Microwave-assisted grafting to MCM-41 silica and its application as catalyst in flow chemistry

Article abstract of DOI:10.1071/CH11125

Finding environmentally gentle methods to graft Lewis acid on the surface of mesoporous materials is a topic of current interest. Herein we describe the optimization of a preparation procedure of a mesoporous silica-supported Er ^III catalyst using the microwave-assisted post-calcination functionalization of Mobil Composition of Matter-41 silica as the key step. The required time for functionalization was reduced from several hours to 10min using sealed-vessel microwave technology. Control experiments using conventional heating at the same temperature demonstrated that the rate increase is owing to a simple thermal/kinetic effect as a result of the higher reaction temperature. The resulting Er^III catalyst was tested for the first time as a catalyst in the continuous flow deprotection of benzaldehyde dimethylacetal and a complete leaching study was performed.

Full text of DOI:10.1071/CH11125

CSIRO PUBLISHING
Aust. J. Chem. 2011, 64, 1522–1529
Full Paper
http://dx.doi.org/10.1071/CH11125
Microwave-Assisted Grafting to MCM-41 Silica
and its Application as Catalyst in Flow Chemistry
A D
,
A
B
Manuela Oliverio,
Antonio Procopio, Toma N. Glasnov,
C
B
Walter Goessler, and C. Oliver Kappe
A
Dipartimento Farmacobiologico,Universita` Magna Graecia Viale Europa,
88100-Germaneto (Cz), Italy.
B
Christian Doppler Laboratory for Microwave Chemistry (CDLMC) and Institute
of Chemistry, Karl-Franzens-University, Heinrichstrasse 28, A-8010 Graz, Austria.
C
Institute for Chemistry Analytical Chemistry, Karl-Franzens University,
]
Universitaetsplatz 1-8010 Graz, Austria.
D
Corresponding author. Email: m.oliverio@unicz.it
Finding environmentally gentle methods to graft Lewis acid on the surface of mesoporous materials is a topic of current
interest. Herein we describe the optimization of a preparation procedure of a mesoporous silica-supported ErIII catalyst
using the microwave-assisted post-calcination functionalization of Mobil Composition of Matter-41 silica as the key step.
The required time for functionalization was reduced from several hours to 10 min using sealed-vessel microwave
technology. Control experiments using conventional heating at the same temperature demonstrated that the rate increase is
owing to a simple thermal/kinetic effect as a result of the higher reaction temperature. The resulting ErIII catalyst was tested
for the first time as a catalyst in the continuous flow deprotection of benzaldehyde dimethylacetal and a complete leaching
study was performed.
Manuscript received: 31 March 2011.
Manuscript accepted: 10 August 2011.
Published onilne: 16 November 2011.
Introduction
involve the use of toxic solvents under conventional heating
conditions.[5a] To the best of our knowledge, there are only two
publications that describe the application of microwave-assisted
procedures to the post-calcination grafting of organic molecules
on the surface of silica-type materials. These involve the
functionalization of controlled-pore glass as support for the
solid phase synthesis of nucleic acids,[5b] and the immobili-
zation of alkaloids on the surface of HPLC-grade silica gel via
click chemistry.[5c] In both of these examples, microwave
irradiation enhanced the rate and the yield of the functionali-
zation compared with the classical procedure, but no rationali-
zation of the observed phenomena was given. More recently,
Garc`ıa and coworkers[6] proposed that microwave heating
increases the silica functionalization rate and yield inducing
autocondensation of polar silylating agents. Compared with the
conventionally heated functionalized silica, a change in the
grafted layer structure was observed.
In two of our recent publications we applied microwave
heating to the synthesis of a new mesoporous silica-supported
ErIII catalyst[7] and have performed an evaluation of microwave
heating for the grafting of different organic moieties on the
surface of the mesoporous silica support.[8] In these studies, we
initially assumed the involvement of a ‘specific microwave
effect’ on the silica functionalization reaction.[4] The aim of
the present work was to re-evaluate the reaction conditions for
the previously described post-calcination functionalization
Heterogeneous catalysis offers several advantages in both
organic synthesis and in industrial processes because catalyst
recovery and recycling is often easier compared with homo-
geneous catalysts, without loss in reaction selectivity and yield.
In addition, heterogeneous catalysts derived from homogeneous
catalysts immobilized on an inorganic support, can be applied in
continuous flow synthesis which is a particularly suitable
technology for performing reactions in a fast, clean and scalable
way.[1] Amorphous silica and mesoporous silica Mobil
Composition of Matter-41 (MCM-41) have been extensively
used as supports for immobilizing homogeneous catalysts and
have recently also been used in continuous flow processes.[2]
The advantages of mesoporous supports are related to their
˚
ordered array of hexagonal channels with pores up to 50 A
diameter. This results in superior access to the catalytically
active sites located on the channels, a large superficial area, high
catalyst loading and minimum diffusional resistance. On the
other hand, these supports are generally temperature and solvent
resistant, and easy to prepare and functionalize.[2]
Alternative heating methods such as microwave dielectric
heating have been used in the past in order to synthesize
functionalized mesoporous MCM-like materials by co-
condensation,[3,4] while until date the reported methodologies
to introduce organic groups on their surface by post-calcination
grafting are often time-consuming, labour-intensive and
Journal compilation Ó CSIRO 2011
www.publish.csiro.au/journals/ajc

MW-Functionalized MCM-41 Silica in Continuous Flow
1523
reactions using a dedicated single-mode microwave reactor with
internal temperature measurement capabilities, in order to more
carefully scrutinize the role of microwave irradiation on these
types of grafting processes.
the toluene and solvent-free experiments. In toluene, the
optimum functionalization level of 1.5 mmol gꢀ1 was obtained
at 1308C, while under solvent-free conditions the loading
was already around 2 mmol gꢀ1 even at 1108C, too high for
catalytic applications. In general, we found that when using
toluene as solvent, the degree of functionalization could be
better controlled than using solvent-free conditions. In this
context, it should be emphasized that proper stirring in these
heterogeneous silica functionalizations is required to ensure
reproducible results. For example, using conical vials for
the Biotage reactor (0.5–2.0 mL),[10a] the agitation induced by
the small stir bar is not optimal and leads to significantly lower
loadings and inhomogeneous samples as obtained in the larger
cylindrical tubes, where appropriate stirring can be ensured.[11]
Finally we studied the dependence of the functionalization
degree on the silica pre-activation step comparing the results
obtained when the same MCM-41 silica sample was pre-treated
with a freshly prepared 25 % (v/v) HCl solution, a one week old
HCl solution and a one month old HCl solution of the same
concentration.
As it can be seen in Fig. 1, no comparison is possible between
samples activated with HCl solutions of different age. It can
therefore be concluded that the activation step and the date of
preparation of the HCl source are crucial for the silica surface
coverage. The experiments described above have shown that the
time required for the functionalization of mesoporous MCM-41
silica with 3-(chloropropyl)triethoxysilane (CPTES) using
toluene as a solvent can be reduced from 8 h under reflux
conditions to 10 min using sealed vessel microwave heating at
1308C. These conditions lead to a medium level of surface
coverage (1.5 mmol gꢀ1), which makes this material suitable for
use as a heterogeneous catalyst. In order to establish if the
acceleration of the grafting performed under microwave condi-
tions is owing to a specific microwave effect as presumed
previously,[8] we have performed careful control experiments
between microwave heating and conventional heating at the
identical reaction temperature of 1308C.
Results and Discussion
In our previous work[8] we performed a rough comparison
between classical[9] and microwave-assisted grafting protocols
involving different silylating agents and MCM-41 silica
supports. Microwave experiments were performed in a multi-
mode system with external IR temperature monitoring.
MCM-41 silica was activated with a 25 % (v/v) HCl solution at
room temperature for 3 h. The acid pretreatment allows to
increase the OH-functionalities on the silica surface.[8,9] The
classical functionalization was time and energy consuming (8 h
at the reflux temperature of toluene, ,1108C) and the loading
was lower or comparable with the loadings achieved after
microwave irradiation within only 15 min. For graftings per-
formed both in toluene and under solvent-free conditions, there
was a strong dependence of the loading on the reaction time,
with loadings ranging from 0.28–1.0 mmol gꢀ1 in toluene up to
1.74 mmol gꢀ1 using solvent-free conditions.[8]
In order to optimize these conditions more carefully and to
re-evaluate the specific role of microwave irradiation on the
grafting process, the functionalization of MCM-41 silica with
3-(chloropropyl)triethoxysilane (CPTES) was chosen as model
reaction (Scheme 1). The grafting process was performed in
single-mode microwave reactors with either external IR
sensors or an internal fibre-optic probe temperature sensor,
the latter allowing for very accurate reaction temperature
measurements.[10]
In a first set of experiments, the microwave-assisted func-
tionalizations were performed in a Biotage Emrys Initiator EXP
2.5 reactor.[10a] In a typical experiment 200 mg of pre-treated
MCM-41 silica (see below) were reacted with a mixture of 1 mL
of CPTES and 2 mL of toluene with magnetic stirring. The
degree of functionalization was determined using Total Organic
Carbon (TOC) measurements in triplicate and is given in mmol
functionality per gram of silica. Using conventional reflux
conditions, the loading after 8 h of reflux heating at 1108C was
,1.5 mmol gꢀ1, which represents an appropriate surface
coverage for catalytic applications.
Initially, the microwave grafting process using CPTES
shown in Scheme 1 was performed at a set temperature of
1508C using either dry toluene as solvent (conditions a) or under
solvent-free conditions (conditions c). Even after only 5 min
reaction time the degree of functionalization was ,1 mmol gꢀ1
using toluene as solvent, and 2 mmol gꢀ1 under solvent-free
conditions. Prolonged exposure to microwave irradiation at
1508C did not further increase the loading for both set of
conditions. In a second series of experiments the functionaliza-
tion was performed for 10 min constant reaction time, varying
the temperature from 110 to 1708C. As expected, the degree of
functionalization increased with higher temperatures for both
HCl 1 month old
HCl 1 week old
HCl fresh
0:00
0:28
0:57
1:26
1:55
2:24
2:52
mmol gϪ1
Fig. 1. Dependence of the loading in the microwave-assisted functionali-
zation (Scheme 1) on the silica pre-activation step using different sources of
25 % (v/v) HCl.
OH
CPTES
OH
Cl
(a) MW, toluene
(b) toluene, reflux
(c) MW, neat
OH
Scheme 1. Functionalization of pre-treated MCM-41 with 3-(chloropropyl)triethoxysilane (CPTES).

1524
M. Oliverio et al.
For this purpose we have used a reactor system that allows us
sample homogeneity. It can therefore be concluded that the
observed rate acceleration seen for this grafting process is the
result of a purely thermal/kinetic phenomenon as a result of the
higher reaction temperature, and can be duplicated in an auto-
clave experiment.[10c]
to perform both types of transformations in the identical reac-
tion vessel and to monitor the internal reaction temperature in
both experiments directly with a fibre-optic probe device.[12]
This set-up can be either immersed into the cavity of a CEM
Discover microwave reactor (CEM Microwave Technology,
Buckingham, UK) or into a preheated and temperature equili-
brated oil bath placed on a magnetic stirrer. In both cases, the
software of the microwave instrument is recording the internal
temperature and similar heating profiles can be obtained (see
below). This system has the advantage that the same reaction
vessel and the same method of temperature measurement are
used. In this way all parameters apart from the mode of heating
are identical and therefore a fair comparison between micro-
wave heating and thermal heating can be made. The reaction
conditions and the temperature profiles recorded for both the
microwave and oil bath experiments at 1308C are shown in
Fig. 2.
Each functionalization has been performed in triplicate
(experiments 1–3) and the results are shown in Table 1. As
can be clearly seen, the TOC measurement results between
microwave-heated and conventionally heated runs are nearly
identical with loadings ranging from 1.88 to 2.33 mmol gꢀ1. All
TOC measurements have been performed in triplicate on the
same sample, and the standard deviations of these three mea-
sures (#0.15 in each case) can be interpreted as a parameter of
Note that the results obtained with the CEM Discover
instrument at 1308C (Table 1, internal fibre-optic temperature
measurement) are in good agreement with the data obtained on
the Biotage Initiator reactor (Fig. 1, external infrared tempera-
ture sensor). The most consistent results were obtained with high
stirring rates, underscoring the importance of proper agitation in
microwave-heated reactions.[11] Heating experiments per-
formed with constant microwave power on suspensions of dried
and pre-treated MCM-41 silica in microwave transparent CCl4
in conjunction with quartz reaction vessels demonstrated that
the silica material itself is only weakly microwave absorbing
and can therefore not be selectively heated by microwave
irradiation (data not shown). This type of experiment has been
used in the past to quickly establish the microwave absorbance
characteristics of materials.[13]
With an optimized microwave-assisted functionalization
procedure in our hands, we again synthesized the Lewis acid
heterogeneous ErIII catalyst that we have already previously
applied in the cyanosilylation of aldehydes and ketones,[7] with
the final aim to use this material as a heterogeneous catalyst in
continuous-flow processes (Scheme 2).
The MCM-Er catalyst was identified by comparison with the
Raman and 13C NMR spectra previously reported,[7] the ErIII
amount present on the silica surface was quantified by induc-
tively coupled plasma mass spectrometer (ICP-MS) (10.2 %,
equivalent to 0.8 mequiv gꢀ1 silica) and the porous structure was
determined by N2 adsorption/desorption isotherms (SBET ¼
575 m2 gꢀ1; pore volume (P/P0 ¼ 0.9) ¼ 0.41 cm3 gꢀ1, see
Accessory Publication). The flow experiments involving
the ErIII silica catalyst were conducted in a high-pressure flow
reactor (ThalesNano X-Cube; Thales Nanotechnology,
Budapest, Hungary) where the heterogeneous Er catalyst is
immobilized in a pre-packed, replaceable stainless steel car-
tridge (60 ꢁ 4 mm i.d., 200 mg MCM-Er).[14] The catalyst
cartridge can be heated to 2008C and the reaction mixture is
pumped through the cartridge using an HPLC pump at pressures
up to 150 bar and flow rates of 0.1 to 3.0 mL minꢀ1. The list of
the cartridges, the amount of ErIII and the support used for each
cartridge is summarized in Table 2. Two cartridges, with two
different amounts of ErIII (0.032 and 0.08 mequiv gꢀ1), were
prepared with this catalyst. A non-swelling material such as
amorphous silica K-60, sieved at 80 mesh, was used as inert
material to dilute the catalyst to the desired concentration.
Samples of ,200 mg of these materials were then packed in
the stainless steel cartridge (C1 and C2, Table 2).
160
140
120
100
80
60
40
20
0
0
2
4
6
8
10
12
Time [min]
MW
CONV
Fig. 2. Temperature profiles for the microwave-assisted (MW) and con-
ventionally heated (CONV) silica functionalization reaction with CPTES
(Scheme 1). Conditions: 200 mg of pre-treated MCM-41 (freshly prepared
25 % (v/v) HCl solution), 1 mL CPTES, 2 mL of dry toluene, 1308C, 10 min
fixed hold time, 10 mL Pyrex vessel. MW: CEM Discover, max. magnetron
output power 150 W, magnetic stirring ‘high’. CONV: magnetic stirring
400 rpm.
To establish whether microwaves could have any effect on
the distribution of the functionalities on the silica surface,
Table 1. Comparison of microwave (MW) and conventional heating (CONV) in the functionalization of
MCM-41 silica with 3-(chloropropyl)triethoxysilane (CPTES) in toluene at 1308CA
Conditions
TOC [mmol gꢀ1
Experiment 1
]
TOC [mmol gꢀ1
Experiment 2
]
TOC [mmol gꢀ1
Experiment 3
]
MW
2.01 ꢂ 0.15
1.94 ꢂ 0.04
2.01 ꢂ 0.04
2.04 ꢂ 0.14
2.08 ꢂ 0.14
CONV
2.33 ꢂ 0.006
AFor conditions, see Figure 2.

MW-Functionalized MCM-41 Silica in Continuous Flow
1525
MCM-41, toluene
SH
(H3CO)3Si
SH
MW, 130ЊC, 10 min
H2O2 (30 %)
rt, 24 h
ErCl3, MeCN
reflux, 24 h
SO3ErCl2
SO3H
MCM-Er
Scheme 2. Preparation of a heterogeneous ErCl3 catalyst (MCM-Er).[7]
Table 2. List of the Er(III) silica catalyst cartridges for the continuous
flow experiments
room temperature. After washing the system with 5 mL of pure
solvent the mixture collected at the exit of the reactor was
analyzed by GC/MS in order to determine the reaction conver-
sion and yield. The results are summarized in Table 3. Using the
C1 cartridge (Table 2) full conversion was achieved up to a flow
rate of 1.4 mL minꢀ1 (Table 3, entry 5). No differences were
recorded using the cartridge C5 (data not shown), demonstrating
that there are no differences in morphology between a micro-
wave and conventionally prepared catalyst.
Subsequently, the reaction was carried out in three additional
solvents; toluene, THF, and CH3CN, different in polarity from
nitromethane (entries 6 to 14, Table 3). Although nitromethane
was the best performing solvent, we recorded full conversion
also in toluene at 508C (entry 7, Table 3) and, using a more
concentrated cartridge, in CH3CN at 758C (entry 14, Table 3),
whereas no conversion was observed in THF even using more
drastic conditions (entries 8 to 11, Table 3).
A comparison between C1/C2 cartridges (made by functio-
nalized silica) and C6/C7 cartridges (filled with a mixture of
ErCl3 and MCM-41 in the same amount present in C1 and C2
respectively) in terms of conversion efficiency was performed
(entries 18 to and 20, Table 3). The cartridges containing the
same amount of ErIII prepared mixed with non-functionalized
MCM-41 silica are less effective compared with their analogues
filled with functionalized silica.
A completely different behaviour was recorded for the C3
cartridge (Table 2) filled with mesoporous MCM-41 silica as
inert material (Table 3, entries 15–17). Due to the small size of
the particles, the recorded inlet pressure flushing the reaction
mixture through this cartridge was very high, reaching almost
the maximum value (150 bar) allowed by the instrument at
1.5 mL minꢀ1 flow rate. For safety reasons we chose a flow rate
of 1.0 mL minꢀ1 to perform our experiments with this material.
An important parameter that has to be evaluated in all flow
chemistry involving heterogeneous metal catalysts is the leach-
ing of the metal from the support into the reaction mixture. In
order to measure the leaching we performed ICP-MS measure-
ments on the amount of ErIII found in the solution collected at the
exit of the system (Table 4). This technique has been found quite
useful in the past to establish if an immobilized catalyst is
leached from the support or not. Leaching is likely to be a
common problem for all metal-catalyzed transformations that
involve a quasi-homogeneous mechanism where catalysis pro-
ceeds through a dissolution/re-adsorption process of the metal
species from the support. Under flow conditions the active metal
catalyst will be leached and transported away from the support,
thereby constantly reducing the activity of the immobilized
catalyst and contaminating the product with metal. In a conven-
tional batch experiment the catalytically active metal will also
Cartridge
Inert Material
Er(III) Amount
[mmol gꢀ1
]
C1
C2
C3
C4
C5
C6
C7
K-60 silica
0.032
0.08
K-60 silica
MCM-41 silica
MCM-41 silica
K-60 silica
0.032
0.08
0.032A
0.032B
0.08B
K-60 silica
K-60 silica
ACatalyst prepared in a conventional way.
BCartridge prepared by mixing of ErCl3, MCM-41 silica and K-60 silica.
leading to possible changes in activity, a small amount of
catalyst was prepared in a conventional way and then mixed
with the amorphous silica K-60 to prepare the last cartridge
(C5 in Table 2). Two other cartridges were prepared by mixing
the catalyst with commercial MCM-41 silica as support (C3 and
C4, Table 2).
Finally, two additional cartridges (C6 and C7, Table 2)
containing the same amount of ErCl3 (respectively as C1 and
C2 cartridges) and non-functionalized MCM-41 silica, were
prepared by mixing these materials with the appropriate amount
of K-60 silica as inert material.
As an initial trial to study the performance of this immobi-
lized ErIII catalyst in continuous-flow mode we studied the
deprotection of benzaldehyde dimethylacetal to benzaldehyde
as a model transformation. Previous work from our laboratory
on the Lewis acid activity of ErIII salts in homogeneous phase
has demonstrated that it is possible to achieve full conversion of
the acetal to the aldehyde within only 50 min simply mixing it
with 1 mol % of Er(OTf)3 in MeNO2 saturated with water.[15]
Gratifyingly, the same result was obtained when the reaction
was performed in the presence of 1 mol % of the heterogeneous
MCM-Er catalyst, as shown in Scheme 3, whereas only a partial
deprotection was recorded when the reaction was performed
both without any catalyst and with the same amount of commer-
cial MCM-41 silica (10 and 20 % yield respectively after 6 h).
In light of this result, we planned to perform the reaction
under continuous-flow conditions using the more diluted car-
tridges C1, C3, and C5. The cartridges were washed with wet
nitromethane and used immediately, considering the optimum
conditions found in the batch experiment (Scheme 3). Thus
1 mL of a solution of benzaldehyde dimethylacetal in nitrometh-
ane saturated with water (7.6 mg mLꢀ1, 0.05 mmol mLꢀ1) was
passed through the MCM-Er cartridges at different flow rates at

1526
M. Oliverio et al.
OMe
O
OMe
H
1 mol % MCM-Er, MeNO2
rt, 50 min
Ͼ99 %
Scheme 3. Deprotection of benzaldehyde dimethylacetal using heterogeneous MCM-Er catalyst (Scheme 2).
Conversion was determined by GC/MS and the product was identified by comparison with the data previously
reported in the literature.[15]
Table 3. Catalytic efficiency of heterogeneous MCM-Er catalyst in the continuous-flow deprotection of benzaldehyde dimethylacetal (Scheme 3)A
Entry
Cart
Solvent
Flow rate
[mL minꢀ1
T
Inlet pressure
[bar]
Conv
[%]B
]
[8C]
1
C1
C1
C1
C1
C1
C1
C1
C1
C1
C1
C2
C1
C2
C2
C3
C3
C3
C6
C7
C7
CH3NO2
CH3NO2
CH3NO2
CH3NO2
CH3NO2
Toluene
Toluene
THF
0.5
1
25
25
25
25
25
25
50
25
50
75
75
90
50
75
25
25
25
50
50
75
2
2
100
100
100
100
90
2
3
1.3
1.4
1.5
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1
2
4
2
5
2
6
2
29
7
2
100
0
8
2
9
THF
2
0
10
11
12
13
14
15
16
17
18
19
20
THF
2
0
THF
2
0
CH3CN
CH3CN
CH3CN
CH3NO2
CH3NO2
CH3NO2
Toluene
Toluene
CH3CN
2
69
2
75
2
100
100
100
100
31
105
125
135
2
1.2
1.3
1.4
1.4
1.4
2
98
2
24
AFor experimental details, see main text.
BConversions were determined by GC/MS.
Table 4. Leaching of Er(III) catalyst in the continuous-flow deprotection of benzaldehyde dimethylacetal (Scheme 3)A,B
Entry
Cart
T [8C]
Solvent
Washing solvent
[mL]B
Reaction mixture
[mL]
Er leaching
[%]
1
C1
C1
C1
C1
C1
C1
C2
C1
C2
C3
C3
C6
C6
C7
C7
C7
25
25
25
50
25
75
75
90
75
25
25
25
50
50
75
75
CH3NO2
CH3NO2
Toluene
Toluene
THF
6
0
6
0
6
0
6
6
0
6
0
6
0
0
6
0
0
6
0
6
0
6
0
0
6
0
6
0
6
6
0
6
0.008
1.2
2
3
0.00077
0.00017
20.4
4
5
6
THF
15.7
7
THF
0.6
8
CH3CN
CH3CN
CH3NO2
CH3NO2
Toluene
Toluene
Toluene
CH3CN
CH3CN
8.36
9
0.87
10
11
12
13
14
15
16
0.0002
0.00026
0.0007
0.00016
0.00003
1.76
1.4
ALeaching analysis was performed by ICP-MS measurements.
BAnalysis performed on 6 ml of washing solvent/reaction mixture flowing at the rate of 1.4 mL minꢀ1 for C1, C2, C6 and C7 cartridges and 1.0 mL minꢀ1
for C3 cartridge.

MW-Functionalized MCM-41 Silica in Continuous Flow
1527
R1
In order to test the strength of the method, we checked the
possibility to cleave some more resistant acetals such as
the 20,30-O-isopropylidene-uridine and 2,3-O-isopropylidene-
D-ribolactone.[17] For solubility reasons, we used nitromethane
and acetonitrile as solvents. The reactions were performed using
both C1 and C2 cartridges with a flow rate ranging from 0.3 to
0.5 mL minꢀ1 and in a temperature region between 25 and
1108C. In all cases only the starting material was found in the
processed samples, confirming that the solvent has critical
importance for the outcome of the reaction.
5 mol-% MCM-Er
toluene
O
100ЊC, 4 h
O
R2
2a,b (80 %)
R1
R2
1a R1 ϭ Ph, R2 ϭ H
1b R1 ϭ Ph, R2 ϭ Ph
Scheme 4. Epoxide rearrangement to aldehydes in heterogeneous phase.
be leached, but at the end of the reaction is likely to be
redeposited/re-adsorbed on the support. Phenomena like this
have been observed for Pd/C or Cu/C catalyzed processes such
as Heck couplings or azide-alkyne cycloaddition reactions.[16]
The amount of catalyst lost during the reaction was calcul-
ated as % weight in comparison to the total amount of ErIII
present in each cartridge, considering that ,23 versus ,57 mg
of MCM-Er catalyst is present in each cartridge, with a total
amount of ,3 versus ,7.5 mg ErIII per C1/C6 versus C2/C7
cartridges. As can be seen from the data presented in Table 4,
there is a significant amount of leaching taking place when
nitromethane (entries 1 and 2, Table 4), THF (entries 5 to 7,
Table 4) and acetonitrile (entries 8 and 9, Table 4) are passed
through the cartridges containing the Er catalyst. The leaching
values are significantly lower using toluene as solvent (entries
3 and 4, Table 4). In general there was an inverse correlation
between the conversion and the leaching: the lower the metal
leaching was in a solvent, the higher the conversion. Note that
despite the high inlet pressure, considerably less Er leaching is
experienced for the cartridges filled with MCM-41 as solid
support (Table 4, entries 10 and 11).
Regarding the C6 and C7 cartridges, the metal leaching was
absolutely comparable to the corresponding cartridges filled
with functionalized silica. In this case, the drop of conversion
(see Table 3) was not directly related to the metal leaching but
rather to the loss of the proximity effect between the metal centre
and the pores within the silica.
As one would expect, the observed leaching was accompa-
nied by a loss of activity in the cartridge. In a preparative
experiment, a 7.6 mg mLꢀ1 solution of benzaldehyde dimethy-
lacetal in aqueous nitromethane was pumped through the
cartridges C1 and C3, continuously measuring the conversion
to benzaldehyde in the collected solution at the exit of the
system. Under these conditions, the conversion drops from 100
to 90 % in only 15 min, and to 70 % in 25 min (see Accessory
Publication).
As reported in the literature,[15] the deprotection reaction of
benzaldehyde dimethylacetal is influenced by the solvent polar-
ity. Polar solvents are the best choice in batch but, because of the
high ErIII leaching associated to these solvents (entries 1, 2, 8, 9,
15, 16, Table 4,) we checked also less polar solvents in flow.
The oxyphilic character of the ErIII could explain the high
leaching/low yields obtained using THF as solvent (entries 5–7,
Table 4).
From a comparison between the leaching data (Table 4) and
the conversion results (Table 3), we decided to process benzal-
dehyde dimethylacetal in toluene as solvent at 508C and
1.4 mL minꢀ1 flow rate as optimized continuous-flow condi-
tions. Considering a leaching of 0.00017 % after 4.28 min
(corresponding to 6 mL of mixture passing through the cartridge
at 1.4 mL minꢀ1), we can realize a scale up to roughly 400 kg of
benzaldehyde dimethylacetal processed before that the catalyst
loss 20 % of its activity.
Finally, we decided to apply the heterogeneous ErIII catalyst
in the rearrangement of epoxides to aldehydes (Scheme 4). We
have previously found that Er(OTf)3 is a very good catalyst for
this interesting transformation in homogeneous phase,
achieving full conversion both for the styrene oxide (1a) and
the trans-stilbene oxide (1b) in 45 min at room temperature in
presence of 1 mol % of Er catalyst.[18] Unfortunately, the reac-
tion requires more drastic conditions when performed using the
heterogeneous MCM-41 Er catalyst (Scheme 2) and only 80 %
conversion was obtained when both starting materials were
reacted in toluene at 1008C for 4 h in the presence of 5 mol %
of MCM-Er (the conversions were determined by GC/MS and
the products were identified by comparison with the data
reported in the literature).[18] Clearly, given the long reaction/
residence times, such a process is not well suited for a continu-
ous flow approach.
Conclusions
Herein we have performed a detailed investigation on the role of
microwave irradiation on the post-calcination functionalization
of mesoporous MCM-41silica with CPTES. Control experi-
ments using microwave and oil bath heating at the same internal
reaction temperature of 1308C have confirmed that the observed
rate enhancements compared with conventional reflux condi-
tions are the results of a purely thermal/kinetic effect related to
the higher reaction temperatures. Apart from temperature and
reaction time, the effects of pre-activation and stirring efficiency
were additionally studied. The optimized silica functionaliza-
tion method was also applied to the synthesis of a heterogeneous
ErIII catalyst and the application of this material in the contin-
uous flow deprotection of benzaldehyde dimethylacetal was
evaluated. The high reaction conversions and rates suggest a
good potential application of these materials to a cleaner con-
tinuous flow technique; at the same time the leaching studies
suggest that the efficiency of the method is strictly related to
both the lowering of the metal leaching and the proximity
between the catalytic active centre and the pores on the silica
surface.
Experimental
All chemicals were obtained from Aldrich Chem. Co (Milan,
Italy), Acros Organics (Geel, Belgium), or Alfa Aesar
(Karlsruhe, Germany) and used as received. Microwave irradi-
ation experiments in the Biotage Initiator EXP 2.5 single-mode
reactor (2.45 GHz, 400 W) used an external IR sensor and fol-
lowed standard procedures.[10a] Similarly, reactions in the CEM
Discover LabMate single-mode instrument (2.45 GHz, 300 W)
were performed using an internal fibre-optic sensor as previ-
ously reported.[11] Continuous-flow reactions were performed
in an X-Cube from Thales Nanotechnology Inc.[19] TOC mea-
surements were performed in triplicate for each sample on a
Analytik Jena multi N/C 2100 analyzer equipped with a

1528
M. Oliverio et al.
HT 1300 oven. Raman spectra were recorded on a Renishaw
inVia microRaman spectrophotometer using a laser power of
1.2 mW and a laser excitation wavelength of 514 nm. FT-IR
spectra were performed on a Jasco 430 instrument 13C HR-MAS
NMR spectra were recorded on a Bruker Avance 500 MHz in-
strument at 298 K. The characterization of the porous structure
was obtained by N2 adsorption/desorption isotherms, measured
at 77 K on a Micromeritics ASAP 2010 volumetric adsorption
analyzer. Before measurements, samples were outgassed at
393 K for 6 h. These isotherms were used to evaluate Brunauer,
Emmett, and Teller (BET) specific surface area, pore volume,
and pore size distributions. The pore size distribution has been
calculated using an algorithm based on the Barrett, Joyner and
Halenda (BJH) theory. Er leaching in the reaction mixtures was
determined with an Agilent 7500ce inductively coupled plasma
mass spectrometer (Agilent, Waldbroinn, Germany). The
effluents from the cartridges were collected and evaporated to
dryness. The residues were quantitatively transferred with nitric
acid into the 12 mL quartz vials of the MLS UIltraClave III. The
temperature was ramped in 30 min to 2508C and kept at this
temperature for a further 30 min. Er was quantitatively deter-
mined at m/z 166 with an ICP-MS. A calibration was performed
with an external calibration curve established from 1.000g Er Lꢀ1
standard (CPI International, Amsterdam, The Netherlands).
Indium was used as an internal standard. GC-MS spectra
were recorded using a Thermo Focus GC coupled with a Thermo
DSQ II (EI, 70 eV). A HP5-MS column (30 m ꢁ 0.250 mm ꢁ
0.025 mm) was used with helium as carrier gas (1 mL minꢀ1
constant flow). The injector temperature was set to 2808C. After
1 min at 508C the temperature was increased in 258C minꢀ1 steps
up to 3008C and kept at 3008C for 4 min. 1H NMR spectra were
recorded on a Bruker 300 MHz instrument. Low resolution mass
spectra were obtained on an Agilent 1100 LC/MS instrument
using atmospheric pressure chemical ionization in positive or
negative mode.
oil bath conditions with internal temperature control were
performed in a similar manner following the general protocols
previously described.[12]
General Procedure for the Synthesis of the Heterogeneous
Catalyst MCM-ErIII(Scheme 2)
Pre-treated MCM-41 (1 g) was reacted with 3-(mercaptopropyl)
trimethoxysilane (5 mL) in dry toluene (10 mL) for 10 min at
1308C in a sealed 10–20 mL microwave process vial equipped
with a stirring bar (Initiator 2.5). The resulting solid was filtered,
washed three times with THF and extracted for 2 h in CH2Cl2/
Et2O mixture using a Soxhlet extractor, then dried under
vacuum and stored overnight at 1008C. The solid materials
possessing a thiol moiety bonded (1) was oxidized with 30 %
H2O2 (20 mL) for 24 h at room temperature. After filtration and
washing with water (3 ꢁ 3 mL) and ethanol (3 ꢁ 3 mL), the solid
material (2) was treated with ErCl3 (0.573 g) in acetonitrile
(10 mL) at reflux temperature for 24 h. After cooling to room
temperature the mixture was filtered, washed with acetonitrile,
extracted for 2 h in a CH2Cl2/Et2O mixture using a Soxhlet
extractor, then dried under vacuum and finally kept at 908C
overnight. The resulting heterogeneous MCM-Er catalyst
obtained was stored under anhydrous conditions. The catalyst
was identified by comparison with the Raman and 13C NMR
data reported in the literature.[7] ICP-MS: (weight %) ¼
10.2 (% Er).
Deprotection of Benzaldehyde Dimethylacetal (Scheme 3)
Deprotection of Benzaldehyde Dimethylacetal
Under Batch Conditions
A solution of benzaldehyde dimethylacetal (30.44 mg,
0.2 mmol) and 1 mol % MCM-Er (3 mg) in 4 mL of MeNO2
saturated with water was stirred at room temperature for 50 min.
The catalyst was removed by filtration and the yield was
determined by GC/MS. The product was identified by compari-
son with the 1H NMR and MS data reported in the literature.[15]
General Procedures for the Microwave-assisted
Functionalization of MCM-41
Deprotection of Benzaldehyde Dimethylacetal
Under Continuous-flow Conditions
Pretreatment of MCM-41 Silica
Commercial MCM-41 (1 g) was treated with 25 mL of 25 %
(v/v) HCl for 3 h at reflux temperature. The temperature was
lowered to room temperature and the mixture was filtered. The
solid material was washed with water and dried overnight at
1008C. The HCl solution was prepared by dilution of the
commercial available HCl solution (33 % v/v) and stored in a
Pyrex volumetric flask (amber glass) at room temperature. The
solution had a shelf life of 1 week.
K-60 silica (310 mg) previously sieved to 80 mesh and dried
at 3508C for 3 h, was mixed with the MCM-Er catalyst (40 mg,
0.032 mmol Er). The powder was packed in a stainless steel
cartridge (60 ꢁ 4 mm i.d.) from ThalesNano. The cartridge was
inserted in the X-cube reactor, warmed to 508C and washed with
toluene (10 mL). Subsequently, 1 mL of a solution of benzalde-
hyde dimethylacetal (7.6 mg mLꢀ1) in toluene was passed
through the cartridge at a 1.4 mL minꢀ1 flow rate. The crude
solution was collected at the bottom of the cartridge after rinsing
with toluene (5 mL). The yield was determined by GC/MS. The
product was identified by comparison with the 1H NMR and MS
data reported in the literature.[ 15]
MW-assisted Functionalization of MCM-41
Pre-treated MCM-41 (200 mg) suspended in 2.0 mL of dry
toluene was reacted with CPTES (1 mL) for 10 min at 1308C in a
sealed 10 mL microwave process vial equipped with a stirring
bar (Initiator 2.5). At the end of the reaction the system was
cooled to room temperature. The solid was filtered, washed
three times with THF and extracted for 2 h in CH2Cl2/Et2O
mixture using a Soxhlet extractor, then dried under vacuum and
stored overnight at 708C. The functionalized mesoporous mate-
rial obtained was stored under anhydrous conditions. The
functionalized silica was identified by comparison with the
FT-IR and 13C NMR data reported in the literature.[8] TOC
measurement (weight %) ¼ 5.46 ꢂ 0.5 (% C). Experiments
performed in a CEM Discover instrument using microwave or
ErIII-Leaching Studies
A cartridge corresponding to each reaction test was prepared as
previously described. Each new cartridge was inserted in the
X-cube reactor, heated to the appropriate temperature and
washed with solvent (10 mL). The flow rate (1.4 mL minꢀ1) was
selected on the flow reactor and the processing was started,
whereby only pure solvent was pumped through the system until
a fraction of 6 mL was collected. At that point the inlet tube was
switched to a 10 mL vial containing 6 mL of a solution of

MW-Functionalized MCM-41 Silica in Continuous Flow
1529
benzaldehyde dimethylacetal (7.6 mg mLꢀ1) in solvent. After
the reaction mixture was consumed, the inlet tubing was moved
back into a vial of pure solvent and further processing was
continued. A fraction of 6 mL was collected (containing prod-
uct). New processing was started once more with the same
procedure. The fraction containing pure solvent and those
containing the product were evaporated under reduced pressure
and the samples were analyzed by ICP-MS.
(b) S. Caddick, R. Fitzmaurice, Tetrahedron 2009, 65, 3325.
doi:10.1016/J.TET.2009.01.105
[5] (a) G. E. Fryxell, Inorg. Chem. Commun. 2006, 9, 1141. doi:10.1016/
J.INOCHE.2006.06.012
(b) S. Padmanabhan, J. E. Coughlin, R. P. Iyer, Tetrahedron Lett. 2005,
46, 343. doi:10.1016/J.TETLET.2004.09.189
(c) K. M. Kacprzak, N. M. Maier, W. Lindner, Tetrahedron Lett. 2006,
47, 8721. doi:10.1016/J.TETLET.2006.10.018
[6] (a) N. Garcia, E. Benito, J. Guzma`n, R. de Francisco, P. Tiemblo,
Langmuir 2010, 26, 5499. doi:10.1021/LA903743N
Rearrangement of Styrene Oxide Under Batch Conditions
(Scheme 4)
(b) N. Fukaya, H. Yamashita, H. Haga, T. Tsuchimoto, S. Onozawa,
T. Sakakura, H. Yasuda, J. Organomet. Chem. 2011, 825.
[7] A. Procopio, G. Das, M. Nardi, M. Oliverio, L. Pasqua, ChemSusChem
2008, 1, 916. doi:10.1002/CSSC.200800183
A solution of styrene oxide 1a (25 mg, 24 mL, 0.2 mmol) and
5 mol % MCM-Er catalyst (12.5 mg) in 0.75 mL of toluene was
stirred at 1008C in a sealed vial for 4 h. The catalyst was removed
by filtration and the conversion was determined by GC/MS to be
80 %. The product was identified by comparison with the
1H NMR and MS data reported in the literature.[ 18]
[8] A. Procopio, M. Oliverio, R. Paonessa, M. Nardi, G. De Luca, Green
Chem. 2009, 11, 770. doi:10.1039/B820417A
[9] (a) J. H. Clark, A. J. Butterworth, S. J. Tavener, A. J. Teasdale,
J. Chem. Technol. Biotechnol. 1997, 68, 367. doi:10.1002/(SICI)1097-
4660(199704)68:4,367::AID-JCTB618.3.0.CO;2-H
(b) D. Brunel, Microporous Mesoporous Mater. 1999, 27, 329.
doi:10.1016/S1387-1811(98)00266-2
Accessory Publication
N2 absorbtion and desorbtion curves for MCM-ErIII pores
measurements and GC-MS profile for the continuous-flow
deprotection of benzaldehyde dimethylacetal in nitromethane
are available on the Journal’s website.
(c) S. Jaenicke, G. K. Chuah, X. H. Lin, X. C. Hu, Microporous
Mesoporous Mater. 2000, 35–36, 143. doi:10.1016/S1387-1811(99)
00215-2
(d) L. R. Hilliard, X. Zhao, W. Tan, Anal. Chim. Acta 2002, 470, 51.
doi:10.1016/S0003-2670(02)00538-X
(e) L. H. N. Arakaki, J. G. P. Esp´ınola, M. G. da Fonseca, S. F. de
Oliveira, A. N. de Sousa, T. Arakaki, C. Airoldi, J. Colloid Interface
Sci. 2004, 273, 211. doi:10.1016/J.JCIS.2004.01.006
Acknowledgements
This work was supported by a grant from the Christian Doppler Society
(CDG). We thank Thales Nanotechnology Inc. for the provision of the
X-Cube reactor.
[10] (a) For a survey of commercially available microwave reactors, see:
C. O. Kappe, D. Dallinger, S. S. Murphree, in Practical Microwave
Synthesis for Organic Chemists – Strategies, Instruments, and Proto-
cols, 2009, ch. 3, pp. 45–85 (Wiley-VCH: Weinheim).
(b) D. Obermayer, C. O. Kappe, Org. Biomol. Chem. 2010, 8, 114.
doi:10.1039/B918407D
References
[1] (a) J. M. Thomas, R. Raja, D. W. Lewis, Angew. Chem. Int. Ed. 2005,
44, 6456. doi:10.1002/ANIE.200462473
(b) A. Corma, H. Garc`ıa, Chem. Rev. 2003, 103, 4307. doi:10.1021/
CR030680Z
(c) D. Obermayer, B. Gutmann, C. O. Kappe, Angew. Chem. Int. Ed.
2009, 48, 8321. doi:10.1002/ANIE.200904185
(c) L. Soldi, W. Ferstl, S. Loebbecke, R. Maggi, C. Malmassarri,
G. Sartori, S. Yada, J. Catal. 2008, 258, 289. doi:10.1016/J.JCAT.
2008.07.005
[11] (a) For previous reports on the importance of stirring in microwave-
assisted reactions, see: M. A. Herrero, J. M. Kremsner, C. O. Kappe,
J. Org. Chem. 2008, 73, 36. doi:10.1021/JO7022697
[2] (a) A. P. Wight, M. E. Davis, Chem. Rev. 2002, 102, 3589.
doi:10.1021/CR010334M
(b) J. D. Moseley, P. Lenden, A. D. Thomson, J. P. Gilday, Tetrahedron
Lett. 2007, 48, 6084. doi:10.1016/J.TETLET.2007.06.147
(c) R. O. M. A. de Souza, O. A. C. Antunes, W. Kroutil, C. O. Kappe,
J. Org. Chem. 2009, 74, 6157. doi:10.1021/JO9010443
(d) M. Irfan, M. Fuchs, T. N. Glasnov, C. O. Kappe, Chemistry 2009, 15,
11608. doi:10.1002/CHEM.200902044
(b) G. J. de A.A. Soler-Illia, C. Sanchez, B. Lebeau, J. Patarin, Chem.
Rev. 2002, 102, 4093. doi:10.1021/CR0200062
(c) A. P. Bhatt, K. Pathak, R. V. Jasra, R. I. Kureshy, N. H. Khan,
S. H. R. Abdi, J. Mol. Catal. A 2006, 244, 110. doi:10.1016/
J.MOLCATA.2005.08.045
[12] M. Hosseini, N. Stiasni, V. Barbieri, C. O. Kappe, J. Org. Chem. 2007,
(d) X.-J. Feng, X.-B. Lu, R. He, Appl. Catal. A Gen. 2004, 272, 347.
doi:10.1016/J.APCATA.2004.06.007
(e) M. Vicevic, K. V. K. Boodhoo, K. Scott, Chem. Eng. J. 2007, 133,
43. doi:10.1016/J.CEJ.2006.12.003
72, 1417. doi:10.1021/JO0624187
[13] (a) T. Razzaq, J. M. Kremsner, C. O. Kappe, J. Org. Chem. 2008, 73,
6321. doi:10.1021/JO8009402
(b) J. M. Kremsner, C. O. Kappe, J. Org. Chem. 2006, 71, 4651.
doi:10.1021/JO060692V
[14] For a detailed description, see: C. Csaja´gi, B. Borcsek, K. Niesz,
[3] (a) S.-E. Park, E. A. Prasetyanto, Top. Catal. 2009, 52, 91.
doi:10.1007/S11244-008-9152-6
¨
I. Kova´cs, Z. Sze´kelyhidi, Z. Bajko´, L. Urge, F. Darvas, Org. Lett.
(b) M. C. A. Fantini, J. R. Matos, L. C. C. da Silva, L. P. Mercuri, G. O.
Chiereci, E. B. Celer, M. Jaroniec, Mater. Sci. Eng. B-Solid State
Mater. Adv. Technol. 2004, 112, 106. doi:10.1016/J.MSEB.2004.
05.013
2008, 10, 1589. doi:10.1021/OL7030894
[15] R. Dalpozzo, A. De Nino, L. Maiuolo, M. Nardi, A. Procopio,
A. Tagarelli, Synthesis 2004, 496.
(c) B. L. Newalkar, H. Katsuki, S. Komarneni, Microporous Mesoporous
Mater. 2004, 73, 161. doi:10.1016/J.MICROMESO.2004.05.002
(d) E. A. Prasetyanto, S.-C. Lee, S.-M. Jeong, S.-E. Park, Chem. Comm.
2008, 1995.
[16] (a) T. N. Glasnov, S. Findenig, C. O. Kappe, Chem. Eur. J. 2009, 15,
1001. doi:10.1002/CHEM.200802200
(b) M. Fuchs, W. Goessler, C. Pilger, C. O. Kappe, Adv. Synth. Catal.
2010, 352, 323. doi:10.1002/ADSE.2009.00726
(e) P. Liu, Z. Su, Mater. Chem. Phys. 2005, 94, 412. doi:10.1016/
J.MATCHEMPHYS.2005.05.023
[17] A. Procopio, M. Gaspari, M. Nardi, M. Oliverio, R. Romeo, Tetrahe-
dron Lett. 2008, 49, 1961. doi:10.1016/J.TETLET.2008.01.089
[18] A. Procopio, R. Dalpozzo, A. De Nino, L. Maiuolo, M. Nardi,
B. Russo, Adv. Synth. Catal. 2005, 347, 1447. doi:10.1002/ADSC.
200505096
[19] The Cube product series are available from Thales Nanotechnology
Inc. (Budapest, Hungary). For further information, please refer to
http://www.thalesnano.com (accessed 3 March 2011).
(f) T.-Z. Ren, Z.-Y. Yuan, B.-L. Su, Coll. Surf. A 2007, 88.
(g) Sujandi, S.-C. Han, D.-S. Han, M.-J. Jin, S.-E. Park, J. Catal.
2006, 243, 410. doi:10.1016/J.JCAT.2006.08.010
[4] (a) For general reviews on microwave chemistry, see: C. O. Kappe,
D. Dallinger, Mol. Divers. 2009, 13, 71. doi:10.1007/S11030-009-
9138-8