Core-and-Shell Nature of Stöber Silica Particles

Abstract

Two different samples of monodisperse Stöber silica particles were examined in the analytical transmission electron microscope, using different imaging modes: bright-field, dark-field, energy-loss and elemental distribution maps. The particles (effective diameters = 141 and 36 nm) are formed by domains of variable O/Si ratio, which is consistent with a variable degree of hydration, and they coexist with particles with a high O/Si ratio measuring a few nanometers only, which appear dispersed in the picture background. Bright-field and energy-loss images of the larger particles show a core-and-shell morphology, and the shells have a higher amount of high-O/Si domains as well as contaminating carbon compounds. On the other hand, the smaller particles (effective diameter = 36 nm) are also formed by distinct domains, but their morphology is neither spherical or core-and-shell. The mechanisms for particle formation presented in the literature are discussed, considering the present findings.

Full text

Article
Core-and-Shell Nature of Stöber Silica Particles
Carlos A. P. Leitea, Elizabeth F. de Souzab and Fernando Galembecka*
aInstituto de Química, Universidade Estadual de Campinas, CP 6154, 13083-970, Campinas - SP, Brazil
bInstituto de Ciências Biológicas e Química, Pontifícia Universidade Católica de Campinas,
CP 1111, 13020-904, Campinas - SP, Brazil
Duas amostras diferentes de partículas monodispersas de sílica de Stöber foram examinadas
por microscopia eletrônica de transmissão analítica, utilizando-se diferentes tipos de imagens:
campo-claro, campo escuro, imagens espectrais e mapas de distribuição elementar. As partículas
(141 e 36 nm de diâmetro efetivo) contêm domínios de relação elementar O/Si e, portanto, de
grau de hidratação muito variável, que coexistem com partículas medindo poucos nanômetros de
diâmetro e com relação O/Si elevada, que aparecem dispersas no fundo das imagens. As imagens
de campo claro e de perda de energia revelam que as partículas maiores possuem uma morfologia
do tipo caroço-casca e as cascas das partículas possuem uma quantidade maior de domínios com
razão O/Si maior, comparada com o interior da partícula, o que é atribuído ao acúmulo de domínios
mais hidratados na casca, e também à presença de compostos de carbono, contaminantes da sílica.
Por outro lado, as partículas menores (diâmetro efetivo = 36 nm) não são esféricas nem têm
morfologia caroço-casca, embora também sejam formadas por domínios de composições químicas
muito diferentes. Os vários mecanismos de formação de partículas apresentados na literatura são
discutidos, considerando-se estes novos resultados.
Two different samples of monodisperse Stöber silica particles were examined in the analytical
transmission electron microscope, using different imaging modes: bright-field, dark-field, energy-
loss and elemental distribution maps. The particles (effective diameters = 141 and 36 nm) are
formed by domains of variable O/Si ratio, which is consistent with a variable degree of hydration,
and they coexist with particles with a high O/Si ratio measuring a few nanometers only, which
appear dispersed in the picture background. Bright-field and energy-loss images of the larger
particles show a core-and-shell morphology, and the shells have a higher amount of high-O/Si
domains as well as contaminating carbon compounds. On the other hand, the smaller particles
(effective diameter = 36 nm) are also formed by distinct domains, but their morphology is neither
spherical or core-and-shell. The mechanisms for particle formation presented in the literature are
discussed, considering the present findings.
Keywords: colloidal silica, transmission electron microscopy, energy-filtered imaging, silica
particle microchemistry
J. Braz. Chem. Soc., Vol. 12, No. 4, 519-525, 2001.
Printed in Brazil 2001 Soc. Bras. Química
0103 - 5053 $6.00+0.00
Introduction
Uniform fine particles are interesting as model systems
in the study of adsorption and catalysis, as well as in the
study of size-dependent solid state properties such as
quantum confinement1 and superplasticity2,3. There have
been significant achievements concerning the preparation
of uniform colloid dispersions, both inorganic4 and organic5.
However, uniformity has often been considered only in
relationship to particle size and shape, and little information
is currently available concerning the uniformity of chemical
composition of particles made out of one only constituent,
such as the Stöber silica particles.
Many years ago, Stöber et al.6 reported an elegant
method of preparing monodisperse spherical silica particles
with sizes covering almost the whole colloidal range, by
tetraethylorthosilicate (TEOS) hydrolysis, in ethanolic
medium in the presence of ammonia. They carried out a
systematic study of the reaction parameters and after some
major changes of the experimental conditions he has
obtained suspensions of quasi-monodisperse silica spheres.
After this pioneering work, Stöber silica particles have been
used as model colloids in a large number of experimental
*e-mail: fernagal@iqm.unicamp.br

520 Leite et al. J. Braz. Chem. Soc.
investigations7-10. Based on the Stöber method, Kaiser11
prepared completely porous silica particles by co-hydro-
lysis and subsequent condensation of tetraethoxysilane and
an n-alkyltrialkoxysilane in a mixture of ethanol, water and
aqueous ammonia. Silica particles almost perfectly
spherical with porous siliceous shells and dense cores were
also obtained by Büchel and co-workers12. Preparation of
spherical silica nanoparticles with a narrower size
distribution than that obtained by hydrolysis of tetraethoxy-
silane in homogeneous alcoholic media has been achieved
in microemulsion systems13-15.
Recently, several investigators have shifted their
attention to the study of the mechanisms of formation and
growth of these particles16. For this purpose various
techniques, such as nuclear magnetic resonance17,
conductimetry18, Raman scattering19, dynamic light
scattering20, transmission electron microscopy21 and small-
angle X-ray scattering22 were applied to investigate both
the chemistry and the physical properties of the particles,
but more noticeably the dynamics of growth23. Despite
these intensive investigations, a clear and complete picture
for the formation of uniform silica particles has not yet
emerged. The elucidation of their domain nanostructure is
also an attractive research objective.
In this laboratory, we have recently observed significant
heterogeneity and domain structure of polymer latex
particles, which affects particle aggregation, macro-
crystallization and film formation24, 25. This information
is not usually available for inorganic fine particles, for
which reason we decided to examine the uniformity of
chemical composition of the Stöber silica particles. To do
this, we have examined two samples of particles of different
sizes, using three different imaging techniques in an
analytical transmission electron microscope.
Experimental
Preparation of the silica particles
Silica particles were prepared by the method of Stöber
et al.6. TEOS, absolute ethanol used as the solvent and
ammonium hydroxide were of analytical reagent grade.
Glassware was cleaned with 10% hydrogen chloride, rinsed
with distilled water and absolute ethanol. The desired
concentrations of ammonium hydroxide, ethanol and water
were mixed in Erlenmeyer flasks with ground stoppers.
Total water contents were calculated by adding up the
fractional amounts introduced by the components.
Subsequently, the TEOS was added and the flasks were
placed in a water bath at a temperature of 25 ± 0.1oC and
under ultrasonic vibration (20 kHz) for 120 minutes. The
total volume of solution in each preparation was 50 mL.
The measured volumes of the components used to prepare
the larger silica particles were 4 mL TEOS, 4 mL saturated
ammonium hydroxide and 50 mL ethanol. The smaller
particles were prepared by mixing 4 mL TEOS, 2 mL
saturated ammonium hydroxide and 50 mL ethanol.
Following the synthesis, ca. 50 mL of each silica dispersion
was stored in a 100 mL glass bottle.
Photon Correlation Spectroscopy (PCS)
Effective diameter was measured in a ZetaPlus
instrument (Brookhaven Instruments) with Bi-MAS
software and a solid state laser (15 mW, λ= 670 nm) as the
radiation source. Samples were contained within 3-mL
dust-free acrylic cuvettes. A volume of 100 µL of the silica
dispersion was added to the cuvette previously filled with
2.5 mL of 10-3 mol L-1 aqueous potassium chloride in order
to give a suitable scattering intensity.
Analytical transmission electron microscopy imaging
A Carl Zeiss CEM 902 transmission electron
microscope, equipped with a Castaing-Henry energy filter
spectrometer within the column, a Proscan Slow Scan CCD
camera and controlled by a microcomputer running the
AnalySis 3.0 system was used. The spectrometer uses
inelastically scattered electrons to form energy-loss and
element-specific images. When the electron beam passes
through the sample, interaction with electrons of different
elements results in characteristic energy losses. A prism-
mirror system deflects electrons with different energies to
different angles so that only electrons with a well defined
energy are selected. If elastic electrons only are chosen
(∆E = 0 eV) a transmission image with reduced chromatic
aberration is obtained. When monochromatic inelastically
scattered electrons are selected, electron spectroscopic
images (ELSI) are formed, in which contrast is dependent
on the local energy-loss spectrum and thus on the
concentration fluctuations of a particular chosen element.
Clear areas in the elemental distribution maps correspond
to element-rich domains. The following procedure is used
to acquire spectral images: a set of 38 to 42 images is
acquired, around the absorption border for each element
of interest. The energy window used is 6 eV, and the energy
steps between images is 2.5 eV. This image set is used to
define the three energy windows used for elemental
mapping. Two images are recorded at energy windows
below the absorption threshold, and they are used for fitting
the background with a chosen function. The third image is
obtained using an energy window set at the absorption band.
The elemental map is obtained by subtracting the

Vol. 12 No. 4, 2001 Core-and-Shell Nature of Stöber Silica Particles 521
background from the image acquired in the third image,
and it is checked for signal saturation, using the R-map
macro from the AnalySis software. Each elemental map is
validated by three independent checks: i) contrast inversion
in the plasmon region; ii) spectral verification and iii)
absence of signal saturation.
For individual silica particle examination, one drop
of the silica dispersion (1% solids content) was deposited
on carbon-coated parlodion films supported in 400 mesh
copper grids (Ted Pella). To make sure that the whole
particles were not excessively thick, they were first
observed using ∆E = 0 eV electrons, then observed again
at ∆E = 15-50 eV. Image contrast inversion was always
obtained, showing that a significant number of electrons
were transmitted throughout the particles26.
This observation is understood, considering that the
80 keV electrons mean free path within these silica
particles is greater than 80 nm for elastic scattering27,
and is estimated as a few hundreds of nanometers, for
inelastic scattering28.
Elemental images were observed for the relevant
elements found in this sample, using the three-window
technique29, with monochromatic electrons corresponding
to the silicon K-edge (1860 eV, 20 eV window), oxygen
K-edge (532 eV, 15 eV window) and carbon K-edge (284
eV, 15 eV).
Image processing was performed in an IBM PC
microcomputer using the Image-Pro Plus 4.0 image
analyzer program (Media Cybernetics).
Results
According to the data from the PCS measurements,
the effective particle diameter of the larger particles is
141 ± 2.5 nm, showing that these particles have a narrow
size distribution. The sample of smaller particles has an
effective particle diameter of 36 ± 1.0 nm.
Different types of images were obtained: Figure 1 shows
images obtained for the larger silica particles: bright-field, dark-
field, energy-filtered image obtained at 20 eV energy loss, and
Si, O and C elemental distribution maps. Energy-loss spectra
were scanned for this sample, from 100 to 2,000 eV, and the
only bands observed are those assigned to C, O and Si, as shown
in the Figure 2. The detection of carbon compounds is rather
unexpected since these were not previously reported in the
abundant literature on these silica particles.
The bright-field, dark-field and energy-filtered images
for the smaller particles are in Figure 3, and the
corresponding elemental maps were published in another
recent paper from this laboratory30.
The observations made from these pictures are summed
up as follows, for the sample of larger silica particles:
ba
fd
c
e
220 nm ba
fd
c
e
ba
fd
c
e
220 nm
Figure 1. (a) Brightfield, (b) dark-field, (c) 20eV energy-loss, (d) Si map, (e) O and (f) C map images taken from the same particles of sample 1 (effective
particle diameter = 141 nm).

522 Leite et al. J. Braz. Chem. Soc.
the atomic ordering of the different particle domains, their
thickness and solid structural characteristics. These silica
particles are non-crystalline, consequently this difference
cannot be assigned to differences in crystalline domain
orientation, but rather to differences in the chemical
composition in different domains. Beyond, the brighter
domains predominate at the particle outer shell while the darker
domains predominate at the particle cores. Interparticle space
is also filled with scattering material and the picture
background shows scattered bright spots, well away from the
major particles. These small bright spots are assigned to either
dry solute or to very small particles with few nanometers.
iii) The energy-filtered image in Figure 1(c) is an energy-
loss image at 20 eV, in the giant resonance (or plasmon, in
conducting solids) region. It presents a similar pattern to the
dark-field image, but with a higher contrast since the particle
cores appear darker in Figure 1c. Moreover, the bright rings
around the particles are thinner in Figure 1c than in Figure
1b, showing that the factors for the contrast in both are not
identical. There is not sufficient information in the literature
to allow us to correlate giant-resonance images with chemical
features of sample domains, but this image confirms the core-
and shell nature of these particles.
bf
200 nm
df
200 nm
bf
200 nm
df
200 nm
Figure 3. (a) Brightfield and (b) dark-field images taken from the smaller
particles (effective particle diameter = 36 nm).
500 520 540 560 580 600
0
500
1000
1500
2000
2500
3000
3500
4000 O
Intensity/a.u.
Energy/eV
1830 1860 1890 1920
0
1000
2000
3000
4000
5000
6000
7000
8000 Si
In te n s it y/a .u .
Energy/eV
260 280 300 320 340 360
0
500
1000
1500
2000
2500
3000 C
In te n s it y/a .u .
Energy/eV
Figure 2. Energy-loss spectra of the 141 nm diameter particles, in the
regions corresponding to the C(K), Si(K) and O(K) thresholds (from top
to botton).
i) The bright-field picture in Figure 1(a) shows that the
particles are rather uniform in size, and their borders are not
strictly circular. Particles in close contact with the neighbors
display a marked deformation approaching the hexagonal
shape, analogous to the deformations observed in polymer
latexes24,25 and evidencing the plasticity of these particles.
In some areas, the interparticle space is darker than the
background, showing the presence of material joining the
particles. Moreover, the background shows scattered darker
areas, made out of small nanometric grains.
ii) The darkfield image in Figure 1(b) shows a clear contrast
within each particle, displaying bright and dark spots extending
for a few nanometers each. Contrast in darkfield images is
due to differences in the electron scattering ability and thus in
(a)
(b)

Vol. 12 No. 4, 2001 Core-and-Shell Nature of Stöber Silica Particles 523
iv) The O and Si elemental maps are similar, but the
particles appear slightly larger in the O maps than in the Si
maps. This is best seen in the line-scans presented in Figure
4. In the O scan we observe sharp peaks and valleys,
evidencing large composition changes among neighboring
domains or voids and pores, extending for 2-10 nm each.
The Si map also displays the same sharp features, but these
are packed into broader bands that correspond to the
individual particles. Consequently, the Si/O atom ratio in
the particles is maximum at the particle centers and it
decreases gradually to the borders. Finally, the C map shows
the particle centers as dark areas, with bright outer rings,
bright interstitial spaces and bright spots in the background,
at positions roughly coinciding with the bright areas in the
background of the dark-field, 20-eV images. Nanometer-
sized spots are also seen scattered in the particle cores.
Observations of similar images for the sample with 36
effective particle diameter did not show a core-and-shell
structure, in the dark-field images, as seen in Figure 3.
Examination of other types of images (Si and O elemental
maps, energy-loss30) did not also give any evidence for a
core-and-shell structure in these small particles.
Discussion
According to Iller31, in most sols that consist of discrete
spherical particles of amorphous silica, the interior of the
particles is made out of anhydrous SiO2 with a density of
2.2 g.cm-3. The silicon atoms located at the surface hold
OH groups that are not lost when the silica is dried to
remove free water. Calculation of the silanol number of
the silica surface by purely geometric considerations and
the density of amorphous silica indicated that there should
be 7.8 silicon atoms.nm-2 at or very near the surface.
The present results show that the silica particles have a
pronounced domain structure but the larger particles
(141nm diameter) have also an accumulation of particles
with a higher O/Si ratio at the particle borders, as compared
to the particle cores. The accumulation of one domain type
at the particle outer layers imparts to the particles a core-
and-shell nature, and the thickness of the shell layer extends
for many nanometers. The particles detected by PCS coexist
with nanometric scattered particles (which appear dispersed
in the pictures background as well as in the particle
interstices) with a Si/O ratio lower than that of the large
particle cores.
Following existing reports on silica particle formation
from TEOS, carbon from the ethoxy groups is transformed
into ethanol, which is easily dialyzed and also lost by
evaporation12. However, Van Helden et al.7 have already
reported the presence of carbon in the composition of
Stöber silica particles. They assumed that the carbon
content determined gravimetrically originates from
unhydrolyzed ethoxy groups and calculated that the
particles contain 91.5% weight silica, 5.25% water and
0.97% ethoxy groups. Therefore, the observation of
carbon compounds in the particle shells as well as in the
background and in the interparticle interstices is in
agreement with their work. The amount of non-dialyzable,
non-volatile carbon compounds found at the particle
surfaces were detected by EELS, which is a technique
endowed with single-atom sensitivity. It is also important
to note that the observed carbon compounds may derive
not only from incomplete ethoxy group hydrolysis
(leading to ethoxylated silicic acid), but also from TEOS
impurities, as well as from some ethanol oxidation
product, or perhaps from some contamination introduced
during the particles fabrication, e.g. from hydrolyzed
fragments from the dialysis bags cellulose. The
elucidation of the origin of the carbon will require further
additional work, well beyond the scope of this paper. On
the other hand, the smaller particles (36 nm diameter)
examined in this work do not have the same structure as
the larger (141 nm) particles.
Si scan
0
20
40
60
80
100
0 200 400 600 800
Pixel number
G
ray level
O scan
0
20
40
60
80
100
0 200 400 600 800
Pixel number
Gray level
Figure 4. Line-scans from the Si (top) and O (bottom) maps in Figure 1.
The width of the line scanned is 16 pixels, and the length is 752 pixels.
The position of the line is shown in Figure 1d.

524 Leite et al. J. Braz. Chem. Soc.
The domain structure of the larger and smaller particles
as well as the core-and-shell nature of the former may be
understood considering the particle synthesis procedure and
the intervening reactions. The form of the resulting
polymers obtained during base catalyzed hydrolysis of
tetraethylorthosilicate in alcoholic solutions is governed
by the relative rates of hydrolysis and condensation. Silanol
groups are formed by hydrolysis of silicon alkoxide
monomers and undergo condensation, according to the
following equations23,32:
≡Si-OR + H2O → ≡Si-OH + ROH (1)
≡Si-OH + ≡Si-OH → ≡Si-O-SI ≡ + H2O (2)
≡Si-OR + ≡Si-OH → ≡Si-O-SI ≡ + ROH (3)
where R represents an alkoxide group. Spherical
particles are obtained when enough ammonia is present in
the initial reaction mixture, and the final particle size
depends mainly on the initial alkoxysilane, water and
ammonia concentrations.
Harris and co-workers33 have investigated the growth
of particles produced by this method. They assumed that
the first hydrolysis of an alkoxide group is the rate-limiting
step in the formation of nuclei. The remaining groups
hydrolyze rapidly, and small nuclei are created from fully
hydrolyzed species. In the early stages of the reaction, these
small nuclei aggregate, forming colloidally stable seed
particles, and in the later stages of the reaction particle
growth occurs mainly by addition of monomer to the
surface, not by aggregation of the small nuclei particles as
claimed by Bogush and Zukoski34. Using the results of
cryo-TEM experiments, Bailey and Mecartney21 postulated
that hydrolyzed monomer polymerizes to form microgel
clusters due to polysilicic acid cross-linking, that collapse
upon reaching a certain size and cross-linking density.
Collapsed particles densified by condensation are
colloidally stable with respect to each other. The denser
seed particles grow by addition of hydrolyzed monomer
or polymer addition to their surfaces. The rate of growth
of the polymers must be slow enough so that after a
sufficient number of seeds has been formed, the polymers
attach to a particle surface before they grow to a large
enough size to collapse and form a seed particle themselves.
Boukari and co-workers16 have proposed that the initial
particles also with a polymeric structure are better described
like mass fractals, in which the nucleating backbones or
seeds are used to build the compact and stable particles
observed later in the growth. Before the equilibrium is
reached, there is a distribution of particles of various sizes
and fractality. Besides that, under conditions in which
condensation is rapid compared to hydrolysis, polysiloxane
chains or rings result while in the reverse case more
extensively crosslinked polymeric clusters are formed.
Therefore, polymers formed in base-catalyzed reactions
are more highly branched clusters35.
In a batch synthesis procedure, the changing medium
may thus allow for changes in the particle structure, due to
changes in the nature of the siloxane chains formed as the
reaction proceeds. However, we note that none of these
authors seemed to be aware of the core-and-shell structure
or of presence of carbon compounds in these particles, as
described in the present paper. Consequently, our data add
two new pieces of information, which will have to be
considered in future work aiming at elucidating silica particle
formation: the particles are formed by differentiated domains,
and there is a detectable amount of non-dialyzable, non-
volatile carbon compounds at the particle surfaces.
Finally, the core-and-shell structure formation is
understood, since in the larger particles the more hydrophilic
chains move to the particle surface, driven by the
minimization of interfacial tension and thus following the
same mechanism proposed by El-Aasser and colleagues36
for core-and-shell formation in polymer colloids. Moreover,
the accumulation of carbon compounds at the particle
surfaces add to differentiation between particle core and
particle shell. However, the amount, nature and significance
of the carbon compounds detected in this work will require
a significant amount of further experimental work.
Conclusions
The larger silica particle (effective diameter = 141 nm)
cores have a domain structure, and the thick outer shells
contain both hitherto unidentified carbon compounds and
a higher silanol content than the particle bulk. On the other
hand, the smaller particles (36 nm) do not show a marked
core-and-shell morphology, but a marked particle bulk
heterogeneity. Very small particles with few nanometers
are observed in the background, with a low Si/O ratio and
a high C/Si ratio. These observations are consistent with
the following picture for particle formation: different kinds
of siloxane chains are formed and aggregate, generating
small particles of non-uniform chemical composition. In
the larger particles, the more hydrophilic chains move to
the outer particle shell, driven by the minimization of
interfacial tension, thus forming a core-and-shell structure.
Acknowledgements
The authors thank FAPESP, Pronex/Finep/MCT
and CNPq.

Vol. 12 No. 4, 2001 Core-and-Shell Nature of Stöber Silica Particles 525
References
1. Henglein, A. Chem. Rev. 1989, 89, 1861.
2. Averback, R. S.; Höfler, H. J.; Tao, R. Mater. Sci.
Eng. 1993, 166, 169.
3. Galembeck, F.; Lima, E. C. O.; Masson, N. C.;
Monteiro, V. A. R.; Souza, E. F. In Fine Particles Science
and Technology: From Micro to Nanoparticles;
Pelizzetti, E., Ed.; Kluwer; Dordrecht, 1996, p.267-279.
4. Matijevic, E. Chem. Mater. 1993, 5, 426.
5. Hammouda, A.; Gulik-Krzywicki, T.; Pileni, M. P.
Langmuir 1995, 11, 3656.
6. Stöber, W.; Fink, A.; Bohn, E. J. Colloid Interface
Sci. 1968, 26, 62.
7. Van Helden, A. K.; Vrij, A. J. Colloid Interface Sci.
1980, 78, 312.
8. Kirkland, J. J. J. Chromatogr. 1979, 185, 273.
9. Kops-Werkhoven, M. M.; Fijnaut, H. M. J. Chem.
Phys. 1981, 74, 1618.
10. Tan, C. G.; Bowen, B. D.; Epstein, N. J. Colloid
Interface Sci. 1987, 118, 290.
11. Kaiser, C.; Büchel, G.; Lüdtke, S.; Lauer, I.; Unger,
K. K. In Characterisations of Porous Solids IV;
McEnaney, B.; Mays, T. J.; Rouquérol, J.; Rodríguez-
Reimoso, F.; Sing, K. S. W.; Unger, K. K., Eds.; Royal
Society of Chemistry; Cambridge, 1997, p. 406-412.
12. Büchel, G.; Unger, K. M.; Matsumoto, A.; Tsutsumi,
K. Adv. Mater. 1998, 10, 1037.
13. Osseo-Asare, K.; Arriagada, F. J. Colloids Surf. 1990,
50, 321.
14. Minehan, W. T.; Messing, G. L. Colloids Surf. 1992,
63, 181.
15. Esquena, J.; Tadros, Th. F.; Kostarelos, K.; Solans, C.
Langmuir 1997, 13, 6400.
16. Boukari, H.; Lin, J. S.; Harris, M. T. Chem. Mater.
1997, 9, 2376.
17. Lee, K. T.; Look, J. L.; Harris, M. T.; McCormick, A.
V. J. Colloid Interface Sci. 1997, 194, 78.
18. Bogush, G. H.; Zukoski, C. F. J. Colloid Interface
Sci. 1991, 142, 1.
19. Matsoukas, T.; Gulari, E. J. Colloid Interface Sci.
1988, 124, 252.
20. Biddle, D.; Walldal, C.; Wall, S. Colloids Surf., A
1996, 118, 89.
21. Bailey, J. K.; Mecartney, M. L. Colloids Surf., A 1992,
63, 151.
22. Konishi, T.; Yamahara, E.; Ise, N. Langmuir 1996,
12, 2608.
23. Boukari, H.; Lin, J. S.; Harris, M. T. J. Colloid
Interface Sci. 1997, 194, 311.
24. Cardoso, A. H.; Leite, C. A. P.; Galembeck, F.
Langmuir 1998, 14, 3187.
25. Galembeck, F., Souza, E.F. In Polymer Interfaces and
Emulsions; Esumi K., Ed.; Marcel Dekker; New York,
1999, p. 119-166.
26. We are grateful to Dr. W. Probst (LEO-Zeiss
Elektronenmikroskopie Gmbh) for this valuable
private communication.
27. Newbury, D. E. In Principles of Analytical Electron
Microscopy; Joy, D. C.; Romig Jr., A. D.; Goldstein,
J. L., Eds.; Plenum Press; New York, 1986.
28. Ibid., p. 20.
29. Reimer, L.; Zepke, U.; Moesch, J.; Schulze-Hillert,
St.; Ross-Messemer, M.; Probst, W.; Weimer, E.
EELS Spectroscopy: A Reference Handbook of
Standard Data for Identification and Interpretation
of Electron Energy Loss Spectra and for Generation
of Electron Spectroscopic Images; Carl Zeiss;
Oberkochen, 1992.
30. Costa, C. A. R.; Leite, C. A. P.; Souza, E. F.;
Galembeck, F. Langmuir 2001, 17, 189.
31. Iler, R. K. The Chemistry of Silica; Wiley; New
York, 1979.
32. Lindberg, R.; Sjöblom, J.; Sundholm, G. Colloids
Surf., A 1995, 99, 79.
33. Harris, M. T.; Brunson, R. R.; Byers, C. H. J.Non-
Cryst. Solids 1990, 121, 397.
34. Bogush, G. H.; Zukoski, C. F. In Ultrastructure
Processing of Advanced Ceramic; Mackenzie, J. D.,
Ed.; Wiley; New York, 1988, p.477.
35. Yamane, M.; Inoue, S.; Yasumori, J. J. Non-Cryst.
Solids 1984, 63, 13.
36. Dimonie, V. L.; El-Aasser, M. S.; Vanderhoff, J. W.
Polym. Mat. Sci. Eng. 1988, 58, 821.
Received: July 11, 2000
Published on the web: May 25, 2001
FAPESP helped in meeting the publication costs of this article.