Dialkylaminopyridine-functionalized mesoporous silica nanosphere as an efficient and highly stable heterogeneous nucleophilic catalyst

Article abstract of DOI:10.1021/ja0524898

A new nucleophilic catalytic system comprised of dialkylaminopyridine- functionalized mesoporous silica nanosphere (DMAP-MSN) has been synthesized and characterized. We have demonstrated that this material is an efficient heterogeneous catalyst for Baylis-Hillman, acylation, and silylation reactions with good reactivity, product selectivity, and recyclability. We envision that this DMAP-functionalized mesoporous silica material can also serve as an effective heterogeneous catalyst for many other catalytic nucleophilic reactions.

Full text of DOI:10.1021/ja0524898

Published on Web 09/03/2005
Dialkylaminopyridine-Functionalized Mesoporous Silica
Nanosphere as an Efficient and Highly Stable Heterogeneous
Nucleophilic Catalyst
Hung-Ting Chen,†,‡ Seong Huh,†,‡ Jerzy W. Wiench,‡ Marek Pruski,‡ and
Victor S.-Y. Lin*,†,‡
Contribution from the Department of Chemistry and United States Department of
Energy Ames Laboratory, Iowa State UniVersity, Ames, Iowa 50011-3111
Received April 16, 2005; E-mail: vsylin@iastate.edu.
Abstract: A new nucleophilic catalytic system comprised of dialkylaminopyridine-functionalized mesoporous
silica nanosphere (DMAP-MSN) has been synthesized and characterized. We have demonstrated that
this material is an efficient heterogeneous catalyst for Baylis-Hillman, acylation, and silylation reactions
with good reactivity, product selectivity, and recyclability. We envision that this DMAP-functionalized
mesoporous silica material can also serve as an effective heterogeneous catalyst for many other catalytic
nucleophilic reactions.
1. Introduction
catalytic reactivity has been attributed to the difficulty of
controlling the degree of functionalization and the complication
due to the mass transport of reactants and products in-and-out
of the catalytic sites within the solid matrices. For example,
the catalytic reactivity of DMAP-functionalized polymers has
been shown to be solvent dependent due to the swelling and
contraction of the polymer matrices in different solvent environ-
ments.19
The MCM- and SBA-type silicas, with their regular structural
characteristics, high surface areas and tunable pore diameters,
represent a promising support for DMAP immobilization.
However, to the best of our knowledge, successful functional-
ization of DMAP on mesoporous silica surfaces has not yet been
reported. This can be attributed to the undesired protonation of
the pyridyl group of various trialkoxysilyl-derivatized dialkyl-
amino pyridines by the surface silanol group during the
conventional postsynthesis grafting reaction.
Recently, we have developed a co-condensation-based syn-
thetic method that provides control of the degree of organic
functionalization and particle morphology.20,21 By utilizing this
method, we report herein the synthesis, characterization, and
catalytic reactivity study of a 4-(dialkylamino)pyridine-func-
tionalized mesoporous silica nanosphere (DMAP-MSN) mate-
rial. Our synthetic approach allows us to circumvent the unde-
sired protonation of DMAP and produce a heterogeneous
catalyst for Baylis-Hillman, acylation, and silylation reactions
with superb reactivity, product selectivity, and recyclability.
Due to its high nucleophilicity, 4-(dimethylamino)pyridine
(DMAP) has been widely utilized as an efficient catalyst for
many important reactions,1-5 such as acylation, silylation,
tritylation, ester rearrangement, polymerization, Darkin-West
reaction, and Baylis-Hillman reaction. Several approaches for
immobilization of this remarkable homogeneous catalyst on
various organic and inorganic solid supports, such as polymers6-16
and sol-gel silica-based materials,17,18 have been reported in
the literature. However, the catalytic reactivities of most of these
systems have been shown to be lower than that of the DMAP
molecule in homogeneous solution reactions. The decrease in
† Department of Chemistry.
‡ Ames Laboratory.
(1) Hoefle, G.; Steglich, W.; Vorbrueggen, H. Angew. Chem. 1978, 90, 602-
615.
(2) Heinrich, M. R.; Klisa, H. S.; Mayr, H.; Steglich, W.; Zipse, H. Angew.
Chem., Int. Ed. Engl. 2003, 42, 4826-4828.
(3) Scriven, E. F. V. Chem. Soc. ReV. 1983, 12, 129-161.
(4) Spivey, A. C.; Arseniyadis, S. Angew. Chem., Int. Ed. Engl. 2004, 43,
5436-5441.
(5) Murugan, R.; Scriven, E. F. V. Aldrichim. Acta 2003, 36, 21-27.
(6) Hierl, M. A.; Gamson, E. P.; Klotz, I. M. J. Am. Chem. Soc. 1979, 101,
6020-6022.
(7) Delaney, E. J.; Wood, L. E.; Klotz, I. M. J. Am. Chem. Soc. 1982, 104,
799-807.
(8) Shinkai, S.; Tsuji, H.; Hara, Y.; Manabe, O. Bull. Chem. Soc. Jpn. 1981,
54, 631-632.
(9) Tomoi, M.; Akada, Y.; Kakiuchi, H. Makromol. Chem., Rapid Commun.
1982, 3, 537-542.
(10) Menger, F. M.; McCann, D. J. J. Org. Chem. 1985, 50, 3928-3930.
(11) Deratani, A.; Darling, G. D.; Frechet, J. M. J. Polymer 1987, 28, 825-
830.
(12) Deratani, A.; Darling, G. D.; Horak, D.; Frechet, J. M. J. Macromolecules
1987, 20, 767-772.
2. Experimental Section
(13) Guendouz, F.; Jacquier, R.; Verducci, J. Tetrahedron 1988, 44, 7095-
All reagents were purchased from commercial sources and used as
received without further purification. Benzene was dried by calcium
7108.
(14) Bergbreiter, D. E.; Li, C. Org. Lett. 2003, 5, 2445-2447.
(15) Bergbreiter, D. E.; Osburn, P. L.; Li, C. Org. Lett. 2002, 4, 737-740.
(16) Corma, A.; Garcia, H.; Leyva, A. Chem. Commun. 2003, 2806-2807.
(17) Rubinsztajn, S.; Zeldin, M.; Fife, W. K. Macromolecules 1990, 23, 4026-
4027.
(19) Benaglia, M.; Puglisi, A.; Cozzi, F. Chem. ReV. 2003, 103, 3401-3429.
(20) Huh, S.; Wiench, J. W.; Trewyn, B. G.; Song, S.; Pruski, M.; Lin, V. S. Y.
Chem. Commun. 2003, 2364-2365.
(21) Huh, S.; Wiench, J. W.; Yoo, J.-C.; Pruski, M.; Lin, V. S. Y. Chem. Mater.
2003, 15, 4247-4256.
(18) Rubinsztajn, S.; Zeldin, M.; Fife, W. K. Macromolecules 1991, 24, 2682-
2688.
9
10.1021/ja0524898 CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 13305-13311
13305

A R T I C L E S
Chen et al.
hydride and used after distillation. Dichloromethane was dried by
aluminum oxide column. All catalytic reactions were performed in a
screw-capped vial.
screw-capped vial at 50 °C for 24 h. After the reaction, the crude
reaction mixture was filtered and washed by copious amount of acetone.
The filtrate was concentrated under vacuum. The corresponding
products were isolated by column chromatography with ethyl acetate/
hexane as eluent.
General Procedure for DMAP-MSN-Catalyzed Acylation Reac-
tions. The mixture of DMAP-MSN (50 mg, 7.5 mol%), an alcohol of
choice (1 mmol), triethylamine (1.5 mmol), and dry benzene (2 mL)
were charged to a screw-capped vial and heated to 60 °C for 20 min.
Acetic anhydride (2 mmol) was added to the solution mixture via
injection. After 2.5 to 24 h, the crude reaction mixture was filtered on
a glass frit, and the catalyst was thoroughly washed with copious amount
of chloroform. The resulting filtrate was evaporated under vacuum.
The corresponding acetates were obtained by column chromatography
with ethyl acetate/hexane as eluent.
General Procedure for DMAP-MSN-Catalyzed Silylation Reac-
tions. An alcohol of choice (0.5 mmol), tert-butyldimethylsilyl chloride
(0.55 mmol), DMAP-MSN (20 mg, 6 mol %), and triethylamine (0.55
mmol) were added to a dichloromethane solution (2 mL) in a screw-
capped vial. The solution mixture was stirred at room temperature for
24 h. After the reaction, the reaction mixture was filtered and washed
by copious amount of chloroform. The filtrate was concentrated under
vacuum. The corresponding silyl ether products were isolated by column
chromatography with ethyl acetate/hexane as eluent.
Preparation of 4-[N-[3-(Triethoxysilyl)propyl]-N-methyl-amino]-
pyridine (DMAP-TES). A solution of 4-(N-methylamino)pyridine (5.0
g, 46.2 mmol) in 70 mL of dry tetrahydrofuran (THF) was added
dropwise to a suspension of sodium hydride (1.77 g, 73.9 mmol) in 30
mL of dry THF under N2 atmosphere at 0 °C ice bath. The solution
was stirred for additional 2 h at room temperature. A solution of
chloropropyltriethoxysilane (11.1 mL, 46.2 mmol) in 10 mL of dry
THF was introduced to the mixture at 0 °C ice bath via injection. After
injection, the solution was stirred further for 15 h at 70 °C. The solution
was filtered, evaporated in vacuo, and chromatographed on silica gel
with eluent (MeOH/CHCl3 ) 1/9 with 5% NEt3), to give pure DMAP-
TES as a liquid (8.85 g, yield ) 61.3%). The 1H and 13C NMR spectra
of the product were found identical with reported data.17,18
Synthesis of Dialkylaminopyridine-Functionalized Mesoporous
Silica Nanosphere Catalyst (DMAP-MSN). The mixture of cetyltri-
methylammonium bromide surfactant (CH3(CH2)15N(CH3)3Br, referred
to as CTAB) (2.0 g, 5.49 mmol), 2.0 M of NaOH (aq) (7.0 mL, 14.0
mmol) and H2O (480 g, 26.67 mol) was heated to 80 °C for 30 min.
To this clear solution, tetraethoxysilane (9.34 g, 44.8 mmol), DMAP-
TES (1.80 mL, 5.74 mmol) was added sequentially and rapidly via
injection. White solids were observed within 85 s upon mixing of the
initial opaque emulsion. The reaction was stirred vigorously at 80 °C
for 2 h followed by a hot filtration of the solution to yield the crude
DMAP-MSN product (white solid). The as-made material was washed
with copious amount of water and methanol and then dried under
vacuum. An acid extraction was performed with a methanolic solution
(100 mL) of concentrated hydrochloric acid (0.6 mL) and the as-made
DMAP-MSN material (1.0 g) at 60 °C for 3 h. The resulting surfactant-
removed DMAP-MSN was filtered, washed with water and methanol,
and dried under vacuum. The neutralization was conducted in saturated
sodium carbonate methanol solution (100 mL) of extracted DMAP-
MSN (1.0 g) at room temperature for 3 h. The neutralized DMAP-
MSN material was isolated by filtration, washed by water and methanol,
and dried under vacuum.
3. Results and Discussion
Synthesis and Characterization of DMAP-MSN Catalyst.
We first synthesized an organosilane, 4-[N-[3-(triethoxysilyl)-
propyl]-N-methyl-amino]pyridine (DMAP-TES), via a procedure
described in the Experimental Section (Scheme 1). To prepare
the DMAP-functionalized mesoporous silica nanosphere (DMAP-
MSN) material, DMAP-TES and tetraethoxysilane (TEOS) were
introduced to a sodium hydroxide aqueous solution with low
concentration of cetyltrimethylammonium bromide (CTAB)
under a reaction condition that we have reported previously.20,21
After an acid extraction of the CTAB surfactant from the as-
made material, the DMAP-MSN material was neutralized by
submerging in a saturated sodium carbonate methanolic solution
at room temperature for 3 h. The resulting solid was isolated
by filtration, washed by water and methanol, and dried under
vacuum.
The XRD measurement of DMAP-MSN showed a large (100)
peak and a weak broad peak representing a higher-order
diffraction (Figure 1a). The observed d100 value was 32.2 Å.
The observed low-intensity broad peak (2θ) at ∼4.5° could be
attributed to the overlapping (110) and (200) diffraction peaks,
typical of a disordered pore structure as we investigated and
reported previously.21 The TEM micrograph of the material also
confirmed its disordered pore structure (Figure 1b). The DMAP-
MSN material exhibited a spherical particle shape with an
average particle diameter of 400 nm as depicted in the SEM
micrograph (Figure 1c). The N2 surface sorption analysis of this
mono-functionalized MSN exhibited a typical type IV isotherm
without any significant hysteresis. The measured BET surface
area of DMAP-MSN is 835 m2/g and the BJH average pore
diameter is around 20 Å.
1
NMR Spectroscopy. The H and 13C NMR spectra of products in
solution were acquired on a Varian VRX 300 spectrometer. Solid-state
13C and 29Si NMR on DMAP-MSN utilized a Chemagnetics Infinity
400 MHz spectrometer equipped with 5 mm (Chemagnetics) and 1.8
mm (A. Samoson22) double tuned probes capable of MAS at 10 and
40 kHz, respectively. The fast MAS probe was used to carry out the
1
two-dimensional (2D) heteronuclear H-13C correlation (HETCOR)
experiment, where a spinning rate of 40 kHz provided sufficient
1
homonuclear H-1H decoupling during proton evolution without any
additional RF irradiation in the 1H channel, as previously described.23
In addition, a single π pulse at the 13C spin frequency was introduced
in the middle of the 1H evolution period to refocus the JC-H coupling.
One of the additional advantages of using the fast MAS is that low
power decoupling could be efficiently used during the acquisition of
13C signal.23,24 Other experimental details are given in section 3 and in
the figure captions, where νR is the MAS rate and νXRF denotes the
1
magnitude of radio frequency field applied to X nuclei (X ) H and
1
13C). The H, 13C, and 29Si chemical shifts are referenced to TMS at
0 ppm.
General Procedure for DMAP-MSN-Catalyzed Baylis-Hillman
Reactions. An enone of choice (0.5 mmol), aldehyde of choice (0.25
mmol), and DMAP-MSN (50 mg, 30 mol%) were added to a mixture
solution of tetrahydrofuran and water (2 mL, volume ratio ) 3/1) in a
Solid-State NMR Studies of DMAP-MSN Catalyst. The
structure of the organic functionality was detailed by comparing
the 13C liquid spectrum of DMAP-TES (Figure 2a) with the
13C MAS spectrum of DMAP-MSN (Figure 2b) acquired using
(22) Samoson, A. In Encyclopedia of Nuclear Magnetic Resonance; Grant, D.
M., Harris, R. K., Ed.; John Wiley & Sons: Chichester, 2002; Vol. 9, pp
59-64.
1
ramped-amplitude cross polarization (CP)25 and with the H-
(23) Trebosc, J.; Wiench, J. W.; Huh, S.; Lin, V. S. Y.; Pruski, M. J. Am. Chem.
Soc. 2005, 127, 3057-3068.
(24) Ernst, M.; Samoson, A.; Meier, B. H. Chem. Phys. Lett. 2001, 348, 293-
302.
(25) Metz, G.; Wu, X.; Smith, S. O. J. Magn. Reson. Series A 1994, 110, 219-
227.
9
13306 J. AM. CHEM. SOC. VOL. 127, NO. 38, 2005

Silica Nanosphere as a Nucleophilic Catalyst
A R T I C L E S
Scheme 1
13C HETCOR spectrum of the same sample (Figure 3). The
assignments of 13C resonances in solution DMAP-TES were
based on the observed chemical shifts (see Table 1). The shifts
obtained in solid-state NMR spectra of DMAP-MSN, which
are also listed in Table 1, match well with the solution data,
and thus confirm the presence of a DMAP ligand on the silica
surface. The additional peak observed at around 48 ppm in
Figure 2b most likely represents the methanol molecules
adsorbed on the surface during the synthesis. The changes in
peak intensities between solution and solid-state spectra reflect
the differences in CP dynamics for individual carbon sites.
The concentration of functional groups in DMAP-MSN was
measured from the relative intensities of Tn and Qn silicon groups
observed by 29Si MAS NMR under direct polarization (DP),
(Figure 2c). Tn and Qn denote silicon sites (tSiO)nSi(OH)4-n-mRm
with m ) 1 and m ) 0, respectively. To improve the sensitivity,
the 29Si DPMAS spectrum was acquired using the Carr-
Purcell-Meiboom-Gill (CPMG) train of echoes during the
acquisition period.23 By co-adding 10 echoes to construct the
enhanced spectrum, a sensitivity gain of approximately 3 was
achieved without any distortions due to transverse relaxation.
We first note that the presence of resonances representing T2
and T3 sites at around -60 and -66 ppm confirms that covalent
bonds exist between DMAP and the surface in this sample. By
deconvolution and integration of the 29Si spectrum, the following
molecular formula was obtained for DMAP-MSN: (SiO2)100
-
(H2O)8(C9H14N2)13. From this formula, the amount of organic
groups can be estimated at 1.2 ((0.1) per nm2 (1.6 ( 0.15
mmol‚g-1) and the number of silanol groups at 2.6 ((0.2) per
nm2.
The NMR spectra also show that the pyridine and amine
nitrogen atoms in DMAP-MSN are not protonated. In Table 1,
we compare the chemical shifts of carbons of 4-(dimethylami-
no)pyridine in CDCl3 solution with the corresponding shifts
observed in monoprotonated species (in the presence of trifluoro-
acetic acid) and diprotonated species (in the presence of tri-
fluoromethanesulfonic acid).26 After a prolonged exposure (3
weeks) of the DMAP-MSN sample to ambient humidity, an
upfield shift of about 10 ppm was observed for C8 and C10
resonances in the 1D and 2D 13C spectra (not shown), which
demonstrates that the pyridine nitrogen is susceptible to pro-
tonation under such conditions.
DMAP-MSN Catalyzed Baylis-Hillman Reaction. Bay-
lis-Hillman reaction is one of the most useful synthetic ways
for forming carbon-carbon bonds between aldehydes and R,â-
Figure 1. (a) XRD spectrum of DMAP-MSN. (b) TEM micrograph of
DMAP-MSN, Scale bar ) 50 nm, (c) SEM micrograph of DMAP-MSN,
Scale bar ) 2.0 µm, (d) TEM micrograph of DMAP-MSN after 10 runs of
acylations, Scale bar ) 50 nm.
(26) Dega-Szafran, Z.; Kania, A.; Nowak-Wydra, B.; Szafran, M. J. Mol. Struct.
1994, 322, 223-232.
9
J. AM. CHEM. SOC. VOL. 127, NO. 38, 2005 13307

A R T I C L E S
Chen et al.
Figure 2. (a) 13C NMR spectrum of DMAP-TES in CDCl3 solution. (b) 13C CPMAS spectrum of DMAP-MSN resulting from 12 000 scans acquired with
a delay of 1 s in a 5 mm probe (νR ) 10 kHz). During each CP period of 1.5 ms, νH was ramped between 16 and 40 kHz (in 11 steps), while νRCF was set
RF
to 36 kHz. The νHRF fields of 83 kHz and 65 kHz were applied to protons during initial excitation and high power decoupling, respectively. (c) 29Si DPMAS
spectrum of DMAP-MSN obtained with the same probe using CPMG acquisition (10 echoes). A total of 600 scans were collected with a delay of 300 s to
allow the complete relaxation of 29Si nuclei.
stoichoimetric or overstoichoimetric amounts of catalysts and
is usually performed in homogeneous solutions. Among various
ketone substrates, cycloalkenones are known for their low
reactivities for undergoing the Baylis-Hillman reaction cata-
lyzed by nucleophilic molecules, such as DABCO (1,4-
diazabicyclo [2,2,2]-octane).29 Furthermore, in the homogeneous
reaction of aryl aldehydes with methyl vinyl ketone (MVK)
catalyzed by DABCO, the formation of undesired products, such
as diadduct (compound 2 in Scheme 1) and Michael adduct
(compound 3), have been reported in the literature.30
DMAP, on the other hand, is able to efficiently catalyze the
Baylis-Hillman reaction of cycloalkenones in homogeneous
solutions.31 To construct a recyclable DMAP catalyst, Corma
and co-workers16 have recently reported on the synthesis of a
Figure 3. 1H-13C HETCOR spectrum of DMAP-MSN measured under
MAS at 40 kHz. During the CP period of 1.5 ms, the νHRF field was ramped
between 100 and 140 kHz (in 15 steps) and νCRF was set to 80 kHz. TPPM
decoupling with νHRF ) 10 kHz was used during the detection of 13C
signals, which were accumulated using a delay of 1 s between scans. The
hypercomplex method was employed to discriminate the sine and cosine
parts in the 1H dimension. The experiment was completed in approximately
80 h, using 10 mg of natural abundance sample.
DMAP-functionalized, Merrifield-type polystyrene resin that
could serve as an active catalyst for the Baylis-Hillman
reaction. However, this polymer-based DMAP catalyst showed
a lower reactivity in protic solvents in comparison with those
of DMAP-catalyzed homogeneous reactions.32,33 The low
reactivity was attributed to the contraction of polymer matrix
in protic solvents.16 Furthermore, a stoichoimetric amount of
the polymer-based DMAP catalyst was needed to obtain decent
yields within reasonable reaction times. In some cases, the
undesired diadduct and Michael adduct were also observed.
Table 1. 13C Chemical Shifts Observed in DMAP-MSN and in the
Reference Compounds (DMAP-TES and 4NMe2-Py)
sample
solvent
C1
C2
C3
C5
C6
C7,C11 C8,C10
In contrast, our DMAP-MSN material showed an excellent
catalytic performance (Table 2) for the Baylis-Hillman reaction
of aryl aldehydes and various R,â-unsaturated ketones. Only a
DMAP-TES CDCl3
DMAP-MSN Solid-state
4NMe2-Pya CDCl3
CF3COOH
7.8 20.1 54.0 37.5 153.5 106.6 149.8
8.8 19.3 53.0 35.8 153.8 106.1 147.8
38.4 153.6 106.0 149.1
40.1 153.6 105.6 139.9
in CDCl3
CF3SO3H
in CD3NO2
(27) Basavaiah, D.; Rao, A. J.; Satyanarayana, T. Chem. ReV. 2003, 103, 811-
891.
47.3 157.8 120.6 145.6
(28) Basavaiah, D.; Rao, P. D.; Hyma, R. S. Tetrahedron 1996, 52, 8001-
8062.
a Data taken from ref 26.
(29) Rezgui, F.; El Gaied, M. M. Tetrahedron Lett. 1998, 39, 5965-5966.
(30) Shi, M.; Li, C.-Q.; Jiang, J.-K. Chem. Commun. 2001, 833-834.
(31) Lee, K. Y.; Gong, J. H.; Kim, J. N. Bull. Korean Chem. Soc. 2002, 23,
659-660.
unsaturated ketones or esters. This reaction can be catalyzed
by organic bases, such as amines, pyridines, and phosphines,
without any assistance of metal.27,28 It typically requires
(32) Aggarwal, V. K.; Dean, D. K.; Mereu, A.; Williams, R. J. Org. Chem.
2002, 67, 510-514.
(33) Yu, C.; Liu, B.; Hu, L. J. Org. Chem. 2001, 66, 5413-5418.
9
13308 J. AM. CHEM. SOC. VOL. 127, NO. 38, 2005

Silica Nanosphere as a Nucleophilic Catalyst
A R T I C L E S
Table 2. DMAP-MSN Catalyzed Baylis-Hillman Reactiona
* Side products 2 and 3 are not observed in all reactions. a Reaction condition: aldehyde (0.25 mmol), R,â-unsaturated ketone (0.5 mmol), and catalyst
(50 mg, 30 mol %) in THF/H2O ) 3:1 (2 mL) at 50 °C for 24 h. b Isolated yield. c Aldehyde/ketone ) 1:4 at 50 °C for 3 d. d 3-(dimethylamino)propyl-
functionalized silica gel. e No reaction.
catalytic amount (30 mol%) of DMAP-MSN was needed for
the complete conversion of cyclic and aliphatic enones to the
corresponding Baylis-Hillman products (entry 1-3 in Table
2). In the case of methyl vinyl ketone (MVK), the desired
product was obtained in quantitative yield (entry 3 in Table 2)
within 24 h. No undesired diadduct and Michael adduct were
detected even at relatively high reaction temperature (50 °C).
Conversely, the same reaction catalyzed by DMAP in homo-
geneous solution gave rise to a mixture of products under the
same reaction condition (Scheme 2a). In addition to the diadduct
and Michael addition side products, the formation of insoluble
brown precipitates was also observed, which could be attributed
to products of the oligomerization of MVK and compound 1
catalyzed by DMAP at the elevated temperature. Using a
physical mixture of DMAP and fumed silica gel as catalysts
yielded a mixture of product as well. Based on these results,
the high yield and good product selectivity of DMAP-MSN
catalyzed reactions could not be due to the silanol groups on
the silica surface. Given the fact that the DMAP catalytic groups
are mainly located in the mesopores of MSN, we suspect that
9
J. AM. CHEM. SOC. VOL. 127, NO. 38, 2005 13309

A R T I C L E S
Scheme 2
Chen et al.
the observed product selectivity could be attributed to the
difference in the rate of diffusion to the “active sites” located
inside the pores between the aldehyde reactant and compound
1, which would serve as the reactant for the undesired side
reactions. To investigate this supposed “matrix effect”, we
carried out two DMAP- and DMAP-MSN-catalyzed Baylis-
Hillman reactions by using compound 1 (1.0 equiv) as reactant
to interact with MVK (2.0 equiv) in the aforementioned THF/
H2O solution at 50 °C for 24 h. Indeed, the reaction catalyzed
by homogeneous DMAP catalyst gave rise to a mixture of
products. On the contrary, no reaction was observed in the case
of DMAP-MSN. The results support the hypothesis that the
MSN matrix could regulate the reaction selectivity by prefer-
entially allowing certain reactants to access the catalytic sites.
The intrinsic steric and electronic properties of the aldehyde
reactants also strongly affect the product yields of the Baylis-
Hillman reaction catalyzed by our DMAP-MSN. In the case of
sterically hindered aryl aldehydes, e.g., o-nitrobenzaldehyde, a
lower product yield of 49% was observed (entry 4 in Table 2).
Compared to the case of p-nitrobenzaldehyde, the reactions of
the less activated or nonactivated aldehydes, such as p-
chlorobenzaldehyde and benzaldehyde, respectively, required
more methyl vinyl ketones and extended reaction time to obtain
the desired products (entry 6, 7 in Table 2). In contrast to the
reactivities of other reported, surface-supported DMAP catalysts,
the DMAP-MSN-catalyzed Baylis-Hillman reaction of a non-
activated aryl aldehyde (benzaldehyde) with MVK gave rise to
a high yield of the desired product with less catalyst loading
(0.3 equiv vs 1.0 equiv) and shorter reaction time (4 days vs
10 days).16 Therefore, our heterogeneous DMAP-MSN catalyst
exhibits higher reactivity than polymer-based DMAP catalyst
(entry 7 in Table 2). Among the examined substrates, the
reactivity of R,â-unsaturated ketones follows the order: methyl
vinyl ketone > cyclopentenone > cyclohexenone. To further
compare the catalytic performance of DMAP-MSN with other
silica-immobilized tertiary amines, we have examined a Baylis-
Hillman reaction catalyzed by 3-(dimethylamino)propyl-func-
tionalized silica gel (DMA-SiO2). A 20% yield (entry 8 in Table
2) was observed, which indicated that the reactivity of DMAP-
MSN (99% yield, entry 3 in Table 2) was indeed superior.
Unlike DMAP-MSN, pure MCM-41 silica showed no reactivity
in catalyzing the Baylis-Hillman reaction (entry 3, 9 in Table
2).
DMAP-MSN Catalyzed Acylation Reaction. Catalytic acy-
lations of various alcohols by DMAP-MSN were investigated,
and the results are outlined in Table 3. As shown in entry 1
and 2, the acylations of secondary alcohols proceeded rapidly
and completed within 2.5 h under our reaction condition. In
the absence of DMAP-MSN catalyst, however, only a small
amount of product (33%) was obtained under the same reaction
condition. As a control, pure inorganic MCM-41 silica, prepared
by a similar procedure, was also used as a catalyst for the
acylations. As shown in Table 3 (entry 3, 4), pure MCM-41
silica does not catalyze acylation. These results clearly indicated
that DMAP-MSN could efficiently catalyze acylation reactions
of various secondary alcohols. Contrary to the reactions with
secondary alcohols, lower yields and longer reaction times were
found in the cases of various tertiary alcohols (entry 5, 6 in
Table 3). Such a pronounced difference could be attributed to
the steric hindrance of carbinol as observed in homogeneous
reactions.34
Recyclability and Stability Studies of DMAP-MSN Cata-
lyst. To investigate the recyclability of DMAP-MSN catalyst,
1-(1-naphthyl)ethanol was chosen as a substrate because the
chemical shift of the R-proton of this molecule could be clearly
1
resolved from other peaks in the H NMR spectrum. After the
reaction reached completion, the catalyst was reused in 10
consecutive cycles, which included recovery by decanting the
aliquots after centrifuge and drying in air followed by the
reaction. No change in reaction yield was observed. The TEM
micrograph of the recovered DMAP-MSN catalyst (Figure 1d)
showed the same clear structure as the freshly synthesized
material. The result indicated that the mesopores of DMAP-
MSN were not destroyed by chemical or thermal decompositions
during the repeated use as a catalyst. To further evaluate the
stability of DMAP-MSN, cyclohexanol (100.2 mmol) was
introduced to a 200 mL benzene solution with acetic anhydride
(200.4 mmol) and DMAP-MSN (20 mg). The reaction was
stirred vigorously at 60 °C for 24 days. The resulting cyclohexyl
1
acetate product was obtained in 100% yield based on the H
NMR spectrum, which gave rise to a turnover number (TON)
of 3340 for 24 days. This result suggested that the catalytic
reactivity of DMAP-MSN could be maintained for an extended
period of time.
(34) Hassner, A.; Krepski, L. R.; Alexanian, V. Tetrahedron 1978, 34, 2069-
2076.
9
13310 J. AM. CHEM. SOC. VOL. 127, NO. 38, 2005

Silica Nanosphere as a Nucleophilic Catalyst
A R T I C L E S
Table 3. DMAP-MSN Catalyzed Acylation of Alcoholsa
a Reaction condition: alcohol (1 mmol), catalyst (50 mg, 7.5 mol %), Ac2O (2 mmol), and NEt3 (1.5 mmol) in dry benzene (2 mL) at 60 °C. b Isolated
yield.
Table 4. DMAP-MSN Catalyzed Silylation of Alcoholsa
a Reaction condition: alcohol (0.5 mmol), catalyst (20 mg, 6 mol %), t-butyldimethylsilyl chloride (0.55 mmol), and NEt3 (0.55 mmol) in dry CH2Cl2
(2 mL) at room temperature for 24 h. b Isolated yield.
4. Conclusion
DMAP-MSN Catalyzed Silylation Reaction. We also car-
ried out several silylation reactions of alcohols by using DMAP-
MSN as a catalyst (Table 4).35 The reaction yield of these
DMAP-MSN-catalyzed silylations appeared to be sensitive to
the steric hindrance of the alcohols. In the case of sterically
hindered alcohols, lower yields (entry 1-3 in Table 4) were
observed in comparison with those of less sterically demanding
alcohols. It is interesting to note that pure MCM-41 silica also
catalyzed the silylation reactions and gave rise to certain
amounts of silyl ethers. The result suggested that the acidic
surface silanol groups of MCM-41 could be responsible for
catalyzing the silylation reactions.
We have successfully synthesized and characterized DMAP
immobilized mesoporous silica catalyst, which showed high
recyclability and high turnover number for Baylis-Hillman,
acylation, and silylation reaction. We expect that this DMAP-
functionalized mesoporous silica material also can serve as an
efficient heterogeneous catalyst for many other catalytic nu-
cleophilic reactions.
Acknowledgment. This research was supported at Ames
Laboratory by the U.S. Department of Energy, Office of Basic
Energy Sciences, through the Catalysis Science Grant No. AL-
03-380-011 and under Contract W-7405-Eng-82.
(35) Chaudhary, S. K.; Hernandez, O. Tetrahedron Lett. 1979, 99-102.
JA0524898
9
J. AM. CHEM. SOC. VOL. 127, NO. 38, 2005 13311