Highly ordered aluminium-planted mesoporous silica as active catalyst for Biginelli reaction and formyl C-H insertion reaction with diazoester

Article abstract of DOI:10.1039/c0cp00565g

Al-planted MCM-41s (Al-M41s) with regular mesoporous structure and Si/Al ratio of 23-32 were successfully prepared by the template-ion exchange method in which the template/Si molar ratio and Si/Al ratio were adjusted at 1.44 and 5-15, and showed much hig

Full text of DOI:10.1039/c0cp00565g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/pccp | Physical Chemistry Chemical Physics
Highly ordered aluminium-planted mesoporous silica as active catalyst
for Biginelli reaction and formyl C–H insertion reaction with diazoesterw
Hiroaki Murata, Haruro Ishitani and Masakazu Iwamoto*
Received 14th May 2010, Accepted 20th September 2010
DOI: 10.1039/c0cp00565g
Al-planted MCM-41s (Al-M41s) with regular mesoporous
structure and Si/Al ratio of 23–32 were successfully prepared
by the template-ion exchange method in which the template/Si
molar ratio and Si/Al ratio were adjusted at 1.44 and 5–15, and
showed much higher catalytic activity for the titled reactions
than the other types of Al-M41s prepared by post-synthesis or
sol–gel methods.
catalytic activity were widely recognized in various reactions.7
For example, the catalytic activity for the dehydrative
dimerization of acetone or the reaction of formaldehyde with
phenol was found to be maximized at Si/Al = 77a or 70.7b The
combination of highly loading of Al into the silica wall with
the maintenance of the regular pore structure is one of the
challenging research topics in the heterogeneous catalysis field.
Here we tried to prepare Al-M41 with ordered mesoporous
structure and high Al contents by applying various prepara-
tion methods of the parent M41s and supporting methods of
Al ions. The sol–gel and post-synthesis methods were used as
well as the TIE method. The effect of the ratio of template/Si
was also investigated in the preparation of the parent M41 on
the basis of the finding that the high template/Si ratio in
sol–gel synthesis resulted in the highly ordered Al-M41,8
though the template/Si ratios were 0.1–0.3 in this experiments8
while was 0.72 in our previous study.4 The catalytic activity for
the Biginelli reaction was mainly employed as the indicator of
the effectiveness of each preparation method. The results
reported here will show that Al-M41 samples with low Si/Al
ratios and high regularity of pores could be prepared and
both factors were indeed important for appearance of high
catalytic activity.
Recently, Al-planted MCM-41 (Al-M41) catalysts prepared
by template-ion exchange (TIE) method1 were found to be
very active for C–H insertion reaction of aldehyde with
diazoester (the formyl C–H insertion reaction, Scheme 1)2
and Biginelli dihydropyrimidinone (DHPM) synthesis (the
Biginelli reaction, Scheme 2) by the present group.3,4 In the
latter reaction, the yield of desired DHPM 5a showed a
volcano-shaped dependence on the Si/Al atomic ratio of
Al-M41 and the highest yield 75% was attained at Si/Al =
35 as shown in Fig. 3, open circle.4 The decrease in the yield of
5a at or below Si/Al = 30 was due to the partial collapse of the
pore structure, that is, the drastic decrements in mesopore
volume evaluated from N2 adsorption isotherms with t-plot
method (Vp), specific surface area calculated with BET method
(SBET), and intensity of XRD diffraction as shown in the
previous study.
The effect of preparation method was first studied. Six kinds
of Al-M41s were prepared according to the previous studies
reporting Al-M41s with high SBET and high Al content
(Table 1).9,10 The sample name Al(x)-M41-y or -[z] in this
paper shows the conditions of initial Si/Al ratio in the
preparation (x), preparation method (y), and template/Si ratio
in preparation of the parent (z). Al-M41-a, -b and -c were
synthesized by post-synthesis methods using Al(Oi-Pr)3,
Al(acac)3 and AlMe3 as Al sources.9 Al-M41-d, -e and -f were
prepared by sol–gel methods with NaAlO2, Al(Oi-Pr)3 and
Al(OH)3.10 In all experiments C12H25NMe3Br was employed
as a template in order to minimize the influence of surfactant
on the catalytic activity. The high SBET and Vp could be
obtained in all experiments as shown in entries 1–6 of
Table 1. All of them showed catalytic activity for the Biginelli
reaction of 3b, 1b and 4a (Scheme 2). The yields of 5a,
however, were lower than that of Al(35)-M41-[0.72] prepared
by the TIE method (entry 7). It follows that the situations of
Al ions loaded by various methods would be strongly depen-
dent on the preparation methods; for example, the sol–gel
Al-M41s (entries 5 and 6 in Table 1) would have Al species
buried in the silica wall, while the TIE method would give Al
species separately supported on the wall, though there was no
direct evidence for the speculation at the present. The prepara-
tion conditions of TIE were therefore investigated in more detail
to enhance the catalytic activity, though the characterization of
respective Al ions remained to be clarified in near future.
The incorporation of Al into silica-based mesoporous
materials was frequently applied to prepare effective acid
catalysts, and several preparation methods such as sol–gel
method,5 post-synthesis,6 and the TIE method1 were reported.
However, the partial destruction of pore structure owing to
too much Al contents and the resulting decrease in the
Scheme 1 Formyl C–H insertion reaction of diazoester.
Scheme 2 Biginelli reaction.
Chemical Resources Laboratory, Tokyo Institute of Technology,
4259-R1-5 Nagatsuta, Midori-ku, Yokohama, 226-8503, Japan.
E-mail: iwamoto@res.titech.ac.jp; Fax: +81 459245228;
Tel: +81 459245225
w Electronic supplementary information (ESI) available: Nitrogen
adsorption isotherms, XRD patterns and 27Al NMR spectra of the
catalysts and correlations between catalytic activity and physico-
chemical properties. See DOI: 10.1039/c0cp00565g
c
14452 Phys. Chem. Chem. Phys., 2010, 12, 14452–14455
This journal is the Owner Societies 2010

View Article Online
Table 1 Physicochemical properties and catalytic activity of Al-M41s prepared by post, sol–gel, or TIE methods
Catalysta
Physicochemical properties
Method Al source Ref. [Al]/mmol gÀ1 Si/Al SBET/m2 gÀ1 Vp mm3 gÀ1 Db/nm Int100c/103 cps
GC
yieldd(%)
Entry
1
2
3
4
5
6
7
Al(30)-M41-a
Al(30)-M41-b
Al(30)-M41-c
Al(25)-M41-d
Al(14)-M41-e
Al(3)-M41-f
Post
Post
Post
Al(Oi-Pr)3 9a
Al(acac)3 9b
0.35
0.35
0.40
25
25
34
20
12
1012
968
1011
969
825
799
896
566
532
606
570
488
453
406
2.2
2.3
2.3
2.5
1.9
2.2
2.2
29
29
28
11
7.2
18
24
38
47
45
48
38
49
75
AlMe3
9c
Sol–gel NaAlO2
Sol–gel Al(Oi-Pr)3 10b 1.2
Sol–gel Al(OH)3 10c 3.3
Al(NO3)3 0.38
10a 0.62
3.2
35
Al(35)-M41-[0.72] TIE
1
a
Numbers in the parentheses indicate initial Si/Al atomic ratios and those in the square brackets initial C12H25NMe3Br/SiO2 molar ratios.
Estimated from the N2 desorption isotherms with the Barrett–Joyner–Halenda (BJH) method. The values estimated from the adsorption
b
c
d
isotherms were almost the same as those listed in the table within the errors of 0.1 nm. Relative intensities of X-ray (100) diffraction. Reaction
conditions: catalyst 50 mg, 3b 1.0 mmol, 1b 1.0 mmol, 4a 0.50 mmol, octane 1.0 mL, 388 K, 10 h.
The correlation of the template/Si molar ratio with the
regularity of pores or the catalytic activity was examined.
Three types of parent M41s were prepared on the conditions
of the template/Si ratios 0.36, 0.72 and 1.44. It should be noted
that the experiment with template/Si ratio of 1.8 did not give
correct results because of insufficient dissolving of template
ion. The TIE treatment was carried out at the initial Si/Al
ratio of 30 (Table 2, entries 1–3). The pore structure and
catalytic activity were measured after the TIE treatment and
calcination at 873 K for 6 h. The SBET and XRD peak intensities
of the samples were in the order Al(30)-M41-[1.44] 4
Al(30)-M41-[0.72] 4 Al(30)-M41-[0.36] (Fig. 1). The catalytic
activity were Al(30)-M41-[1.44]
= Al(30)-M41-[0.72] 4
Al(30)-M41-[0.36] within the experimental errors. The M41
prepared at template/Si = 1.44 was mainly employed as the
parent for the following study.
Fig. 1 X-Ray diffraction patterns of Al(30)-M41s prepared at various
template/Si molar ratios.
The effect of the concentration of Al(NO3)3 in the TIE
treatment was next studied. As shown in Table 2, entries 4–6,
the amounts of Al added in the solution little affected the Si/Al
ratios of the resulting Al-M41s, while surprisingly increased
greatly the Vp values in spite of the approximately constant
The small change in SBET and great increase in Vp indicate the
increase in internal surface area of pores, which was confirmed
in the separated experiments as shown later. The second point
is the question which the increment in Vp resulted from the
high Al concentration or the low pH of the solution, since the
pH values in entries 4–6 were lower than those of entries 1–3.
This subject would also be studied later. Before the detailed
structural studies, the catalytic activities for the Biginelli
reaction were compared. The yields of DHPM on
Al(5)-M41-[0.72], Al(15)-M41-[1.44] and Al(5)-M41-[1.44]
were 75, 88 Æ 2 and 87 Æ 1%, respectively. Clearly the
S
BET. The Vp values of Al(5)-M41-[0.72], Al(15)-M41-[1.44]
and Al(5)-M41-[1.44] were 601–639 mm3 À1, which can be
g
compared with the values of 340–379 of entries 1–3. This is
clearly shown in Fig. 2 where the amounts of nitrogen
adsorbed at P/P0 = 0.23 onto the Al(5)-M41s were much
greater than those of Al(30)-M41s (Fig. 2). The results were
sufficiently reproducible within the respective error shown
in the table.11 Two findings/discussion should be noted here.
Table 2 Effects of template/Si molar ratio and initial Si/Al atomic ratio on physicochemical properties and catalytic activity
Entry Al-M41a
pHb
[Al]/mmol gÀ1 Si/Al
SBET/m2 gÀ1 Vp/mm3 gÀ1 Dc/nm
Int100d/103 cps GC yielde (%)
1
2
3
4
5
6
7
Al(30)-M41-[0.36]
6.9
7.0
0.38
0.44
30
29
731
791
340
379
2.3
2.0
15
22
29
66
Al(30)-M41-[0.72]
Al(30)-M41-[1.44]
Al(5)-M41-[0.72]
Al(15)-M41-[1.44]
Al(5)-M41-[1.44]
7.2 Æ 0.4 0.45 Æ 0.03
29 Æ 0 832 Æ 38
375 Æ 14
2.1 Æ 0.0 28 Æ 2
60 Æ 1
3.2 0.32
3.4 Æ 0.1 0.44 Æ 0.02
3.0 Æ 0.1 0.48 Æ 0.05
38 1009
601
2.2 25
75
29 Æ 0 892 Æ 34
26 Æ 3 1057 Æ 16
57 Æ 4 899 Æ 32
639 Æ 15
617 Æ 9
598 Æ 41
2.3 Æ 0.1 26 Æ 2
2.2 Æ 0.1 31 Æ 2
2.2 Æ 0.2 31 Æ 3
88 Æ 2
87 Æ 1
66 Æ 5
Al(30)-M41-[1.44]-pH 3.0 Æ 0.1 0.26 Æ 0.03
a
b
Numbers in the parentheses indicate initial Si/Al atomic ratios and those in the square brackets initial C12H25NMe3Br/SiO2 molar ratios. pH
c
values of TIE solution. Estimated from the N2 desorption isotherms with the BJH method. The values estimated from the adsorption isotherms
d
e
were almost the same as those listed in the table within the errors of 0.1 nm. Relative intensities of X-ray (100) diffraction. Reaction conditions:
Al-M41 50 mg, 3b 1.0 mmol, 1b 1.0 mmol, 4a 0.50 mmol, octane 1.0 mL, 388 K, 10 h.
c
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 14452–14455 14453

View Article Online
hand, the Vp values of Al(30)-M41-[1.44]-pH was comparable
to those of entries 4–6, clearly indicating that the treatment of
the parent M41 with an acidic solution resulted in the higher
regularity of pore structure. The yield of DHPM 5a, however,
was lower than those in entries 4–6. This would be due to the
low Al contents as shown in the next paragraph. All results
obtained on Al-M41s newly prepared in the present work and
those reported in the previous work4 were plotted in Fig. 3 as a
function of the Si/Al ratio.11 Closed circles and open triangles
show the results on Al-M41-[1.44] and Al-M41-[1.44]-pH.
Open circles are those on Al-M41-[0.72]. It follows that the
volcano-shaped correlation between the Si/Al ratio and
the catalytic activity, the solid line, could be improved to the
level of the dotted line through the present preparation
method.
To reveal the factors controlling the catalytic activity, the
yields of 5a were plotted as a function of the Vp values, the
intensity of (100) XRD peak (Int100), and the SBET values.12
No correlation between the Int100 or SBET values and the
yields could be observed. The long-range regularity of
Al-M41, measured by Int100, would not be significant for the
appearance of catalytic activity. The total surface areas SBET
would not be the good index of the present catalysis because
the values are the sum of outer and inner surface areas of
mesoporous structure. In contrast, the very good relationship
between the Vp values and the yield of 5a could be recognized.
Fig. 4 shows the dependence of the yield of 5a on the Si/Al
ratio and the VP value. The Al-M41s having given the 85%
or more yields were plotted by closed circles in the figure.
They were located at the positions of low Si/Al ratios and
large mesopore volumes. The results indicate that the reaction
took place in the pores of M41 and the important factor for
the high catalytic activity would be maintenance of pore
structures and the high Al contents. This was further
confirmed by the Sint of Al-M41s which could be calculated
from the difference between SBET and the external surface area
(Sext) determined by the t-plot method. Similar correlation
could be observed (Fig. S11 in ESI).w The reason why
high template/Si ratio and high Al concentration was essential
for the high catalytic activity should be resolved in the near
future.
Fig. 2 N2 adsorption–desorption isotherms of Al(30)-M41-[0.72],
Al(30)-M41-[1.44], Al(5)-M41-[0.72] and Al(5)-M41-[1.44].
preparation with high-Al-solutions was very effective for
enhancement of the catalytic activity, though the amounts of
Al loaded on the silica surface did not increase.
The reason for the increments in pore volume and catalytic
activity in entries 4–6, Table 2 was studied. The pH value of
TIE solution was adjusted at 3.0 by addition of an aq. HNO3
solution in entry 7 of Table 2 where the initial Si/Al ratio was
adjusted at 30. The Si/Al ratio of the resultant sample was
57 Æ 4. The higher Si/Al values than those of the usual
treatment would be due to the presence of much amount of
proton as already reported in the literature.1 On the other
Fig. 3 Dependence of yield of 5a on Si/Al ratio of Al-M41 catalysts
prepared by the TIE method. Open circle, Al-M41-[0.72]. Closed
circle, Al-M41-[1.44]. Open triangle, Al(30)-M41-[1.44]-pH.
Fig. 4 Effect of Si/Al ratio and mesopore volume Vp on the catalytic
activity of the Al-M41s.
c
14454 Phys. Chem. Chem. Phys., 2010, 12, 14452–14455
This journal is the Owner Societies 2010

View Article Online
Notes and references
1 (a) M. Iwamoto and Y. Tanaka, Catal. Surv. Jpn., 2001, 5, 25;
(b) M. Yonemitsu, Y. Tanaka and M. Iwamoto, J. Catal., 1998,
178, 207; (c) M. Yonemitsu, Y. Tanaka and M. Iwamoto, Chem.
Mater., 1997, 9, 2679.
2 H. Murata, H. Ishitani and M. Iwamoto, Tetrahedron Lett., 2008,
49, 4788.
3 (a) C. O. Kappe, in Multicomponent Reactions, ed. J. Zhu and
H. Bienaymie, Wiley-VHC, Weinheim, 2005, pp. 95–120;
(b) C. O. Kappe and A. Stadler, Org. React., 2004, 63, 1;
(c) C. O. Kappe, Tetrahedron, 1993, 49, 6937; (d) L.-Z. Gong,
X.-H. Chen and X.-Y. Xu, Chem.–Eur. J., 2007, 13, 8920.
4 H. Murata, H. Ishitani and M. Iwamoto, Org. Biomol. Chem.,
2010, 8, 1202.
5 (a) R. Ryoo, S. Jun, J. M. Kim and M. J. Kim, Chem. Commun.,
1997, 2225; (b) M. V. Landau, E. Dafa, M. L. Kaliya, T. Sea and
M. Herskowitt, Microporous Mesoporous Mater., 2001, 49, 65;
(c) A. Goldbourt, M. V. Landau and S. Vega, J. Phys. Chem. B,
2003, 107, 724; (d) S. Ito, H. Yamaguchi, Y. Kubota and
M. Asami, Tetrahedron Lett., 2009, 50, 2967; (e) K. Iwatani,
T. Sakakura and H. Yasuda, Catal. Commun., 2009, 10, 1990.
Fig. 5 Yields of the formyl C–H insertion reaction on Al(5)-M41-
[1.44] (closed circle) and Al(30)-M41-[0.72] (open circle). Reaction
conditions: Al-M41 20 mg, 1a 8.0 mmol, 2a 12 mmol, CH2Cl2 40
mL, 298 K. Determined by GC analysis.
6 (a) A. Corma, V. Fornes, M. T. Navarro and J. Perez-Paariente,
´ ´
J. Catal., 1994, 148, 569; (b) Z. Luan, H. He, W. Zhou,
C.-F. Cheng and J. Klinowski, J. Chem. Soc., Faraday Trans.,
1995, 91, 2955; (c) J. Weglarski, J. Datka, H. He and J. Klinowski,
J. Chem. Soc., Faraday Trans., 1996, 92, 5161; (d) R. Mokaya,
W. Jones, Z. Luan, M. D. Alba and J. Klinowski, Catal. Lett.,
1996, 37, 113; (e) Z. Luan, C.-F. Cheng, W. Zhou and
J. Klinowski, J. Phys. Chem., 1995, 99, 1018; (f) R. B. Borade
and A. Clearfield, Catal. Lett., 1995, 31, 262; (g) T. Kugita,
M. Ezawa, T. Owada, Y. Tomita and S. Namba, Microporous
Mesoporous Mater., 2001, 44–45, 531.
7 (a) H. Kosslick, G. Lischke, B. Parlitz, W. Storek and R. Fricke, Appl.
Catal., A, 1999, 184, 49; (b) S. K. Jana, T. Kugita and S. Namba, Appl.
Catal., A, 2004, 266, 245; (c) M. M. L. R. Carrott, F. L. Conceic¸ ao,
J. M. Lopes, P. J. M. Carrott, C. Bernardes, J. Rocha and
F. R. Riberio, Microporous Mesoporous Mater., 2006, 92, 270.
8 G. A. Eimer, L. B. Pierella, G. A. Monti and O. A. Anunziata,
Catal. Commun., 2003, 4, 118.
The effect of coordination of Al ions was studied by
using the solid-state 27Al-MAS NMR technique because tetra-
hedrally coordinated Al (AlO4) ion is often suggested as an
important acidic site.13 Both of AlO4 and AlO6 species could
be observed on the present samples but no significant correla-
tion with the catalytic activity was recognized (Fig. S12 in
ESI).w The results observed in Table 1 and Fig. 4 clearly show
the importance of coordination state of aluminium on the
silica wall but they were still unclear.
The catalytic activity of the present Al-M41 for the formyl
C–H insertion reaction2 were compared with the previous
catalyst in Fig. 5, in which the reaction of n-heptaldehyde
(1a) with ethyl diazo acetate (2a) was selected as the model
reaction. The reaction rate on the Al(5)-M41-[1.44] catalyst
was two times greater than that on the reported one
(Al(30)-M41-[0.72]) as shown in Fig. 5. The amount of
Al(5)-M41-[1.44] employed was only 2.5 mg per mmol of 1a
and the yield reached to 100%. As long as we know, the
amount of catalyst used in the figure is the least among those
of the reported heterogeneous catalysts.14,15
9 (a) R. Mokaya and W. Jones, Chem. Commun., 1997, 2185;
(b) A. Ortlar, M. Wark, G. Schulz, E. Kloff, J. Rathousky´ and
A. Zukal, Stud. Surf. Sci. Catal., 1998, 117, 357; (c) Y. Oumi,
H. Takagi, S. Sumiya, R. Mizuno, T. Uozumi and T. Sano,
Microporous Mesoporous Mater., 2001, 44–45, 267.
10 (a) C. Blanco, C. Pesquera and F. Gonzalez, Stud. Surf. Sci. Catal.,
´
2004, 154, 432; (b) C. Perego, S. Amarilli, A. Carati, C. Flego,
G. Pazzuconi, C. Rizzo and G. Bellussi, Microporous Mesoporous
Mater., 1999, 27, 345; (c) S. Biz and M. G. White, Microporous
Mesoporous Mater., 2000, 40, 159.
In summary, the enhancement of catalytic activity of
Al-M41 was achieved by improvement of the TIE method.
The parent M41 should be prepared at template/Si = 1.44.
The TIE treatment in a low pH solution with high Al contents
was also the key. The resulting Al-M41s with the Si/Al ratio of
20–30 have the large pore volumes and showed high catalytic
activity for the Biginelli reaction and the formyl C–H insertion
reaction.
11 The details are described in Table S1, ESIw.
12 See Fig. S8–10, ESIw.
13 (a) A. Corma, Chem. Rev., 1997, 97, 2373; (b) F. S. Xiao, Top.
Catal., 2005, 35, 9; (c) Z. Y. Wu, H. J. Wang, T. T. Zhuang,
L. B. Sun, Y. M. Wang and J. H. Zhu, Adv. Funct. Mater., 2008,
18, 82; (d) A. Yin, X. G. Guo, W. L. Dai and K. Fan, J. Phys.
Chem. C, 2010, 114, 8523.
14 The amount of solid catalysts used in literatures was more than 10
mg-cat mmol-aldehydeÀ1 2,15
.
15 (a) D. D. Dhavale, P. N. Patil and R. S. Mali, J. Chem. Res.,
Synop., 1994, 152; (b) B. S. Balaji and B. M. Chanda, Tetrahedron,
1998, 54, 13237; (c) P. Phukan, J. M. Mohan and A. Sudalai,
J. Chem. Soc., Perkin Trans. 1, 1999, 3685.
The authors gratefully acknowledge financial support
from the ministry of Education, Culture, Sports, Science and
Technology of Japan.
c
This journal is the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 14452–14455 14455