A new ordered mesoporous silica material (COK-19) with cubic symmetry is synthesized by silicate polycondensation in a citric acid/citrate buffered micellar solution of Pluronic F127 triblock copolymer near neutral pH. SAXS, nitrogen adsorption, TEM, and electron tomography reveal the final material has a cubic close packed symmetry (Fm3̅m) with isolated spherical mesopores interconnected through micropores. Heating of the synthesis medium from room temperature to 70 °C results in a mesopore size increase from 7.0 to 11.2 nm. Stepwise addition of the silicate source allows isolation of a sequence of intermediates that upon characterization with small-angle X-ray scattering uncovers the formation process via formation and aggregation of individual silica-covered Pluronic micelles.
A new ordered mesoporoussilica material (COK-19) with cubic symmetry is synthesized by silicate polycondensation in a citric acid/citrate buffered micellar solution of Pluronic F127 triblock copolymer near neutral pH. SAXS, nitrogen adsorption, TEM, and electron tomography reveal the final material has a cubic close packed symmetry (Fm3̅m) with isolated spherical mesopores interconnected through micropores. Heating of the synthesis medium from room temperature to 70 °C results in a mesopore size increase from 7.0 to 11.2 nm. Stepwise addition of the silicate source allows isolation of a sequence of intermediates that upon characterization with small-angle X-ray scattering uncovers the formation process via formation and aggregation of individual silica-covered Pluronic micelles.
Ordered mesoporoussilica (OMS) materials have arrays of identical
mesopores separated by amorphous silica walls. The uniformity of the
mesopores and the walls is particularly attractive for a variety of
applications including catalysis, pharmaceutical formulation, and
microelectronics.[1−9] OMS materials are created by a supramolecular assembly process of
silicate molecules and surfactant. The cooperative self-assembly of
individual surfactant and silicate molecules is one of the proposed
formation mechanisms.[10−13] Silica precipitation on prearranged surfactant micelles, fulfilling
a role of liquid crystal templates, is an alternative model.[13−16] After evacuation of the surfactant molecules from the assembled
material, mesopores separated by silica walls remain. Initially, OMS
materials were synthesized from strongly alkaline or strongly acidic
synthesis mixtures. Later, synthesis protocols were developed enabling
synthesis at less extreme pH levels.[17−21] Particularly, pH-buffered solutions provide attractive
options for mastering the formation process.COK-12 is an example
of an ordered mesoporoussilica material prepared
under slightly acidic, buffered conditions.[19,20] The synthesis of COK-12 starts from a citric acid/citrate buffered
solution of amphiphilic triblock copolymer P123 (EO20PO70EO20) which self-assembles into spherical micelles.
Hydrophobic poly(propylene oxide) (PPO) chains form the hydrophobic
core of the micelles, surrounded by a more hydrophilic layer of poly(ethylene
oxide) (PEO) chains.[22−24] Silica preferentially infiltrates the outer PEO zone
of the micelles, diminishing the mobility of the PEO chains and the
sterical stabilization of the micelle dispersion. The infiltration
of silica in the PEO layers triggers coalescence of the original micelles
to form cylinders, which subsequently are arranged in a hexagonal
packing. In this way, Pluronic P123 favors formation of hexagonally
ordered platelet shaped OMS materials.[19] Its hexagonal mesostructure assembles spontaneously upon adding
basic sodium silicate solution to a citric acid/citrate buffered Pluronic
P123 triblock copolymer solution.[20] This
formation mechanism was investigated by identifying the intermediate
structures, which were prepared by limiting sodium silicate addition
to the buffered P123 micelle suspension (Figure ).[20]
Figure 1
Schematic representation
of the different stages in the formation
of COK-12, a hexagonally ordered mesoporous silica. Reprinted with
permission from ref (20). Copyright 2011 American Chemical Society.
Schematic representation
of the different stages in the formation
of COK-12, a hexagonally ordered mesoporoussilica. Reprinted with
permission from ref (20). Copyright 2011 American Chemical Society.F127 (EO102PO70EO102) is
a related
triblock copolymer with substantially longer PEO blocks compared to
P123 (EP20PO70EO20). Under dilute
aqueous conditions, F127 polymer molecules self-assemble to spherical
micelles with a hydrophobic core similar in size to that of spherical
P123 micelles but with a thicker hydrophilic palisade layer due to
the substantially longer PEO blocks.[12] F127
self-assembles into micelles with a hydrodynamic radius of approximately
8.0 to 12 nm and a core radius of 5.9 nm.[12,22,24] Previously, F127 was used for the synthesis
of ordered mesoporoussilica materials adopting body centered Im3m symmetry, such as SBA-16[17,25] and face centered Fm3̅m symmetry
in the instance of FDU-12[13,26−28] and KIT-5.[18,29] KIT-5 and FDU-12 are both synthesized
using F127 as a template under different synthesis conditions and
the use of different additives.[18,28,30] However, the SAXS diffraction patterns for both Fm3̅m materials differ substantially due to
differences in pore configuration.[30]Intrigued by the documented influence of the length of the PEO
chains in the Pluronic template on the symmetry of the final silica
material (cubic with F127, versus hexagonal with P123), we performed
synthesis experiments with F127 in citric acid/citrate buffer and
staged silicate addition to investigate the intermediates of this
OMS formation process. Synthesis conditions leading to formation of
a highly ordered new cubic phase designated as COK-19 were identified.
Experimental Section
Synthesis
A typical
buffered F127 solution was prepared
by dissolving 2.6 g of Pluronic F127 (BASF) in 107.5 g of deionized
water. To this solution, 3.684 g of citric acid monohydrate (Sigma-Aldrich)
and 2.540 g of trisodium citrate dihydrate (Sigma-Aldrich) were added.
This buffered surfactant solution was stirred overnight to dissolve
all components.For the OMS material synthesis, 10.4 g of sodium
silicate solution (27 wt % SiO2, extra pure, Merck) diluted
in 30 g of water was added to the buffered F127 solution at room temperature,
while stirring vigorously. Precipitation of a white solid occurred
instantaneously, and after 5 min of stirring, the suspension was kept
under quiescent conditions to age for 24 h. Aging was done either
at room temperature, 50 °C, or 70 °C. As-synthesized materials
were filtered, washed, and dried at 50 °C. Finally, the powders
were calcined in air at 350 °C for 24 h using a heating rate
of 0.5 °C min–1.For the investigation
of the assembly process with SAXS, flasks
were filled with 3 g aliquots of buffered F127 solution. To these
solutions, 0.15, 0.30, 0.40, 0.50, 0.60, 0.65, 0.7, 0.8, 0.90, and
1.00 mL of the diluted sodium silicate solution were added while stirring
vigorously. These quantities correspond to 15, 30, 40, 50, 60, 65,
70, 80, 90, and 100% of the total sodium silicate amount of the typical
COK-19 synthesis. SAXS patterns were recorded 5 min after mixing.
pH increased from 3.6 in the buffered Pluronic solution to 3.8, 4.0,
4.1, 4.3, 4.4, 4.5, 4.5, 4.7, 4.8, and 5.0, respectively.
Characterization
Dynamic Light Scattering (DLS) analyses
of suspensions were conducted in polystyrene cuvettes at 25 °C
on a 90Plus instrument (Brookhaven Instruments Corporation), at a
scattering angle of 90° using a 659 nm laser. The fluctuations
in the scattered laser light intensity were correlated between 5 μs
and 1 s. Correlation functions were analyzed with Igor Pro 6.2, using
the Clementine package for modeling decay kinetics based on the Maximum
Entropy method. The decay times were converted to hydrodynamic radii
with the Stokes–Einstein relation, yielding intensity weighted
size distributions. Viscosity values used in the Stokes–Einstein
relation were adjusted by measuring viscosities using a rolling ball
viscometer (Anton Paar).Small-Angle X-ray Scattering (SAXS)
of powder samples was recorded at room temperature using a SAXSess
mc2 instrument (Anton Paar) with line-collimated CuKα radiation and a 2D imaging plate detector in combination
with a temperature controlled RotorCell sample stage. SAXS patterns
were normalized to scattering intensity at higher angles. Background
subtraction and desmearing were performed with the SAXSquant software.To investigate the formation mechanism, the scattering profiles
for synthesis mixtures with varying silicate concentrations were recorded
using a 1 mm quartz capillary flow cell sample stage mounted in the
SAXSess mc2 instrument. The temperature was kept at 20
°C using a Peltier device (Anton Paar, TCS 120). SAXS patterns
for the different mixtures were recorded over a time frame of 1 h.
Ultrapure mQ water was used as a background. Modeling of the desmeared
scattering curves was performed using Igor Pro 6.2 in combination
with the NCNR SANS & USANS Data Reduction and Analysis package
(v7.20).[31] The recorded SAXS patterns were
fitted with a combined model, based on spherical core–shell
particles and small dense particles with a Gaussian size distribution.[32] Synthesis mixtures containing larger amounts
of sodium silicate were fitted with the same model, with pseudo-Voigt
peaks superimposed. The pseudo-Voigt functions were introduced to
best describe the peak shapes and are a combination of Lorentzian
and Gaussian peak shapes.[33]Nitrogen
sorption isotherms were recorded on an Autosorb-1 apparatus
(Quantachrome) at −196.8 °C. Prior to measurement, calcined
samples were degassed at 150 °C for 13 h under reduced pressure.
BET surface area was estimated in the P/P0 region 0.05–0.3.
Micropore volumes were estimated using the t-plot method. The pore
size distribution was determined using NLDFT analysis. NLDFT models
pore sizes are based on the nitrogen adsorption on a silica material
with cylindrical pores smaller than 5 nm and spherical pores larger
than 5 nm.Scanning Electron Microscopy (SEM) and Scanning Transmission
Electron
Microscopy (STEM) images were recorded on a Nova NanoSEM450 (FEI).
For SEM, samples were dispersed on carbon tape and measured without
coating, using low voltages (0.5–1 kV) to reduce sample charging.
In STEM mode, the device operated at 30 kV. STEM samples were prepared
on 50 nm 300-mech carbon-coated copper grids.High-Resolution
Scanning Transmission Electron Microscopy (HRSTEM)
images of the COK-19 crystals were acquired using an FEI Tecnai G2
operated at an accelerating voltage of 200 kV. The images were collected
using a high angle annular dark field (HAADF) detector. The samples
were crushed and dispersed in ethanol before a droplet of the dispersion
was added to a carbon coated copper grid. For electron tomography,
bright field TEM images was collected on an FEI Tecnai G2, operated
at 200 kV. The tilt series of the sample aged at 70 °C covered
angles between −72° and 76° and the series from the
sample aged at room temperatures −76° to 76°. Both
series were collected with a tilt increment of 2°. The tomographic
reconstructions were performed using a SIRT algorithm, and the visualization
was done with AMIRA.29Si magic angle spinning (MAS)
NMR spectra were recorded
using an AMX300 spectrometer (Bruker) operating at 100 MHz. 944 scans
were accumulated with a recycle delay of 120s. Samples were packed
in 4 mm zirconia rotors. Tetramethylsiliane was used as a chemical
shift reference.
Results and Discussion
Characterization of OMS
Material
The addition of sodium
silicate to the buffered Pluronic F127 solution led to the instantaneous
formation of a white suspension. SAXS patterns of as-made materials
indicated highly ordered materials (see Supporting
Information, Figure S1). Drying and subsequent removal of the
organic template by calcination resulted in porous material, coined
COK-19. The calcined sample, prepared and aged at room temperature,
shows well resolved diffraction lines indexed in an Fm3̅m symmetry as (111), (200), (220), and (311)
with a unit cell parameter a = 21.9 nm (Figure ). However, SAXS
data alone are insufficient to determine meso-ordering in the structure.
Figure 2
Small
angle X-ray scattering patterns of COK-19 aged at room temperature,
50 or 70 °C and calcined at 350 °C. Diffraction peak positions
are indicated by | .
Small
angle X-ray scattering patterns of COK-19 aged at room temperature,
50 or 70 °C and calcined at 350 °C. Diffraction peak positions
are indicated by | .Pore structure and long-range ordering were further investigated
using transmission electron microscopy. High angle annular dark field
(HAADF) STEM images reveal the structure of an ordered mesoporous
material consistent with the Fm3̅m structure (Figure ). The images further show the presence of some stacking faults in
COK-19 samples. As layers of close packed spheres are forming a three-dimensional
structure, the layers can combine along the [111]c direction
with two different sequences.[34,35] ABCA... stacking leads
to cubic close packing (ccp), and ABA... stacking leads to hexagonal
close packing (hcp).[36] A HAADF STEM image
of a COK-19 particle aged at room temperature shows a large domain
of cubic ordering, with some stacking faults near the rim of the particle
(Figure A). An image
from another particle exhibited a larger amount of stacking faults,
suggesting some nonuniformity throughout the sample (see Supporting Information, Figure S2).
Figure 3
HAADF-STEM
images of COK-19 (a) and COK-19-70C crystal (b, c).
The COK-19-70C particle is rotated around the [111]c axis
(marked by a white arrow). The particle is viewed along [−110]c (b) and [−211]c (c). The red line indicates
the stacking (dis)order. Fourier transforms are shown as insets in
a and b where streaks are characteristic for materials with stacking
disorder.
HAADF-STEM
images of COK-19 (a) and COK-19-70C crystal (b, c).
The COK-19-70C particle is rotated around the [111]c axis
(marked by a white arrow). The particle is viewed along [−110]c (b) and [−211]c (c). The red line indicates
the stacking (dis)order. Fourier transforms are shown as insets in
a and b where streaks are characteristic for materials with stacking
disorder.Scanning electron microscopy revealed
the morphology of the calcined
COK-19 as aggregates of truncated octahedrons with a mean diameter
of approximately 500 nm to 3 μm (Figure ). 29Si MAS NMR analysis of the
as-synthesized COK-19 material was performed to analyze the degree
of condensation (see Supporting Information, Figure S8). The NMR spectra show the chemical environment around
the silicon atom as Si(OSi)(OH)4–. Q represents Q2, Q3, and Q4 signals at −92,
−101, and −110 ppm respectively, with an intensity distribution
of 9%, 42%, and 48%, indicating a condensed silicate network even
at room temperature.
Figure 4
SEM micrographs of calcined COK-19.
SEM micrographs of calcined COK-19.Freshly synthesized COK-19 suspensions were aged at 50 and
70 °C
instead of room temperature for 24 h. Low angle diffractograms of
calcined COK-19 materials aged at 50 and 70 °C resulted in less
ordered material and caused the unit cell to increase to 25.9 and
27.2 nm, respectively (Figure ). The structure loss was also apparent from HAADF images
for COK-19-70C, aged postsynthesis at 70 °C (Figure B). This material has more
stacking faults, which gives rise to streaks in the Fourier transform
of the image. TEM is a method to study the local structure of selected
particles. By combining the information from TEM with the results
from SAXS, which provide averaged information from the entire sample,
it can be confirmed that the sample aged at room temperature possesses
a higher quality cubic ordering, with stacking faults increasing with
temperature.Heating causes considerable swelling of Pluronic
micelles.[3,22,37] Aging above
room temperature
of the COK-19 aggregates formed at room temperature introduced stacking
faults in the material. The introduction of stacking disorders upon
heating suggests that, in the initial colloids, the interaction between
Pluronic core–silica shell particles is still largely physical,
enabling alteration of stacking.Besides line broadening through
general lattice disorder, stacking
faults in the [111]c direction also affect the SAXS pattern.
Hcp intergrowths in a ccp structure influence line width and positions
of certain reflections.[34,36] These stacking disorders,
also known as intergrowths, occur more frequently for COK-19-70C as
was seen with HAADF-STEM. COK-19-70C was aged at higher temperatures,
where energetic differences between hcp and ccp packings become less
deterministic. The increasing fraction of stacking disorder with temperature
also explains the decreasing resolution of the low angle reflections
in the SAXS curve with increasing aging temperature (Figure ).[34]COK-19 displayed a nitrogen physisorption isotherm with characteristics
of both micro- and mesopores (Figure A, Table ). The isotherm exhibits a hysteresis loop, characteristic for cage-type
mesopores. The cage diameter is 7.0 nm, and the mesopore volume is
0.09 cm3 g–1. The total pore volume of
0.29 cm3 g–1 is mostly microporous (0.2
cm3 g–1).
Figure 5
Adsorption (●)
and desorption (○) branches of nitrogen
physisorption isotherms of calcined (A) COK-19, (B) COK-19-50C, and
(C) COK-19-70C. The insets show the pore size distribution based on
NLDFT calculations.
Table 1
Synthesis,
Structural, and Porous
Properties of COK-19 Material Formed at Room Temperature, Aged at
Different Temperatures, and Other Fm3̅m Structured Mesoporous Materials
molar
composition and synthesis condition
structural
and porous properties
samples
Si
template
swelling agent
KCl
citrate
HCl
H2O
Ta (°C)
ab (nm)
dporec (nm)
Wtd (nm)
BET (m2g–1)
Vte (cm3g–1)
Vmif (cm3g–1)
materials prepared
with F127 as a template, without swelling agents
COK-19
1
0.0044
0
0
0.56
0
163
18/18
22
7
8.5
514
0.29
0.2i
COK-19–50C
1
0.0044
0
0
0.56
0
163
18/50
25.9
9.4
8.9
1187
0.78
0.46j
COK-19–70C
1
0.0044
0
0
0.56
0
163
18/70
27.2
11.2
8
1035
0.88
0.25j
KIT-5[18]
1
0.0035
0
0
0
0.88
119
45/45
17
6.8
5.2
390
0.21
KIT-5[18,29]
1
0.0035
0
0
0
0.88
119
45/100
21.5
9
6.2
750
0.45
materials prepared
with F127 as a template, KCl, and with swelling agents
LP-FDU-12[26]
1
0.0040
0.51 (TMB)
3.51
0
6.13
170
15/170
32.9
18
5.3
426
0.87
FDU-12[27]
1
0.0040
2.13 (Xylene)
1.75
0
6.03
168
14/100
40.3
19.4
9.1
378
0.41
0.12
materials prepared
with other templates
SBA-12[43]
1
0.0585 (Brij76g)
0
0
0
2.35
53
RT/100
8.2
2.4
3.4
610
0.29
FDU-1[41]
1
0.0074 (B50–6600h)
0
0
0
6
155
25/100
23.6
12
4.7
820
0.68
0.25k
Temperature of synthesis/(hydro)thermal
treatment at elevated temperature.
Unit cell parameter.
Pore diameter based on the NLDFT
analysis for adsorption or as represented in the references.
Wall thickness is determined based
on pore diameter and unit cell parameter of a Fm3̅m structured material: Wt =
(a/√2) – dp.
Total pore volume.
Volume of micropores and small mesopore,
calculated using the t-plot method (see the Supporting
Information Figure S9).
Brij76 (C18EO10).
B50–6600 (EO39BO47EO39).
Negligible
volume of small mesopores.
Small mesopores mainly.
Micropore volume including the
volume of the interconnecting pores, if the latter are of a diameter
below 4 nm.[41]
Adsorption (●)
and desorption (○) branches of nitrogen
physisorption isotherms of calcined (A) COK-19, (B) COK-19-50C, and
(C) COK-19-70C. The insets show the pore size distribution based on
NLDFT calculations.Temperature of synthesis/(hydro)thermal
treatment at elevated temperature.Unit cell parameter.Pore diameter based on the NLDFT
analysis for adsorption or as represented in the references.Wall thickness is determined based
on pore diameter and unit cell parameter of a Fm3̅m structured material: Wt =
(a/√2) – dp.Total pore volume.Volume of micropores and small mesopore,
calculated using the t-plot method (see the Supporting
Information Figure S9).Brij76 (C18EO10).B50–6600 (EO39BO47EO39).Negligible
volume of small mesopores.Small mesopores mainly.Micropore volume including the
volume of the interconnecting pores, if the latter are of a diameter
below 4 nm.[41]Increasing the aging temperature to 50 °C resulted
in a larger
hysteresis loop (Figure B). The COK-19-50C material has a mesopore size of 9.4 nm and a mesopore
volume of 0.32 cm3 g–1. Such a temperature
induced increase of mesopore diameter and volume is commonly observed
in Pluronic based synthesis and has been ascribed to an increased
hydrophobicity of the EO moieties, expanding the size of the micellar
core.[37−41] COK-19-50C contains a volume of 0.46 cm3 g–1 of smaller mesopores (∼2.5 nm) and almost no micropores,
resulting in a total pore volume of 0.78 cm3 g–1 (Table ). The significant
increase of pore volume compared to the sample aged at room temperature
and the presence of small mesopores can be interpreted as follows.
Owing to the swelling of the individual micelles, upon stacking some
spaces are left between the core–shell Pluronic-silica particles,
giving rise to small mesopores. The pore size distribution is shown
in the inset of Figure B.Incrementing the aging temperature further to 70 °C
increased
the mesopore size to 11.2 nm and the mesopore volume to 0.63 cm3 g–1. The size of the small mesopores was
also increased to 3.2 nm.From the pore size distribution (Table ) and the unit cell
parameter a derived from SAXS, the wall thickness
was estimated to be approximately
8–9 nm for all materials, aged at different temperatures. These
large wall thicknesses, together with a high degree of silica condensation,
according to 29Si NMR, are expected to contribute high
mechanical stability to these materials.[42]In order to investigate the pore structure of the material
in more
detail, electron tomography was performed on the sample aged at room
temperature as well as at 70 °C (COK-19-70C). From the tomographic
reconstructions, an arrangement of the pores consistent with the proposed
structure model can be resolved in 3D (Figure , Figures S3–S7 and a rotating 3D
representation in the Supporting Information). The mesopores in the sample aged at room temperature are smaller
and more separated by pore walls compared to the pores in the sample
aged at 70 °C. Tomography shows the COK-19 material has isolated
spherical mesopores, separated by a thick silica wall, whereas the
material aged at 70 °C shows some of these mesopores are interconnected.
Moreover, these mesopores are stacked in a cubic order. When temperature
is increased during the aging process, pore sizes increase and stacking
is less ordered, causing some of these pores to connect. Electron
tomography thus corroborates characterization with SAXS and nitrogen
sorption.
Figure 6
Pore structure of COK-19 (A, C) and COK-19-70C (B, D) revealed
by orthoslices through the electron tomography reconstructions, A
and B, and as segmentations of the pores, shown in purple, C and D.
All four views are along a [111] direction, and the arrangement of
pores is consistent with cubic close packing. In COK-19, the pores
are well separated, whereas in the COK-19-70C sample some of the pores
are interconnected. A rotating representation of the pore structure
is provided in the Supporting Information.
Pore structure of COK-19 (A, C) and COK-19-70C (B, D) revealed
by orthoslices through the electron tomography reconstructions, A
and B, and as segmentations of the pores, shown in purple, C and D.
All four views are along a [111] direction, and the arrangement of
pores is consistent with cubic close packing. In COK-19, the pores
are well separated, whereas in the COK-19-70C sample some of the pores
are interconnected. A rotating representation of the pore structure
is provided in the Supporting Information.Table gives an
overview of Fm3̅m ordered
mesoporous materials reported in literature, synthesized using F127
(KIT-5 and FDU-12) or other amphiphilic surfactant templates such
as Brij76 (SBA-12) and B-50–6600 (FDU-1). COK-19 is crystallized
in circum neutral conditions, while the materials from literature
are typically synthesized in hydrochloric acid environment. The materials
from literature have been reported to be intergrown structures, combining
cubic with hexagonal symmetry,[41,43] similar to the COK-19-50C
and −70C. Aging at increased temperature not only increases
the fraction of stacking disorders, it also dramatically increases
the overall porosity. The total porous volume of COK-19 is in the
range of 0.29 to 0.88 cm3 g–1, of which
up to 0.46 cm3 g–1 can be located in
the micropores and small mesopores. Among all of the reported mesoporoussilica materials, FDU-12 is the only one exhibiting similar wall thicknesses,
although the unit cell dimensions and pore sizes are approximately
the double of COK-19 as a result of swelling agents used in the synthesis.
Besides these Fm3̅m type materials
synthesized using amphiphilic surfactants, also cationic surfactants
have been used.[36,44]
Formation Mechanism
For a deeper understanding of the
material and its properties, the formation process of this material
was investigated by DLS and SAXS characterization of intermediates
in the COK-19 synthesis. A similar experimental setup was used to
elucidate the formation mechanism of COK-12.[20] A series of samples was prepared by adding incremental amounts of
sodium silicate solution to buffered F127 solution. At silicate additions
of less than 70% of the nominal quantity, the synthesis mixture was
optically transparent, while at higher silicate contents the mixture
became turbid (Figure ). At 70% silica addition a viscous gel was obtained, and eventually
at 90% and 100% silica addition a suspension of particles was obtained.
In Figure , the onset
of sedimentation is already visible in the sample containing the total
silica amount.
Figure 7
Photograph of the series of synthesis mixtures containing
citrate
buffered F127 surfactant solution and 0%, 15%, 30%, 40%, 50%, 60%,
65%, 70%, 80%, 90%, and 100% of silica addition. Photograph was taken
5 min after silicate addition.
Photograph of the series of synthesis mixtures containing
citrate
buffered F127 surfactant solution and 0%, 15%, 30%, 40%, 50%, 60%,
65%, 70%, 80%, 90%, and 100% of silica addition. Photograph was taken
5 min after silicate addition.In the citrate buffered solution, Pluronic F127 formed micelles
with a hydrodynamic radius of 6.6 nm. It is commonly known that triblock
copolymers such as P123 and F127 form spherical micelles in dilute
aqueous solutions.[23] These micelles are
composed of a dense core of hydrophobic PPO moieties, which are sterically
stabilized by a diffuse shell of PEO chains.[23] When up to 50% of the total silica amount is added to the buffered
F127 solution, the radius of the particles increases slightly with
a few nanometers (Table , Figure S10).
Table 2
Size Estimation
from SAXS and DLS
of F127-Silica Core-Shell Micelles Prepared Using Different Percentages
of Silica Addition
DLS
SAXSPolydisperse Core–Shell Particles
%
Si
Rh (nm)
Rcore (nm)
tSiO2 (nm)
Rtotal (nm)
0%
6.6
15%
8.4
5.6
1.0
6.7
30%
9.0
5.2
3.1
8.3
40%
9.2
5.1
3.6
8.7
50%
9.4
4.6
4.1
8.7
60%
13.2
4.8
4.0
8.8
65%
36.6
4.9
4.2
9.1
70%
4.8
4.3
9.1
80%
6.2
4.6
10.8
90%
7.7
3.9
11.6
100%
4.7
4.0
8.7
Scanning
transmission electron microscopy (STEM) of an evaporated
synthesis mixture with 50% silicate addition confirmed the formation
of spherical nanoparticles. The bright field STEM micrograph (Figure ) shows loosely aggregated,
monodisperse spherical nanoparticles with radii of ca. 8.5 nm. Increasing
the silica concentration further led to a further increase in particle
size. However, DLS was limited to samples with 65% of the total silica
amount. Higher silica amounts led to a sharp transition of more aggregated
particles, which could not be monitored with DLS. Using SAXS, these
nanostructures could be analyzed by modeling the scattering patterns
with spherical core–shell models, giving additional information
on shape, inner structure, and aggregation state.
Figure 8
Bright field STEM image
of F127 micelles, to which only 50% of
the nominal silicate quantity was added.
Bright field STEM image
of F127 micelles, to which only 50% of
the nominal silicate quantity was added.In the initial buffered F127 solution, the scattering contrast
between the aqueous environment and the F127 micelles was insufficient
to enable recording a SAXS pattern. SAXS patterns of synthesis mixtures
with 15% up to 100% of the nominal silicate content are presented
in Figure . At silicate
contents below 65%, the scattering curves could be described by a
spherical core–shell model, in which the electron density of
the core was lower than in the shell (details provided in the Supporting Information). These differences in
contrast were interpreted as a core consisting of organics surrounded
by a shell containing silica. Regarding contrast, it is noteworthy
that due to dilution in an aqueous solution, the presence of the organic
F127 surfactant inside the pores, and an incomplete silica condensation
in the final material, scattering contrast is suboptimal. After filtration
and removal of the organic template through calcination, diffraction
lines in the SAXS pattern are sharper and more pronounced (Figure ).
Figure 9
SAXS patterns of synthesis
mixtures consisting of buffered F127
surfactant solutions to which an increasing amount of sodium silicate
was added: (a) 15%, (b) 30%, (c) 40%, (d) 50%, (e) 60%, (f) 65%, (g)
70%, (h) 80%, (i) 90%, and (j) 100% of the standard synthesis. Curves
were shifted vertically for clarity.
SAXS patterns of synthesis
mixtures consisting of buffered F127
surfactant solutions to which an increasing amount of sodium silicate
was added: (a) 15%, (b) 30%, (c) 40%, (d) 50%, (e) 60%, (f) 65%, (g)
70%, (h) 80%, (i) 90%, and (j) 100% of the standard synthesis. Curves
were shifted vertically for clarity.On analogy with previously investigated P123 micelles,[20] the organic core is interpreted as the PPO part
of the F127 micelle, while the shell consists of the PEO chains and
infiltrated silica. An overview of estimated core radii and shell
thicknesses depending on silica addition is presented in Table . The overall particle
radii for core–shell particles, with increasing sodium silicate
content, were consistent with the size estimation from DLS. DLS generally
yields a slightly larger size estimate than SAXS. This discrepancy
is caused by the presence of a solvation layer, reducing the diffusivity
and increasing the size of particles in DLS measurements.SAXS
patterns and model fittings are shown in Figures and 10 and Figures
S11–S21 in the Supporting
Information. The high q part of the residual
(q values > 0.2 A–1) additionally
displayed the fingerprint of subnanometer particles. Combination of
such interpretation with the core–shell micelle description
led to a size estimation of nanoparticles of approximately 0.6 nm.
Previously, similarly sized silica nanoparticles have been observed
in SAXS studies of zeolite formation processes and have been identified
as partially condensed silicate oligomer aggregates from which crystallization
originates.[45−47] In the present system, it seems these silicate oligomers are the
micelle infiltrating species. This interpretation was supported by
the observation that the scattering intensity at q values > 0.2 A–1 assigned to silica nanoparticles
did not change with increasing silica content. Silica nanoparticles
formed in solution infiltrate the PEO fuzzy layer of F127 micelles
and precipitate forming a capsule. Similar spherical silica infiltrated
micelles were observed using P123 triblock copolymer in the formation
of hexagonally ordered mesoporoussilica materials.[12,20] As silica is added, more silica is concentrated in the hydrophilic
PEO layer, increasing the shell thickness, slightly compressing the
PPO core (Table ).
Figure 10
SAXS
patterns of F127 surfactant solutions to which (A) 15%, (B)
40%, (C) 60%, (D) 70%, (E) 80%, and (F) 100% of the total sodium silicate
amount in a standard synthesis was added and corresponding fitting
with core–shell micelle model and unassociated silica nanoparticles.
Patterns A, B, and C were fitted with a combined model for polydisperse
core–shell particles and smaller, dense spherical particles.
Patterns D, E, and F were fitted with the same model superimposed
with pseudo-Voigt peaks (supplementary modeling information can be
found in the Supporting Information).
SAXS
patterns of F127 surfactant solutions to which (A) 15%, (B)
40%, (C) 60%, (D) 70%, (E) 80%, and (F) 100% of the total sodium silicate
amount in a standard synthesis was added and corresponding fitting
with core–shell micelle model and unassociated silica nanoparticles.
Patterns A, B, and C were fitted with a combined model for polydisperse
core–shell particles and smaller, dense spherical particles.
Patterns D, E, and F were fitted with the same model superimposed
with pseudo-Voigt peaks (supplementary modeling information can be
found in the Supporting Information).The addition of 70% and higher
percentages of the silica caused
spontaneous aggregation of the silica coated F127 micelles. In SAXS,
a structuring factor appeared in the small angle scattering patterns,
which was modeled using pseudo-Voigt peaks. The introduction of pseudo-Voigt
peaks in the model corresponds to the appearance of Bragg diffraction.
Suspensions with 80% up to 100% silicate addition required three pseudo-Voigt
peaks to adequately describe the SAXS patterns. The scattering intensity
at q values > 0.2 A–1 remained
constant, whereas, relatively, the amount of scattering from core–shell
particles decreased. This suggests selective involvement of the core–shell
micellar structures in the assembly process.From this series
of SAXS experiments and the previously illustrated
COK-12 formation mechanism,[20] the following
formation mechanism is proposed (Figure ). When the highly alkaline sodium silicate
solution is added to the citrate buffered F127 solution, silicate
anions are protonated and form neutral silicate oligomers and orthosilicic
acid Si(OH)4. These small oligomers condense to larger
oligomers with radii of 0.6 nm. The silicate oligomers interact strongly
with the PEO tails of the Pluronic micelles, condensing in between
the diffuse layer of PEO chains.[12,48,49]
Figure 11
Schematic representation of COK-19 formation mechanism,
showing
the effect of increased silica concentration. Silica (blue) is precipitated
on the PEO chains of the Pluronic F127 micelles (brown). The silica
layer around the F127 micelles will increase in thickness until sterical
stabilization is overcome and spherical silica coated micelles start
to aggregate, ultimately stacking in a cubically ordered structure.
Schematic representation of COK-19 formation mechanism,
showing
the effect of increased silica concentration. Silica (blue) is precipitated
on the PEO chains of the Pluronic F127 micelles (brown). The silica
layer around the F127 micelles will increase in thickness until sterical
stabilization is overcome and spherical silica coated micelles start
to aggregate, ultimately stacking in a cubically ordered structure.The synthesis was performed at
pH 5, close to the isoelectric point
of F127. The absence of charge, together with the large salt concentration,
implies stabilization of the F127 micelles must occur through sterical
effects of the PEO chains. With increasing sodium silicate concentration,
larger amounts of silica are incorporated in the PEO layer, increasing
the shell thickness around the micelles. It is interesting to note
that in the COK-12 formation, sterical stabilization was overcome
after adding only 15% of the nominal silica amount, transforming the
coated, spherical micelles into elongated cylinders. In contrast,
the silica coated Pluronic F127 micelles omit to form a cylindrical
intermediate and form a regular arrangement of close packed silica
coated spheres. As the main difference between both syntheses is the
length of the PEO chains of the Pluronic surfactants (EO20 versus EO102 for, respectively, P123 and F127), the difference
in sterical stabilization and consequently the different self-assembly
process must be attributed to the increasing chain length.
Conclusions
By replacing the P123 Pluronic with the F127 as a template in the
formation of mesoporoussilica, a new ordered material was discovered.
This COK-19 material has a Fm3̅m symmetry with hcp intergrowths, which increase with increasing temperature.
The material contains spherical mesopores of 7–11 nm, interconnected
via micropores and small mesopores in its 8–9-nm-thick silica
walls, which could be tuned via varying aging temperature.Intermediates
in the formation process of the COK-19 material could
be investigated by limiting the silicon source and modeling the SAXS
pattern. Uncharged F127 micelles are wrapped with silica by accumulation
of the silicate oligomers in the hydrophilic PEO layer of the micelles.
Sterical stabilization decreased with increasing amounts of silica,
until F127 core–silica shell particles self-assemble in a cubic
close packed organization, ultimately forming the COK-19 material.
Authors: S Förster; A Timmann; M Konrad; C Schellbach; A Meyer; S S Funari; P Mulvaney; R Knott Journal: J Phys Chem B Date: 2005-02-03 Impact factor: 2.991
Authors: Michiel Van Speybroeck; Valéry Barillaro; Thao Do Thi; Randy Mellaerts; Johan Martens; Jan Van Humbeeck; Jan Vermant; Pieter Annaert; Guy Van den Mooter; Patrick Augustijns Journal: J Pharm Sci Date: 2009-08 Impact factor: 3.534
Authors: Jasper Jammaer; Titus S van Erp; Alexander Aerts; Christine E A Kirschhock; Johan A Martens Journal: J Am Chem Soc Date: 2011-08-10 Impact factor: 15.419
Authors: Randy Mellaerts; Raf Mols; Jasper A G Jammaer; Caroline A Aerts; Pieter Annaert; Jan Van Humbeeck; Guy Van den Mooter; Patrick Augustijns; Johan A Martens Journal: Eur J Pharm Biopharm Date: 2007-11-23 Impact factor: 5.571
Authors: Jivaldo R Matos; Michal Kruk; Lucildes P Mercuri; Mietek Jaroniec; Lan Zhao; Tomoaki Kamiyama; Osamu Terasaki; Thomas J Pinnavaia; Yu Liu Journal: J Am Chem Soc Date: 2003-01-22 Impact factor: 15.419
Authors: Aleksandr A Goncharenko; Ilya A Tarasyuk; Yuriy S Marfin; Kirill V Grzhegorzhevskii; Albert R Muslimov; Andrey B Bondarenko; Maxim D Lebedev; Ilya A Kuz'min; Artur S Vashurin; Kirill V Lepik; Alexander S Timin; Evgeniy V Rumyantsev Journal: Molecules Date: 2020-08-21 Impact factor: 4.411