is due to
the different reaction space into the zeolite: H–Y has larger reaction
space (super cage) than H-Beta. However, H-MOR (1-dimensional
12 MR: SAR = 30) and H-ZSM-5 (3-dimensional 10-MR: SAR = 40)
zeolites exhibited much lower activities: 7% and 6% yields of 2a,
respectively (runs 8 and 9). It is notable that, although H–Y zeolite
has the lowest acidity among the used zeolites, the catalytic activity
The influences of solvents are summarized in Table 2. Among
used solvents, nitromethane was found to be the best to afford the
2a in 82% yield with excellent conversion of 1a. A polar solvent

214
K. Komura et al. / Applied Catalysis A: General 388 (2010) 211–215
Fig. 1. Effect of reaction temperature in the aza-Henry reaction over H–Y zeolite.
Fig. 2. Reusability of H–Y zeolite in the aza-Henry reaction of 1a with nitromethane.
such as MeOH gave 2a in 65% yield; however, EtOH, 2-propanol,
toluene, chloroform, and THF gave much lower yields, and no reac-
tion occurred in ethyl acetate and DMF. Interestingly, the addition
of water did not remarkably affect the catalysis, showing the same
product distribution of 2a and 2b in nitromethane (run 10). Con-
sequently, such results indicate that nitromethane is the most
appropriate solvent in the aza-Henry reaction over H–Y zeolite.
We then examined the influence of the reaction temperature
shown in Fig. 1. The gradual increasing in yield of 2a was observed
at 30 ◦C, and the reaction was reached in 55% yield after 12 h. While
the reaction was carried out at 60 ◦C, the yield of 2a was increased
with the reaction period, and the good yield (81%) was obtained for
9 h. However, the reaction was finished after 3 h with a steep yield
gradient at 90 ◦C, affording 2a in 83% yield. Although the reaction
was rapidly finished at 90 ◦C, the yield of the desired product 2a was
almost same as that at 60 ◦C. These results indicate that the reaction
of 1a over H–Y zeolite that take place at 60 ◦C for 9 h should provide
suitable results.
electron deficient carbon atom of an imine. Therefore, one can con-
sider two types of reaction pathways catalyzed by zeolite: one
is the nitromethane activated path and the other is the imine
activated path. Recently, we [17] and Onaka et al. [19,20,21]
independently reported the availability of solid-state 13C MAS
NMR technique of the occluded organic molecule into the zeo-
lite to discuss the catalysis associating with the activation manner.
In order to clarify which substrates are activated by H–Y zeo-
lite, we examined solid-state 13C MAS NMR measurements of
the aldimine or the nitromethane occluded H–Y zeolite (abbrevi-
ated as imine@H–Y and CH3NO2@H–Y, respectively). The spectra
of a liquid nitromethane and a solid aldimine were compared;
no significant distinct peak was observed in the spectrum of
CH3NO2@H–Y (Figure not shown), while in Fig. 3, quite differ-
ent peak patterns in 13C NMR charts were obviously detected in
the imine@H–Y. Due to the drastic peak change, we cannot fully
assign their patterns; however, it is evident that the H–Y zeolite
Table 3 summarizes the aza-Henry reaction over H–Y zeolite
using various aldimines in nitromethane. The reaction of N-(p-
anisyl)benzaldimines derived from benzaldehyde and p-anisidine
gave the corresponding ␤-nitroamine in 81% yield (run 1). N-(p-
Anisyl)phenylaldimines with p-substituted methyl, methoxy and
chloro groups also gave good yields (runs 2–4) irrespective of their
electron-donating or withdrawing natures. However, the reaction
of N-(p-anisyl)-p-nitrobenzaldehyde imine gave the mixture of
␤-nitroamine and the ␤-nitrostyrene derivative with almost the
same ratio. This is because that a nitro substituent at para posi-
tion is the strongest electron-withdrawing nature, which leading
to the elimination to form ␤-nitrostyrene derivative (entry 5).
H–Y zeolite also catalyzed the reaction of N-(p-anisyl) heterocyclic
aldimines, derived from 2-furfural and 2-thiophenaldehyde, and
showed moderate yields in 47% and 44%, respectively.
Recycle experiments in the reaction of 1a in nitromethane were
examined over H–Y zeolite; results are shown in Fig. 2. The yield of
2a remained satisfactory until the 3rd use: 80% (1st), 83% (2nd), 83%
(3rd); however, it decreased slightly at 4th use (74%). H–Y zeolite
after 4th use was rejuvenated by the calcination at 550 ◦C for 4 h
under air flow, resulting in 85% yield of 2a. This suggests that H–Y
zeolite is a promising solid catalyst for aza-Henry reaction with
excellent reusability.
Fig. 3. 13C MAS NMR spectra of imine 1a (down) and imine 1a occlude into H–Y
zeolite (imine@H–Y) (up), respectively.
The aza-Henry reaction is one of the typical nucleophilic addi-
tion reactions: nitroalkanes work as a nucleophile to attack the

K. Komura et al. / Applied Catalysis A: General 388 (2010) 211–215
215
In conclusion, we have developed the heterogeneous aza-Henry
reaction over zeolite catalyst. Corresponding ␤-nitroamines can be
obtained in good yields over H–Y zeolite catalyst, and H–Y zeolite
showed the excellent reusability and recyclability. Thus, H–Y zeo-
lite is a versatile heterogeneous catalyst. Further research is now
ongoing, and it will be report in near future.
References
[1] K. Juhl, N. Gathergood, K.A. Jørgensen, Angew. Chem. Int. Ed. 40 (2001)
2995–2997.
[2] B. Westermann, Angew. Chem. Int. Ed. 42 (2003) 151–153.
[3] M.A. Sturgess, D.J. Yarberry, Tetrahedron Lett. 34 (1993) 4743–4746.
[4] K. Imagawa, E. Hata, T. Yamada, T. Mukaiyama, Chem. Lett. (1996) 291–292.
[5] D. Enders, J. Wiedemann, Synthesis (1996) 1443–1450.
[6] D. Lucet, L. Toupet, T.L. Gall, C. Mioskowski, J. Org. Chem. 62 (1997) 2682–2683.
[7] R. Ballini, M. Petrini, Tetrahedron Lett. 40 (1999) 4449–4452.
[8] E. Foresti, G. Palmieri, M. Petrini, R. Profeta, Org. Biomol. Chem. 1 (2003)
4275–4281.
Fig. 4. Plausible reaction mechanism of the aza-Henry reaction over zeolite catalyst.
[9] H. Adams, J.C. Anderson, S. Peace, A.M.K. Pennell, J. Org. Chem. 63 (1998)
9932–9934.
[10] J.C. Anderson, S. Peace, S. Pih, Synlett (2000) 850–852.
[11] C. Qian, F. Gao, R. Chen, Tetrahedron Lett. 42 (2001) 4673–4675.
[12] N. Nishiwaki, K.R. Knudsen, K.V. Gothelf, K.A. Jørgensen, Angew. Chem. Int. Ed.
40 (2001) 2992–2994.
[13] K. Yamada, S.J. Harwood, H. Gröger, M. Shibasaki, Angew. Chem. Int. Ed. 38
(1999) 3504–3506.
[14] M.E. Davis, Micropor. Mesopor. Mater. 21 (1998) 173–182.
[15] K. Komura, J. Tsutsui, R. Hongo, Y. Sugi, Stud. Surf. Sci. Catal. 170B (2007)
1111–1119.
[16] J. Horniakova, K. Komura, H. Osaki, Y. Kubota, Y. Sugi, Catal. Lett. 102 (2005)
191–196.
[17] K. Komura, R. Hongo, J. Tsutsui, Y. Sugi, Catal. Lett. 128 (2009) 203–209.
[18] S. Kobayashi, M. Sugiura, H. Kitagawa, W.W.-L. Lam, Chem. Rev. 102 (2002)
2227–2302.
[19] T. Okachi, M. Onaka, J. Am. Chem. Soc. 126 (2004) 2306–2307.
[20] S. Imachi, M. Onaka, Chem. Lett. (2005) 708–709.
[21] K. Kobayashi, Y. Igura, S. Imachi, Y. Masui, M. Onaka, Chem. Lett. (2007) 60–61.
would affect the interaction with imine more effectively than with
nitromethane.
Based upon the above results and the considerations of the
catalytic mechanism by a typical Brønsted and/or Lewis acid in
organic synthesis, we speculate a plausible reaction mechanism
shown in Fig. 4. Because the acid sites in the framework of H–Y
zeolite should be responsible for catalysis, acid sites of H–Y zeo-
lite would activate the imine by the interaction with lone pairs
of nitrogen atoms to enhance the electron deficiency of carbon
atoms, then a reactive tautomer of nitromethane attacks to the
imine, resulting in yield of a product, accompanying with regen-
erating the acid site again. However, it is necessary to carry
out more experiments to prove our speculated reaction mecha-
nism.