Bifunctional heterogeneous catalysis of silica-alumina-supported tertiary amines with controlled acid-base interactions for efficient 1,4-addition reactions

Article abstract of DOI:10.1002/chem.200901380

We report the first tunable bifunctional surface of silica-aluminasupported tertiary amines (SA-NEt₂) active for catalytic 1,4-addition reactions of nitroalkanes and thiols to electron-deficient alkenes. The 1,4-addition reaction of nitroalkanes to electron-deficient alkenes is one of the most useful carbon-carbon bond-forming reactions and applicable toward a wide range of organic syntheses. The reaction between nitroethane and methyl vinyl ketone scarcely proceeded with either SA or homogeneous amines, and a mixture of SA and amines showed very low catalytic activity. In addition, undesirable side reactions occurred in the case of a strong base like sodium ethoxide employed as a catalytic reagent. Only the present SA-supported amine (SA-NEt₂) catalyst enabled selective formation of a double-alkylated product without promotions of side reactions such as an intramolecular cyclization reaction. The heterogeneous SA-NEt₂ catalyst was easily recovered from the reaction mixture by simple filtration and reusable with retention of its catalytic activity and selectivity. Fur-thermore, the SA-NEt₂ catalyst system was applicable to the addition reaction of other nitroalkanes and thiols to various electron-deficient alkenes. The solid-state magic-angle spinning (MAS) NMR spectroscopic analyses, including variable-contact-time ¹³C cross-polarization (CP)/MAS NMR spectroscopy, revealed that acid-base interactions between surface acid sites and immobilized amines can be controlled by pretreatment of SA at different temperatures. The catalytic activities for these addition reactions were strongly affected by the surface acid-base interactions. 2009 Wiley-VCH Verlag GmbH & Co. KGaA.

Full text of DOI:10.1002/chem.200901380

FULL PAPER
DOI: 10.1002/chem.200901380
Bifunctional Heterogeneous Catalysis of Silica–Alumina-Supported Tertiary
Amines with Controlled Acid–Base Interactions for Efficient 1,4-Addition
Reactions
Ken Motokura,[a, b] Satoka Tanaka,[a] Mizuki Tada,[a, c] and Yasuhiro Iwasawa*[a, d]
Abstract: We report the first tunable
bifunctional surface of silica–alumina-
supported tertiary amines (SA–NEt2)
active for catalytic 1,4-addition reac-
tions of nitroalkanes and thiols to elec-
tron-deficient alkenes. The 1,4-addition
reaction of nitroalkanes to electron-de-
ficient alkenes is one of the most
useful carbon–carbon bond-forming re-
actions and applicable toward a wide
range of organic syntheses. The reac-
tion between nitroethane and methyl
vinyl ketone scarcely proceeded with
either SA or homogeneous amines, and
a mixture of SA and amines showed
very low catalytic activity. In addition,
undesirable side reactions occurred in
the case of a strong base like sodium
ethoxide employed as a catalytic re-
agent. Only the present SA-supported
amine (SA–NEt2) catalyst enabled se-
lective formation of a double-alkylated
product without promotions of side re-
actions such as an intramolecular cycli-
zation reaction. The heterogeneous
SA–NEt2 catalyst was easily recovered
from the reaction mixture by simple fil-
tration and reusable with retention of
its catalytic activity and selectivity. Fur-
thermore, the SA–NEt2 catalyst system
was applicable to the addition reaction
of other nitroalkanes and thiols to vari-
ous electron-deficient alkenes. The
solid-state magic-angle spinning (MAS)
NMR spectroscopic analyses, including
variable-contact-time 13C cross-polar-
AHCTUNGTRENGNiUN zation (CP)/MAS NMR spectroscopy,
revealed that acid–base interactions be-
tween surface acid sites and immobi-
lized amines can be controlled by pre-
treatment of SA at different tempera-
tures. The catalytic activities for these
addition reactions were strongly affect-
ed by the surface acid–base interac-
tions.
Keywords: acid–base interactions ·
À
amines · C C coupling · heteroge-
neous catalysis · silica–alumina
Introduction
Much attention has been paid to “double catalytic activa-
tion” of two substrates in a single reaction step for efficient
organic synthesis. Because acidic and basic functions can ac-
tivate electrophiles and nucleophiles, respectively, to make
various carbon–carbon bond-forming reactions possible,
many acid–base pairs have been examined in homogeneous
reaction systems.[1] However, the use of both strong acidic
and basic species in a homogeneous single reactor is impos-
sible due to the neutralization that forms inactive salts. In
addition, recovery and reuse of both catalytic species are
often problematic in homogeneous systems. To overcome
these problems in homogeneous systems, several acid–base
bifunctional heterogeneous catalysts have been devel-
oped.[2,3] These heterogeneous catalyst systems make the co-
existence of acid and base functions without complete neu-
tralization possible.
[a] Dr. K. Motokura, Dr. S. Tanaka, Dr. M. Tada, Prof. Dr. Y. Iwasawa
Department of Chemistry
Graduate School of Science, The University of Tokyo
Hongo, Bunkyo-ku, Tokyo 113-0033 (Japan)
E-mail: iwasawa@chem.s.u-tokyo.ac.jp
[b] Dr. K. Motokura
Present Address: Department of Environmental Chemistry and Engi-
neering
Interdisciplinary Graduate School of Science and Engineering
Tokyo Institute of Technology
4259 Nagatsuta-cho, Midori-ku, Yokohama, 226-8502 (Japan)
[c] Dr. M. Tada
Institute for Molecular Science
38 Nishigo-Naka, Myodaiji, Okazaki 444-8585 (Japan)
[d] Prof. Dr. Y. Iwasawa
Another feature of acid–base heterogeneous catalysts is
their positioning control of acidic and basic sites, which en-
AHCTUNGTREGaNNUN bles not only the avoidance of the acid–base neutralization
Department of Applied Physics and Chemistry
The University of Electro-Communications
Chofu, Tokyo 182-8585 (Japan)
Chem. Eur. J. 2009, 15, 10871 – 10879
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
10871

but also the creation of highly functionalized surfaces for se-
lective reactions. We also expected that even unstable sub-
strates, which are easily dimerized or oligomerized by a
strong acid or base, can be used in the weak acid–base bi-
functional catalyst system. However, there are few reports
on such selective reactions by acid–base bifunctional hetero-
geneous catalysts.[4] In addition, appropriate spectroscopic
techniques to estimate the acid–base interaction on solid
surfaces have not been reported until now.
Our strategy for the design of acid–base bifunctional het-
erogeneous catalysts is to use acidic silica–alumina (SA) as
a support on which basic alkyl amines are immobilized by
silane-coupling reactions. The bifunctional SA surfaces may
Scheme 1. Preparation of SAACTHNUTRGNEG(NU 500)–NEt2.
À
exhibit high catalytic activities for C C bond-forming reac-
tions without undesired neutralization at the surface. Re-
cently, we have reported the high performance of the SA-
supported tertiary and primary amine catalysts (SA–NR2)
for cyano-ethoxycarbonylation, the Michael reaction of ni-
triles, and the nitro-aldol reaction.[3] During the course of
our research on SA–NEt2 catalyses, we have found that the
acid–base interaction between the SA surface and the im-
mobilized amine group strongly affects its catalytic activities.
It can be said that controlling the acid–base interaction to
avoid neutralization is a key to improving the SA–NEt2 cat-
alyst for its application toward selective organic transforma-
tions.
In this study, we report the first tunable bifunctional sur-
face active for catalytic 1,4-addition reactions of nitroal-
kanes and thiols to electron-deficient alkenes. The 1,4-addi-
tion reaction of nitroalkanes to electron-deficient alkenes is
one of the most useful carbon–carbon bond-forming reac-
tions and applicable toward a wide range of organic synthe-
ses.[5] We also found that the surface density of silanol
groups on SA plays a pivotal role in controlling the acid–
base interactions of the SA–NEt2 catalysts. This acid–base
interaction was estimated by using several solid-state NMR
spectroscopic techniques, such as variable-contact-time
measurements in 13C cross-polarization (CP)/magic-angle
spinning (MAS) NMR spectroscopy.
Results and Discussion
Figure 1. A) 13C NMR spectrum of DAPS in CDCl3. Solid-state 13C CP/
MAS NMR spectra of B) SA(500)–NEt2, C) SA(400)–NEt2, D) SA(200)–
NEt2, E) SA(120)–NEt2, and F) triethylamine adsorbed on SA(500).
G
E
ACHTUNGTRENNUNG
N
ACHTUNGTRENNUNG
Preparation and characterization of catalysts: SA (silica
66.5, alumina 25.1%; 380 m2 gÀ1) was pretreated at 120–
5008C under vacuum before use as a support for amines.
The treated samples are denoted as SA(treatment tempera-
ture: T). Tertiary amines were immobilized on the SA(T) by
treatment of the SA(T) with a solution of 3-(diethylamino)-
propyltrimethoxysilane (DAPS) in toluene under reflux for
24 h (Scheme 1). Then the immobilized SA was filtered and
washed with dichloromethane, followed by drying under
vacuum. The solid-state 13C CP/MAS NMR spectrum for
SA(T)–NEt2 (Figure 1B–E) was similar to that of parent
DAPS (Figure 1A), thereby indicating the retention of the
DAPS structure after the DAPS immobilization. A stronger
signal for methoxy groups (see Figure 1, peak f: d=47 ppm)
in SA(500)–NEt2 (Figure 1B) than that for the SA(120)–
NEt2 (Figure 1E) indicates that the quantity of the methoxy
groups that remained in the SA(500)–NEt2 is larger than
that in the SA
(120)–NEt2. This is in agreement with the 29Si
MAS NMR spectroscopic data for SA(T)–NEt2, in which
G
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
the peak intensity of the T1 site (T1 =Si
ACHTNUTRGNEUGN(OCH3)2) at d=
À49 ppm is larger than that of the T2 site (T2 =Si
ACHTUNGRTEN(NUNG OCH3)) at
d=À56 ppm in SA
(500)–NEt2 (Figure 2A),[6] but the peak
ACHTUNGTRENNUNG
intensity of the T2 site increases and that of the T1 site de-
10872
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10871 – 10879

Silica–Alumina-Supported Tertiary Amines
FULL PAPER
Table 1. 1,4-Addition of 1 to 2 by various catalysts.[a]
Entry Catalyst
Amine
loading
A
Conversion Yield Initial rate
[c,d]
of 1 [%][c]
of 3
[mmolhÀ1
ACHTUNGTERNNUNG ]
[mmolgÀ1
]
[%][c]
[b]
1
2
3
4
SA
SA
SA
SA
SA
N
99
99
99
99
99
84
74
80
82
86
93
49
0.28
0.36
0.41
0.53
0.53
0.16
5[e]
6
SiO
NEt2
SiO2A(500)–
NEt2
H-USY–NEt2 0.45
SA
DAPS
NEt3
(120)–
0.73
7
E
0.69
76
14
0.13
8
37
22
8
16
99
99
0
33
0
8
0.01
0.02
0.03
<0.01
0.27
n.d.[h]
–
9[f]
10
11
12[f]
13
14
–
–
–
–
–
–
2
Figure 2. Solid-state 29Si MAS NMR spectra of A) SA
ACTHNUGTREN(GUN 500)–NEt2, B) SA-
ACHTUNGTRENNUNG(400)–NEt2, C) SAACHTUNGTRENNUNG(200)–NEt2, and D) SAACHTNUGTRENN(UGN 120)–NEt2.
SA
NaOEt
none
28
1[g]
0
[a] Reaction conditions:
1 (1 mmol), 2 (3 mmol), catalyst (4.5ꢁ
10À2 mmol), toluene (1 mL), 508C, 20 h. [b] Determined by elemental
1
creases with a decrease in the pretreatment temperature of
SA (Figure 2B–D) because the number of surface silanol
groups decreases when the calcination temperature is in-
creased.[7] These results suggest that the DAPS precursor is
immobilized on the SA surface by Si-O-Si covalent bonding.
The amine loading in the SA(T)–NEt2 was determined typi-
analysis. [c] Determined by H NMR spectroscopy and GC using an inter-
nal standard technique. Based on 1. Other byproducts were not detected
except for 4 in the case of supported catalysts. [d] Initial formation rate
of the addition products (3+4). [e] 24 h. [f] 0.08 g of SA was used. [g] A
byproduct was formed from 3 by a cyclization reaction. [h] n.d.=not de-
termined.
cally as 0.56 mmolgÀ1 for SA
ACTHNUTRGNEU(GN 500)–NEt2 by elemental analy-
sis. These loading amounts decreased when the calcination
temperature was increased because of a decrease in the
tion (entries 6–8). Monoalkylated nitroethane (4), which is
regarded as an intermediate, was also detected during the
double addition reaction. It is to be noted that this reaction
À
amount of surface Si OH groups of SA by dehydration
(Table 1). A Lewis base of N,N-dimethylaminopyridine
could adsorb on the SA
G
scarcely proceeded with either SA
amines (entries 9–11). A mixture of SA
ACHTUNGTRENaNUGN mine showed catalytic activity, but the yield of 3 was very
ACHTUNGTRENNUNG
nm2), thereby indicating the retention of acid sites after the
AHCTUNGTRENNUNG
amine immobilization. In the case of SA
G
2.25 mmolgÀ1 of pyridine derivative was adsorbed, which
low (entry 12). In the case of a strong base like sodium eth-
oxide as a homogeneous catalyst, the selectivity toward 3
was as low as 1–7% during the whole course of the reaction
due to further conversion of 3 toward an undesired cycliza-
tion product (entry 13, Scheme 2).[8] In addition, there were
several reports of a double addition reaction of simple nitro-
alkanes to MVK; however, the selectivities were usually less
than 50%.[9] It is to be noted that the present report is the
À
also suggests that the amount of acidic Si OH was de-
creased by the thermal pretreatment of SA.
Catalysis of SA(T)-supported amines for 1,4-addition of ni-
troalkanes: The 1,4-addition reaction of nitroalkanes to elec-
tron-deficient alkenes is traditionally promoted by homoge-
neous strong bases. However, there are several problems,
such as difficulties in their handling in air due to moisture
and carbon dioxide sensitivities and unavoidable undesirable
side reactions. To develop an effective 1,4-addition catalyst
system that overcomes the problems in carbon–carbon
bond-forming processes, we used the SA–NEt2 as heteroge-
neous catalysts and examined the 1,4-addition reaction of ni-
troethane (1) to methyl vinyl ketone (MVK; 2). As shown
in Table 1, the catalyst SAACHTNUTRGENNG(U 500)–NEt2 was found to be re-
markably active to afford the double-alkylated product 3 in
93% yield (Table 1, entry 5). Decreasing the pretreatment
temperature of SA reduced the catalytic activity of SA(T)–
NEt2 (entries 1–4). Amine catalysts on other supports such
as silica and zeolite were much less effective for this reac-
Scheme 2. The formation of the cyclization product from 1 and 2.
Chem. Eur. J. 2009, 15, 10871 – 10879
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10873

Y. Iwasawa et al.
first one for selective formation of the double-alkylated ni-
troalkane product.
The general applicability of the SAACTHNUTRGNEG(NU 500)–NEt2 catalyst to
1,4-addition reactions of nitroalkanes to electron-deficient
alkenes is listed in Table 2. The use of a nitroalkane in an
amount three times larger than that of an alkene induced se-
lective formation of a monoalkylated product. Various nitro-
alkanes such as 1-nitropropane, 1-nitrohexane, 2-nitroetha-
nol, and nitrocyclohexane reacted with 2 to give the corre-
sponding addition products in good to excellent yields (62–
alyst was easily recovered from the reaction mixture by
simple filtration, and reusable with retention of its catalytic
activity and selectivity (entry 2). However, the product yield
decreased in the third use (64%) compared with the second
use, and the preservation of amine loading after the catalytic
reaction was confirmed by elemental analysis and NMR
spectroscopy. The catalyst might be deactivated after the
second use by the adsorption of some organic molecules on
active sites.
96%; Table 2, entries 1–7). The SA
G
Relationship between surface acid–base interaction and cat-
also active for other alkenes. The 1,4-addition reactions of 1
to not only acyclic 2 and ethyl vinyl ketone but also cyclic
a,b-unsaturated ketones proceeded smoothly and afforded
the corresponding 4-nitroketones (entries 8–10). Acrolein
was also found to be a good acceptor to form 4-nitro-1-hep-
tylaldehyde in 92% yield (entry 11). The reactions of acrylo-
nitrile did not proceed effectively (entry 12). After the cata-
alytic performances: After the immobilization of tertiary
amine groups onto SAACTHNUTRGNEUNG
(120), the 13C NMR spectroscopic
signal of the terminal carbon (see Figure 1, peak a: d=
9.5 ppm) shifted upfield compared with that of the amine
precursor (d=11.8 ppm; Figure 1A and C). The signal for
the mixture of SA
large upfield shift (Figure 1F).[3] On the other hand, a signal
for SA(500)–NEt2 was observed at d=11.0 ppm, which
ACHTUNGTRENN(UNG 500) and triethylamine also showed a
lytic addition reaction of 1 to 2, the used SA
G
ACHTUNGTRENNUNG
showed only a very small shift
from the precursor (Figure 1B).
The sharpness of the carbon
signals increased with increas-
ing pretreatment temperature,
thereby suggesting an increase
of molecular mobility. Thus,
contact-time array 13C CP/MAS
NMR spectroscopic measure-
Table 2. 1,4-Addition of nitroalkanes to alkenes by SAACTHNUTRGNEUNG
(500)–NEt2.[a]
Entry
Nitroalkane
Alkene
Product
Yield [%][b]
1
85
83
2[c]
1
2
2
4
ments of SA
ACHTUNGTRNEUN(NG 500)–NEt2 and
SA(120)–NEt2 were conducted
AHCTUNGTRENNUNG
3
87
to determine the molecular mo-
bility of immobilized amines, as
shown in Figures 3 and 4, re-
spectively. This variable-con-
tact-time 13C CP/MAS NMR
4
2
96
spectroscopy is
a
technique
5[d]
2
2
62
96
used to determine the molecu-
lar motion of solid materials.[10]
Since CP is a measure of the ef-
ficiency of magnetization trans-
fer by the dipolar coupling
6[e]
1
from H to 13C, it is most effi-
7
2
78
cient for the static H–13C dipo-
1
lar interactions. As a result, the
less mobile carbon groups ex-
hibit the faster cross-polariza-
tion rate, and the NMR spec-
troscopic signal intensity be-
comes stronger with relatively
short CP contact time. This
signal intensity becomes weak
with prolonged contact time
due to attenuation of the trans-
ferred magnetization. On the
other hand, the signal intensity
of a molecule with high mobili-
ty shows a strong intensity with
8[f]
9[g]
1
1
87
83
10[g,h]
1
93
11[i]
1
1
92
24
12[d]
[a] Reaction conditions: nitroalkane (3 mmol), alkene (1 mmol), SAACHTNUTGRNEUNG
(500)–NEt2 (4.5ꢁ10À2 mmol), toluene
(1 mL), 508C, 24 h. [b] Determined by H NMR spectroscopy and GC. Based on alkene. [c] Reuse experiment.
[d] 70 h. [e] 12 h. [f] 808C. [g] 1008C. [h] 50 h. [i] Room temperature, 1 h.
1
10874
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10871 – 10879

Silica–Alumina-Supported Tertiary Amines
FULL PAPER
Figure 5. Dependence of the normalized intensity of variable-contact-
time 13C CP/MAS NMR spectra for the terminal carbon (a) on contact
time: A) SAACHTNUGRTNENUG(500)–NEt2 (*) and B) SAACHTUNTGERN(NUGN 120)–NEt2 (+).
Table 3. Contact times with highest peak intensities derived from con-
tact-time-array solid-state 13C CP/MAS NMR spectroscopic measure-
ments.
Figure 3. Contact-time-array 13C CP/MAS NMR spectra for SA
NEt2. The values on the spectra are contact times [ms].
ACHTUNGTRNE(NUNG 500)–
Sample
Contact time with highest peak intensity [ms]
Terminal carbon
Carbon next to nitrogen
(a: d=8–11 ppm)
(b: d=48 ppm)
SA
SA
SA
U
5.0
1.0
1.0
ꢀ1.5–3.0
0.5
ꢀ0.9–2.0
the highest intensity for SA
time. Thus, the cross-polarization time constant (TCH) values
of SA(500)–NEt2 are much longer than those of SA(120)–
NEt2, thereby indicating the much higher mobility of the ter-
tiary amine group in SA(500)–NEt2. The strong acid–base
interaction between the amine nitrogen and the SA acid site
in SA(120)–NEt2 suppressed the tertiary amine mobility
compared with SA(500)–NEt2. Based on the chemical shifts
ACHTUNGRTEN(NUNG 120)–NEt2 was at 1.0 ms contact
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
of the terminal carbon atoms and molecular motility of the
immobilized amines, it seems that the acid–base interaction
in SAACTHNUGTRENNG(U 500)–NEt2 is much weaker than that in SAACHTUNGTRENNUNG(120)–
[11,12]
NEt2
in spite of the increasing acid strength of the SA
surface by the pretreatment at 5008C.[13,14]
It can be expected that the amines are immobilized on a
Figure 4. Contact-time-array 13C CP/MAS NMR spectra for SA
ACTHNUTRGNE(NUG 120)–
NEt2. The values on the spectra are contact times [ms].
neighboring position of
a Brønsted acid site on SA
(SiO(H+)Al) because of the following amine-immobilization
mechanism: 1) the nitrogen atom of free DAPS interacts
with the strong Brønsted acid (SiO(H+)Al) on the SA sur-
relatively long contact time.[10] Figure 5 shows the normal-
ized intensity of 13C CP/MAS NMR spectra versus contact
time curves for amine terminal carbon atoms (a: d=9–
face, and 2) one Si OMe of DAPS reacts with a surface Si
OH species near the strong acid site to form a covalent Si-
O-Si (surface) bond (Scheme 1).[3] The calcination of SA at
5008C reduced the amount of surface silanol groups, thus re-
sulting in a high dispersion of silanols on the SA surface,
and as a result, DAPS may be immobilized on positions far
away from a strong acid site (Scheme 1). On the other hand,
À
À
11 ppm) on SAACHTUNGTRENNUNG(500)–NEt2 and SAACHTUNGTRNEN(GUN 120)–NEt2. Contact
times with the highest peak intensities of the terminal
carbon atom and the carbon atom next to the nitrogen atom
(b: d=48 ppm) derived from the variable-contact-time
measurements are summarized in Table 3. Variable-contact-
in the case of a SAACTHNUTRGNEUNG(120) support with a higher concentration
time data for SA
the terminal carbon with 5.0 ms contact time. In contrast,
(500)–NEt2 showed the highest intensity of
of acid sites, the amine–acid site distance can become close,
thereby suggesting strong acid–base interaction (Scheme 3).
Chem. Eur. J. 2009, 15, 10871 – 10879
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10875

Y. Iwasawa et al.
Scheme 3. Preparation of SA–NEt2 with untreated or low-temperature
calcined SA.
IR spectra for SAACHTNUTRGENNUG(500)–NEt2 and SAAHCTUNTGREN(NUGN 120)–NEt2 revealed
À
that there were no clear signals assignable to n˜ACTHUNTGRNEUNG(N H),
À
whereas strong signals for n˜
A
G
(500) around 2500–2800 cmÀ1
(Figure 6).[15] 13C CP/MAS NMR spectra for triethylamine
adsorbed on SA(500) showed the largest upfield shift of the
ACHTUNGTRENNUNG
terminal carbon signal (Figure 1F) than those for immobi-
lized amines on SA (Figure 1B–E), thereby suggesting the
strongest acid-site–amine-base interaction. It has been re-
ported that the acid–base neutralization is suppressed at the
immobilized surface.[2a,f,g,3] In conclusion, the proposed sur-
Scheme 4. Proposed surface structures for A) SA
(120)–NEt2, and C) SA(500) with triethylamine.
ACHTUNGTRNE(NGNU 500)–NEt2, B) SA-
N
ACHTUNGTRENNUNG
face structures of SA
ethylamine on SA(500) with acid–base interactions are illus-
trated in Scheme 4. Thus, it is suggested that both surface
acidic and basic species in SA(500)–NEt2 without acid–base
ACHUTGTNRENN(GU 500)–NEt2, SAHCATUNGTREN(NUNG 120)–NEt2, and tri-
U
ACHTUNGTRENNUNG
neutralization act as effective catalytic sites, which agrees
A
with the highest catalytic activity of SA
the performances listed in Table 1.
ACHTUNGTRNE(NUNG 500)–NEt2 among
In the SAACHTUNTRGENUG(N 500)–NEt2-catalyzed 1,4-addition reaction of 1
to 2, nonpolar solvents such as toluene and heptane were
good solvents, and the strongly electron-donating DMF sol-
vent gave a poor result.[16] In comparison, a typical solid
base of MgO showed high catalytic performances with polar
solvents. These results suggest that the SAACTHNUTRGEN(UNG 500)–NEt2-cata-
lyzed reaction mechanism involves not only proton abstrac-
tion by basic amine groups from nitroalkane but also activa-
tion of carbonyl compounds by surface strong acid sites of
SA (Scheme 5), which are easily deactivated by electron-do-
nating solvents.
Scheme 5. Proposed catalytic cycle for the 1,4-addition reaction with SA-
AHCTUNGTRE(NGNUN 500)–NEt2.
This acid–base bifunctional catalysis of SAACTHNUTRGENUG(N 500)–NEt2 was
Figure 6. FTIR spectra of A) SA
ACHTUNGTRENNUNG
ACHUTGTNRENN(GU 500)–NEt2, B) SAACHTUNGTREN(NGUN 120)–NEt2, C) SA-
also highlighted by the 1,4-addition reaction of thiol to an
electron-deficient alkene (Table 4). The reaction of 1-hexa-
decanethiol to 2-cyclohexen-1-one was promoted by acid
room temperature in a transmission mode. The solid was pressed into a
disk with KBr and subjected to IR measurements.
10876
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10871 – 10879

Silica–Alumina-Supported Tertiary Amines
FULL PAPER
Table 4. 1,4-Addition of 1-hexadecanethiol to 2-cyclohexen-1-one.[a]
Entry
Catalyst
Conversion [%][b]
Selectivity [%][b]
1
SA
SA
SA
N
97
70
60
trace
>99
>99
>99
–
2[c]
3[c]
4
NEt3
[a] Reaction conditions: 1-hexadecanethiol (1 mmol), 2-cyclohexen-1-one
(1.5 mmol), amine (4.5ꢁ10À2 mmol), toluene (1 mL), RT, 7 h. [b] Deter-
1
mined by H NMR spectroscopy. Based on the thiol. [c] 0.08 g of SA was
used.
sites on SAACHTUNGTRENNUNG
(500) (Table 4, entry 2).[17] The acidic perfor-
mance decreased by exposure to triethylamine due to deac-
tivation of acid sites (entry 3). The reaction hardly occurred
with only triethylamine (entry 4). However, the amine im-
mobilization of the SAACTHNUGRTENUNG(500) promoted excellent catalytic
performance to form the addition product in 97% yield
(entry 1). These results strongly indicate a significant contri-
bution of the weak interaction between the surface acidic
site and the NEt2 group to the catalytic function of SA-
Figure 7. Contact-time-array 13C CP/MAS NMR spectra for 1-hexa-
A
ACHTUNGTRNEN(UNG 500)–NEt2. Values with spectra are contact
ACHTUNGTRENNUNG(500)–NEt2, in which the amine group activates the nucleo-
AHCTUNGTRENNUNG
phile by proton abstraction,
thereby resulting in cooperative
promotion of addition reac-
tions. In fact, activation of the
SH group by the tertiary amine
for the 1,4-addition was consid-
ered to be the reason for the
improved performance. Shorter
contact-time values of the high-
est peak intensity of 1-hexa-
Scheme 6. Nitro-aldol reaction with SA(T)–NEt2 catalysts.
A
ACHTUNGTRNE(NUNG 500)–
A
atom in the case of a 1,2-addition is approximately 2.4 ꢂ.
The far positioning between amines and acid sites in SA-
(500)–NEt2 is not so effective for a 1,2-addition reaction,
such as a nitro-aldol reaction, regardless of its higher nu-
cleophilicity and acidity than SA(120)–NEt2.
in solid-state 13C CP/MAS NMR spectroscopic analysis
(Figure 7 and Table 3), which indicates the interaction be-
tween SH groups and immobilized tertiary amines.
AHCTUNGTRENNUNG
We also examined the 1,2-addition reaction of nitrome-
A
ACHTUNGTRENNUNG
thane to benzaldehyde with SA
NEt2 (Scheme 6). The catalytic activity and selectivity of
SA(500)–NEt2 are similar to those of SA(120)–NEt2 in spite
ACHUTGTNRENU(NG 500)–NEt2 and SAACHTUNGTERN(NUGN 120)–
N
A
Conclusion
of its better performance in 1,4-addition reactions. The
À
amount of weak acidic Si OH groups, which are amine-im-
Silica–alumina-supported tertiary amine catalysts were
found to be efficient heterogeneous acid–base bifunctional
catalysts for 1,4-addition reactions of nitroalkanes and thiol
to electron-deficient alkenes. The interaction between sur-
face acidic sites and basic amine groups can be controlled
by the pretreatment temperature of SA. The detailed
amine–acid interactions, which strongly affect catalytic per-
formance, are determined by variable-contact-time-array
measurements of solid-state 13C CP/MAS NMR spectrosco-
py. This simple preparation procedure of heterogeneous
acid–base bifunctional catalysts (Scheme 1) can be widely
applied to the design of new heterogeneous organic catalysts
copromoted by oxide support surfaces.
mobilizing sites, on SA
G
G
be 5 and 2 nm2, respectively.[7] It can be expected that the
amines are immobilized on the neighboring position of the
strong acid site (SiO(H+)Al) on SA,[3] thus the averaged dis-
tance between the amine-immobilizing site and the strong
acid site on SAACHTUNGTRENNUNG(120) and SAHCATUNGTRNE(NUGN 500) are calculated to be 4.1
and 5.7 ꢂ, respectively. These distances are not too different
from those in the 1,4-addition reactions of nucleophiles to
a,b-unsaturated carbonyl compounds, in which an estimated
distance from the carbonyl oxygen in electrophiles to the
carbon atom in nucleophiles is approximately 4.4 ꢂ. Howev-
er, an estimated distance from the oxygen to the carbon
Chem. Eur. J. 2009, 15, 10871 – 10879
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10877

Y. Iwasawa et al.
Acknowledgements
Experimental Section
This work was supported by JSPS (grant nos. 19760541, 21760626) and
the Global COE Program for Chemistry Innovation at the UT.
1H and 13C NMR spectra were obtained using JNM-AL400 spectrometers
at 400 MHz in CDCl3 with TMS as an internal standard. Solid-state 13C
and 29Si MAS NMR spectra (MAS rate=4 kHz) were recorded using a
Chemagnetics CMX-300 spectrometer operating at 75.5 and 59.7 MHz,
respectively. 13C MAS NMR spectra with cross-polarization (CP) were
acquired with a contact time of 5.0 ms. A single-pulse detection method
with hydrogen decoupling was used in 29Si NMR spectroscopic analyses,
in which the pulse duration was 1.5 ms. The rotor spin rate was 4 kHz,
with a delay time of 15 and 20 s for 13C and 29Si, respectively. Hexa-
[1] a) H. Grçger, Chem. Eur. J. 2001, 7, 5246; b) S. Kanemasa, K. Ito,
Eur. J. Org. Chem. 2004, 4741; c) J.-A. Ma, D. Cahard, Angew.
Chem. Int. Ed. 2004, 43, 4556; d) M. Kanai, N. Kato, E. Ichikawa,
M. Shibasaki, Pure Appl. Chem. 2005, 77, 2047; e) Y. Wang, H. Li,
Y.-Q. Wang, Y. Liu, B. M. Foxman, L. Deng, J. Am. Chem. Soc.
2007, 129, 6364.
ACHTUNGTRENNUNG
methylbenzene (13C: d=17.17 and 176.46 ppm) and TMS (29Si: d=
0 ppm) were used as external standards for the calibration of chemical
shifts. The accumulation numbers were fixed at about 20000 (13C) and
10000 (29Si). The variable-contact-time 13C CP/MAS NMR spectra were
measured with a rotor spin rate of 4 kHz, 2 s of delay time, and an accu-
mulation number of 2000. Infrared spectra were obtained using a JASCO
FTIR-410. Analytical GLC and GLC–mass were performed using a Shi-
madzu GC-2010 with a flame ionization detector.
[2] For a review, see: a) E. L. Margelefsky, R. K. Zeidan, M. E. Davis,
Chem. Soc. Rev. 2008, 37, 1118. See also: b) M. J. Climent, A.
Corma, V. Fornꢄs, R. Guil-Lopez, S. Iborra, Adv. Synth. Catal.,
2002, 344, 1090; c) S. Huh, H.-T. Chen, J. W. Wiench, M. Pruski,
V. S.-Y. Lin, Angew. Chem. 2005, 117, 1860; Angew. Chem. Int. Ed.
2005, 44, 1826; d) J. D. Bass, A. Solovyov, A. J. Pascall, A. Katz, J.
Am. Chem. Soc. 2006, 128, 3737; e) R. K. Zeidan, S.-J. Hwang, M. E.
Davis, Angew. Chem. 2006, 118, 6480; Angew. Chem. Int. Ed. 2006,
45, 6332; f) T. Takahashi, T. Watahiki, S. Kitazume, H. Yasuda, T.
Sakakura, Chem. Commun. 2006, 1664; g) R. K. Zeidan, M. E.
Davis, J. Catal. 2007, 247, 379.
[3] a) K. Motokura, M. Tada, Y. Iwasawa, J. Am. Chem. Soc. 2007, 129,
9540; b) K. Motokura, M. Tomita, M. Tada„ Y. Iwasawa, Chem. Eur.
J. 2008, 14, 4017; c) K. Motokura, M. Tada, Y. Iwasawa, Angew.
Chem. 2008, 120, 9370; Angew. Chem. Int. Ed. 2008, 47, 9230; d) K.
Motokura, M. Tada and Y. Iwasawa, J. Am. Chem. Soc. 2009, 131,
7944.
Unless otherwise noted, materials were purchased from Wako Pure
Chemicals, Tokyo Kasei Co., and Aldrich Inc., and were used after ap-
propriate purification. Amorphous silica–alumina (SA) was purchased
from Nikki Chemical Co., Japan as N633HN (SiO2 66.5, Al2O3 25.1%;
380 m2g)À1. Silica (Aerosilꢃ 300, 300 m2 gÀ1) and H-USY (Tosoh Co.,
Japan, HSZ-330HUA, Si/Al=3.17, 626 m2 gÀ1) were used. The identities
of products were confirmed by comparison with reported mass and NMR
spectroscopic data.
Preparation of SA–NEt2: SA was pretreated at 5008C for 3 h under re-
duced pressure. SA (1.0 g) was added to a solution of 3-(diethylamino)-
propyltrimethoxysilane in toluene (2 mmol, 20 mL) and heated at reflux
for 24 h. Then the solvent was removed by filtration and the functional-
ized SA was washed with dichloromethane, followed by drying under
[4] Cooperative heterogeneous catalysis of thiol/sulfonic acid pairing
with controlled distances was reported, see: E. L. Margelefsky, R. K.
Zeidan, V. Dufaud, M. E. Davis, J. Am. Chem. Soc. 2007, 129,
13691.
vacuum, which afforded SAACHTNUGTRNEUNG(500)–NEt2. Elemental analysis (%) of SA–
[5] For other reports on catalyzed 1,4-addition reaction of nitroalkanes,
see: a) B. M. Choudary, M. L. Kantam, C. R. V. Reddy, K. K. Rao, F.
Figueras, J. Mol. Catal. A: Chem. 1999, 146, 279; b) P. B. Kisanga, P.
Ilankumaran, B. M. Fetterly, J. G. Verkade, J. Org. Chem. 2002, 67,
3555; c) T. Jackson, J. H. Clark, D. J. Macquarrie, J. H. Brophy,
Green Chem. 2004, 6, 193; d) B. C. Ranu, S. Banerjee, Org. Lett.
2005, 7, 3049.
[6] a) E. J. R. Sudhçlter, R. Huis, G. R. Hays, N. C. M. Alma, J. Colloid
Interface Sci. 1985, 103, 554; b) C. E. Fowler, S. L. Burkett, S. Mann,
Chem. Commun. 1997, 1769.
NEt2: C 7.54, H 1.94, N 0.79. Other solid-supported amines (support–
NR2) were also synthesized by means of a similar procedure with differ-
ent pretreatment temperatures. Elemental analysis (%) for SA
NEt2: C 8.40, H 2.16, N 1.27; SA(200)–NEt2: C 8.35, H 2.15, N 1.03; SA-
(400)–NEt2, C 7.61, H 2.18, N 0.78; SiO2A(120)–NEt2: C 6.58, H 1.42, N
(500)–NEt2: C 7.31, H 1.51, N 0.96; USY–NEt2: C 7.38, H 2.19,
ACHTUNGTRNE(NUNG 120)–
AHCTUNGTRENNUNG
A
CHTUNGTRENNUNG
0.99; SiO
2ACHTUNGTRENNUNG
N 0.63.
1,4-Addition reaction of nitroalkanes to olefins using SA
ACHTUNGTRNE(NUNG 500)–NEt2: SA-
ACHTUNGTRENNUNG
(3.0 mmol), and methyl vinyl ketone (1.0 mmol) were placed into a Pyrex
glass reactor. The resulting mixture was vigorously stirred at 508C under
an N2 atmosphere. After 24 h, the catalyst was separated by filtration and
the 1H NMR spectroscopic and GC–MS analysis of the filtrate showed
85% yield of the addition product 5-nitro-2-hexanone.
[7] Weak acidic silanol groups on SA are readily desorbed by calcina-
tion. Removal of approximately 70% of silanols on the silica phase
by treatment at 5008C was estimated, see: Y. Iwasawa, Tailored
Metal Catalysts, Reidel, Dordrecht (The Netherlands), 1986. The cal-
À
culated amount of Si OH groups on SA
and 2 OHnmÀ2, respectively.
ACHTUGTNREN(UNG 120) and SACAHTUNGTREN(NGUN 500) were 5
1,4-Addition reaction of a thiol to an olefin using SA
ACHTUNGTRNE(NUNG 500)–NEt2: SA-
ACHTUNGTRENNUNG
[8] A side product, 1-(2-hydroxy-2,5-dimethyl-5-nitrocyclohexyl)-1-etha-
none, was formed through an intramolecular cyclization reaction
with a strong base, see: R. Ballini, L. Barboni, G. Bosica, P. Fili-
ppone and S. Peretti, Tetrahedron 2000, 56, 4095.
[9] a) M. Wada, A. Tsuboim K. Nishimura, T. Erabi, Nippon Kagaku
Kaishi 1987, 7, 1284; b) M. V. Berrocal, M. V. Gil, E. Roman, J. A.
Serrano, M. B. Hursthouse, M. E. Light, Tetrahedron Lett. 2005, 46,
3673.
thiol (1.0 mmol), and 2-cyclohexen-1-one (1.5 mmol) were placed into a
Pyrex glass reactor. The resulting mixture was vigorously stirred at room
temperature under an N2 atmosphere. After 7 h, the catalyst was separat-
ed by filtration and the 1H NMR spectroscopic analysis of the filtrate
showed 97% yield of the addition product.
(500)–NEt2 (8.0ꢁ10À2 g,
Preparation of thiol-adsorbed SAACHTNUGRTNENUG(500)–NEt2: SAACHTUGNTRENNUGN
4.5ꢁ10À2 mmol), toluene (1 mL), and 1-hexadecanethiol (4.5ꢁ
10À2 mmol) were placed into a Pyrex glass reactor. The resulting mixture
was vigorously stirred at room temperature. After 1 h, the solvent was
evaporated and the solid was dried under vacuum, thus giving the thiol-
[10] a) L.-Q. Wang, J. Liu, G. J. Exarhos, K. Y. Flanigan, R. Bordia, J.
Phys. Chem. B 2000, 104, 2810; b) S. K. Sahoo, D. W. Kim, J. Kumar,
A. Blumstein, A. L. Cholli, Macromolecules 2003, 36, 2777; c) H.-M.
Kao, T.-T. Hung, G. T. K. Fey, Maclomolecule 2007, 40, 8673.
[11] The 13C NMR spectroscopic signal of the terminal carbon atom of
the ethyl amine function is shifted upfield by the interaction be-
tween the protonic acid site and the nitrogen atom, for example d=
11.8 (free NEt3) to 8.84 ppm (NEt3·HCl).
adsorbed SAACHTUNGTRENNUNG
(500)–NEt2 for the solid-state 13C CP/MAS NMR spectro-
scopic measurements.
[12] The deference between the T1 and T2 silicon atom does not have
much effect on the molecular mobility of the amino group. The
amino group on SA
the reaction of DAPS with the SAACTHNUGNRTE(NUG 120) surface at room temperature
(120) with a T1 silicon atom can be prepared by
ACHTUNGTRENNUNG
10878
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10871 – 10879

Silica–Alumina-Supported Tertiary Amines
FULL PAPER
(see ref. [3]). The TCH value of the amino group with T1 silicon was
[16] The results of the solvent effect on the SAACTHNGUTERNN(UG 500)–NEt2-catalyzed 1,4-
similar to that of SA
[13] G. Crꢄpeau, V. Montouillout, A. Vimont, L. Mariey, T. Cseri, F.
Maugꢄ, J. Phys. Chem. B 2006, 110, 15172.
[14] The strong acid–base interaction between Brønsted acid sites on the
SA surface and a nitrogen atom of the amino moiety before immo-
bilization of DAPS prevents the loss of the strong acid function by
the reaction with alkoxysilane; see ref. [3].
(120)–NEt2.
addition of 1 to 2 to form 3 are as follows: toluene, 93%; n-heptane,
76%; 1,4-dioxane, 70%; DMF, 40%.
[17] For the acid-catalyzed addition of thiols, see: a) S. K. Garg, R.
Kumar, A. K. Chakraborti, Tetrahedron Lett. 2005, 46, 1721; b) A. T.
Khan, S. Ghosh, L. H. Choudhury, Eur. J. Org. Chem. 2006, 2226;
c) G. L. Khatik, G. Sharma, R. Kumar, A. K. Chakraborti, Tetrahe-
dron 2007, 63, 1200.
[15] S. C. Popescu, S. Thomson and R. F. Howe, Phys. Chem. Chem.
Phys. 2001, 3, 111.
Received: May 23, 2009
Published online: September 11, 2009
Chem. Eur. J. 2009, 15, 10871 – 10879
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10879