Continuous synthesis of nanostructured silica based materials in a gas-liquid segmented flow tubular reactor

Article abstract of DOI:10.1039/c5nj03033a

A continuous synthesis of several spherical silica structures-by means of a gas-liquid segmented flow tubular reactor-is reported. Specifically, as proof of concept, we showed that 300-400 nm mesoporous core-shell spheres (SiO₂@mSiO₂), mesoporous spheres (mSiO₂) as well gold-encapsulated spheres (Au@SiO₂) can be effectively produced in a continuous manner in a tubular reactor. Thus, the successful conversion of classical batch methods to continuous processes opens new possibilities for the up-scaled synthesis of advanced nanostructured materials.

Full text of DOI:10.1039/c5nj03033a

NJC
View Article Online
View Journal | View Issue
PAPER
Continuous synthesis of nanostructured silica
based materials in a gas–liquid segmented flow
tubular reactor†
Cite this: New J. Chem., 2016,
40, 4361
Johannes Knossalla, Stefano Mezzavilla and Ferdi Schuth*
¨
A continuous synthesis of several spherical silica structures – by means of a gas–liquid segmented flow
tubular reactor – is reported. Specifically, as proof of concept, we showed that 300–400 nm meso-
porous core–shell spheres (SiO2@mSiO2), mesoporous spheres (mSiO2) as well gold-encapsulated spheres
(Au@SiO2) can be effectively produced in a continuous manner in a tubular reactor. Thus, the successful
conversion of classical batch methods to continuous processes opens new possibilities for the up-scaled
synthesis of advanced nanostructured materials.
Received (in Montpellier, France)
30th October 2015,
Accepted 14th March 2016
DOI: 10.1039/c5nj03033a
www.rsc.org/njc
devices; for this reason, the typical production volumes achiev-
able through such devices are higher compared to the so called
Introduction
The synthesis of advanced porous materials is receiving great ‘‘lab on a chip’’ technologies.6 On a lab scale, several (nano-
scientific attention, especially due to their application as structured) materials, i.e. monodisperse titania,4c,5e,7 silica7a,8
catalysts and support materials.1 In the last few decades, large and carbonate4e beads, have been produced by means of TRs.
families of different materials have been synthesized and Additionally, through the use of multi-stage TRs, the syntheses
studied, although very often the possibility of tailoring the of more complex nanostructures such as CdSe/ZnS, Au/Ag,
material features – by tuning the morphology, particle size, SiO2/TiO2 or Au/Ag/Au were described.9
and porosity, for instance – remains a major challenge. Con-
Silica spheres – or the more advanced dense silica core
ventionally, batch processes are applied to synthesize porous mesoporous silica shell (SiO2@mSiO2) spheres – represent
nanostructured materials.2 However, in many cases, the batch one of the most popular examples of colloidal nanostructured
process presents several disadvantages, i.e. inhomogeneous materials produced via sol–gel methods. The synthesis of
mixing, temperature gradients and reproducibility issues,3 monodisperse silica spheres was firstly reported in 1956 by
which hinder the scaling-up of the synthetic procedure. In Kolbe.2d Stober2e further optimized this system, developing the
¨
order to overcome these problems, a continuous process is synthesis of spherical silica particles with diameters up to the
often desirable.
micrometer range (1.5 mm) under controlled reaction conditions
Tubular reactors (TRs) and segmented flow tubular reactors via the ammonia-catalyzed condensation of silicon alkoxides in
(SFTRs) have been frequently employed to produce, in a conti- several water/alcohol mixtures. Giesche8a,10 carried out the syn-
nuous manner, nano/microcrystals and colloids.4 The basic thesis of particles with sizes up to 3.0 mm in a continuous
principle of a TR set-up can be seen as the continuous mixing manner, via a two-step-growth process. Applying a similar
of (mostly liquids) reagents, which are afterwards pumped method, Kaiser2c reported the batch synthesis of fully meso-
through a tube until the reaction is completed. In the case of porous silica spheres. Bu¨chel and Unger2a described the syn-
SFTRs, an additional immiscible phase (typically a gas) is thesis of solid core mesoporous shell silica spheres through a
injected to form a regular gas–liquid slug pattern, thus ensuring two-step batch process. These core–shell spheres have found, in
a better homogeneity within each liquid droplet.3a,5 The tube the last decades, a large variety of applications and they are
diameter commonly employed in TR set-ups is larger (41 mm) currently investigated in several areas including catalysis11 and
than the channels present in microtubular reactors or microfluidic drug delivery.12 Moreover, they are frequently used as hard
template materials to prepare hollow carbon spheres13 through
¨
¨
Max-Planck-Institut fur Kohlenforschung, 45470 Mulheim an der Ruhr, Germany.
nanocasting methods.
E-mail: schueth@mpi-muelheim.mpg.de
Our group has recently used SiO2@mSiO2 silica spheres
(prepared via a classical batch process) as exotemplates for
the synthesis of hollow graphitic spheres, which were sub-
sequently employed as catalyst support for the oxygen reduction
† Electronic supplementary information (ESI) available: Summarizing specifica-
tion of the mesoporous silica; calculation of the theoretical maximal tube
diameter; investigations on the residence time and variation of ammonium
hydroxide concentration in a TR. See DOI: 10.1039/c5nj03033a
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem., 2016, 40, 4361--4366 | 4361

View Article Online
Paper
NJC
reaction in PEM fuel cells.2b,14 In order to increase the production spheres with a diameter of 320–340 nm, solution 1 contained
volumes of silica SiO2@mSiO2 spheres, and, more in general, as a 9.70 mL of ethanol, 4.47 mL of water and 0.83 mL of NH3(aq).
proof-of-concept, we developed a method for the conversion of the The required silica source is provided via solution 2, which
previously used batch synthesis to a continuous synthesis. In the contained 0.79 mL of TEOS, 0.31 mL of OTMS and 13.90 mL
following we demonstrate, through the synthesis of a series of of ethanol. Au@SiO2 particles were synthesized according to the
representative silica-based structures, that several different batch description of Arnal et al.18a,c Solution 1 contained 24 mL of the
synthesis protocols could be transferred – by means of a simple as prepared aqueous gold colloids (14 nm; 0.01 mol LÀ1), which
SFTR – to continuous synthetic processes.
was further diluted with 51.0 mL of ethanol and 3.36 mL of
NH3(aq). Solution 2 contained 2.9 mL of TEOS and 75.5 mL of
ethanol. All particles were dried overnight at 75 1C; the porogen
containing materials were calcined for 6 h at 550 1C with a
Experimental section
Synthesis set-up
heating rate of 2 K minÀ1
.
The corresponding batch syntheses are performed following
the procedures reported in the literature (ref. Bu¨chel, Unger et al.2a
for SiO2 and SiO2@mSiO2, ref. Kaiser and Unger2c for mSiO2).
Two liquid solutions were combined in the first T-junction
(+ID 0.33 mm) with constant flow rates achieved via a
precision-pump and a peristaltic pump (P-500, Pharmacia GE
Health care; Minipuls 3, Gilson). The formation of a segmented
flow was achieved by the injection (with a constant flow) of
Transmission electron spectroscopy
nitrogen gas (mass flow controller: EL-FLOWs F-201D-FAC-33-V, The TEM images presented in this work were obtained via a
Bronkhorst), via an additional T-junction (+ID 3 mm). The gas Hitachi H-7100 instrument (100 kV). All samples were placed
inlet pressure was set to 1.5 bars. For the standard synthesis, the on a Lacy carbon film supported by a copper grid. The dried
flow rates were set to 60 mL hÀ1 for the pumps and 120 mL hÀ1 solid samples were deposited on the Lacey grid without further
for the gas. The aging was performed in the tubular reactor treatment.
consisting of a PTFE tube (+ 3 mm  4 mm; 60 m in length).
Nitrogen physisorption
Syntheses of silica spheres, SiO2@mSiO2, Au@SiO2 and mSiO2
The nitrogen sorption measurements were carried out on a
¨
The synthesis via the modified Stober process was performed
Micrometrics ASAP 2010 instrument. Prior to analysis the
samples were activated under vacuum for at least 6 h at
200 1C. The measurements were performed at 77.35 K using a
static-volumetric method. The empty volume was determined
with He. The BET surface area was calculated from the adsorp-
tion data in the relative pressure interval from 0.04 to 0.2; the
pore size distribution was evaluated implementing the classical
Barrett–Joyner–Halenda (BJH) method, applied to the adsorption
branch of the isotherm. The total pore volume was estimated from
the amount of nitrogen adsorbed at a relative pressure of 0.97.
by employing the pumps with two different solutions, as
illustrated in Fig. 1. Solution 1 contained ultrapure (Milli-Q
water; 18.2 MO cm at 25 1C) water and the ammonium hydroxide
solution (NH3(aq); Sigma-Aldrich Co.; 500 mL; 28.0–30.0%),
whereas solution 2 contained the silica precursor. Typically,
¨
for the synthesis of Stober silica spheres, solution 1 consisted
of 12.48 mL of ethanol (99.7%; denatured with 2.6% toluene),
5.68 mL of water and 0.55 mL of NH3(aq). Solution 2 was
provided with 17.52 mL of denatured ethanol and 1.19 mL of
TEOS (Sigma-Aldrich Co.; 250 mL; 499.0%). For the shell
¨
deposition, 30 mL of the previously prepared Stober solution
plus 2.53 mL of ethanol, 7.84 mL of water and 0.76 mL of
NH3(aq) were supplied as solution 1. A mixture of 1.62 mL
of TEOS, 0.64 mL of OTMS (Sigma-Aldrich; 90%) and 38.87 mL
of ethanol was employed as solution 2. To prepare mSiO2
Results and discussion
Synthesis set-up description
The SFTR setup (Fig. 1) employed in this work consists of three
sections. (1) Initially, two different liquid solutions were pumped
through a static mixer (T-junction 1) to mix the reactants with
solvents and catalysts.
¨
For the synthesis of Stober spheres, solution 1 contained
water, part of the total volume of ethanol and ammonia
(catalyst), whereas solution 2 contained the silica precursor:
tetraethylorthosilicate (TEOS) and/or octadecyltrimethoxysilane
(OTMS), typically diluted in ethanol. For the synthesis of other
silica structures, slight modifications to this configuration were
applied. The composition of the two solutions for each experiment
is reported in detail in the experimental part. The aging tube was
cleaned by pumping a solution of 1 M NaOH through the tube for
B1 h, followed by rinsing the tube with ethanol.
Fig. 1 Scheme of the SFTR set-up employed for all the syntheses
described in this work. The gas–liquid segmentation present in the tube,
as well as the SiO2@mSiO2 progressive shell growth, is schematically
highlighted in the magnified red rectangle.
The flow rate of the two pumps was always kept equal in order
to maximize the mixing efficiency in the first Tee (T-junction 1).
4362 | New J. Chem., 2016, 40, 4361--4366
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

View Article Online
NJC
Paper
(2) A second T-junction (T-junction 2) is placed along the tube
and N2 is injected to form the segmented gas–liquid slug
pattern. (3) The liquid–gas slugs are continuously pumped
through a long polytetrafluoroethylene tube for a fixed residence
time, till the resulting colloidal suspension is collected at the
outlet. The segmentation created in the second T-junction is
known to give a homogeneous velocity distribution profile
within each slug, significantly different from the parabolic
profile usually observed in a laminar flow.5e Additionally, the
segmentation provides an important internal mixing. The
presence of two vortices assures, besides the diffusion driven
mixing, a continuous homogenization of the reaction mixture
inside each liquid segment, as described in detail by Taylor.5a
Although a turbulent flow could potentially offer a comparable
internal mixing, in our case, such a flow regime was not
achievable because of flow rate restrictions of the employed
pumps. A dimensionless number introduced by Bretherton5b is
commonly used to identify the characteristics of a gas–liquid
segmented flow in a capillary and to estimate the threshold – in
this case the maximum tube diameter – above which the
formation of segmented flow would not be attainable anymore.
If a tube with a diameter larger than the threshold value is
used, the buoyancy forces would overcome the surface tension
between the liquid droplet and the gas bubble, resulting in a
flow where the gas bubbles would just bypass the liquid slugs.
In our case (for the calculation details see the ESI†) such a
critical situation is reached at diameters higher than 3.1 mm.
In order to confirm the calculated threshold value, a PTFE tube
with an inner diameter of 4 mm was employed as the aging
tube. As predicted by the estimate, it was not possible to obtain
a segmented flow with the employed reaction mixture of
ethanol, water and ammonium hydroxide solution. Therefore,
an inner tube diameter of 3 mm was applied in our system,
which worked well in sustaining a segmented flow.
Continuous synthesis of mesoporous silica material
Fig. 2 (A) TEM micrographs displaying the different synthesized core–
shell particles and the respective shell thickness Æ standard deviation
under laminar flow (green; c and d), segmented flow conditions (red; a
and b) as well as in the batch process (blue; e and f). (B) Shell thickness
distribution histograms for core–shell spheres prepared under laminar
¨
The continuous synthesis of silica spheres via the Stober
method in a TR has already been described by Giesche,8a and
more recently by Gutierrez et al. and Khan et al., who performed
the same synthesis in a microfluidic reactor.3a,8b In our case,
¨
the Stober synthesis is the first step in the synthesis of (green), segmented (red) conditions and in a batch process (blue).
SiO2@mSiO2 spheres and therefore was the first synthesis we
optimized for our setup. In agreement with what had previously
been reported by Gutierrez,8b we found the segmented-flow residence time for both experiments (considering that the gas
regime crucial to achieve the desirable monodispersity. In inlet was sealed for the laminar-flow synthesis), the pump flow
addition, we could control the size of the spheres by adjusting rates were adjusted accordingly.
the ammonia content in the reaction mixture, as typically
The difference in monodispersity between the particles prepared
done in the batch synthesis. Moreover, we optimized the total applying these two different flow regimes can be appreciated
residence time to assure the completion of the reaction. analysing the TEM images (Fig. 2a), where a significant shell
All these results are summarized in the ESI.†
thickness heterogeneity among the SiO2@mSiO2-laminar particles
For the continuous synthesis of SiO2@mSiO2 core–shell is observed. The average shell thicknesses (and corresponding
¨
spheres, 220 nm large Stober spheres were used as core. Also standard deviations) of different SiO2@mSiO2 spheres, evaluated
for the shell growth, segmentation of the flow is necessary to through the analysis of the TEM images, are reported in Fig. 2,
achieve highly monodisperse silica core–shell particles. It is together with the shell thickness typically obtained through the
important to note that both types of syntheses were performed equivalent batch synthesis (SiO2@mSiO2-batch). Interestingly,
under similar conditions; in order to maintain the same the standard deviations relative to SiO2@mSiO2-batch and
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem., 2016, 40, 4361--4366 | 4363

View Article Online
Paper
NJC
SiO2@mSiO2-segmented particles are in good agreement as can
be seen from the TEM micrographs as well as in the histogram
(Fig. 2). Thus, the advantages of the tubular reactor, such as
higher productivity and continuous operation can be capitalized
on without compromising the product quality. In contrast, the
standard deviation of SiO2@mSiO2-laminar spheres is twofold
larger (as visible in the sparse distribution histogram in Fig. 2b).
Therefore it is clear that internal mixing provided by segmenta-
tion is crucial to achieve the desired core–shell structure. Based
on these results, the segmented flow pattern was applied for all
the following continuous experiments.
As a next step, the possibility of controlling the final shell
thickness was investigated. Fig. 3 shows SiO2@mSiO2 spheres
obtained by varying the total amount of reactants (TEOS and
OTMS) fed into the reactor. Thanks to an increase of the
final volume fraction of TEOS + OTMS from 1.06% to 2.5%
(see Table S1, ESI†) it was possible to increase the shell
thickness from 18 Æ 3 nm to 60 Æ 7 nm. To further prove the
flexibility of this approach, we extended the synthesis of core–
shell spheres to the preparation of completely mesoporous
silica spheres through modification of the method (batch
synthesis) firstly described by Kaiser.2c By directly mixing TEOS
and OTMS (without the presence of non-porous silica core),
fully mesoporous silica particles (mSiO2) with an average
diameter of 341 Æ 53 nm were obtained.
Once separated and dried, all the silica core–shell particles
were calcined (550 1C, 6 h) to combust the porogen (octadecyl
alky chain of OTMS) and thus form porosity within the silica
shell. The textural properties of different SiO2@mSiO2 spheres
were evaluated by means of N2 physisorption (Fig. 3).
The increase in adsorbed volume is directly proportional to
the thickness of the mesoporous shell, or, in other words, to the
ratio of the shell volume over the volume of the non-porous
core. The fully mesoporous spheres (mSiO2) thus represent
the limiting case (maximum mesoporosity achievable). The
measured BET surface areas ranged from 125 m2 gÀ1 for the
18 nm shell, 282 m2 gÀ1 for the particles with 60 nm thick
shells, and up to 626 m2 gÀ1 for the fully mesoporous spheres
(Table S1, ESI†). All the samples, except the SiO2@mSiO2
spheres with a 18 nm thin shell, have pore sizes of 3–4 nm,
as expected for a mesoporous shell formed at a TEOS/OTMS
ratio equal to 2.5 (vol/vol).15 The deviation from this value for
the SiO2@mSiO2-18 sample might be caused by the very thin
shell, which might hinder the complete formation of a structurally
stable mesoporous framework. The pore size distribution curves
were calculated by applying the BJH (Barrett–Joyner–Halenda)
method to the adsorption isotherm branch. The desorption
branch, which is recommended by the IUPAC for the pore-size
Fig. 3 Textural and physical properties of different silica spheres. (A)
Adsorption isotherms and corresponding BJH pore size distribution curves
(B) for SiO2@mSiO2-18 (blue), SiO2@mSiO2-60 (red) and the mSiO2
(green). (C) TEM images of SiO2@mSiO2, mSiO2 and Au@SiO2.
analysis,17 might be influenced by the presence of measurement (inner diameter 3 mm and 120 min residence time), a productivity
artefacts, caused by the ink-bottle shape of the mesopores or of 0.95 g hÀ1 was achieved (based on sample mass after calcina-
capillary instability, and could therefore not be used.
tion). An increase in the production rate could only be achieved
by elongating the tube and thus increasing the flow rate, or by
shortening the residence time to a minimum of around 90 min
(Fig. S1, ESI†), which would yield 1.25 g hÀ1. An increase of the
Alternative continuous synthesis approaches to achieve higher
yields
At a typical concentration for the SiO2@mSiO2-segmented inner diameter of values higher than 3 mm is not feasible, since
silica particles of 0.009 g mLÀ1 and under standard conditions the segmentation of the flow would break down (see ESI†).
4364 | New J. Chem., 2016, 40, 4361--4366
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

View Article Online
NJC
Paper
A segmented flow pattern for larger diameters could possibly and Enustun.20 These nanoparticles have a diameter of 14 nm,
be maintained in a droplet flow reactor, in which the reaction as can be seen from the TEM images, Fig. 3C, right panels. The
mixture is sequentially distributed as small droplets in an gold colloidal suspension, after being diluted with an ethanol/
immiscible liquid carrier phase, instead of achieving segmen- water mixture, was continuously mixed with the solution con-
tation with a gaseous phase. However, this would require a fully taining the silica precursor, and pumped through the aging
new optimization protocol, if the synthesis is possible at all. tube to finally grow the silica shell around the Au colloids. The
TEOS is soluble in most oil phases, and thus problems by overall good quality of Au@SiO2 core–shell particles is visible in
partial or full solubilisation of the TEOS in oil segments are to Fig. 3 and is in complete agreement with what was obtained
be expected. Nevertheless, a droplet flow reactor could have under batch conditions: every silica sphere contains a single Au
some advantages, as higher possible tube diameters and thus colloidal particle, and a surrounding silica shell of 40 nm Æ 5.6,
higher productivity and the possibility of separating the reac- supporting again the notion that continuous processing is
tion phase fully from the tube walls, which could reduce the possible without compromising the product quality. Although
risk of fouling and could lead to enhanced operation times.16 not reported here, the shell thickness can potentially be adjusted
However, relatively long stable operation can also be achieved by varying the TEOS or ammonia concentrations, as typically
by the gas segmented flow. Long-term experiments over 8 h done for batch processes. In addition, it would be possible to
were conducted as part of this study for the synthesis of SiO2 and grow an additional mesoporous shell on Au@SiO2 particles,19b
SiO2@mSiO2. In both cases continuous operation was possible applying the same procedure described above for SiO2@mSiO2.
without problems, while the material quality was maintained.
However, silica residues became visible during the cleaning
Conclusions
process, which might cause trouble on a longer timescale.
Zhang et al.4g discussed the upscaling of the continuous
In this work we have described the continuous synthesis of
synthesis of nanocrystals in droplet flow reactors with the
several representative types of silica-based colloidal particles in
immiscible liquid phase and the liquid reactant phase with
an SFTR. In particular, we showed that the continuous synthesis
of silica spheres (Stober process), SiO2@mSiO2, mSiO2 and
an inner diameter of up to 3 mm. Restrictions in terms of quality
¨
were observed for reactors with inner diameters of 3 mm, which
could be overcome by pinching the tube to improve the mixing
characteristics. The additional mixing contributions due to
shearing and reorientation of the droplets led to the formation
of particles with qualities as obtained with smaller diameter
reactors. Nevertheless, the narrowing of the tube, below 3 mm,
was necessary for increasing the diameters indicating restric-
tions in a similar diameter range as for the setup herein. In the
case of segmentation with a gaseous phase, operation of the tube
diameter of up to 3 mm is possible, thanks to the additional
mixing contribution induced by friction between the tube wall
and the reaction phase.5a
Au@SiO2, is possible by adapting the methods commonly used
in batch processes. Our results prove that several features of the
silica spheres, like particle size, shell thickness and meso-
porosity, are completely controllable by adjusting the synthetic
parameters. We believe that this concept could potentially be
extended to other materials (non silica-based sol–gel syntheses)
as well as to other cases (by replacing the Au colloidal particles
with other metals or metal oxides, for instance). Additionally,
two (or multiple) synthesis steps, as the synthesis of SiO2 and
SiO2@mSiO2, could easily be combined in a single set-up con-
sisting of two sequential stages. These considerations clearly
point out the versatility of the applied system towards the
continuous synthesis of advanced materials.
Higher production yields might be achieved following the
tubular reactor setup of Giesche8a for the synthesis of multi-
step growth of silica beads. Here the reagents are initially mixed
via a vibrating magnetic stirring bar inside a Y-tube. The Acknowledgements
following aging is performed in a ‘‘double helix’’-like twisted
We are grateful for the basic funding provided by Max-Planck-
Institut fu¨r Kohlenforschung.
tube, which was required to prevent accumulation of the silica
particles. However, whether transfer of this concept to the
synthesis of SiO2@mSiO2 particles is possible is still an open
question.
Notes and references
Continuous synthesis of Au@SiO2
1 (a) F. Fajula, A. Galarneau and F. D. Renzo, Microporous
¨
´
Mesoporous Mater., 2005, 82, 227–239; (b) R. Martın-Aranda
A modified Stober process is typically also employed in the
ˇ
synthesis of M@SiO2 core–shell particles, where M can be a
metal (Au, Ag and Pt, among others) or a metal oxide (Fe2O3).18
The resulting spheres can be further processed to obtain, for
instance, yolk–shell particles.14,19 To extend our proof of con-
cept, we used our SFTR to cover gold colloidal nanoparticles
with a dense (non-porous) silica shell. Initially, PVP-capped
(polyvinylpyrrolidone) gold nanoparticles were synthesized in a
batch process, following the synthesis described by Turkevich
and J. Cejka, Top. Catal., 2010, 53, 141–153; (c) C. Perego
and R. Millini, Chem. Soc. Rev., 2013, 42, 3956–3976.
2 (a) G. Bu¨chel, K. K. Unger, A. Matsumoto and K. Tsutsumi,
Adv. Mater., 1998, 10, 1036–1038; (b) C. Galeano, J. C. Meier,
V. Peinecke, H. Bongard, I. Katsounaros, A. A. Topalov, A. Lu,
K. J. Mayrhofer and F. Schuth, J. Am. Chem. Soc., 2012, 134,
20457–20465; (c) C. Kaiser and K. K. Unger, DE 19530031, 1997;
¨
(d) G. Kolbe, PhD thesis, Friedrich-Schiller-Universitat,
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem., 2016, 40, 4361--4366 | 4365

View Article Online
Paper
NJC
Jena, 1956; (e) W. Stober, A. Fink and E. Bohn, J. Colloid 11 G. H. Jeong, E. G. Kim, S. B. Kim, E. D. Park and S. W. Kim,
Interface Sci., 1968, 26, 62–69. Microporous Mesoporous Mater., 2011, 144, 134–139.
3 (a) S. A. Khan, A. Gu¨nther, M. A. Schmidt and K. F. Jensen, 12 (a) Y. J. Wang, V. Bansal, A. N. Zelikin and F. Caruso,
Langmuir, 2004, 20, 8604–8611; (b) Y. Song, J. Hormes and
C. S. Kumar, Small, 2008, 4, 698–711.
4 (a) S. Guillemet-Fritsch, M. Aoun-Habbache, J. Sarrias,
Nano Lett., 2008, 8, 1741–1745; (b) Y. Chen, H. Chen,
M. Ma, F. Chen, L. Guo, L. Zhang and J. Shi, J. Mater. Chem.,
2011, 21, 5290–5298; (c) A. M. El-Toni, M. A. Ibrahim,
J. P. Labis, A. Khan and M. Alhoshan, Int. J. Mol. Sci.,
2013, 14, 11496–11509; (d) X. C. Chen, Colloids Surf., A,
2013, 428, 79–85.
ˆ
A. Rousset, N. Jongen, M. Donnet, P. Bowen and J. Lemaıtre,
Solid State Ionics, 2004, 171, 135–140; (b) P. Bowen, M. Donnet,
A. Testino, M. Viviani, M. Buscaglia, V. Buscaglia and P. Nanni,
Key Eng. Mater., 2001, 206, 21–24; (c) M. Kubo and 13 (a) P. Valle-Vigon, M. Sevilla and A. B. Fuertes, Chem. Mater.,
T. Yonemoto, Chem. Eng. Res. Des., 1999, 77, 335–341;
2010, 22, 2526–2533; (b) S. B. Yoon, K. Sohn, J. Y. Kim,
C. H. Shin, J. S. Yu and T. Hyeon, Adv. Mater., 2002, 14,
19–21.
ˆ
(d) A. Aimable, N. Jongen, A. Testino, M. Donnet, J. Lemaıtre,
H. Hofmann and P. Bowen, Chem. Eng. Technol., 2011, 34,
344–352; (e) R. Vacassy, J. Lemaitre, H. Hofmann and 14 C. Galeano, C. Baldizzone, H. Bongard, B. Spliethoff,
J. Gerlings, AiChE J., 2000, 46, 1241–1252; ( f ) P. Moschou,
C. Weidenthaler, J. C. Meier, K. J. J. Mayrhofer and
H. J. M. de Croon Mart, J. van der Schaaf and C. Schouten Jaap,
F. Schu¨th, Adv. Funct. Mater., 2014, 24, 220–232.
Rev. Chem. Eng., 2014, 30, 127; (g) L. Zhang and Y. Xia, 15 W. R. Zhao, J. L. Shi, H. R. Chen and L. X. Zhang, J. Mater.
Adv. Mater., 2014, 26, 2600–2606. Res., 2006, 21, 3080–3089.
5 (a) G. I. Taylor, J. Fluid Mech., 1961, 10, 161–165; 16 A. M. Nightingale, S. H. Krishnadasan, D. Berhanu, X. Niu,
F. P. Bretherton, J. Fluid Mech., 1961, 10, 166–188;
(b) M. N. Kashid, F. Platte, D. W. Agar and S. Turek,
C. Drury, R. McIntyre, E. Valsami-Jones and J. C. deMello,
Lab Chip, 2011, 11, 1221–1227.
J. Comput. Appl. Math., 2007, 203, 487–497; (c) M. T. Kreutzer, 17 (a) K. S. W. Sing, R. A. Everett, R. A. W. Haul, L. Mouscou,
F. Kapteijn, J. A. Moulijn and J. J. Heiszwolf, Chem. Eng. Sci.,
2005, 60, 5895–5916; (d) M. Kubo, T. Kawakatsu and
T. Yonemoto, Chem. Eng. Res. Des., 1998, 76, 669–676.
6 (a) S. J. Trietsch, T. Hankemeier and H. J. van der Linden,
Chemom. Intell. Lab. Syst., 2011, 108, 64–75; (b) R. L. Hartman,
R. A. Pierotti, T. Rouquerol and T. Siemienniewska, Pure
Appl. Chem., 1985, 57, 603–619; (b) M. Thommes, B. Smarsly,
M. Groenewolt, P. I. Ravikovitch and A. V. Neimark,
Langmuir, 2005, 22, 756–764; (c) M. Thommes, Chem. Ing.
Tech., 2010, 82, 1059–1073.
J. P. McMullen and K. F. Jensen, Angew. Chem., Int. Ed., 2011, 18 (a) S. H. Joo, J. Y. Park, C.-K. Tsung, Y. Yamada, P. Yang and
50, 7502–7519; (c) A. Rios, M. Zougagh and M. Avila, Anal.
Chim. Acta, 2012, 740, 1–11; (d) A. M. Nightingale and
J. C. Demello, Adv. Mater., 2013, 25, 1813–1821.
G. A. Somorjai, Nat. Mater., 2009, 8, 126–131; (b) Y. Lu,
Y. Yin, B. T. Mayers and Y. Xia, Nano Lett., 2002, 2, 183–186;
´
(c) L. M. Liz-Marzan, M. Giersig and P. Mulvaney, Langmuir,
7 (a) T. Ogihara, M. Ilzuka, T. Yanagawa, N. Ogata and K. Yoshida,
1996, 12, 4329–4335.
J. Mater. Sci., 1992, 27, 55–62; (b) T. Ogihara, M. Ikeda, M. Kato 19 (a) P. M. Arnal, M. Comotti and F. Schuth, Angew. Chem.,
and N. Mizutani, J. Am. Ceram. Soc., 1989, 72, 1598–1601.
8 (a) H. Giesche, J. Eur. Ceram. Soc., 1994, 14, 205–214;
(b) L. Gutierrez, L. Gomez, S. Irusta, M. Arruebo and
J. Santamaria, Chem. Eng. J., 2011, 171, 674–683.
Int. Ed., 2006, 45, 8224–8227; (b) C. Galeano, R. Guttel,
M. Paul, P. Arnal, A. H. Lu and F. Schuth, Chemistry, 2011,
17, 8434–8439; (c) P. M. Arnal, M. Comotti and F. Schu¨th,
Angew. Chem., 2006, 118, 8404–8407.
9 L. Gomez, M. Arruebo, V. Sebastian, L. Gutierrez and 20 (a) B. V. Enustun and J. Turkevich, J. Am. Chem. Soc., 1963,
J. Santamaria, J. Mater. Chem., 2012, 22, 21420.
10 H. Giesche, J. Eur. Ceram. Soc., 1994, 14, 189–204.
85, 3317–3328; (b) J. Turkevich, P. C. Stevenson and
J. Hillier, Discuss. Faraday Soc., 1951, 11, 55–75.
4366 | New J. Chem., 2016, 40, 4361--4366
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016