Catalytic properties of dendron-OMS hybrids

Article abstract of DOI:10.1016/j.jcat.2009.10.009

The synthesis and catalytic testing of several dendron-ordered mesoporous silica hybrids are reported. These materials are active in both the nitroaldol (Henry) reaction and the transesterification of glyceryl tributyrate to afford methyl esters. In both reactions it is observed that dendrons terminated with primary amines are more catalytically active than samples containing dendrons terminated with secondary amines. On a mmol nitrogen per gram of silica basis, the first generation dendrons are the most active for both chemistries, and the SBA-15 samples display a higher activity than the MCM-41 samples. The pore-size effect observed is consistent with significant diffusion resistance in the MCM-41 samples. The activity trend observed in the SBA-15 materials is consistent with decreased cooperative effects between the amines and surface silanols as the dendrons become larger. Clear trends are observed indicating that higher generation dendrons are more selective to alcohol formation in the Henry reaction. The dendron catalysts are much more active and stable than simple amines attached to silica in the transesterification of triglycerides. Preliminary results shown indicate that these materials can also catalyze more demanding chemistries, an example of which is the Aldol condensation of 5-(hydroxymethyl)furfural and acetone. The results shown indicate that dendron-OMS hybrids can serve as effective solid base catalysts for a diverse range of chemistries.

Full text of DOI:10.1016/j.jcat.2009.10.009

Journal of Catalysis 269 (2010) 15–25
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Catalytic properties of dendron–OMS hybrids
*
Qingqing Wang, Victor Varela Guerrero, Anirban Ghosh, Seunguk Yeu, Jonathan D. Lunn, Daniel F. Shantz
Artie McFerrin Department of Chemical Engineering, Texas A&M University, College Station, TX 77843-3122, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis and catalytic testing of several dendron–ordered mesoporous silica hybrids are reported.
These materials are active in both the nitroaldol (Henry) reaction and the transesterification of glyceryl
tributyrate to afford methyl esters. In both reactions it is observed that dendrons terminated with pri-
mary amines are more catalytically active than samples containing dendrons terminated with secondary
amines. On a mmol nitrogen per gram of silica basis, the first generation dendrons are the most active for
both chemistries, and the SBA-15 samples display a higher activity than the MCM-41 samples. The pore-
size effect observed is consistent with significant diffusion resistance in the MCM-41 samples. The activ-
ity trend observed in the SBA-15 materials is consistent with decreased cooperative effects between the
amines and surface silanols as the dendrons become larger. Clear trends are observed indicating that
higher generation dendrons are more selective to alcohol formation in the Henry reaction. The dendron
catalysts are much more active and stable than simple amines attached to silica in the transesterification
of triglycerides. Preliminary results shown indicate that these materials can also catalyze more demand-
ing chemistries, an example of which is the Aldol condensation of 5-(hydroxymethyl)furfural and ace-
tone. The results shown indicate that dendron–OMS hybrids can serve as effective solid base catalysts
for a diverse range of chemistries.
Received 2 September 2009
Revised 8 October 2009
Accepted 8 October 2009
Available online 14 November 2009
Keywords:
Ordered mesoporous silica
Organocatalysts
Dendrons
Aldol reactions
Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction
the Stein Lab is also particularly noteworthy in this area [10]. More
recently periodic mesoporous organosilicas (PMOs) have been
Ordered mesoporous silicas (OMS) have attracted considerable
interest in the last 15 years as potential adsorbents, catalysts/cata-
lyst supports, sensors, etc. [1–8]. These materials possess several
highly desirable properties: pore topologies that possess long-
range structural ordering, uniform pore sizes in the 2–10 nm range,
and high surface areas. While the pore structure often possesses
long-range ordering the oxide phase is in fact amorphous. Another
aspect of considerable interest in the current work is that the sila-
nol groups present can be readily functionalized using well-known
silane chemistry [3]. Thus these OMS phases represent a model
substrate for fabricating well-defined organic–inorganic hybrid
materials.
Numerous laboratories have reported on the synthesis of organ-
ic-functionalized OMS materials. Early works investigated materi-
als containing simple chemical groups (e.g. amines, vinyl groups)
and the properties of the resulting material [9–12]. Many works
have compared the properties of OMS phases functionalized via
post-synthetic grafting versus those made via co-condensation
wherein the organic group is incorporated directly during synthe-
sis. A fairly recent review by Hoffman et al. [2] has summarized
many aspects of functionalized OMS materials. An early paper from
developed, which represent composite materials with extremely
high organic loadings.
Numerous laboratories have reported the synthesis and cata-
lytic testing of OMS materials with organic groups attached. This
literature has expanded enormously in recent years, and several
reviews are available for the interested reader [1,4,13]. Given the
scope of the current work, the area of amine–OMS materials will
be briefly summarized. Many laboratories, including those of Brun-
el, Lin, Katz, Shimizu, Yoshitake, and Corma, have investigated the
activity of amines on OMS phases for a variety of reactions includ-
ing Aldol chemistry and Michael additions. [14–27]. The second
body of work has focused on proline-derivatized and alkaloid-
derivatized surfaces with the aim of making solid base catalysts
for the production of chiral molecules [27,28]. A few laboratories,
most notably the laboratories of Davis and Lin, have also investi-
gated the synthesis of OMS materials containing multiple func-
tional groups and their catalytic efficacy [20,21,29–32].
The fabrication of OMS-hybrid materials with high densities of
organic functional groups is difficult in the sense that the distribu-
tion of catalytically active sites often becomes heterogeneous.
Work by Jones’ lab [33–35] and Katz’s lab [14,15,36] has shown
that surface patterning and molecular imprinting techniques to
achieve higher densities of homogeneous organic functional
groups can circumvent this issue. Another possible approach to
* Corresponding author. Fax: +1 979 845 6446.
E-mail address: Shantz@chemail.tamu.edu (D.F. Shantz).
0021-9517/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2009.10.009

16
Q. Wang et al. / Journal of Catalysis 269 (2010) 15–25
obtain high organic group densities is the use of surface-tethered
dendrimers. Linden’s and Jones’ laboratories have reported ring-
opening polymerization of aziridine groups to generate hyper-
branched polymers, although their catalytic properties are yet to
be reported [37–40]. Recently our group has reported the immobi-
lization of melamine-based dendrimers on the surfaces of amine-
functionalized SBA-15 materials [41–43]. At approximately the
same time, collaborative work between the Alper and Sayari labo-
ratories reported similar investigations of PAMAM-based dendrons
in large pore MCM-41 materials [44,45]. The current investigation,
building off our previous reports, describes the catalytic testing of
these materials in the Henry (nitroaldol) reaction, the synthesis of
methyl esters from triglycerides, and the cross aldol reaction be-
tween acetone and 5-hydroxymethylfurfural.
the method reported by Zhao et al. [48]. As an example for the syn-
thesis of MCM-41, 7.9 g of sodium silicate solution was mixed with
45.4 mL deionized water. NaOH (0.27 g) was added to the solution
followed by the addition of 7.8 mL of 1 M H2SO4. CTAB (7.29 g) was
dissolved in the solution and stirred for 15 min at room tempera-
ture. The mixture was then placed in an oven at 100 °C for 24 h un-
der static conditions. After 24 h the sample was removed from the
oven, allowed to cool sufficiently so that it could be easily handled,
and titrated to a pH of approximately 10 using 1 M H2SO4. The
sample was then placed back in the oven at 100 °C. The titration
step was performed two additional times in regular 24-h intervals.
The total heating period was 96 h. For the SBA-15 synthesis, 4.0 g
of Pluronic P123 was dissolved in 60 mL of 4 M HCl and 85 mL of
deionized water by stirring for 5 h at room temperature. Then,
8.5 g of TEOS was added to that solution and stirred for 24 h at
35 °C. The mixture was then aged at 80 °C for 24 h without stirring.
After completion of the reaction, the solid products were filtered,
washed with deionized water, and air-dried overnight. The solid
products were calcined to remove CTAB or Pluronic. The calcina-
tion procedure was as follows: the air-dried samples were heated
from room temperature to 100 °C at a rate of 1 °C/min; held at
100 °C for 2 h; increased from 100 to 500 °C at a rate of 1 °C/min;
and then held at 500 °C for 8 h.
2. Materials and methods
2.1. General
Sodium silicate (PQ Brand N, SiO2 28.7%, SiO2/Na2O = 3.22),
cetyltrimethylammonium bromide (CTAB, Fisher Chemical, high
purity grade), H2SO4 (Sigma–Aldrich, 95–98% ACS reagent), sodium
hydroxide (NaOH, Mallinckrodt Chemicals, pellet), tetraethoxysi-
lane (TEOS, P99%, Fluka), Pluronic P123 (EO20PO70EO20
,
MW = 5800, BASF), and HCl (Sigma–Aldrich, reagent grade, 37%)
were used in the OMS synthesis as received. 3-(Aminopropyl)tri-
ethoxysilane (APTES, 99%, Sigma–Aldrich), piperazine (Fluka,
P98%), diisopropylethylamine (DIPEA, 99%, Alfa Aesar), 4-(amino-
methyl)piperidine (AMP, P98%, TCI America), and cyanuric chlo-
ride (CC, 99%, ACROS) were used in the dendrimer synthesis as
received. Tetrahydrofuran (THF, Sigma–Aldrich, P99%) and tolu-
ene (Sigma–Aldrich, P99.5%) were dried using standard methods
using an MBraun solvent purification system. Methanol and
dichloromethane (DCM) (ACS solvent grade) were purchased from
EMD and used as received. Nitromethane (Acros, reagent ACS) and
nitrobenzaldehyde (Fluka, P99%, HPLC) were used as received for
the Henry reaction. Glyceryl tributyrate (99%) was obtained from
Sigma–Aldrich and was used as received in the transesterification
reaction. All NMR analyses were performed in deuterated chloro-
form (Cambridge Isotope, 99.8% D, 1% V/V TMS).
2.3. Synthesis of MCM-41/SBA-15–dendron hybrids
The synthesis method for generating the functionalized OMS
materials is shown in Scheme 1. The amine-functionalized MCM-
41 and SBA-15 samples were prepared using post-synthetic graft-
ing. Unless noted otherwise the target loading of organic is
0.2 mmequiv./g SiO2. One gram of calcined MCM-41 or SBA-15
was placed in a round-bottomed flask, and dried at 100 °C under
vacuum for 1 h. Then, 100 mL of anhydrous toluene was added into
the flask under nitrogen. An aliquot of APTES 46 lL (0.2 mmol) was
added to the solution under nitrogen. This mixture was stirred
overnight in a closed flask at room temperature. The product was
collected by filtration, washed sequentially with 50 mL of toluene,
50 mL of methanol, and 500 mL of deionized water, and air-dried.
The synthesis of the dendron–SBA-15/MCM-41 composites was
performed as reported previously by our laboratory [41,42]. As
an example for the case of the dendrons containing 4-(amino-
methyl)piperidine as the linkers (i.e. AMP dendrons), a 0.3 M cyan-
uric chloride (CC) solution was prepared by adding 1.25 g of CC and
2.5 mL of diisopropylethylamine (DIPEA) (8 mmol) to 25 mL of
THF. A 0.4 M 4-(aminomethyl)piperidine (AMP) solution was
2.2. Synthesis of MCM-41 and SBA-15
MCM-41 was synthesized using the reported procedure of Edler
and White [46,47], and the SBA-15 samples were synthesized using
Scheme 1. Dendron synthesis from OMS surface illustrated with PIP.

Q. Wang et al. / Journal of Catalysis 269 (2010) 15–25
17
prepared by adding 1.25 g of AMP to 25 mL of THF. One gram of
amine-functionalized MCM-41 or SBA-15 was placed in a 30 mL
vial, and 25 mL of the prepared CC solution was added. The vial
was shaken for 24 h at room temperature. The solution was fil-
tered, and rinsed with 50 mL portions of methanol, dichlorometh-
ane, and THF sequentially. The silica was transferred back to a
clean vial, and 25 mL of the linker molecule solution (AMP or
PIP) was added, and the vial was again shaken for 24 h. The mate-
rial was filtered and rinsed as described above. This process was re-
peated to increase the dendron generation. Note that in all the
results shown for a series of samples of different dendron genera-
tion, they are all derived from the same parent sample. For clarity,
samples with dendrons made using 4-(aminomethyl)piperidine as
the diamine linker are denoted as Gx AMP dendrons, and samples
with dendrons made using piperazine as the diamines linker are
denoted as Gx PIP dendrons where x is the dendron generation.
as-plot over the range 0 6 as 6 0.6, and the mesopore volumes
were determined from the y-intercept of the tangent line taken
over the range 1.5 6 as 6 2.0. The mesopore size distributions were
calculated from the adsorption branch of the isotherms using the
Barret–Joyner–Halenda (BJH) method with a modified equation
for the statistical film thickness. Solid-state NMR experiments
were performed at 9.4 T on a Bruker Avance. 13C{1H} CP-MAS were
performed at 100.61 MHz using a 4 mm probe with ZrO2 rotors, a
spinning rate of 9 kHz, a contact time of 2 ms, a 1H 90° pulse length
of 2.5 ls, and a recycle delay of 5 s. Chemical shifts were refer-
enced to tetramethylsilane. Elemental analysis (Si, C, H, and N anal-
ysis) was performed by the Galbraith laboratories.
2.5. Catalytic testing
Scheme 2 shows the three reactions investigated in this work.
The nitroaldol reaction (Henry reaction) was performed as follows.
One hundred milligrams of the catalyst was placed in a 10 mL
Schlenk tube with a small stir bar, and heated in an oven at
100 °C overnight. Upon cooling, 0.38 g of nitrobenzaldehyde
(2.5 mmol) and 1.35 mL of nitromethane (25 mmol) were intro-
duced into the tube. The mixture in the tube was kept at 40 °C un-
2.4. Analytical
Powder X-ray diffraction (PXRD) measurements were per-
formed using a Bruker-AXS D8 powder diffractometer with Cu
K
a radiation over a range of 0.5–10° 2h. Infrared spectroscopy
was recorded using a Nexus 670 FT-IR spectrometer from Thermo
Nicolet. Thermal gravimetric analyses (TGA) were performed using
a TG 209C Iris instrument from Netzsch over a temperature range
of 25–500 °C using oxygen and nitrogen as carrier gases and a tem-
perature ramping rate of 1 °C/min. Nitrogen-adsorption experi-
ments were performed on a Micromeritics ASAP 2010 micropore
system using approximately 0.06 g of sample. The samples were
degassed under vacuum at room temperature for 2 h, then at
100 °C for 24 h before analysis. The mesopore volumes and surface
der stirring. At different time intervals 100 lL of the solution was
transferred into a small glass tube, 1 mL of deuterated chloroform
was added to the tube, and then the solid was removed from the
solution using a centrifuge. The liquid was analyzed by 1H NMR.
The conversion was calculated from the relative intensity of the
nitrobenzaldehyde peak (d 9.9–10.1 ppm, 1H) and 1-(4-nitro-
phenyl)-2-nitroethanol peak (d 5.58–5.64 ppm, 1H).
The transesterification reaction of glyceryl tributyrate with
methanol was performed as follows: to a 50 mL flask equipped
with a stir bar and a condenser, glyceryl tributyrate (GTB) and
methanol were added in a 1:10 weight ratio (0.5 g GTB:5 g MeOH).
areas were determined using the
as-method [49,50]. The surface
areas were determined from the slope of the linear portion of the
Scheme 2. Reactions investigated in the current work. From top to bottom: Henry reaction, methyl ester synthesis from triglycerides, and Aldol condensation of 5-
(hydroxymethyl)furfural (HMF) with acetone.
Scheme 3. Molecular structures of GTB and the methyl ester product, and chemical shifts of the different protons as determined by 1H NMR.

18
Q. Wang et al. / Journal of Catalysis 269 (2010) 15–25
The catalyst (0.175 g) was then added. All reactions were run at re-
flux (ꢀ65 °C) for 6 h. At the end of the reaction period the flask was
placed in an ice bath to quench the reaction, and the solids were
separated by centrifugation and the liquid phase was isolated.
For kinetics measurements of conversion versus time, aliquots
were taken out at the time intervals noted. The conversion of the
glyceryl tributyrate to methyl ester was determined from the rela-
tive intensities of the methyl ester protons (d = 3.6 ppm, group A in
Scheme 3) and methylene protons adjacent to the ester group
(d = 2.2 ppm, group 3 and group B in Scheme 3) as has been done
previously [51]. For the recycle studies, the recovered catalyst
was rinsed with 20 mL of methanol three times, dried overnight
at 40 °C, and the testing was repeated.
ring. At different time intervals 100 lL of the solution was trans-
ferred into a small glass tube, 1 mL of deuterated chloroform was
added to the tube, and then the solid was removed from the solu-
tion using a centrifuge. The liquid was analyzed by 1H NMR. The
conversion was calculated from the relative intensity of the furfu-
ral (d 9.6 ppm, 1H) and the product peak (d 4.5–4.6 ppm, 2H).
The catalytic data presented for the Henry reaction and transe-
sterification reaction are presented as both absolute conversion of
reactant and amount of substrate consumed (or product formed)
per mmol of nitrogen per gram of silica. The latter is an attempt
to normalize the reactivity data to an effective mmol of amine ba-
sis, as different samples have different amine contents.
The Aldol condensation of 5-hydroxymethylfurfural (HMF) with
acetone was also investigated. One hundred milligrams of the cat-
alyst was placed in a 10 mL Schlenk tube with a small stir bar, and
heated in an oven at 100 °C overnight. Upon cooling, 0.32 g of HMF
(2.5 mmol) and 1.83 mL of acetone (25 mmol) were introduced
into the tube. The mixture in the tube was kept at 40 °C under stir-
3. Results and discussion
3.1. Hybrid characterization
Fig. 1 shows the powder XRD patterns of representative parent
MCM-41 and SBA-15 materials used in all subsequent work. As can
Fig. 1. PXRD patterns of (left) MCM-41 and (right) SBA-15.
Fig. 2. Nitrogen adsorption isotherms of (left) MCM-41 and (right) SBA-15. Adsorption branches are represented by open circles; desorption branches by solid circles.

Q. Wang et al. / Journal of Catalysis 269 (2010) 15–25
19
Table 1
be observed the syntheses lead to highly ordered OMS materials. In
the case of the MCM-41 samples four reflections are clearly ob-
served and the fifth is weakly visible. For the SBA-15 materials
there are three reflections. The PXRD patterns of all samples are in-
cluded in the Supplementary material. Consistent with previous
work the higher order reflections become weaker as the organic
content of the hybrid material increases and no appreciable change
in the peak positions is observed. The PXRD results show that the
parent materials are well-defined MCM-41 and SBA-15 materials.
Nitrogen adsorption was used to quantify the change in poros-
ity of all samples. The adsorption isotherms of the parent OMS
materials are shown in Fig. 2 and data for all of the samples are
summarized in Table 1. The isotherms shown in Fig. 2 are consis-
tent with those reported in the previous work. The nominal pore
sizes of the parent MCM-41 and SBA-15 are 3.5 and 7.8 nm, respec-
tively. Several observations can be made from the data summa-
rized in Table 1. First, in the case of the AMP-MCM-41 samples it
Summary of adsorption data of the OMS-hybrid materials.
Sample
S(
a
s) (m2/g)
vmeso
(a
s) (cm3/g)
dBJH (nm)
MCM-41
906
814
597
ꢀ0
0.68
0.61
0.40
ꢀ0
3.5
3.4
3.0
NH2-MCM-41
G1-AMP-MCM-41
G2-AMP-MCM-41
G3-AMP-MCM-41
G1-PIP-MCM-41
G2-PIP-MCM-41
G3-PIP-MCM-41
SBA-15
ꢀ0
ꢀ0
665
408
ꢀ0
0.45
0.22
ꢀ0
3.0
3.0
766
523
401
337
320
407
339
312
0.78
0.70
0.56
0.48
0.43
0.53
0.43
0.32
7.8
7.9
7.9
7.9
7.5
7.8
7.5
6.4
NH2-SBA-15
G1-AMP-SBA-15
G2-AMP-SBA-15
G3-AMP-SBA-15
G1-PIP-SBA-15
G2-PIP-SBA-15
G3-PIP-SBA-15
Fig. 3. Plots comparing elemental analysis and TGA results along with theoretical weight losses assuming 100% conversion at every step.

20
Q. Wang et al. / Journal of Catalysis 269 (2010) 15–25
appears that for both dendrons investigated the resulting G2 and
G3 hybrids are non-porous, at least based on nitrogen porosimetry.
Given the relatively small (3.5 nm) pore size of the parent OMS this
does not seem entirely unreasonable. The G1-PIP dendron should
be modestly smaller as this sample appears to have both a slightly
higher surface area and pore volume (ꢀ10% larger) than the AMP
sample. Also note that while the G2-PIP sample has some porosity
based on nitrogen adsorption, the G3 sample does not. The SBA-15
samples on the whole show similar trends to the materials re-
ported previously by our laboratory at low amine loading. How-
ever, in contrast to our previous work with a nominal amine
loading of 0.5 mmequiv./g, here we observe a systematic decrease
in pore volume from amine-functionalized SBA-15 to G1 to G2
samples. The pore sizes, as manifested by the position and width
of the capillary condensation hysteresis loop, do not change appre-
ciably. The change in pore volumes from G2 to G3 is smaller than
would be expected for both samples; this will be discussed in more
detail below in the context of the TGA and EA data. One possible
explanation for this is pore mouth blockage/preferential formation
of dendrimers at the pore mouth due to preferential attachment of
the silane at the pore mouth opening. At the most cautious level,
the adsorption data clearly show that a significant portion of the
organic content attached to the surface is in the mesopores given
G2 dendrons to G3 dendrons, particularly for the AMP diamine. It
also shows that conversion of the aminosilica to G1 dendrons is
less efficient on SBA-15 than on MCM-41. We have observed this
previously, and one possible explanation consistent with our prior
work is that some of the amines preferentially graft into the
micropores of SBA-15 and thus are inaccessible for further func-
tionalization. If the chemistry went perfectly (i.e. 100% conversion
at every step) the ratio of nitrogen should be 8:22:50 (1:2.75:6.25);
as can be seen this is not observed. The motivation for showing
Fig. 4 is that all the catalytic data will be presented as mmol of sub-
strate consumed (or product formed) per mmol of nitrogen per
gram of silica. Also, for the catalytic data shown the activity is nor-
malized per mmol of nitrogen and not per mmol of periphery
amines. In other words, the authors did not attempt to ascertain
which fraction of the nitrogen atoms participate in the reaction.
While it is not expected that the nitrogen centers on the cyanuric
the pronounced decreases in the pore volume. The as-plots for all
the samples are included in the Supplementary material.
The amount of organic content occluded in the materials was
determined by elemental analysis and thermogravimetric analysis.
These results are summarized in Fig. 3, along with calculations
showing ideal weight loadings if every step proceeded at 100% effi-
ciency. In general, the elemental analysis and TGA data are in very
good agreement. However, the organic content of the amine–OMS
samples is much higher (2.5–3-fold times larger) than would be
predicted theoretically as are the G1 samples. NMR measurements
do not show the presence of alkoxy groups. The TGA/EA results also
indicate, consistent with the nitrogen adsorption data, that the
reaction of converting the G2-MCM-41 materials to G3-MCM-41
materials is generally inefficient. The data shown in Fig. 3 are in-
cluded in the Supplementary material. Fig. 4 shows the mmol of
nitrogen per gram of silica for each of the samples as determined
by elemental analysis. Several observations can be made from
these data. The results shown here reinforce the conclusions from
Fig. 3. These include that the chemistry is inefficient in converting
Fig. 5. 13C{1H} CP MAS NMR spectra of (from top to bottom): G3-PIP-MCM-41, G3-
AMP-MCM-41, G3-PIP-SBA-15, and G3-AMP-SBA-15.
Fig. 4. Nitrogen loading for the different samples as determined by elemental analysis. Note that for the G2-MCM-41 samples the mmol N/g-SiO2 are essentially identical (i.e.
the data markers overlap).

Q. Wang et al. / Journal of Catalysis 269 (2010) 15–25
21
chloride, for instance, participate in the reaction it is unclear if
(whether) the interior aliphatic amines do. Thus, given the consis-
tency between the TGA and elemental analysis results the authors
thought it appropriate to normalize based on the total nitrogen
content.
Evidence for organic group incorporation is obtained from
13C{1H} CP MAS NMR (Fig. 5). The signal at 163 ppm is attributed
to the aromatic carbons. The two resonances at approximately 40
and 25 ppm for the AMP-dendron hybrid samples are due to the
aliphatic carbons of the 4-(aminomethyl)piperidine group. For
the PIP-dendron hybrid samples the signal at 163 ppm is also
due to the aromatic carbons from CC groups. The other strong sig-
nal at approximately 40 ppm is from the aliphatic carbons on
piperazine groups.
test reaction to study the catalytic properties of our dendron–
OMS materials. Fig. 6 shows the conversion versus time and mmol
of aldehyde consumed per mmol nitrogen per gram silica for the
MCM-41-dendron composites. All samples (on a percent conver-
sion basis) appear to possess reasonable activity for this reaction.
When the conversion is normalized to the mmol of nitrogen per
gram of silica several trends emerge. First, it becomes clear for both
samples that the G1 dendron hybrids are more active. Second, the
AMP samples (on per mmol nitrogen basis) are also more active.
Also noteworthy is that on a per site basis the G2 and G3 samples
appear essentially identical. There are a few possible explanations
for the decrease in activity of the higher generation dendrons. The
first is that there are diffusion resistances present in the MCM-41
composites. The second is that the intrinsic activity of the sites de-
creases with increasing number densities of amines. The third, re-
lated to the second reason, is that as the dendron fragment gets
larger there are less amine–silanol interactions. Based on the re-
sults mentioned below for the SBA-15 samples, we discount that
3.2. Catalytic testing, nitroaldol reaction
As several laboratories have investigated the nitroaldol reaction
over amine–OMS materials we thought this was an appropriate
Fig. 6. Conversion of aldehyde (left axis) and mmol aldehyde consumed per mmol nitrogen per gram of silica (right axis) versus time for AMP (left) and PIP (right) dendrons
supported on MCM-41 in the nitroaldol reaction. Dendron generation denoted, solid symbols correspond to right axis, open symbols to left axis.
Fig. 7. Conversion of aldehyde (left axis) and mmol aldehyde consumed per mmol nitrogen per gram of silica (right axis) versus time for AMP (left) and PIP (right) dendrons
supported on SBA-15 in the nitroaldol reaction. Dendron generation denoted, solid symbols correspond to right axis, open symbols to left axis.

22
Q. Wang et al. / Journal of Catalysis 269 (2010) 15–25
the decreased activities observed can be solely attributed to
diffusion effects.
ity in contrast to the MCM-41 G2/G3 samples. This likely indicates
that for the MCM-41 samples significant diffusion resistances are
present. The increase in the G1-AMP-SBA-15 reactivity as com-
pared to the G1-AMP-MCM-41 sample also supports the presence
of diffusion resistances in the MCM-41 composites.
Different generations of dendrimers, with AMP and PIP func-
tional groups, on SBA-15 were also investigated (Fig. 7). Again,
on a percent conversion basis all samples appear active and com-
parable. When normalized to the nitrogen loading, however, some
clear differences can be observed. First, these materials are more
active on a per site basis than the MCM-41 samples, and there
are clear differences in activity between G2 and G3 samples. This
point can likely be assigned to the absence (or decrease) of diffu-
sion resistances. The AMP samples are again more active than
the PIP samples, and again the G1 samples are much more active
than the G2 and G3 samples. Taking the results in Figs. 6 and 7 to-
gether, a few observations can be made. First, it appears that upon
going to larger pored silicas the intrinsic reactivity of the dendri-
mers goes down with increasing dendron generation. The authors
ascribe this to decreased amine–silanol interactions. The SBA-15
G2/G3 samples show statistically significant differences in reactiv-
Figs. 6 and 7 show that these materials are active catalysts for
the Henry reaction. In addition, some clear trends can be discerned
in the selectivity of the dendron composites (i.e. alcohol versus al-
kene formation). Fig. 8 shows the average selectivity for the sam-
ples as a function of dendron generation and amine (i.e. AMP
versus PIP) identity. As can be seen in all cases, the selectivity to
the nitroalcohol is very high (over 70%), and it can be observed that
the nitroalcohol selectivity increases with increasing dendrimer
generation and in going from primary amines to secondary amines.
The error bars shown indicate one standard deviation from the
average selectivity over the reaction run. That the selectivity to
the alcohol is highest with the PIP samples and that it increases
with increasing dendrimer generation are consistent with previous
Fig. 8. Selectivity to the nitroalcohol product for the AMP (left) and PIP (right) dendrons.
Fig. 9. Conversion of triglyceride (left axis) and mmol methyl ester formed per mmol nitrogen per gram of silica (right axis) versus time for AMP (left) and PIP (right)
dendrons supported on MCM-41. Dendron generation denoted, solid symbols correspond to right axis, open symbols to left axis.

Q. Wang et al. / Journal of Catalysis 269 (2010) 15–25
23
work by Katz as well as work from Asefa’s laboratory [36,52,53].
Katz invokes a bifunctional surface to generate the alkene [36],
and in the case of the G3 dendrons one might expect that many
of the amines are far enough from the surface that amine–silanol
cooperativity effects are not observed. It is also worth noting that
the ‘initial’ selectivity to the alcohol (at 15 or 30 min into the reac-
tion) is always higher than the average selectivity shown in Fig. 8.
This would seem to indicate that it is possible for the alcohol to un-
dergo subsequent dehydration to the olefin product. Plots of selec-
tivity versus time for all samples can be found in the
Supplementary material.
The primary amine-terminated dendrons are slightly more active
than the secondary amine-terminated dendrons based on the re-
sults shown in Fig. 9. Fig. 10 shows the same chemistry performed
over SBA-15 dendron hybrids. The results shown in Fig. 10 support
the idea of diffusion resistances being present in the MCM-41
materials. The amount of methyl ester formed, particularly on a
per mmol nitrogen basis, is significantly higher in the SBA-15 sam-
ples than in the MCM-41 samples. One can also see from the re-
sults shown in Fig. 10 that the primary amines are significantly
more active than the secondary amines in forming methyl esters.
While the conversions observed over these materials are
encouraging the most significant result is their stability. Fig. 11
shows the conversion of triglyceride after 6 h as a function of cat-
alyst recycle number. As can be seen there are only very modest
decreases of conversion with multiple recycles. This is in contrast
to previous work from our laboratory where simple aminosilane
groups attached to OMS were used as catalysts for this reaction
[59]. Thus the materials formed possess reasonable activity and
stability. This is attributed to the dendron structure inhibiting
the solvent from cleaving the silane attachment to the surface.
3.3. Transesterification reaction
Biodiesel, which is derived from triglycerides by transesterifica-
tion with methanol, has attracted considerable attention during
the past decade as a renewable, biodegradable, and nontoxic fuel
[54–58]. Fig. 9 shows conversion of triglyceride to methyl ester
over dendron-MCM-41 hybrids at 338 K. The conversion is highest
over the G1 samples, again consistent with diffusion resistances.
Fig. 10. Conversion of triglyceride (left axis) and mmol methyl ester formed per mmol nitrogen per gram of silica (right axis) versus time for AMP (left) and PIP (right)
dendrons supported on SBA-15. Dendron generation denoted, solid symbols correspond to right axis, open symbols to left axis.
Fig. 11. Conversion after 6 h on stream versus recycle number for AMP MCM-41 (left) and AMP SBA-15 (right) dendron hybrids.

24
Q. Wang et al. / Journal of Catalysis 269 (2010) 15–25
Fig. 12. Conversion of HMF over G1-PIP MCM-41.
Ongoing work is studying this in more detail as it relates to run-
ning reactions at higher temperatures.
(A-1638), Sabic Americas, CONACyT (V.V.G.), and LG Chemical
(S.Y.).
3.4. Aldol coupling of HMF
Appendix A. Supplementary material
The final reaction investigated was the condensation of 5-
(hydroxymethyl)furfural (HMF) with acetone. This reaction repre-
sents a route to large alkanes in a biorefinery as HMF is formed via
biomass hydrolysis and Aldol chemistry is often held up as a route
to larger hydrocarbons from biomass precursors [60,61]. Fig. 12
shows the conversions versus time in the cross aldol condensation
of HMF with acetone catalyzed by G1-PIP MCM-41. The conversion
is 20% after 6 h of reaction and 80% after 32 h. The preliminary re-
sult shows that the dendron–OMS hybrids are active in this reac-
tion; however, the reaction rate is lower than those in the
previously discussed reactions. Ongoing work is investigating
these reactions under higher temperature conditions and will be
reported elsewhere.
Diffraction patterns, nitrogen adsorption isotherms and
as-anal-
ysis of all samples, summary of TGA/EA calculations for all sam-
ples. This information can be downloaded free of charge from the
internet. Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.jcat.2009.10.009.
References
[1] D.M. Ford, E.E. Simanek, D.F. Shantz, Nanotechnology 16 (7) (2005) S458–S475.
[2] F. Hoffmann, M. Cornelius, J. Morell, M. Froba, Angew. Chem., Int. Ed. 45 (20)
(2006) 3216–3251.
[3] K. Moller, T. Bein, Chem. Mater. 10 (10) (1998) 2950–2963.
[4] A. Taguchi, F. Schüth, Micropor. Mesopor. Mater. 77 (2004) 1–45.
[5] U. Ciesla, F. Schüth, Micropor. Mesopor. Mater. 27 (1999) 131–149.
[6] A. Corma, Chem. Rev. 97 (1997) 2373–2419.
[7] M.E. Davis, Nature 417 (2002) 813–821.
[8] F. Schüth, W. Schmidt, Adv. Mater. 14 (2002) 629–638.
[9] M.H. Lim, C.F. Blanford, A. Stein, J. Am. Chem. Soc. 119 (1997) 4090–4091.
[10] M.H. Lim, A. Stein, Chem. Mater. 11 (1999) 3285–3295.
[11] S.L. Burkett, S.D. Sims, S. Mann, Chem. Commun. (1996) 1367–1368.
[12] D.J. Macquarrie, Chem. Commun. (16) (1996) 1961–1962.
[13] C. Li, Catal. Rev. 46 (2004) 419–492.
[14] J.D. Bass, S.L. Anderson, A. Katz, Angew. Chem., Int. Ed. 42 (42) (2003) 5219–
5222.
[15] J.D. Bass, A. Katz, Chem. Mater. 15 (2003) 2757–2763.
[16] J.D. Bass, A. Katz, Chem. Mater. 18 (6) (2006) 1611–1620.
[17] H.-T. Chen, S. Huh, J.W. Wiench, M. Pruski, V.S.Y. Lin, J. Am. Chem. Soc. 127
(2005) 13305–13311.
[18] J.L. Defreese, A. Katz, Chem. Mater. 17 (2005).
[19] G. Demicheli, R. Maggi, A. Mazzacani, P. Righi, G. Sartori, F. Bigi, Tetrahedron
Lett. 42 (2001) 2401–2403.
[20] S. Huh, H.-T. Chen, J.W. Wiench, M. Pruski, V.S.Y. Lin, J. Am. Chem. Soc. 126
(2004) 1010–1011.
[21] S. Huh, H.-T. Chen, J.W. Wiench, M. Pruski, V.S.Y. Lin, Angew. Chem., Int. Ed. 44
(2005) 1826–1830.
4. Conclusions
The synthesis and catalytic testing of several dendron–OMS
materials are reported. These materials show potential for a wide
range of chemical reactions including Aldol chemistry and transe-
sterification reactions. The results observed in the nitroaldol reac-
tion are qualitatively consistent with the previous literature in
terms of product selectivity, and the dendrons display good activ-
ity for this reaction. The dendron–OMS materials are not only ac-
tive in the transesterification of triglycerides but also very stable.
While possessing at best modest activity for the aldol reaction be-
tween HMF and acetone at 313 K, the results indicate that the
OMS–dendron materials have potential as solid base catalysts for
a range of reactions. Ongoing work is exploring additional chemis-
tries, studying the thermal/solvothermal stability of the compos-
ites more carefully, and exploring multifunctional dendrons.
[22] Y. Kubota, K. Goto, S. Miyata, Y. Goto, Y. Fukushima, Y. Sugi, Chem. Lett. 32
(2003).
[23] D.J. Macquarrie, R. Maggi, A. Mazzacani, M.J. Sabater, G. Sartori, R. Sartori, Appl.
Catal. A 246 (2003) 183–188.
[24] K. Shimizu, H. Suzuki, E. Hayashi, T. Kodama, Y. Tsuchiya, H. Hagiwara, Y.
Kitayama, Chem. Commun. (10) (2002) 1068–1069.
[25] K.-I. Shimizu, E. Hayashi, T. Inokuchi, T. Kodama, H. Hagiwara, Y. Kitayama,
Tetrahedron Lett. 43 (2002).
Acknowledgments
The authors acknowledge the National Science Foundation
(CTS-0624813) for financial support of this work. Partial financial
support was also provided by the Robert A. Welch Foundation
[26] H. Yoshitake, E. Koiso, T. Tatsumi, H. Horie, H. Yoshimura, Chem. Lett. 33
(2004) 872–873.

Q. Wang et al. / Journal of Catalysis 269 (2010) 15–25
25
[27] A. Corma, S. Iborra, I. Rodriguez, M. Iglesias, F. Sánchez, Catal. Lett. 82 (2002)
237–242.
[28] D. Dhar, I. Beadham, S. Chandrasekaran, Proc. Indian Acad. Sci. (Chem. Sci.) 115
(2003) 365–372.
[29] V. Dufaud, M.E. Davis, J. Am. Chem. Soc. 125 (31) (2003) 9403–9413.
[30] R.K. Zeidan, V. Dufaud, M.E. Davis, J. Catal. 239 (2006) 299–306.
[31] E.L. Margelefsky, A. Bendjeriou, R.K. Zeidan, V. Dufaud, M.E. Davis, J. Am. Chem.
Soc. 130 (2008) 13442–13449.
[44] J.P.K. Reynhardt, Y. Yang, A. Sayari, H. Alper, Chem. Mater. 16 (2004) 4095–4102.
[45] J.P.K. Reynhardt, Y. Yang, A. Sayari, H. Alper, Adv. Funct. Mater. 15 (2005)
1641–1646.
[46] P.J. Branton, K.S.W. Sing, J.W. White, J. Chem. Soc., Faraday Trans. 93 (1997)
2337–2340.
[47] K.J. Edler, J.W. White, Chem. Mater. 9 (5) (1997) 1226–1233.
[48] D.Y. Zhao, Q.S. Huo, J.L. Feng, B.F. Chmelka, G.D. Stucky, J. Am. Chem. Soc. 120
(24) (1998) 6024–6036.
[32] E.L. Margelefsky, R.K. Zeidan, V. Dufaud, M.E. Davis, J. Am. Chem. Soc. 129
(2007) 13691–13697.
[33] J.C. Hicks, R. Dabestani, A.C.I. Buchanan, C.W. Jones, Chem. Mater. 18 (2006)
5022–5032.
[34] J.C. Hicks, C.W. Jones, Langmuir 21 (2006) 2676–2681.
[35] M.W. McKittrick, C.W. Jones, Chem. Mater. 15 (2003) 1132–1139.
[36] J.D. Bass, A. Solovyov, A.J. Pascall, A. Katz, J. Am. Chem. Soc. 128 (11) (2006)
3737–3747.
[37] J.M. Rosenholm, A. Penninkangas, M. Linden, Chem. Commun. (37) (2006)
3909–3911.
[38] J.M. Rosenholm, A. Duchanoy, M. Linden, Chem. Mater. 20 (3) (2008) 1126–
1133.
[39] K.Q. Yu, C.W. Jones, J. Catal. 222 (2) (2004) 558–564.
[40] J.C. Hicks, J.H. Drese, D.J. Fauth, M.L. Gray, G.G. Qi, C.W. Jones, J. Am. Chem. Soc.
130 (2008) 2902–2903.
[41] E.J. Acosta, C.S. Carr, E.E. Simanek, D.F. Shantz, Adv. Mater. 16 (12) (2004) 985–
989.
[42] S. Yoo, J.D. Lunn, S. Gonzalez, J.A. Ristich, E.E. Simanek, D.F. Shantz, Chem.
Mater. 18 (2006) 2935–2942.
[49] M. Jaroniec, M. Kruk, J.P. Olivier, Langmuir 15 (16) (1999) 5410–5413.
[50] F. Rouquerol, J. Roquerol, K. Sing, Adsorption by Powders and Porous Solids,
Academic, San Diego, 1999.
[51] M. Morgenstern, J. Cline, S. Meyer, S. Cataldo, Energy Fuel 20 (2006) 1350–
1353.
[52] A. Anan, R. Vathyam, K.K. Sharma, T. Asefa, Catal. Lett. 126 (1–2) (2008) 142–
148.
[53] K.K. Sharma, T. Asefa, Angew. Chem., Int. Ed. 46 (16) (2007) 2879–2882.
[54] A. Cauvel, G. Renard, D. Brunel, J. Org. Chem. 62 (3) (1997) 749–751.
[55] M. Di Serio, M. Ledda, M. Cozzolino, G. Minutillo, R. Tesser, E. Santacesaria, Ind.
Eng. Chem. Res. 45 (9) (2006) 3009–3014.
[56] M. Di Serio, R. Tesser, L. Pengmei, E. Santacesaria, Energy Fuel 22 (2008) 207–
217.
[57] S. Gryglewicz, Bioresour. Technol. 70 (1999) 249–253.
[58] F. Ma, M.A. Hanna, Bioresour. Technol. 70 (1999) 1–15.
[59] V. Varela Guerrero, D.F. Shantz, Ind. Eng. Chem. Res. (ASAP article).
[60] G.W. Huber, J.N. Chheda, C.J. Barrett, J.A. Dumesic, Science 308 (2005) 1446–
1450.
[61] J.N. Chheda, G.W. Huber, J.A. Dumesic, Angew. Chem., Int. Ed. 46 (2007) 7164–
7183.
[43] S. Yoo, S. Yeu, R.L. Sherman, E.E. Simanek, D.F. Shantz, D.M. Ford, J. Membr. Sci.
334 (2009) 16–22.