Substrate inhibition in the heterogeneous catalyzed aldol condensation: A mechanistic study of supported organocatalysts

Article abstract of DOI:10.1016/j.jcat.2012.04.005

In this study, we demonstrate how materials science can be combined with the established methods of organic chemistry to find mechanistic bottlenecks and redesign heterogeneous catalysts for improved performance. By using solid-state NMR, infrared spectroscopy, surface and kinetic analysis, we prove the existence of a substrate inhibition in the aldol condensation catalyzed by heterogeneous amines. We show that modifying the structure of the supported amines according to the proposed mechanism dramatically enhances the activity of the heterogeneous catalyst. We also provide evidence that the reaction benefits significantly from the surface chemistry of the silica support, which plays the role of a co-catalyst, giving activities up to two orders of magnitude larger than those of homogeneous amines. This study confirms that the optimization of a heterogeneous catalyst depends as much on obtaining organic mechanistic information as it does on controlling the structure of the support.

Full text of DOI:10.1016/j.jcat.2012.04.005

Journal of Catalysis 291 (2012) 63–68
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Substrate inhibition in the heterogeneous catalyzed aldol condensation:
A mechanistic study of supported organocatalysts
Kapil Kandel a,b, Stacey M. Althaus a,b, Chorthip Peeraphatdit a,b, Takeshi Kobayashi a, Brian G. Trewyn b,
Marek Pruski a,b, Igor I. Slowing a,
⇑
a US Department of Energy, Ames Laboratory, Ames, IA 50011-3020, USA
b Department of Chemistry, Iowa State University, Ames, IA 50011-3111, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
In this study, we demonstrate how materials science can be combined with the established methods of
organic chemistry to find mechanistic bottlenecks and redesign heterogeneous catalysts for improved
performance. By using solid-state NMR, infrared spectroscopy, surface and kinetic analysis, we prove
the existence of a substrate inhibition in the aldol condensation catalyzed by heterogeneous amines.
We show that modifying the structure of the supported amines according to the proposed mechanism
dramatically enhances the activity of the heterogeneous catalyst. We also provide evidence that the reac-
tion benefits significantly from the surface chemistry of the silica support, which plays the role of a co-
catalyst, giving activities up to two orders of magnitude larger than those of homogeneous amines. This
study confirms that the optimization of a heterogeneous catalyst depends as much on obtaining organic
mechanistic information as it does on controlling the structure of the support.
Received 19 January 2012
Revised 30 March 2012
Accepted 7 April 2012
Available online 23 May 2012
Keywords:
Mesoporous silica nanoparticles
Heterogeneous catalysis
Aldol condensation
Substrate inhibition
Cooperative catalysis
Solid-state NMR
Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction
and are then hydrolyzed to release the product [12,13]. Homoge-
neous catalysis of this reaction has been accomplished by strong
One of the characteristics that distinguish homogeneous from
heterogeneous catalysis is the intricacy of the reaction environ-
ment. The uniform nature of homogeneous catalysts facilitates
identification of intermediates and enables understanding of the
transformations in terms of reaction mechanisms. This permits
optimization of the activity by slight variations to the molecular
structure of the catalyst [1]. Unraveling of reaction mechanisms
in heterogeneous catalysis poses additional challenges due to the
different environments that the active sites can encounter on a so-
lid support. Therefore, optimization of heterogeneous catalysts is
often performed by selection or design of supports rather than
by modifying the structure of the catalytic groups [2–5]. In addi-
tion, homogeneous catalysts typically exhibit superior selectivity
and kinetics. Despite these disadvantages, heterogeneous catalysts
are valued because they allow easy separation of products and can
be reused for extended periods of time [6].
acids or bases, combining nucleophilic addition with enolization
[14,15], and recently by proline and catalytic antibodies [16,17].
Many heterogeneous catalysts have been developed to promote
this reaction, and among those with organic groups as active sites,
the most commonly used are aminoalkyls [18–24].
While supported aminoalkyls promote the aldol condensation,
their catalytic efficiency is relatively low [20,21,23–27]. A way to
solve this problem could be by introducing a secondary functional
group in the material [28]. Davis and co-workers adopted this strat-
egy and synthesized a bifunctional catalyst by introducing amine
and sulfonic acid groups on mesoporous silica, which dramatically
increased the activity due to cooperativity between both groups
[25,26]. Solin and collaborators, as well as Thiel and co-workers, ob-
tained similar results, using different combinations of alkylamines
and acidic groups in mesoporous silica supports [23,27]. However,
the poor catalytic activity of monofunctional amine on silica, which
remains commonly used for the aldol and similar types of condensa-
tion [20,22,24,29–32], is still not well understood.
Herein, we investigate in detail the mechanistic causes of the
poor catalytic activity of amine-functionalized mesoporous silica
toward the aldol condensation. Based on this understanding, we
demonstrate that the performance can be dramatically improved
by the proper choice of the catalytic groups. Furthermore, we
report activities that surpass those observed in the homogeneously
Given its importance as a CAC bond-forming reaction, the aldol
condensation has been a common target for catalyst design [7–11].
This reaction is performed in organisms by aldolases, which acti-
vate donor ketones with the amino group of a highly conserved ly-
sine, to give enamines. The enamines attack aldehyde acceptors
⇑
Corresponding author.
E-mail address: islowing@iastate.edu (I.I. Slowing).
0021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.04.005

64
K. Kandel et al. / Journal of Catalysis 291 (2012) 63–68
catalyzed reactions and demonstrate that these enhanced activities
arise from the cooperative interactions between organocatalysts
and the support.
ated at 200 kV (Fig. S1). For the TEM measurements, an aliquot of
the powder was sonicated in methanol for 15 min. The small-angle
powder X-ray diffraction patterns were obtained with a Rigaku Ul-
tima IV diffractometer using Cu target at 40 kV and 44 mA (Fig. S2).
The Cu Kb radiation was removed using a monochromator. A single
drop of this suspension was placed on a lacey carbon-coated cop-
per TEM grid and dried in air. Fourier transform infrared (FT-IR)
spectra were recorded on Nicolet Nexus 470. Solid-state NMR
experiments are described separately below. Loading of the cata-
lytic groups was measured by elemental analysis in a Perkin Elmer
2100 Series II CHN/S Analyzer, with acetanilide as calibration stan-
dard, and combustion and reduction temperatures of 925 °C and
640 °C. The expected precision and accuracy of the analysis was
0.3% for each element, and each material was tested by triplicate.
2. Materials and methods
2.1. Materials
Cetyltrimethylammonium bromide (CTAB), mesitylene, p-nitro-
benzaldehyde, hexamethyldisilazane (HMDS) and dimethyl sul-
fone were purchased from Sigma–Aldrich. Tetraethylorthosilicate
(TEOS), 3-aminopropyl trimethoxysilane, [3-(methylamino)propyl]
trimethoxysilane and [3-(N,N-dimethylamino)propyl] trimethox-
ysilane were purchased from Gelest. All reagents were used as re-
ceived without further purification.
2.6. General procedure for aldol condensation reaction
2.2. Synthesis of amine-functionalized mesoporous silica nanoparticles
All catalytic reactions were performed in screw-cap vials. p-
Nitrobenzaldehyde (0.39 mmol) was dissolved in acetone
(1.5 mL). To this solution, a suspension of the catalyst containing
0.0117 mmol of the amine group (corresponding to 3 mol% of the
p-nitrobenzaldehyde) in hexane (1.5 mL) was added. The solution
was stirred at 60 °C for specified times and set on ice to quench
the reaction. The catalyst was separated by centrifugation, and
the supernatant was concentrated under reduced pressure. The
yield of the products was determined by 1H NMR, using dimethyl
sulfone as an internal standard.
The functionalized mesoporous silica nanoparticles (MSNs)
with particle sizes ranging from 80 to 150 nm (Fig. S1) were pre-
pared following a previously published method [33]. CTAB (1.0 g,
2.7 mmol) was dissolved in nanopure water (480 g, 26.7 mol), fol-
lowed by the addition of NaOH solution (2.0 M, 3.5 mL, 7.0 mmol).
The mixture was heated to 80 °C for 1 h. To this clear solution,
TEOS (4.7 g, 23 mmol) was added dropwise, followed by immedi-
ate addition of 3-aminopropyl trimethoxysilane (for AP-MSN)
(1.0 mL, 5.7 mmol) and [3-(methylamino)propyl] trimethoxysilane
(for MAP-MSN) (1.0 mL, 5.0 mmol). The reaction mixtures were
stirred vigorously at 80 °C for 2 h and then filtered to yield white
functionalized solids. The as-synthesized materials were washed
with copious amounts of water and methanol and then dried under
vacuum. The final catalysts were obtained by removing the CTAB
surfactant via Soxhlet extraction with methanol for 24 h, followed
by drying overnight under vacuum. These samples were labeled
AP-MSN-2.8 and MAP-MSN-2.6, based on the average pore size
(2.8 nm and 2.6 nm, respectively).
2.7. Solid-state NMR
Solid-state NMR experiments utilizing 13C cross-polarization un-
der magic angle spinning (CPMAS) were performed to determine the
structures of surface-bound species and intermediates in AP-MSN.
The NMR technique was also used as an additional tool to evaluate
the loading of the functional groups on the MSN surface by means
of 29Si spectra, measured using direct polarization (DP) under magic
angle spinning (MAS) with Carr-Purcell-Meiboom-Gill (CPMG) refo-
cusing of 29Si magnetization [37], as previously described
(Table S1) [38].
2.3. Synthesis of pore-expanded functionalized mesoporous silica
nanoparticles
The experiments were performed at 9.4 T on a Chemagnetics
Infinity 400 spectrometer equipped with a 5-mm MAS probe oper-
ated at 400.0 MHz (1H) and 79.4 MHz (29Si) and at 14.1 T on a Var-
ian NMR System 600 spectrometer equipped with a 1.6-mm
FastMAS™ probe operated at 599.6 MHz (1H) and 150.8 MHz
The functionalized MSN materials with larger pores (AP-MSN-
3.6 and MAP-MSN-3.5) were prepared following the same proce-
dure as described above, except for the initial step that involved
adding the pore expander agent mesitylene (1.73 g, 14.4 mmol)
to the original CTAB solution, as previously published [34]. A third
pore-expanded catalyst was also prepared functionalized with [3-
(N,N-dimethylaminopropyl)] trimethoxysilane (1.0 mL, 4.6 mmol)
and labeled DMAP-MSN-3.2.
(
13C). Experimental parameters are given below using the follow-
ing symbols: mR denotes the MAS rate,
mRF(X) the magnitude of
radiofrequency magnetic field (RF) applied to X spins,
s
CP the mix-
ing time during cross-polarization, NCPMG the number of CPMG
echoes, sRD the recycle delay, NS the number of scans and AT the
total acquisition time.
2.4. Passivation of MAP-MSN-3.5 material
29Si DPMAS with CPMG:
RF(1H) = 45 kHz, NCPMG = 10,
13C CPMAS:
R = 40 kHz, mRF
CP = 60 kHz,
RF(1H) during SPINAL-64 decoupling = 12 kHz, sCP
3 ms,
m
R = 10 kHz, mRF
RD = 300 s, NS = 296 and AT = 25 h.
13C) = 140 kHz, RF(1H) during
(
29Si) = 50 kHz,
m
s
The silylation was performed by placing 1.0 g of MAP-MSN-3.5
material in a hexane solution of hexamethyldisilazane (HMDS)
(100 mL hexane, 10 mmol HMDS) [35,36]. The mixture was re-
fluxed for 24 h; the resulting solution was then filtered, washed
three times with hexane and dried under vacuum.
m
(
m
m
=
s
RD = 2 s, NS = 26,400 and AT = 15 h.
The chemical shifts of 29Si, 13C and 1H are reported using the d
scale and are referenced to tetramethylsilane (TMS) at 0 ppm.
2.5. Characterization
3. Results and discussion
The surface areas and pore size distributions of the catalysts
were measured by nitrogen sorption isotherms in a Micromeritics
Tristar analyzer and calculated by the Brunauer–Emmett–Teller
(BET) and Barrett–Joyner–Halenda (BJH) methods, respectively
(Table S1). The transmission electron microscopy (TEM) examina-
tion was completed on a Tecnai G2 F20 electron microscope oper-
3.1. Catalytic activity of homogeneous and heterogeneous
propylamine
To set a reference, we measured the activity of propylamine as a
homogenous catalyst for the cross-aldol condensation between p-

K. Kandel et al. / Journal of Catalysis 291 (2012) 63–68
65
Scheme 1. Cross-aldol condensation between p-nitrobenzaldehyde 1 and acetone.
nitrobenzaldehyde 1 and acetone at 60 °C in hexane (Scheme 1).
imal rate (Vmax) of 0.594 mmol hÀ1 (r2 = 0.9992) (Fig. S3). These
Catalytic activity was determined by measuring the formation of
estimates are clearly larger than the experimental values and
the aldol 2 and the
a,b-unsaturated carbonyl 3 products. Consis-
therefore indicate a strong substrate inhibition [39,40]. For
tent with the report by Davis and co-workers, homogeneous pro-
pylamine displayed poor activity with only 4.5% conversion after
2 h [25]. Interestingly, while Davis observed a fourfold increase
in yield upon supporting the amine on mesoporous silica [25],
the activity of our 3-aminopropyl mesoporous silica nanoparticle
catalyst with 2.8 nm pores (AP-MSN-2.8) was even lower than that
of the homogeneous amine, giving only 2% conversion after 2 h.
Considering that the pores of AP-MSN-2.8 were smaller than
those previously used (6.3 nm) [25,26], we examined the activity
of an AP-MSN-3.6 catalyst (with 3.6-nm-wide pores). Remarkably,
we found that this small increase in pore size, from 2.8 nm to
3.6 nm, led to a 20-fold increase in activity: the 2-h yield rose from
2% to 47%. The apparent pseudo-first-order rate constant of AP-
MSN-3.6 (k = 3.7 Â 10À1 hÀ1) was two orders of magnitude larger
than that of the homogeneous propylamine (k = 2.6 Â 10À3 hÀ1).
This large change in activity suggested that the rate of the reaction
was limited by the molecular diffusion within the narrower pores.
However, the longest dimensions of the reactants and products
(0.4 nm for acetone, 0.6 nm for 1 and 1 nm for 2 or 3) were small
compared to the pore diameters of AP-MSN-2.8 and AP-MSN-3.6.
Therefore, diffusion could not be the only factor limiting the cata-
lytic activity.
instance, the catalytic sites of the material could be blocked by
the formation of a stable Schiff base, which has been reported to
inhibit the aldol condensation by catalytic antibodies (Scheme 2)
[41,42].
3.2. Detection of inhibition intermediate
Although it has been suggested that an imine could form during
the aldol condensation catalyzed by AP-MSN, no direct evidence
has been provided for its existence [23]. When comparing the
infrared and NMR spectra of AP-MSN-2.8 before and after the reac-
tion, we confirmed the formation of imine intermediate 4, which
was stable even after washing and drying the material. While the
CAC stretching e2g band (1606 cmÀ1) and the symmetric and
asymmetric stretching bands of ANO2 (1345 cmÀ1, 1537 cmÀ1) of
1 could be clearly identified in AP-MSN-2.8 following the reaction,
the stretching frequency of C@O (1706 cmÀ1) was no longer visible,
but was replaced with a signal at 1646 cmÀ1 corresponding to the
C@N stretching of the imine (Fig. 2a). Solid-state 13C NMR spectros-
copy unambiguously confirmed the formation of 4 (Fig. 2b). The
resonance of carbon ‘c’ in AP-MSN-2.8 decreased considerably after
the reaction, appeared to a resonance at 160 ppm corresponding to
the C@N carbon (‘d’) in 4, whereas no signal due to the carbonyl
carbon of 1 (190 ppm) was observable. A strong downfield shift
of the resonance of C-3 in AP-MSN (resonance ‘c’ shifting to ‘c⁄’)
was also observed after the reaction with 1 (Fig. 2b). This suggested
a chemical transformation of the aminopropyl group rather than a
mere physisorption of 1 to the surface of the particles. A fraction of
unreacted aminopropyl was still visible in the sample as shown by
resonance ‘c’ in the used catalyst. Comparison of nitrogen content
of the material before and after the reaction by elemental analysis
revealed that approximately 70% of amine groups formed the
imine. Although the 13C CPMAS spectra in Fig. 2b are not strictly
quantitative, the intensity ratio of resonances ‘c⁄’ and ‘c’ is in
approximate agreement with the elemental analysis.
When comparing the properties of AP-MSN-2.8 before and after
the reaction, we discovered that despite the surface area of the
material remaining relatively constant (906 m2/g before and
894 m2/g after the reaction), the pore size decreased significantly
to 2.0 nm. In addition, we observed an inhibition of the reaction
kinetics at high concentrations of substrate 1 (Fig. 1). The Linewe-
aver–Burk plot of the data obtained at concentrations lower than
100 mM gave a Michaelis–Menten constant (KM) of 273 and a max-
10
8
Treatment of the poisoned AP-MSN catalyst with 0.01 M HCl for
24 h at room temperature led to hydrolysis of the Schiff base, as
evidenced by disappearance of the signals of 4 in the infrared
and NMR spectra of the treated material. The regeneration of AP-
MSN catalyst was also confirmed by elemental analysis: the num-
ber of mmol of nitrogen per gram of material varied from 1.0 be-
fore reaction to 1.7 after formation of 4, to 1.16 after treatment
with dilute acid.
The relatively large size of the Schiff base 4 (about 1 nm) ex-
plained the reduction in the pore size of AP-MSN by 0.8 nm, as well
as the dramatic effect of the small increase in pore size on the reac-
tion yield. The inhibition at high concentrations of 1 suggested that
the mechanism of the reaction is unlikely Mannich type, but
should involve either enamine or enolate intermediates.
6
4
2
0
0
50
100
150
[1] / mM
200
250
Fig. 1. Effect of substrate concentration on the rate of AP-MSN-3.6-catalyzed cross-
aldol condensation. The drop in rate at high concentrations of 1 suggests substrate
inhibition of the reaction.
O
N
H2N
+
3.3. Structural modification of the catalytic group
O2N
O2N
1
4
Based on the hypothesis that the low activity of AP-MSN was
caused by the formation of a stable Schiff base, we functionalized
the MSNs with a secondary amine, which is unable to form imines
Scheme 2. Formation of a Schiff base between p-nitrobenzaldehyde substrate 1
and the aminopropyl group of AP-MSN.

66
K. Kandel et al. / Journal of Catalysis 291 (2012) 63–68
Fig. 3. Kinetics of aldol condensation between 1 and acetone catalyzed by AP-MSN-
3.6 and MAP-MSN-3.5 in hexane at 60 °C with 3 mol% catalyst.
catalysts, giving higher conversion in only 2 h than the bifunctional
catalysts gave over 20 h of reaction [25–27].
3.4. Cooperative effects of the support
Having established the role of the catalytic groups, we focused
on the role of the support. As in the case of AP-MSN-3.6, the activ-
ity of the heterogeneous MAP-MSN-3.5 catalyst (k = 1.35 hÀ1) is
much higher than that of the corresponding homogeneous catalyst
N-methyl-propylamine (k = 0.056 hÀ1) (Fig. S5). These unusual re-
sults contradict the general observation that homogeneous cataly-
Fig. 2. Infrared (a) and 13C CPMAS NMR (b) spectra of AP-MSN-2.8 before (black)
and after (blue) reaction with 1. Infrared spectrum of 1 (red) is included as a
reference. (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)
[43]. Two catalysts containing 3-(N-methylamino)propyl (MAP)
with different pore sizes were prepared (MAP-MSN-2.6 and MAP-
MSN-3.5) and tested for the reaction. As expected, no imine was
detected by infrared and NMR analyses of MAP-MSNs after the
reaction. Remarkably, even the narrow-pore MAP-MSN-2.6 dou-
bled the activity of the wide-pore AP-MSN-3.6, yielding 93% con-
version after 2 h. Also, no inhibition of reaction kinetics was
observed at high concentrations of 1 (Fig. S4). The conversion using
MAP-MSN-3.5 was further elevated to 97%. These results suggested
that imine formation with AP-MSN catalysts lowered the activity
by blocking diffusion and by reducing the number of active sites.
We also noted that the apparent rate constant of MAP-MSN-3.5
(k = 1.35 hÀ1) was over three times larger than that of AP-MSN-3.6
(k = 0.37 hÀ1) (Figs. 3 and S5). We considered the possibility that
the increase in catalytic activity could be due to MAP-MSN being
more nucleophilic or basic than AP-MSN. This could imply a mech-
anism involving enolates rather than enamines. Although unlikely,
due to the high pKb of amines, we tested this hypothesis by prepar-
ing a new material functionalized with a tertiary amine. The mate-
rial containing 3-(N,N-dimethylamino)propyl group (DMAP-MSN-
3.2) failed to catalyze the reaction, proving that the conversion is
not promoted by simple deprotonation. It must be pointed out
that, being a tertiary amine, DMAP cannot form an enamine, which
is consistent with the reaction proceeding through an enamine
pathway.
Fig. 4. 29Si DPMAS spectra of MAP-MSN-3.5 before (bottom) and after (top)
blocking silanol groups with HMDS. Appearance of M sites due to the attached
silane matches the conversion of the Q2 and Q3 sites to Q3 and Q4, respectively [50].
100
80
60
40
20
0
a
b
c
d
Fig. 5. Effect of proximity between silanol and amine groups on the conversion of
p-nitrobenzaldehyde. Catalytic activities are compared for: (a) homogeneous N-
methyl-propylamine, (b) homogeneous N-methyl-propylamine + heterogeneous
MSN, (c) silanol-passivated HMDS-MAP-MSN-3.5 and (d) heterogeneous MAP-
MSN-3.5.
We finally noted that the MAP-MSN catalysts are significantly
more active than the previously reported amphoteric bifunctional

K. Kandel et al. / Journal of Catalysis 291 (2012) 63–68
67
Scheme 3. Possible pathway of proton transfer assisted by silanol groups.
sis is much faster than heterogeneous catalysis [44] and suggest
that the support plays an active role in the reaction mechanism. In-
deed, the weakly acidic silanol groups on the surface of silica have
been previously recognized as capable of assisting various reac-
tions [23,45–49].
To test the participation of silanol groups in the catalytic pro-
cess, we treated MAP-MSN-3.5 with hexamethyldisilazane
(HMDS). 29Si NMR showed that this reduced the number of silanols
by 39% (from 3.8 to 2.3 mmol/g, Fig. 4). When using this passivated
catalyst, the yield of the reaction dropped by 34% (from the original
97% to 63%). Furthermore, addition of non-functionalized MSN to
homogeneous N-methyl-propylamine increased the conversion to
51% compared to the 10% yield observed when using only the
homogeneous catalyst (Fig. 5). These results confirm not only that
the silanol groups play an active role in the catalytic process, but
also that their proximity to the amine sites is beneficial, which sug-
gests a cooperativity between both groups.
We also observed that the immobilization on silica led to a lar-
ger increase in the activity of the primary amine than the second-
ary amine (khet/khomo AP = 142, khet/khomo MAP = 23). This significant
difference suggests that silanol groups may play yet another role in
the catalysis. As mentioned above, the low catalytic activity of pro-
pylamine and AP-MSNs is attributed to the formation of a stable
imine. Since imine hydrolysis is catalyzed by acids [61], it is likely
that the weakly acidic silanols assist the hydrolysis of a fraction of
the imines, thus giving an additional advantage to AP-MSN com-
pared to homogeneous propylamine. This would also explain, in
part, the enhanced activity observed upon co-functionalization of
aminopropyl mesoporous silica with acidic groups, as previously
reported by Davis and Solin [25–27].
4. Conclusions
In conclusion, the poor catalytic activity of aminopropyl-func-
tionalized mesoporous silica for the aldol condensation arises from
a substrate inhibition taking place by the formation of a stable
Schiff base, which not only eliminates active sites but also blocks
diffusion in pores 2.8 nm or smaller. This inhibition can be partially
reduced by increasing the pore size of the support or eliminated by
modifying the structure of the amine from primary to secondary.
The silanol groups in the support assist the catalytic activity of
immobilized amines by offering binding sites for the reactants in
close proximity to the amines, providing pathways for proton
transfer throughout all the steps of the reaction, and facilitating
the departure of water during the formation of intermediates. This
cooperation between the silica support and the amines dramati-
cally improves the activity of the heterogeneous catalysts in com-
parison with the homogeneous catalysts in solution.
The role of silanols in the reaction can be explained by the fact
that carbonyl compounds adsorb on the surface of silica via
hydrogen bonding [26,51–56]. We confirmed this interaction by
measuring NMR spectrum of 13C-labeled acetone set in contact
with non-functionalized MSN (Fig. S6), which exhibited a down-
field shift of the carbonyl carbon signal compared to that of neat
acetone (ꢀ213 ppm versus 206 ppm). Our earlier solid-state NMR
and theoretical studies demonstrated that surface silanols on silica
also interact with the amine functionalities [57]. These findings
suggest that silanol groups play a key role in bringing all reactants
and the catalytic group together for the reaction to take place. In
contrast, the probability of bringing acetone, the aldehyde and
the amine catalyst together is dramatically decreased in the homo-
geneous medium. Similarly, the probability of encounter must be
lower if the amine is not covalently attached to the silica support.
This explains the observed activity trend: homogeneous MAP <
MAP + MSN < MAP-MSN (Fig. 5). In addition, when introducing
In summary, we have shown that heterogeneous organocata-
lysts are not only amenable to conventional mechanistic studies,
but that the understanding achieved through these types of studies
can guide their rational design to significantly improve their
activity.
DMSO (hydrogen bond-acceptor,
a = 0.00, b = 0.76) [23,58] to the
reaction, we observed a significant drop in the conversion cata-
lyzed by MAP-MSN-3.5, from 97% to 55%. This drop can be attrib-
uted to the competition of DMSO with the reactants for
hydrogen-binding the silanol groups.
Acknowledgments
The formation of hydrogen bonds between silanols and carbon-
yls may also contribute to the activation of the latter for nucleo-
philic attack by the amine and may assist in the formation of the
reaction intermediate by facilitating the departure of carbonyl
oxygen as water (Scheme 3). The formation of the intermediate
enamine involves a series of proton transfers, which may be diffi-
cult to achieve in a non-polar medium. The mildly acidic silanol
groups could assist these transfers by aligning with acetone and
amine groups in six-membered ring-like arrangements, as in the
Zimmerman–Traxler model (Scheme 3) [59,60]. For these interme-
diates to form, the silanol groups should be as close as 5–6 Å from
each other. As mentioned earlier, 29Si NMR spectroscopy of MAP-
MSN revealed a silanol content of 3.8 mmol/g (Fig. 4), which at a
surface area of 937 m2/g gives a silanol density of 2.4 groups/
nm2. This density satisfies the inter-silanol distance required for
the cyclic model. A third silanol group could also be closely located
to this intermediate, providing a site for hydrogen bonding of 1, to
complete the reaction by a similar proton transfer process.
This research is supported by the US Department of Energy, Of-
fice of Basic Energy Sciences, Division of Chemical Sciences, Geosci-
ences, and Biosciences through the Ames Laboratory. The Ames
Laboratory is operated for the US Department of Energy by Iowa
State University under Contract No. DE-AC02-07CH11358.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2012.04.005.
References
[1] B. Cornils, W.A. Herrmann, J. Catal. 216 (2003) 23.
[2] F. Rodríguez-Reinoso, Carbon 36 (1998) 159.
[3] X. Yu, X. Yu, S. Wu, B. Liu, H. Liu, J. Guan, Q. Kan, J. Solid State Chem. 184 (2011)
289.
[4] C.M.A. Parlett, D.W. Bruce, N.S. Hondow, A.F. Lee, K. Wilson, ACS Catal. (2011)
636.

68
K. Kandel et al. / Journal of Catalysis 291 (2012) 63–68
[5] J.N. Stuecker, J.E. Miller, R.E. Ferrizz, J.E. Mudd, J. Cesarano, Ind. Eng. Chem. Res.
43 (2003) 51.
[6] A. Corma, H. Garcia, Top. Catal. 48 (2008) 8.
[34] I.I. Slowing, B.G. Trewyn, V.S.-Y. Lin, J. Am. Chem. Soc. 129 (2007) 8845.
[35] H. Yang, G. Zhang, X. Hong, Y. Zhu, Micropor. Mesopor. Mater. 68 (2004) 119.
[36] D.J. Kim, B.C. Dunn, P. Cole, G. Turpin, R.D. Ernst, R.J. Pugmire, M. Kang, J.M.
Kim, E.M. Eyring, Chem. Commun. (2005) 1462.
[7] H. Hattori, Chem. Rev. 95 (1995) 537.
[8] K.K. Rao, M. Gravelle, J.S. Valente, F. Figueras, J. Catal. 173 (1998) 115.
[9] H. Matsuhashi, H. Miyazaki, Y. Kawamura, H. Nakamura, K. Arata, Chem. Mater.
13 (2001) 3038.
[10] D. Tichit, B. Coq, S. Cerneaux, R. Durand, Catal. Today 75 (2002) 197.
[11] T.R. Gaydhankar, P.N. Joshi, P. Kalita, R. Kumar, J. Mol. Catal. A: Chem. 265
(2007) 306.
[12] F. Tanaka, R. Fuller, H. Shim, R.A. Lerner, C.F. Barbas III, J. Mol. Biol. 335 (2004)
1007.
[13] F. Tanaka, R. Fuller, C.F. Barbas III, Biochemistry 44 (2005) 7583.
[14] M. Chatterjee, K. Matsushima, Y. Ikushima, M. Sato, T. Yokoyama, H.
Kawanami, T. Suzuki, Green Chem. 12 (2010) 779.
[15] S. Mukherjee, J.W. Yang, S. Hoffmann, B. List, Chem. Rev. 107 (2007) 5471.
[16] B. List, R.A. Lerner, C.F. Barbas III, J. Am. Chem. Soc. 122 (2000) 2395.
[17] S.M. Dean, W.A. Greenberg, C.-H. Wong, Adv. Synth. Catal. 349 (2007) 1308.
[18] L. Sujandi, S.-E. Park, D.-S. Han, S.-C. Han, M.-J. Jin, T. Ohsuna, Chem. Commun.
(2006) 4131.
[19] S. Luo, J. Li, L. Zhang, H. Xu, J.-P. Cheng, Chem. Eur. J. 14 (2008) 1273.
[20] S. Luo, X. Zheng, J.-P. Cheng, Chem. Commun. (2008) 5719.
[21] T.M. Suzuki, M. Yamamoto, K. Fukumoto, Y. Akimoto, K. Yano, J. Catal. 251
(2007) 249.
[22] K.K. Sharma, T. Asefa, Angew. Chem. 119 (2007) 2937.
[23] S. Shylesh, A. Wagner, A. Seifert, S. Ernst, W.R. Thiel, Chem. Eur. J. 15 (2009)
7052.
[24] B. Liu, S. Wu, X. Yu, J. Guan, Q. Kan, J. Coll. Interface Sci. 362 (2011) 625.
[25] R.K. Zeidan, S.-J. Hwang, M.E. Davis, Angew. Chem. Int. Ed. 45 (2006) 6332.
[26] R.K. Zeidan, M.E. Davis, J. Catal. 247 (2007) 379.
[37] S. Meiboom, D. Gill, Rev. Sci. Instrum. 29 (1958) 688.
[38] J.W. Wiench, V.S.-Y. Lin, M. Pruski, J. Magn. Reson. 193 (2008) 233.
[39] D.E. Purich, Enzyme Kinetics: Catalysis and Control, Elsevier, 2002.
[40] K.B. Taylor, Enzyme Kinetics and Mechanisms, Kluwer Academic, 2002.
[41] J. Wagner, R.A. Lerner, C.F. Barbas III, Science 270 (1995) 1797.
[42] T. Hoffmann, G. Zhong, B. List, D. Shabat, J. Anderson, S. Gramatikova, R.A.
Lerner, C.F. Barbas III, J. Am. Chem. Soc. 120 (1998) 2768.
[43] M. Wiesner, G.g. Upert, G. Angelici, H. Wennemers, J. Am. Chem. Soc 132
(2009) 6.
[44] D.J. Cole-Hamilton, Science 299 (2003) 1702.
[45] J.D. Bass, A. Solovyov, A.J. Pascall, A. Katz, J. Am. Chem. Soc. 128 (2006) 3737.
[46] S. Huh, H.-T. Chen, J.W. Wiench, M. Pruski, V.S.-Y. Lin, Angew. Chem. Int. Ed. 44
(2005) 1826.
[47] L. Forni, G. Fornasari, G. Giordano, C. Lucarelli, A. Katovic, F. Trifiro, C. Perri, J.B.
Nagy, Phys. Chem. Chem. Phys. 6 (2004) 1842.
[48] L. Forni, G. Fornasari, G. Giordano, C. Lucarelli, A. Katovic, F. Triflrò, C. Perri, J.B.
Nagy, Effect of exchange procedure and crystal size on high silica MFI zeolite
as catalyst for vapor phase Beckmann rearrangement, in: M.C.E. van Steen,
L.H. Callanan (Eds.), Stud. Surf. Sci. Catal., Elsevier, 2004, p. 2823.
[49] L.-P.B. Beaulieu, L.B. Delvos, A.B. Charette, Org. Lett. 12 (2010) 1348.
[50] D.W. Sindorf, G.E. Maciel, J. Phys. Chem. 87 (1983) 5516.
[51] V.H. Pan, T. Tao, J.-W. Zhou, G.E. Maciel, J. Phys. Chem. B 103 (1999) 6930.
[52] M.I. Zaki, M.A. Hasan, F.A. Al-Sagheer, L. Pasupulety, Langmuir 16 (1999) 430.
[53] J.H. Drese, A.D. Talley, C.W. Jones, ChemSusChem 4 (2011) 379.
[54] V.Y. Borovkov, A.V. Zaiko, V.B. Kazansky, W.K. Hall, J. Catal. 75 (1982) 219.
[55] I.D. Gay, J. Phys. Chem. 78 (1974) 38.
[27] N. Solin, L. Han, S. Che, O. Terasaki, Catal. Commun. 10 (2009) 1386.
[28] J. Alauzun, A. Mehdi, C. Reye, R.J.P. Corriu, J. Am. Chem. Soc. 128 (2006) 8718.
[29] B.M. Choudary, M.L. Kantam, P. Sreekanth, T. Bandopadhyay, F. Figueras, A.
Tuel, J. Mol. Catal. A: Chem. 142 (1999) 361.
[30] X. Wang, K.S.K. Lin, J.C.C. Chan, S. Cheng, Chem. Commun. (2004) 2762.
[31] J. Mondal, A. Modak, A. Bhaumik, J. Mol. Catal. A: Chem. 335 (2011) 236.
[32] A. Anan, K.K. Sharma, T. Asefa, J. Mol. Catal. A: Chem. 288 (2008) 1.
[33] S. Huh, J.W. Wiench, J.-C. Yoo, M. Pruski, V.S.-Y. Lin, Chem. Mater. 15 (2003)
4247.
[56] T. Bernstein, D. Michel, H. Pfeifer, P. Fink, J. Coll. Interface Sci. 84 (1981) 310.
[57] S. Nedd, T. Kobayashi, C.-H. Tsai, I.I. Slowing, M. Pruski, M.S. Gordon, J. Phys.
Chem. C 115 (2011) 16333.
[58] M.B. Schmid, K. Zeitler, R.M. Gschwind, J. Am. Chem. Soc. 133 (2011) 7065.
[59] H.E. Zimmerman, M.D. Traxler, J. Am. Chem. Soc. 79 (1957) 1920.
[60] S. Bahmanyar, K.N. Houk, J. Am. Chem. Soc. 123 (2001) 11273.
[61] A.C. Dash, B. Dash, D. Panda, J. Org. Chem. 50 (1985) 2905.