Key Engineering Materials Vol. 391 (2009) pp 45-78
online at http://www.scientific.net
© (2009) Trans Tech Publications, Switzerland
NanoStructured Sonogels
Nicolás de la Rosa-Foxa)*, Víctor Morales-Flóreza), Manuel Piñero b)
and Luis Esquivias c).
a)
Departamento de Física Materia Condensada. Facultad de Ciencias
b)
Departamento de Física Aplicada. CASEM.
Universidad de Cádiz. 11510 P uerto Real (Cádiz). SPAIN.
c)
Departamento de Física de la Materia Condensada. Facultad de Físicas.
Instituto de Ciencias de los Materiales de Sevilla (CSIC).
Universidad de Sevilla. 41012 Sevilla. SPAIN
a
b
c
d
nicolas.rosafox@uca.es, victor.morales@uca.es, manolo.piniero@uca.es, luisesquivias@us.es
Keywords: Solution, Sol, Gel, Precipitate, Xerogel, Aerogel.
Abstract
Acoustic cavitation effects in sol- gel liquid processing permits to obtain nanostructured materials,
with size-dependent properties. The so-called “hot spots” produce very high temperatures and
pressures which act as nanoreactors. Ultrasounds force the dissolution and the reaction stars. The
products (alcohol, water and silanol) help to continue the dissolution, being catalyst content,
temperature bath and alkyl group length dependent. Popular choices used in the preparation of
silica-based gels are tetramethoxysilane (TMOS), Si(OCH3 )4 , and tetraethoxysilane (TEOS),
Si(OC2 H5 )4 . The resultant “sonogels” are denser gels with finer and homogeneous porosity than
those of classic ones. They have a high surface/volume ratio and are built by small particles (1 nm
radius) and a high cross- linked network with low –OH surface coverage radicals. In this way a
cluster model is presented based on randomly-packed spheres in several hierarchical levels that
represent the real sonoaerogel. Organic modified silicates (ORMOSIL) were obtained by
supercritical drying in ethanol of the corresponding alcogel producing a hybrid organic/inorganic
aerogel. The new material takes the advantages of the organic polymers as flexibility, low density,
toughness and formability whereas the inorganic part contributes with surface hardness, modulus
strength, transparency and high refractive index. The sonocatalytic method has proven to be
adequate to prepare silica matrices for fine and uniform dispersion of CdS and PbS quantum dots
(QDs), which show exciton quantum confinement. We present results of characterization of these
materials, such as nitrogen physisorption, small angle X-ray/neutrons scattering, electron
microscopy, uniaxial compression and nanoindentation. Finally these materials find application as
biomaterials for tissue engineering and for CO 2 sequestration by means the carbonation reaction.
1.Sol-gel nanomaterials processing.
In many fields of the natural sciences the boundary between two subjects is difficult to establish,
and the frontier between extended solids and nanostructured materials is no exception. The
difference between the macro and nano scales is both quantitative and qualitative. It is well
established that a material can be considered “nano” when the grain size scale extends from the
molecular level (0.1 nm) up to around 100 nm; that is, from a physico-chemical point of view, it
extends from the scale of the chemical bond, where its behaviour is governed by quantum
mechanics, to the mesoscopic systems (100 nm), at which specific models still are needed, before
passing to the macroscopic level.
*
To who m all correspondence should be addressed: nicolas.rosafox@uca.es
All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of the
publisher: Trans Tech Publications Ltd, Switzerland, www.ttp.net. (ID: 150.214.245.142-07/07/08,18:04:03)
46
Progress in Sol-Gel Production
Nanostructured materials take advantage of properties that derive from confinement effects,
larger interface-to-volume ratios, links between wave phenomena and structural features, and where
new atomic and macromolecular structures can be generated. The criteria currently used to identify
a nanostructured material is the existence of a size-dependent physico-chemical property in such
materials, but it is not safe to extrapolate from this up or down in scale. From current work on the
development of nanostructured materials the question arises of how the various different properties
change as the microstructural scale descends to nanometre dimensions.
A nanostructured material is thus recognized because it is formed from the assemb ly of either
nanoparticles, nanocrystals or nanolayers, which can be adapted to specific requirements in a
dispersion or a coating, functional nanostructures, compacted materials, biological systems, etc.
Manufacturing processes have been designed to take advantage of the following effects:
a) New physical, chemical or biological properties derived from the grain size scaling.
b) New phenomena due to the reduced grain size, where interaction length scales become
comparable to the size of the particle, crystal or grain microstructure.
c) The generation of new atomic, molecular and macromolecular structures in materials.
d) The significant increase of the degree of complexity and speed of processes in particulate
systems.
All of these characteristics or effects give rise to novel mechanical, optical, electric, magnetic,
thermal, chemical and biological properties, but only a small proportion of these properties have
been fully identified and quantified [1].
In this chapter we describe one of the strategies for building nanomaterials, known as the sol-gel
method. This process is widely used for the preparation of multicomponent nanostructured
materials by the hydrolysis and polycondensation of metal alkoxides. From a structural point of
view a gel can be considered, in the simplest picture, as a giant molecule which has been formed as
a consequence of growth by condensation of polymers or aggregation of particles, but no la tent heat
is evolved. This giant molecule extends to the walls of the vessel that contains it, and the coherent
solid 3-D network inside the fluid phase is known as a gel.
The primary purpose of this method is to produce very specific composite materials, and among
the most important of these are nanomaterials. The main processes involved in this method begin at
the molecular level; hence it is possible to act on the structure at the very initial stages. Applications
of the method include the tailoring of mechanical, optical, electronic, and chemical properties for
sensors, bio-composites, structural and other materials, which are used in all kinds of manufacturing
industry [2].
The new features in coatings and films include ferroelectric, magnetic and photo chromic
properties, electrical conductors and materials with large third-order optical nonlinear coefficients
[3, 4]. Another active field is the use of organic- inorganic hybrid materials (OIHM) developed from
the pioneering work of H. Schmidt in 1985 [5]. He called some of these materials ORMOSILs
(ORganic MOdified SILicate) and others ORMOCERs (ORganic MOdified CERamic). The
mechanical properties are dependent on the organic content and rubber- like elasticity can replace
brittleness [6]. Advantage is taken of the inorganic component that governs the degree of hardness,
brittleness and transparency whereas the density, porosity and thermal stability are governed by the
organic polymer. Using the classification given by Mackenzie [7], we are now in the second
generation of gels, including new aerogels with improved mechanical properties, termed
“aerormosils” [8]. These include hard coatings for plastic ophthalmic lenses in which the
transparency is maintained by the small size of the colloidal particles, and the Ormosil coating
reduces the permeation by water, oxygen and numerous other molecules [9]. Considerable progress
has been made in the field of nano-building blocks by the co-polymerization of different oxide
species [10, 11].
Key Engineering Materials Vol. 391
47
2. Ultrasonically-induced nanomaterials.
Some of the most important recent developments in sonochemistry have been its application in
the synthesis and modification of both organic and inorganic materials. High-power ultrasound can
induce a wide range of chemical and physical conseq uences. The chemical effects of ultrasound can
be grouped in three main areas: homogeneous sonochemistry of liquids, heterogeneous
sonochemistry of liquid- liquid or liquid-solid systems, and sonocatalysis (which overlaps the other
two areas). Applications of ultrasound to materials chemistry are found in all of these categories
[12]. Physical effects of high- intensity ultrasound, which often have chemical consequences,
include enhanced mass transport, emulsification, bulk thermal heating, and a variety of e ffects on
solids.
In all of these cases the process is conducted by means of acoustic cavitation effects. When the
sound wave propagates in a medium it travels via a series of compression and rarefaction stages
induced in the molecules through which it passes. At sufficiently high intensity the rarefaction cycle
may exceed the attractive forces of the molecules and cavitation bubbles will form. Such generated
bubbles grow by a process known as rectified diffusion i.e. small amounts of vapour (or gas) from
the medium enter the bubble during its expansion stage but this vapour is not fully expelled during
the compression step. The bubbles continue to grow over a few cycles until reaching an equilibrium
size for the ultrasound frequency applied. It is the fate of these bubbles to collapse in succeeding
compression cycles, at which point they act as a “hot spot” [13], as shown in Figure 1.
150
Bubble Radius (m)
Collapse
100
Growth
Cooling
50
"hot spot"
Cavitation seeds
0
0
100
200
300
400
500
600
Time (ms)
Figure 1: Ultrasonic cavitation can produce bubbles in liquids. Along this process the bubbles oscillate induced by the
pressure cycle of the sound wave. Eventually they undergo a violent collapse, which generates hot -spots of extremely
high pressures (1000 At m) and temperatures (5000K), in lifet imes of less than 2 s.
Thus the chemical consequences of the use of high-power ultrasound do not arise from an
interaction of acoustic waves and matter at a molecular or atomic level. Instead, in liquids submitted
to high-power ultrasound, acoustic cavitation (a process which takes place in three steps: the
formation, growth and collapse of bubbles) provides the primary mechanism for sonochemical
effects [14, 15]. During cavitation, bubble collapse produces intense local heating, high pressures,
and very short lifetimes; these transient, short-lived localized hot spots provide enough high-energy
to generate the chemical reactions. As described in detail by Suslick et al. [16], these hot spots can
reach temperatures as high as 5000K, pressures of about 1000 atm, and heating and cooling rates
above 1010 K/s, all in a time scale of the order of picoseconds [17].
48
Progress in Sol-Gel Production
Thus, cavitation hot spots can be used as chemical nanoreactors because they serve as a means of
concentrating the diffuse energy of sound into a unique set of conditions to produce unusual
materials from dissolved (and generally volatile) precursors.
3. Sonogels
As already stated, the sol- gel process is based on the possibility of forming the disordered oxide
network by various reactions in a liquid phase, followed by solvent elimination. One of the
approaches to forming the initial gel is the “alkoxide method”. The precursors used in this case are
metal alkoxides M(OR)n , where M is a metal and R an alkyl group. These compounds react with
water, in the presence of any acid or basic catalyst, and undergo hydrolysis and polycondensation
reactions, which lead to the formation of the metal oxide. The polymerization leads first to a
colloidal dispersion of particles, the aggregation of which will form the gel when the percolation
level is accomplished, forming a coherent solid structure immersed in a fluid medium.
To obtain nanostructured metal oxides using the action of high power ultrasounds to promote
the chemical reactions, the “sonogel” method can be used. This procedure avoids the need to use a
common solvent for the alkoxide and water mixture [18, 19]. The acoustic wave effect on the liquid
mixture acts in two ways: firstly, the alkoxide/water mixture is emulsified and diffusion between
the two phases takes place; secondly, chemical reactions take place in the small bubbles (hot spots)
generated by cavitation. The two acoustic processes take place inside the bubbles dispersed
throughout the liquid, and the rapid compression/decompression of the ultrasonic wave (20 kHz)
causes the bubble radius to oscillate around some equilibrium size and finally to collapse.
When high-power ultrasound is used (100 W output power), extremely hot spots are formed in
the liquid in its three discrete stages: nucleation, growth, and implosive collapse [20, 21]. The
extreme conditions generated on bubble collapse lead to the production of excited states, breakage
of chemical bond, and formation of free radicals. Thus, three areas can be identified in a cavitation
system. The center of the hot spot is where the primary chemistry involved in atomic and radical
recombination takes place [33]. The surrounding heterogeneous liquid is relatively unaffected,
although active molecular species generated inside the bubble may diffuse out and react with
reagents in the liquid. At this point, the interfacial region has very large gradients of temperature,
pressure, surface tension, electrical field, shear stresses and rapid motion of molecules, leading to
efficient mixing [22, 23]. It is interesting to note that the presence of foreign pa rticles enhances the
cavitation because they act as nucleation sites.
The temperature inside the hot-spot increases suddenly as a consequence of the adiabatic bubble
collapse, and has been estimated at several thousand degrees Kelvin [24, 25]. The alkoxide-water
mixture involved in sonogels can be classified, under Luche‟s rule [26], as a heterogeneous system
in which ionic reactions are stimulated inside the cavitation bubbles; therefore the product of the
reaction will be the same as it would be in the absence of ultrasound (as by classic or conventional
processes). The application of ultrasound to the synthesis of organosilane and organosiloxane
precursors has been studied by Boudjouk [27] and Price et al. [28].
3.1. Pure Silica Sonogels
This section accounts for the systematic studies carried out on silicon alkoxide+water mixtures.
The aim of such work was to elucidate the differences between sonogels and gels obtained by the
conventional (classic) method. As the first parameter involved is the absence of alcohol, one can
expect to obtain denser sono-sols due to a lower dilution and, in consequence, a fast polymerization
process.
Key Engineering Materials Vol. 391
49
Tarasevich first reported observation of the TEOS/water mixture reaction under the action of
high-power ultrasound in 1984 [29]. He reported that the reaction shows an intense increase in
temperature with the release of alcoholic vapours. He obtained a homogeneous and transparent
solution in a few minutes. Later, Zarzycki‟s group in France and Esquivias‟ group in Spain began
intensive collaborative work to establish the consequence of ultrasound influence on the textural
characteristics of “sonogels” [30, 31, 32] and derived materials [27, 28]. The combined effect of
sonocatalysis and organic- inorganic hybrids has opened new routes to nanostructured materials
[15].
Pure silica sonogels have also been studied in depth by Donatti, Vollet and Ibañez Ruiz, in
Brazil. They report a systematic study on the hydrolysis of TEOS/water and TMOS/water liquid
mixtures under the action of ultrasounds, proposing a dissolution reaction model based on dynamic
calorimetric measurements [33, 34, 35]. The results are thermally activated hydrolysis with
apparent activation energy of 36 kJ/mol. From this, the action of ultrasound can be explained as
follows: ultrasound forces the dissolution and the reaction begins; the products (alcohol, water and
silanol) help to continue the dissolution, with the reaction being dependent on catalyst content, bath
temperature and alkyl group length [36, 37].
3.1.1. Sonocatalysis.
The corresponding mixture of the silicon alkoxide and acidified water (pH < 1) was placed in a
double-volume beaker. No reaction was observed when a neutral or basic catalyst was used. A twophase system, as shown schematically in Figure 2, could be discerned in the beaker. At this point
the tip of the ultrasonic device was immersed some millimetres into the liquid. The ultrasonic waves
(20 kHz, 60-100 W) are then produced by an electrostrictive device equipped with a stainless steel
horn terminated with a titanium tip
Popular choices used in the preparation of silica-based gels are tetramethoxysilane, Si(OCH3 )4 , and
tetraethoxysilane, Si(OC 2 H5 )4 , known as TMOS and TEOS, respectively. Since these compounds
and water are immiscible, a common solvent (usually methyl or ethyl alcohol) has to be added to
obtain an initially- homogeneous liquid. However, no solvent is necessary when high-power
ultrasound is applied to the liquid mixture. The chemical reactions involved can be summarised as
follows:
50
Progress in Sol-Gel Production
PDMS
TEOS
H 2O
Inmiscible starting
precursors
Ultrasounds
exposure
hydrolysis Si (OR) 4 4 H 2 O
)))us
Transparent hybrid
organic-inorganic sonosol
Si (OH ) 4 4 ROH
polycondensation a) Si (OH ) 4 Si (OR) 4
)))us 2SiO2 4 ROH
b) Si (OH ) 4 Si (OH ) 4
)))us 2SiO2 4 H 2 O
Figure 2: Step procedure for h igh-power u ltrasounds treatment of an alko xide -PDMS-water mixture.
where R is –CH3 in the case of TMOS, and –CH2 -CH3 for TEOS. Hydrolysis and
polycondensation can be accelerated or slowed down by employing an appropriate acid catalyst.
These reactions proceed simultaneously and are generally incomplete. Depending on the amount of
water present, hydrolysis may go to completion or stop while the metal is only partially hydrolyzed.
On the other hand, the polymerization can be made reversible because silica solubility increases for
small particles or small negative curvature (spheres or “necks”). This process gives a strengthened
network because the small particles dissolve and are deposited on the larger ones and in the “necks”
between two larger particles.
When several different cations are used to form mixed-oxide networks, an initial complexing step
may be required. When the alkoxide precursors have different hydrolysis rates (e.g. Al, Zr, Y or Ti,
compared with Si), prehydrolysis of the alkoxysilane is preferred [27]. After a complex sequence of
polymerization, sol formation and gelation, a microporous gel with high surface area is formed,
constituted by small particles; its formula is approximated by:
(MO)x (M’O)x’(OH)y (OR)z
The radicals -OH and -OR account for reaction by-products that can, reasonably easily, result in
a system that yields a complex three-dimensional oxide -M-O-M‟-O-M- polymer which preconfigures the network of corresponding oxide glasses.
Key Engineering Materials Vol. 391
51
In the case of copolymerization between (organic) PDMS and (inorganic) TEOS, the reaction
proceeds as follows, in order to form the hybrid organic- inorganic network:
The n value determines the polymer chain length and leads to the desired the physical properties
by the interconnecting of silica particles through the polymer chain cross- links.
Me
Me
2 Si(OH)4
+ HO
OH
HO
Si
OH
O
Si
O
Si
Me
Me
Me
Me
Si
Me
O
OH
n
OH
Si
Me
O
n
Si
OH + 2H2 O
OH
Continuing with the effects of ultrasound, the ultrasonic energy delivered to the solution can be
estimated by the temperature increase of a fixed volume (V) of water during the sonic treatment,
and consequently can be calculated by the relationship:
dQ
dT
W
mc
dt
dt
The mass of water and its specific heat are m and c respectively. Then the energy delivered by the
ultrasound will be:
W t
Us
K t ( J cm 3 )
V
where t is the sonic treatment time in minutes and K is a constant characteristic of the apparatus.
Many factors can affect these K values, such as room temperature, thermostatic bath temperature,
catalyst content, molar ratios of water/alkoxide or solvent/alkoxide, beaker diameter and volume,
depth of horn tip immersion in the liquid solution and tip diameter, among others.
Once the liquid mixture reacts and a homogeneous solution is observed, the energy dose is taken
as the threshold of the hydrolysis reaction. Also there is an energy dose at which the liquid gels “in
situ”; that is, gelation takes place in the beaker during the ultrasonic treatment. At this point, a
transparent soft solid gel, with the tip hole on top, can be observed. Between these two limits one
can tune the gelation at any particular dose within the energy range. As a reference, some of these
energy doses, Us (min) and Us (max), are shown in Table 1 for different precursors and contents. In
this way, if the minimum dose is related to the hydrolysis reaction, as can be seen in Table 1, the
expected decrease of the hydrolysis rate due to the dilution is compensated by the action of
ultrasound. At the other limit, the maximum dose that can be re lated to the polycondensation rate
produces greater reactivity of the TMOS. However, this effect is reduced by the presence of the
PDMS organic polymer. This fact could be a consequence of the different functionality of the
precursors, f=4 for TEOS and TMOS and only f=2 for PDMS.
Once the sonosol has been exposed to the corresponding ultrasonic energy dose, the liquid
sonosol is kept in a hermetic container and allowed to gel at the chosen temperature. Since the
52
Progress in Sol-Gel Production
sonosol is a low- viscosity liquid, it can be cast in a mould with a selected shape. In all cases, for the
silicon alkoxides, a homogeneous and transparent solution is obtained. This property is due to the
fine dispersion of the colloidal particles formed, which begin to aggregate to produce the so-called
nanostructured sonogel.
Table 1. Th reshold and limiting of the ultrasonic energy dose.
3.1.2. Sonogel Gelation
The point at which the sol passes from a viscous fluid to an elastic solid is taken as the gelation
point. At this point the viscosity increases abruptly by several orders of magnitude. The gelation
time is taken as the time interval between the end of the ultrasonic treatment and the abovementioned transition. This transition can be estimated visually when the solution surface is no
longer horizontal when the container is tilted.
Figure 3 shows the results of the gelation time tG as a function of Us for a molar ratio of
H2 O/TEOS=4, at several mixture temperatures. In fact, the temperature, as well as the increase in
the ultrasonic energy dose, both activate the gelation process. As can be observed, two regimes are
present, separated by a level of Us = 600 J cm-3 . This point indicates the rapid increase in the
polycondensation rate. As a reference for classic gels, the gelation time is of the order of several
days for similar compositions, and is dependent on solvent content.
From the experimental values of Figure 3 it is possible to establish the thermal behaviour of the
gelation process. The plot of tG vs. 1/T is depicted in the inset of Figure 3 for different ultrasonic
energy doses. Assuming an Arrhenius behaviour, the corresponding linear fit gives an activation
energy of 50–70 kJ mol-1 TEOS, which is of the same order of magnitude as the results reported by
Tiller et al. [38], 45 kJ mol-1 TEOS. The pH dependence of the activation energy, that is, the
activation energy obtained for sonogels, agrees with the results reported by Coudurier et al. [39], of
61 kJ mol-1 for the condensation process, and 63 kJ mol-1 for the aggregation process. This indicates
that the polycondensation in pure sonogels is achieved mostly in the early stages, producing a more
reticulated structure than in classic processes.
Key Engineering Materials Vol. 391
53
US
Figure 3: Evolution of the gelation time fo r water/TEOS = 4 pure sonogels as a function of the ultrasonic energy dose
evaluated at different temperatures. Inset evolution of the gelation time as a function of 1/T for d ifferent ultrasonic
energy doses, fro m top to the bottom [32].
By using a simple model of the polycondensation ratio (the prod uct of a second order
consecutive reaction [40, 41]), the concentration of polycondensed species results in a sigmoidallike growth. The shape for every particular case depends not only on the standard parameters for
gel preparation (pH, water ratio, precursor, etc.) but also on the ultrasound power of supplied by the
device and the time applied, this is to say energy. Figure 4 are plot of the polycondensation ratio
evolution c(t) during the whole process for three different energies furnished. During insonation the
concentration of polycondensed species follows an evolution
c(t) Ultrasound
assisted
reactions
c(t)
Silent
reactions
tG1
Ultrasound
assisted
reactions
Silent
reactions
c(t) Ultrasound
assisted
reactions
tG2
cc
cc
t
cc
t
Silent
reactions
tG3 = 0
t
Figure 4: Polycondensation ratio evolution c(t) during the whole process for three different ultrasound energies
supplied.
Similar results are obtained for a sol of 90% wt TEOS + 10% wt PDMS and a molar ratio
H2 O/TEOS = 2, in which the two regimes can also be observed. However in this case hybrid
materials are separated by a level of Us = 450 J cm-3 , accounted for by the influence of the PDMS
organic polymer. This effect may be due to the chain breaking as a consequence of the ultrasound
cavitation, preventing the formation of rings. In fact, in the first regime, pure sonogels are obtained
twice as fast, and in the second, three times as fast as the hybrid gels. Since the reactions in the
PDMS/TEOS hybrid system are more complex than those in pure silicon alkoxides, the activation
energy is difficult to calculate. Mackenzie and Hoshino report a value of 60 kJ mol-1 for a 20% wt
PDMS Ormosil [42], which is similar to that of pure silica sonogels, revealing the relatively minor
influence of the temperature in this process, compared with that of the ultrasound applied.
54
Progress in Sol-Gel Production
3.1.3 Sonogel drying
Two methods are used to dry the wet sonogel:
1) A process of slow evaporation or the use of a c hemical additive (DCCA), which give a
very shrunk solid called a sono-xerogel, in which the syneresis and permeability collapse the gel
microstructure.
2) Venting off a supercritical solvent, which produces a solid known as a sono-aerogel, in
which the original microstructure of the gel is preserved.
There are several strategies for preventing theifferential stresses between adjacent pores that
provoke fracture of the solid if the tensile strength of the material is exceeded : one is to add a
chemical additive (DCCA) to the precursor liquid mixture, before gelling, to control the drying
[43].
The additive most commonly used as a DCCA is the protic solvent formamide (HCONH 2 ) [44];
the bonding of hydrogen to hydroxyl ions reduces the catalytic activity and influences both the
hydrolysis and condensation reactions. The influence of formamide as the DCCA is a consequence
of its low vapour pressure and surface tension. The reduction in the capillary pressure is
accomplished by the formation of a film on the pore wall, reducing the contact angle and the solvent
evaporates very slowly, providing a plasticizing effect. The success of this method can be attributed
to the coarsening of the microstructure and the strengthening of the gel network. Other solvents that
are effective in this application include dimethyl formamide (DMF) [45], oxalic acid and glycerol
[46], among others.
The other strategy for maintaining the original microstructure of the gel as it exists at the
gelation point is by supercritical drying. When the critical point of one liquid is surpassed, there are
no distinctions between liquid and vapour phases and their densities become equal (as a
supercritical fluid); then capillary pressure stops and the permeability decreases. To achieve this,
the wet gel is placed in an autoclave and the temperature and pressure are raised up to the critical
point of the corresponding solvent, taking care not to cross the liquid-vapour boundary; then the
solvent is vented isothermally. The critical points of various solvents are shown in Table 2. The gel
obtained is a solid network with the pores filled by air, termed an aerogel.
Table 2. Crit ical points of some common solvents
The experimental process is crucial for maintaining the solid in one monolithic piece, so the
ramp or gradient applied to raise the temperature must be as slow as possible (< 1ºC/min) because
of the differences in the thermal expansion coefficients of the liquid and solid. This difference gives
rise to pressure gradients and can cause the solid gel to crack. It is also common practice to apply
the extra pressure of an inert gas from the beginning, in order to avoid crossing the liquid-vapour
boundary. In the case of a volatile solvent, one disadvantage is the high temperature necessary
(Table 2); the use of CO 2 permits the process to be performed at near ambient temperature. This
process was first applied by T. Woignier [47] and Tewari et al. [48]. Supercritical drying needs the
pore liquid (alcohol+water) to be replaced by liquid CO 2 ; a complete solvent exchange is necessary
because capillary compression is produced by the immiscible boundary between water and CO 2 .
Key Engineering Materials Vol. 391
55
The main application found this method is in the preparation of organic- inorganic hybrid aerogels
because it prevents degradation of the organic polymer by temperature.
Another approach is to freeze the pore liquid and sublime the resulting solid under vacuum,
which is termed freezing-drying [49]; though this is widely used in the preparation of foods, it does
not permit the preparation of monolithic pieces.
3.2. NanoStructure of dry silica sonogels.
The special characteristics of sonogels after drying is that present a particulate structure 50,
contrary to gels obtained by hydrolysis of metallorganic compounds under acid catalys t without
applying ultrasound, which are filamentous. We have shown that can be represented of a hierarchic
arrangement of agglomerated of elementary particles of 1- 2 nm size forming agglomerates 4- 6
nm size and, in some cases, aggregates of these aggregates level [51, 52].
Sonogels, in the form of xerogels and aerogels, are examples of solids with both microporous
and mesoporous structure, respectively. For our samples these features can be seen in Figure 8; that
is, a type I isotherm for the xerogel sample and type IV isotherm for the aerogel. There is adsorption
at low pressure in both cases, but not further in the case of the xerogel, and capillary condensation
with hysteresis in the desorption branch in the case of the aerogel. In line with the previously
reported results, the xerogel is an entanglement of elementary particles of 2.8 nm radius forming a
microporous network of 1.1 nm pore radius, as indicated in the t-plot in the inset of Figure 8. The
aerogel, however, is formed by particles of 1.7 nm radius forming a mesoporous network of 2.3 nm
pore radius. The pore size distribution of these samples shows a narrow peak in the micro- and the
mesopore regions, respectively. The results are explained by the differences between the collapsed
microstructure of the xerogel and the original nanostructure sonogel in the case of the aerogel.
800
3
Vads (cm /g)
200
3 -1
Vads (cm g )
600
1500
150
1000
100
500
400
50
0
4
200
6
8
10
12
14
16
18
0
20
t (Å)
0
0,0
0,2
0,4
0,6
0,8
P/P0
Figure 8: N2 -physisorption isotherms from xerogel (type I-squares) and aerogel (type IV-ciscles) made of silica,
full symbols is the adsorption branch and open symbols to the desorption one. Inset shows the t-plot of these gels and
the continous line corresponds to a nonporous silica sample.
3.3. Cluster Model: random-packing of spheres.
Given the above results, some attempts have been made to depict such hierarchies on several
levels, using models constructed with the Monte-Carlo technique [59, 60, 61]. The approach is
based on comparing the pore size distribution of sonogels with that of a random close-packed hard
sphere model studied by Bernal and Mason [53], Scott [54] and Finney [55], and developed for this
application by Zarzycki [56] and Rodríguez-Ortega and Esquivias [59, 60, 57]. Essentially, the
structural approach consists in building models of the solid phase of a gel by depicting its structure
56
Progress in Sol-Gel Production
as a collection of packed spherical particles. Then, we created a catalogue of pore size distributions
[60] according to different conditions of particle coordination number and compaction according to
the features obtained from its pore space.
The pore size distribution (PSD) of the models are represented as a function of the variable K = r/R,
where r is the radius of the largest inscribed sphere in the interstice and R the radius of the particles.
Finally, pore distributions can be compared to the experimental meas urement. A logarithmic scale
for K is used to make easier to fit the experimental data, by simply sliding it along the K axis until
the position considered to give the best fit is reached. Given the maximum, the fit allows the
particle size of a mono-disperse system to be calculated from the maxima of the pore size
distributions. From the distribution of each of these contributions to the experimental distribution,
successive sizes and local densities of hierarchic distribution may be deduced [66. Data on the pore
volumes associated with different hierarchical levels, size of aggregates, the local density of the i-th
aggregation level, and packing of the successive levels can be obtained. Some of these models and
their pore size distributions (PSD) are shown in Fig. 9.
Figure 9: Mesopore fractions of SG2(sono-aerogel TEOS) and STM S (sono-aerogel TMOS) fro m Hg porosimetry
fitted with specific built models (continuous line). The insets correspond to the N2 physisorption which fit with the
models. Bi-d imensional representations of such models (SG2 in the middle and STM S in the right) are also included
[52].
It is generally accepted that the Cluster-Cluster Aggregation regimes (DLCA – DiffusionLimited Cluster Aggregation, RLCA – Reaction- Limited Cluster Aggregation) describe quite well
the typical structures obtained via sol-gel. One of the goals pursued most by researchers is to
reproduce the formation and growth processes of aerogels, using the RLCA or DLCA algorithms or
some modifications of these [58, 59]. Scherer and col. [60] used structures generated with modified
DLCA algorithms, characterising them by their fractal dimension, to achieve the power law
exponent, and they have presented several models to explain the relationship between structure a nd
mechanical properties [61, 62]. Since then, Woignier and col. have introduced a new technique for
characterising these porous systems [63, 64], and conclude that pore size distribution and hydroxyl
content are relevant for understanding the mechanical p roperties of these materials [65]. In a
previous study, Woignier and Phallipou proposed one approach starting from a cubic structural
Key Engineering Materials Vol. 391
57
model [66] and for a rigid assembly of cohesive spheres [67]. The Cluster Model that is introduced
here has also been applied as an initial approach to the study of mechanical properties [68].
The diagram shown as Figure 10 explains the process of construction. The algorithm works as
follows: first we place one elementary sphere in the centre of our system. Then we randomly p lace
as many other spheres as are needed to cover fully the surface of the first one; this produces the first
random shell. Every sphere has to be in contact with at least one other; that is, the sphere centres
must be at a previously defined distance from each other. One can build as many shells of
randomly-placed spheres as are required. This shell then constitutes the basic aggregate of the first
hierarchical level, and its diameter is measured. This diameter will be taken as the diameter of a
secondary sphere.
Figure 10: a) Diagram of the Cluster Model algorith m. b) Sketch of a Cluster Model, corresponding to the simulat ion
box cropped fro m system 2 of table 5 [70].
The next level in the hierarchy is constructed in the same way, treating the ba sic aggregate as if
it were an elementary sphere to construct the second level with secondary spheres. After
constructing this new aggregate, each secondary sphere forms a new basic aggregate, to obtain a
two- level hierarchically-ordered assembly of randomly-packed spheres. This process can be
repeated as many times as required. Typical values of our models are 1000 to 60,000 particles
organised in 2 to 4 shells of randomly-packed spheres and 2 or 3 hierarchical levels; their contact
distance L usually is found in the interval 0.85d < L < 1.0d, d being the particle diameter.
A sketch of a cluster model can be seen in Figure 10-b rendered using the free software
POVRay [69]. The system illustrated corresponds to the simulation box cropped from model #2,
and has around 3000 particles [70]. We have built several cluster models that represent the
microstructure of various real systems. Table 5 gives the characterization parameters of some real
systems and their corresponding models. Systems #1 and #2 correspond to two different silica
aerogels prepared from TEOS, and were characterized by the analysis of the N 2
adsorption/desorption isotherms [71]. These results confirm the good agreement between the
Cluster Model and the actual nanostructured sonogel.
58
Progress in Sol-Gel Production
System 1
(pure silica aerogel)
Apparent density: 0.83 g/cm 3
Specific surface: 387-407 m2/g
Specific porous volume: 0.73-0.74 cm3/g
Model 1
Apparent density: 0.80 g/cm 3
Specific surface: 384 m 2/g
Specific porous volume: 0.72 cm 3/g
System 2
(pure silica aerogel)
Elemental Sphere radius: 1.2 nm
First aggregate radius: 4.5 nm
Specific surface: 640 m 2/g
Model 2
Elemental Sphere radius: 1.1 nm
First aggregate radius: 4.5 nm
Specific surface: 612 m 2/g
Table 5: Structural parameters of several real systems and the g eometric and structural parameters of their
corresponding counterpart cluster models.
4. Nanocomposites from sonogels.
Nanocomposites are materials in which the constituents are mixed on a nanometer scale to give
properties that are superior to conventional microscale composites. They can be synthesized by a
variety of techniques including the sol- gel process and, more specifically, by using ultrasound. This
produces cavitation in the liquid sol, which in general leads to the matrix phase, as an additional
parameter for controlling the mixing at the nanometer length scale with the secondary phase.
Considerable research has been devoted to this type of material since 1989, covering different
phases, designs, morphologies and technological uses, to produce ceramic [72, 73] and
optoelectronic [74, 75] materials and polymers [76]. Hybrid aerogel nanocomposites made from
sonogels, with the incorporation of a second and active phase, have found applications as bioactive
materials [77, 78] and for CO 2 sequestration [79, 80].
4.1. Nanocrystals embedded in a sonogel matrix.
Adopting the sonosol liquid stage at the early stages of the process also favours the formation of
nanocrystals with uniformity in composition, shape, size, internal structure and surface che mistry,
which are essential characteristics for designing and controlling the behaviour of materials. In this
field, materials containing semiconductor nanocrystallites immersed in a dielectric matrix have
attracted great interest because they display third-order optical non- linearities and dramatic changes
in the optical absorption spectra with respect to the conventional bulk semiconductor [81]. A II-VI
semiconductor such as CdS has been studied in detail because it precipitates easily in a silica gel
matrix to give good optical transmission. The sonocatalytic method has been proven suitable for
preparing silica matrices for fine and uniform dispersion of CdS quantum dots (QDs) [82]. The
silica sono- xerogels were prepared by ultrasound-assisted hydrolysis of a mixture of
tetramethoxysilane (TMOS):acid water:Formamide, in the molar ratio 1:10:3, using an ultrasound
dose of 64 J·cm-3 . Different amounts of Cd(NO 3 )2 (1, 3 and 10 % weight related to the total silica
content) were added under mechanical agitation to the sono-solution before gelation. Once the
resulting gels had been left to age at room temperature, H2 S gas was diffused by thermal
decomposition of thioacetamide (TAA). Small CdS crystals are then produced by precipitation
inside the silica matrix.
The quantum confinement behaviour is manifested by the characteristic blue-shift in the UV-Vis
spectrum (Figure 11-a left axis) in relation to the bulk optical absorption band of CdS with a
bandgap energy of 2.53 eV (490 nm wavelength). This behaviour is consistent with the Éfros-Éfros
model [3] which gives the absorption threshold energy E dependent on the inverse square of
particle mean radius R as follows:
Key Engineering Materials Vol. 391
E Eg
59
2 2
2R2
5
3
6x10
CdS1
CdS3
CdS10
CdO
5
5x10
5
4x10
2
5
-d
3x10
5
2x10
1
Photoluminiscence (cps)
where ħ is the Planck constant, is the electron effective reduced mass and Eg is the bandgap
energy for the bulk crystal. These energy values can be obtained from the position of the optical
absorption band of the different samples.
5
1x10
0
0
300
400
500
600
700
800
900
(nm)
Figure 11: Optical density (left axis) of the outlined samples together the CdO one included as transparent reference
(bottom). Photoluminescence (PL) signal (right axis) of the same samples [82bis].
Figure 11 also includes the photoluminescence (PL) yield (right axis) in which two bands can be
observed. The higher-energy band is related to the intrinsic recombination mechanism. There may
also be weak transitions for dots with RaB=3.2 nm in CdS (aB being the bulk exciton Bohr radius),
because kinetic energy terms are dominant over Coulomb effects, and this produces the observed
line broadening. On the other hand, the lower-energy band is attributed to the recombination of
trapped carriers. A red-shift of the PL emission relative to the absorption band can also be observed,
which is crystal size-dependent: the red-shift increases as the particle size decreases. The origin of
this effect seems to be in a distortion of the crystal lattice after the transformation of the CdO
(cubic) into CdS (hexagonal) at 150ºC, which provokes compressive strains. This redshift/distortion influences the intrinsic band due to the sulfur ion vacancies acting as potential hole
traps. The observable differences in the PL spectra also inform about the crystal surface traps; the
CdS3 sample seems to have lower trap states than the CdS1. However, the CdS10 sample shows an
overlapping of the two bands indicating a significant distortion of the lattice. In the light of these
size-dependent effects (blue- and red-shifts), the lack of correlation between crystal size and CdS
content must be attributed to the slight difference of the silica network pore structure which affects
diffusion of the H2 S gas.
The PL yields were measured by using the third harmonic (3.49 eV) of a pulsed Nd:YAG laser
(6 ns pulse at 10 Hz repetition rate). Using a cylindrical lens, the laser beam is focused on the edge
of the sample to form a narrow rectangular strip (50 m to 2 mm long and 20 m wide) [82, 83].
The amplified luminescence (AL) at right angles to the direction of the strip is measured for
different strip lengths. In all cases the PL yield measurements were taken at atmospheric pressure in
air at room temperature. Use of the variable strip length (VSL) method is reported in [84]; in this
60
Progress in Sol-Gel Production
method the PL yield along the focus axis is related to the optical gain by the relationship:
I
I AL SP e gL 1 ,
g
where IAL and I SP are the amplified and spontaneous emission, respectively. The net gain coefficient
is g and L is the stripe length. The PL yield increases in a superlinear fashion for the intrinsic band,
accounting for a net gain coefficient as can be seen in Figure 12. The gain is spectrally broad with a
steeper decrease on the high-energy side and a long tail stretching to lower energies; in all cases the
optical gain reveals the stimulated emission from the CdS nanocrystals as QDs.
CdS1
CdS3
-1
gain coeff. (cm )
100
50
0
400
450
500
550
600
650
700
(nm)
Figure 12: Optical gain spectra for the filtered intrinsic band (Schott BG-14) of the outlined samples.
The TEM micrograph in Figure 13 confirms the fine and homogeneous distribution of the CdS
nanocrystals immersed in the porous silica matrix, with an average crystal size of 4.6 nm, as
indicated by their size distribution in the inset of Figure 13.
Figure 13: TEM micrograph of CdS/SiO2 co mposite corresponding to a 10 wt % Cd S doped SiO2 gel. The
corresponding particle size distribution is shown in the inset [74].
Key Engineering Materials Vol. 391
61
The electronic and optical properties of IV-VI semiconductors such as PbS have also been
topics for extensive research in the last three decades. Their narrow band-gap permits strongly
quantum-confined excitons (electron- hole pairs) to be created when microcrystallites are smaller
than the bulk exciton Bohr radius (aB=18 nm for PbS) [4,85, 86]. This is the case for PbS
crystallites presenting a particle size smaller than 18 nm, with a threshold of the optical absorption
band at 0.41 eV for the bulk material. Many approaches have been explored for the formation of
small PbS nanocrystals and their incorporation into solid and transparent matrices [87, 88, 89], and
one approach is to use the sol- gel process combined with colloidal chemistry [90, 91, 92]. This
technique prevents agglomeration and allows control of the particle size and size distribution;
although the small precipitates produced are unstable, surface-capping methods [93] may help to
overcome this drawback.
Gel matrices and films have been used as hosts for semiconductor quantum dots since more than
ten years ago, mainly because low temperatures are required to create the network. However, they
present a high porosity which makes them especially active for humid condensation and oxidation,
thus favouring subsequent crystal growth. In spite of these inconveniences, this topic arouses
interest and recently several studies have been published [94, 95, 96, 97, 98] . Oxidation could be
limited by sealing pores and thus reducing the spec ific surface area. One of the methods for
achieving this is by using organic polymers in the gel formation; this allows mechanically- improved
silica gel matrices to be synthesised via copolymerisation with silicon precursors, named ormosils.
These OIHM present properties that make them especially attractive for optical applications [99,
100].
Using this procedure PbS quantum dots stabilised with surface capping agents (SCA),
incorporated into rigid transparent SiO 2 sono-ormosils, have been successfully synthesized for nonlinear optical applications. Table 6 gives the textural parameters evaluated from N 2 isotherms.
Sample
SB ET
(m2 ·g -1 )
Vp
(cm-3 ·g -1 )
* ρa
(g·cm-3 )
**ρ
Geom. (g·cm-3 )
Undoped
Matrix
616
0.3237
1.28
1.29 ± 0.05
677
0.4830
1.07
1.03 ± 0.05
PbS
542
0.4676
1.09
1.08 ± 0.05
Content
414
0.2101
1.56
1.55 ± 0.05
* 1/pa = Vp + 1/ps
** parallelepiped casting
Table 6.- Specific surface (SBET), Pore volume (Vp ), and apparent density (ρa) calculated from N2 physisorption data.
The last column shows the apparent density (ρ) values evaluated from geometrical measurement.
A decrease in the specific surface area with an increase in the PbS content is observed. This
behaviour could be due to differences in pore sizes as a consequence of the steric effects derived
from the non-bonded 3- mercaptopropyl groups, since they are electrostatically opposed to the silica
network. This assumption is confirmed by the good agreement existing between the apparent
densities calculated from the porous volume and from the geometrical dimensions. Bulk densities
(s ), were calculated taking into account the corresponding percentages by weight in the composite
(SiO 2 )1-x-(PbS)x , by using 2.2 gcm-3 for vitreous silica and 7.5 gcm-3 for PbS. Finally, the average
pore size values in the last column corroborate the fine porosity of the host silica matrix.
An HRTEM micrograph from PbS/SiO 2 sample is shown as Figure 14-a. Spherical and/or cubic
morphology of particles, and lattice planes in some cases, can be clearly seen. A mean diameter of
around 10.5 nm with a standard deviation of 1.2 nm was obtained from the distribution plotted in
62
Progress in Sol-Gel Production
Figure 14-b. In such a situation, the SCA molecules passivate the Pb surface sites on the subcritical
PbS crystals and prevent their oxidation. Surplus uncoordinated S sites act as potential surface traps
which produce important steric effects on the fine silica network. The final size of the PbS particle
is then controlled by adjusting the relative concentrations of SC A and sulphide [96, 101]. Moreover,
the ormosil matrix surrounding the PbS crystals prevents their subsequent growth and oxidation.
Figure 14: (a) HRTEM micrograph of the PbS/SiO2 sample (b) Corresponding particle size d iameter d istribution [76].
In the structural analysis an additional measurement was made by means of Small Angle
Neutron Scattering (SANS), at room temperature, using a neutron wavelength of 6.02 Å, at the
Berlin Neutron Scattering Center (HMI) in Berlin. The corrected SANS intensity c urves are shown
in Fig. 20 [102, 103]. In all cases, two regions can be discerned, separated by the well-defined
crossover point around q=0.1 nm-1. For high q- values a wide plateau is apparent, which is
characteristic of fine well-shaped scatters homogeneously distributed, typical of a material with fine
porosity. In contrast, the increased signal at low q- values is characteristic of larger polydisperse
particles, as could be produced by the polymer cross- links or/and the PbS nanocrystals (6.5 nm
size). The scattering from the undoped matrix is also included, inset of Fig. 20. In the absence of
crystals, a smooth broad peak develops at q = 0.5 nm-1 , which accounts for the short-range order of
the homogeneous pore distribution.
It is well-known that crystals in gels grow by a diffusion-controlled process. In such a situation,
the concentration gradient around the crystal depletes a surrounding volume larger than the crystal
in which no other crystals are present [104]. In accordance with this understanding, the
experimental intensities were fitted by the two-correlation model proposed by Debye and Bueche
[105, 106] where the scattering cross-section behaves as:
q2 a 22
A1
d
(q)
A 2 exp
d
4
1 q2 a12 2
(1)
where the correlation lengths a1 and a2 represent the medium-range and short-range fluctuations of
the scattering length density, respectively. The parameters A1 and A2 are related to the
corresponding volume fraction of each phase. Eq.(1) can be considered as the form factor P(q) of
the heterogeneities, then the total scattering cross-section becomes I(q)=P(q)S(q), and S(q)
being the particle number density and the structure factor, respectively. The structure factor S(q)
was evaluated assuming the Percus-Yevick hard-sphere model, and using the expression of
Ashcroft and Lekner [107]. This interpretation of the SANS patterns indicates the existence of a
depleted region around each PbS crystal, where the growth of another crystal is inhibited [111].
Key Engineering Materials Vol. 391
63
-1
-1
d/d (cm str )
10
1
-1
d/d (cm str )
10
-1
undoped matrix
0,1q(nm-11)
x3
1
x1.7
x1
PbS content
0,1
-1
q(nm )
1
Figure 15. SANS scattered intensities fro m increasing PbS content samples. The solid lines correspond to the nonlinear leat-square fit fro m I(q). Inset shows the undoped matrix samp le. Sample -detector distance was set at 1, 4 and 16
m with a λ = 0.6 n m as neutron wavelength. Scattering vector modulus is q = (4 π sin θ/2)/ λ, θ being the scattering
angle. So me contributions of the incoherent scattering comong fro m the ormosil matrix can be noticed at the end of the
curves
(q >2 n m-1 ), this fact mask probably the Porod behavior in this q-reg ion (I α q -4 ) [111]
The structure factor accounts for the interparticle interference terms which could be produced
either by crystals or pores. The results of such a fitting are shown in Table 7 and the fitted curves
from I(q) are drawn in Figure 15. The undoped matrix does not fit (inset Figure 15) this twofold
correlated structure, as can be deduced from the noisy signal at low-q. Thus the observed increase
of intensity must be due to the local heterogeneities produced by the polymer cross- links between
chains [108] with the silica network at the boundary of the SCA molecules surrounding the PbS
nanocrystals. From Table 7, it can be stated that the crossover point at q=0.1 nm-1 resolves the
twofold structure for the doped samples. Thus, A1 and a1 grow with the PbS content. These values
are surprisingly high compared with those of A2 and a2 , in spite of the higher scattering length
density of the matrix/pores (short-range) relative to that of the PbS/matrix (medium-range). These
results indicate a dot density higher than those for QDs in Schott filters [111] due to the smaller
depleted region. We attribute the underestimation of a 2 values with regard to the pore size
calculated from physisorption (see Table 6), to the incoherent scattering contribution for q>1 nm-1 ,
in the absence of which the “knee” of the fitted curves at q=2 nm-1 would shift to lower values, thus
indicating larger pores.
64
Progress in Sol-Gel Production
PbS-ormosil
A1 (cm-1)
a 1(nm)
TEM (nm )
PbS
content
9.8
81.1
104.2
22.5
35.6
44.0
6.5±0.5
10.5±0.5
13.5±0.5
depleted
region (nm)
8
12
15
A2 (cm-1)
0.52
0.49
0.56
a2(nm)
0.50
0.38
0.85
Table 7.- Structural parameters fro m the fitted experimental intensities by using the two-correlation function to the
Debye-Bueche model accord ing to Eq. (14).
These results are promising to obtain strong quantum confinement (R<a B ) of the PbS quantum
dots from which large non- linear optical behavior may be expected.
4.2. Hybrid organic/inorganic sonogels.
Organic modified silicates (ORMOSIL) can be obtained by supercritical drying in ethanol of the
corresponding alcogel, producing a hybrid organic/inorganic aerogel. The new material obtained
takes advantage of the properties of organic polymers, including flexibility, low density, toughness
and formability, whereas the inorganic component contributes properties such as surface hardness,
modulus strength, transparency and high refractive index.
These hybrid aerogels can be classified as Type C Ormosils, according to the Mackenzie
classification; that is, the organic and the inorganic parts are chemically bonded, via covalent or
iono-covalent bonds. The stronger nature of the covalent bond gives considerable improvement of
the mechanical properties of the composites. One of the best methods for obtaining organicinorganic hybrid materials (OIHM) by this approach is the combined reaction of a silanolterminated polymer (PDMS), as the organic component, with a silicon alkoxide (namely, TEOS or
TMOS) as the inorganic component. In the silanol-terminated polymers, the non-bridging oxygen
Si-CH3 groups are dominant, depending on the average polymer molecular weight. The end chain
groups –OH or –OR radicals link the polymer with the silica network via the copolymerisation
between PDMS and TEOS, as stated in paragraph 3.1.1.
Type C ormosils comprising TEOS and PDMS can be also promoted by high power ultrasound,
when they are known as “sono-ormosils”; these are denser and have enhanced mechanical
properties. As a comparison, in Table 8, the Vickers hardness value of some representative
materials is shown for comparison with this new sono-ormosil. This new material falls between the
softest glasses and the hardest transparent plastics. With higher PDMS content the ormosil behaves
as a more rubbery material.
Key Engineering Materials Vol. 391
65
N2
PDMS10
pure silica
0
200
400
600
T (ºC)
800
O2
DTG (ºC/min)
DTG (ºC/min)
This proposed route for obtaining these materials incorporates an organic phase in the inorganic
precursor sol, in combination with the assistance of high-power ultrasound [34, 109, 110,111].
Given that in the sol- gel process little or no heating is required, organic molecules with low thermal
stability can be incorporated into an inorganic ceramic or glassy host. Thus the “sonogel” route is
another approach to modification of the sol- gel process. Ormosils can also be excellent matrices for
non- linear optical materials due to their high transparency, inertness, mechanical strength and ease
of preparation [112]. Similarly semiconductor quantum dots have been dispersed in ormosils [157]
and dye lasers are enhanced [113].
A new route is now being exploited in this field of OIHMs. The process consists of drying the
wet sono-ormosil at the supercritical conditions of pressure and temperature of the added liquid, to
obtain a sono-ormosil aerogel, also known as an aer-ormosil. Aerogels have been successfully
obtained from a TEOS/PDMS sono-ormosil. Results seem to confirm the hybrid characteristics of
these aerogels; that is, the organic groups are retained in the sample that can still be considered as
an OIHM. We have studied the pyrolysis process in producing these aerogels from sonogels (550 J
cm-3 ); Figure 16 shows the DTG results for the sample aerogel in air and in inert atmospheres. As
can be seen, the organics decompose at 230ºC (DTG in air), and then there is a further gradual
weight loss between 500-600ºC, explained by the remaining hydroxyl groups producing water as
by-product. In contrast, the pyrolysis under an inert atmosphere (DTG in inert gas) shows a
different behaviour; up to 400ºC the aerogel is therma lly stable, and between 400-600ºC
carbonisation produces a redistribution between Si-O and Si-C bonds, but no destruction of the
organics groups, as reported by Zang and Pantano [114]. The transition to an inorganic material
occurs between 700-800ºC, with decomposition of the methyl groups. Above 1000ºC Babonneau et
al. [115] report, from 29 Si MAS-NMR experiments, the presence of an oxycarbide phase and free
carbon in a DMDES/TEOS ormosil.
1000
PDMS10
pure silica
100
200
300
400
500
T (ºC)
Figure 16: DTG diferential weight loss curves for a sono-aerogel made of TEOS (green line) and a hybrid
TEOS/PDMS (red line), in N2 and O2 atmospheres.
66
Progress in Sol-Gel Production
These new solids also present variation in several physical properties; as an example, the methyl
group on the polymer chain makes these aerogels hydrophobic as is apparent from the photograph
of Figure 17, from which a contact angle greater than 120º can be estimated.
Figure 17: Water drop in the top of a hybrid T EOS/PDMS aerogel p iece, contact
angle125º [115bis].
4.2.1. Mechanical behaviour.
Mechanical test were performed in an AG-I Autograph from Shimadzu. For the experiments an
uniaxial compression device was used, equipped with a load cell of 5kN. In all the experiments the
strain rate was fixed at 0.5 mm/s at room temperature. For this mechanical test samples were cast as
cylinders of 20 mm length and 10 mm diameter.
30
1
2
3
1,84
1,83
1,82
log [Er(t) (MPa)]
(MPa)
20
10
1,94
1,81
1,80
1,79
1,78
1,77
0
0,0
0,1
0,2
(%)
0,3
0,4
1,76
1,93
1
2
3
log [t(s)]
Figure 18: Stress-strain curves obtained on pure silica aerogel (circles) and on organic/inorganic aerogel (t riangles),
last point correspond to the fracture. Inset shows the log-log plot of the relaxation modulus for the same samples
[115b is].
The results of the stress-strain experiments are shown in Figure 18; the incorporation of organic
polymer chains into the inorganic structure of silica gel dramatically changes the mechanical
behaviour. First, the PDMS aerogel (PDMS 40 % by wt.: bottom curve) presents an elastomeric
behaviour indicating a softer solid network than that of the pure silica gel (top curve); in other
words, the polymeric chains are acting as springs. As the stress increases, the PDMS aerogel begins
to deform less and less, indicating that the solid network is becoming stiffer; then the action of the
polymer stops (at 25% strain) and the inorganic silica network begins to act, and ends with a
similar slope (elastic modulus) as the pure silica aerogel (top curve). The upward curvature
Key Engineering Materials Vol. 391
67
indicates a continuous increase in its elastic modulus, which is characteristic of elastomers that
loose their stiffness as the polymer chains deform. First the polymer chain deforms like a spring by
reducing the Si-O-Si bond angle, giving large strain with low stress. In a second state, the chain
penetrates into the micropores of the inorganic silica nodules behaving like a dashpot to dampen the
small strains induced as the stress increases, and finally the failure of the network provokes the
fracture. Thus the solid presents first an elastic behaviour and then, in a second step, becomes like a
viscous liquid, giving a total mechanical behaviour of a visco-elastic solid. A similar elastomer
signature can be observed in the other samples as the polymer content is increased. The
experimental values from the stress-strain curves are given in the Table 9.
PDMS
(wt. %)
0
10
20
40
50
Max.
Load
(N)
391.10
444.60
1159.90
1508.20
1252.50
Compressive
strength
(Nmm-2)
8.00
9.60
23.10
28.60
23.70
Strain at
fracture
(%)
8.40
13.40
34.70
36.80
42.90
Young´s Modulus
50%
(Nmm-2)
113.80
105.70
55.10
41.50
39.40
Young´sModulus
90%
(Nmm-2)
101.00
89.90
65.60
78.75
55.92
Table 9. Mechanical parameters results fro m uniaxial co mp ression test.
As can be seen, this polymeric aerogel exhibits a complex behaviour on deformation, flowing
like a liquid or responding elastically like a solid, depending on the time scale of the dynamics of
the molecules. This mechanical duality is called visco-elasticity [116, 117] and can be characterized
by the relaxation modulus, the stress time evolution at fixed strain. Such relaxation behaviour is
illustrated in the inset of Figure 18 for a TEOS/PDMS aerogel (PDMS 40 % by wt); under small
deformation (=15%), a biexponential decay reveals two mechanisms in the relaxation process: a
fast mechanism corresponding to the elastic polymer chains (like springs) and another slow
mechanism due to the stiff porous silica matrix (like a dashpot).
The technique known as nanoindentation enables the mechanics of nanomaterials to be probed
[118] and permits direct measurements of physical property of heterogeneous materials with close
spatial resolution. This technique has traditionally been used to characterize elasto-plastic hard
materials, although many groups have used the technique on soft tissues, such as demineralized
dentin [119] and vascular tissues [120]. The problem of suitable analytical techniques comes from
the influence of interfacial adhesive forces, which can affect the indentation contact area and has
not been validated for very soft materials, with an elastic modulus below 5 MPa. Some authors have
demonstrated the validity of nanoindentation in measuring the elastic modulus of pure PDMS with
different degrees of crosslinking [121].
In this context several studies have concentrated on the sol-gel method and specifically on aerogels
[122]. This technique seems to be very useful for the study of such complex structures as hybrid
organic/inorganic aerogels, which are composed of an inorganic phase covalently bonded to an
organic polymer chain.
In a standard load-depth curve, the initial slope on the unloading branch (stiffness-S) is related
to the reduced elastic modulus (Er) by the following relationship:
2
dP
S
Er A
dh m ax
68
Progress in Sol-Gel Production
in which A is the area of the imprint on the sample, in this case a Berkovich pyramidal tip
(A=24.5h2 , h being the depth). The slope S is calculated after least square fitting by the power law
function proposed by Oliver and Pharr [123, 124] in the form: P h m , where contains
geometric constants, the elastic modulus of both sample and indenter, and Poisson‟s ratio, and m is
the power law exponent related to the geometry of the indenter (for a cone m=2). The hardness is
calculated simply by H P
A
and the elastic recovery parameter as ERP
h m ax h f
, hf being the
h m ax
non-recovered depth of the indenter inside the aerogel sample.
Figure 19 shows a typical load-depth hysteresis curve, in which the dramatic difference between
the two samples (pure silica and PDMS hybrid aerogel) is apparent; as a first indication, the load
necessary to provoke a displacement of 1 m is 33% greater for the pure silica sample compared
with the hybrid aerogel with a polymer content of 10 % by wt. Thus the influence of the polymer
chains is mainly on the stiffness of the silica; there is a decrease in the hardness from 226 MPa for
the pure silica to 24 MPa for the hybrid aerogel sample, and consequently a better elastic recovery
indicated by a greater depth interval (hf, hmax ).
2,0
Young's modulus (MPa)
200
Load (mN)
1,5
loading
1,0
unloading
0,5
180
E soft
E stiff
E medium
E uniaxial
E average
160
140
120
100
80
60
40
20
0
10
20
30
40
50
PDMS content (wt%)
0,0
0
500
1000
1500
2000
2500
3000
Depth (nm)
Figure 19: Depth-load hysteresis curves for the pure silica aerogel (red) and hybrid TEOS/PDMS 10 wt.% samp les.
Maximu m load 1.5 mN with a penetration rate of 5 μN/s. Inset average Young‟s modulus as a function of the PDMS
content for different sites and macroscopic (black rho mb) [115bis]
The inset of Figure 19 shows a plot of the Young‟s modulus versus polymer content. The
macroscopic Young‟s modulus from uniaxial compression (solid rhombus) is also included; this
value comes close to that of the softer sites indicating that the polymer phase controls the
mechanical behaviour at the macroscopic level. The silica clusters, bonded by the spring- like
elastomeric polymer chains, act like a dashpot to give the final elastic modulus of a soft material.
Key Engineering Materials Vol. 391
69
Grid maps were recorded of 10X10 indentations, each 20 m apart. The total load-depth curves
show a fan feature indicating stiffest sites against softest ones. The former are dominated by the
inorganic silica clusters and the latter by the organic polymer. This picture is apparent in the map of
the elastic recovery parameter depicted in Figure 20a, in which elastic sites (white) correspond to
those with low reduced modulus, and plastic sites (black) correspond to the hard regions of the
aerogel.
1450
a)
b)
2300
1400
h(nm)
1350
h(nm)
2250
1300
2200
1250
Pure silica aerogel - Load 1mN
Hybrid aerogel 50% wt PDMS - Load 0.8mN
2150
1200
0
1000
2000
3000
t (s)
Figure 20: 10x10 grid maps for the PDMS 50 wt.% sample: a) for the elastic recovery parameter, each indentation is
separated 20 μm and b) creep curves compared with the pure silica aerogel sample [115bis].
In order to study the mechanical properties of these organic- inorganic aerogels in greater depth,
we have performed creep tests on hybrid aerogels, as the time dependence of the strain at fixed
stress [125]. Assays of this type have been used widely on different kinds of materials and, more
specifically, on polymers [126]. From these tests the creep curve shows an instantaneous elastic
deformation followed by a transitory state of retarded deformation, known as primary creep. During
this period, the material can be understood to be undergoing a reinforcement process by
deformation [127]. Then, the system reaches the stage of secondary creep, a steady state where
equilibrium is found within the different mechanisms of strain and recovery. Creep curves from
pure silica and hybrid PDMS aerogels are shown in Figure 20b. They depict the typical saturation
shape with an instant deformation, and then the primary regime of retarded strain and secondary
creep of steady-state linear regime. It can be seen that the depth of the retarded strain is much
greater in the hybrid than in the pure silica aerogel.
4.2.2. Structural aspects.
SANS measurements were carried out at room temperature on the V4 workstation at the Berlin
Neutron Scattering Center of the HMI (Berlin). A neutron wavelength of 0.602 nm, at three sampledetector distances 1, 4 and 16 m, was used to cover a q-range from 0.036 to 3.6 nm-1 . Data were
corrected using the software package available at HMI [128].
Figure 21 shows the SANS experimental intensities of hybrid aerogels with different polymer
content. Several features can be discerned; first, an increase of the intensity is apparent towards the
low q-side. Then wide plateaus appear with Guinier radii ranging from 5 to 8 nm as the polymer
content is increased from 10 to 50 % by wt, as displayed in the inset of Figure 21. The plateaus are
followed by power law decay, with a slope getting steeper in line with the polymer content, up to a
maximum value of -2.5. The final parts of the curves are dominated by the incoherent scattering due
to the 1 H atoms of the organic polymer in the methyl radicals.
70
Progress in Sol-Gel Production
10%
Onuki+RC fit of ajustes_B
20%
Onuki+RC fit of ajustes_D
30%
Onuki+RC fit of ajustes_F
40%
Onuki+RC fit of ajustes_H
50%
Onuki+RC fit of ajustes_J
10
1
0,1
Gyration radius (nm)
-1
d(q)/d (cm )
factor
10% 1
20% 0.9
30% 0.6
40% 0.3
50% 0.2
8
Rg(Gui)
Rg(RC)
7
6
5
10
20
30
40
50
PDMS wt.%
0,01
-1
q0,1
(nm )
1
Figure 21: Plot of experimental SANS intensities versus the transfer mo mentum q, for different PDMS aerogel hybrid
samples (Curves are shifted). Continuous line is the least-squares fit to the Gaussian random coil model. The
corresponding Gu inier radius (red square) and random coil gyration rad ius (green square) as a function of the polymer
content is included in the inset [128b is]
In Figure 21 a first shoulder at low-q (0.04 nm-1 ) is produced by the long range characteristic length
governed by the “frozen-in” elastic constraints, which act through topologically-connected polymer
cross- links. A shorter range correlation length at q=0.3 nm-1 in the plateau corresponds to the size of
the so called “blobs”, which can be considered to be composed of a nucleus of the silica cluster
surrounded by the polymer distributed as a random coil (RC); the average size of the “blob” can be
identified with the Guinier radius. This supposition is supported by the agreement between the
results of non- linear fitting of the experimental curves and a mixed Gaussian random coil form
factor (continuous line in Figure 21) [129, 130]. The wide Guinier region indicates the
homogeneous distribution of these “blobs” as scatterers. As the polymer content is increased, the
“blobs” grow as a consequence of the entanglement of the polymer chains.
The aggregates of the entangled polymer chains surrounding the silica clusters, the “blobs”,
form mass fractal units, their dimension D=2.5 being characteristic of a cross- linked random coil
structure. Although the aggregate does not strictly fulfil the conditions of self-similarity on different
length scales to be considered a fractal object, it can be viewed as an object that does not fill the
space totally. Moreover, this result suggests a kinetic growth mechanism based on the diffusion
limited aggregation (DLA) of the monomer-cluster that has a fractal dimension of 2.5.
4.3. Silica ae rogel nanocomposites.
Finally we present here some applications of the nanostructured sonogels described, in which a
second component added in the liquid will act as the active phase.
4.3.1 Bioactivity
Hybrid organic/inorganic materials (OIHMs) are being used for implants since they are tolerated
by the human organism, which creates a fibrous tissue when they are embedded in the body.
However, they do not become bonded to the bone unless they are bioactive. In such case, a layer of
Key Engineering Materials Vol. 391
71
hydroxycarbonate of apatite (HCA) grows and wraps around the material when it is immersed in
blood plasma. An HCA layer is also formed when bioactive materials are soaked in solutions
mimicking the features of plasma. To perform fully as an implant, these materials need to present
mechanical behaviour very similar to that of human bone.
Recently, several mixtures obtained by ultrasonic agitation of colloidal silica with a sol solution
containing tetraethoxysilane (TEOS) have been used to form crack- free monoliths. We have used
the combination of colloid-polymer as precursor of the inorganic part of the OIHM with the aim of
tailoring the porosity to control the density and mechanical strength in the range of those of human
bone. Finally, in vitro bioactivity of this material has been promoted by adding colloidal silica
particles to the initial sol. In ormosils, it has been shown that Ca ++ cations, together with unreacted
silanols, form the bioactive hydroxyapatite layer in a simulated body fluid [131].
Si
0d
O
3d
Ca
0.04%
10 m
Si
O
Ca
P
0.14% 0.14%
10 m
1d
Si
7d
Si
O
10 m
P
Ca
0.12% 0.10%
O
10 m
P
Ca
0.75% 1.40%
Figure 22: SEM micrographs of aerogel sample and the correspondent EDX spectra before and after be soaked for
different times (0, 1, 3 and 7 days) in SBF. The ato mic weight percentages of Ca and P in samples after each immersion
time are also included [77]
In general terms, those materials coated with a layer of amorphous calcium phosphate that
crystallises into HCA after being soaked in SBF (Simulated Body Fluid) are categorized as
bioactive. The biologically active HCA layer facilitates interaction between the material and
biological entities, and it has been proposed as a first phase in the sequence of reactions that result
in the creation of a mechanically strong bond between the bioactive materials and the living tissues
[132]. To confirm the formation of calcium phosphate on the hybrids after the in vitro assays, the
variations in their surfaces with the time of soaking in SBF were also studied by SEM and EDX.
Figure 22 presents the SEM images and the correspo nding EDX spectra of an aerogel composite
after 0, 1, 3 and 7 days of immersion in SBF. After 1 day, the SEM micrograph shows no
significant differences with respect to the initial sample (t=0). However, after 3 days in SBF, the
surface appears cracked, with deep channels forming. After 7 days, some kind of material appears
to have formed inside the channels. The corresponding EDX spectra show that the proportion of Ca
and P on the material surface increases slowly during the first 3 days of assay, up to 0.14 mol-% for
the two elements. However, the EDX spectrum of the sample immersed for 7 days shows the
presence of significant amounts of P (0.75 mol-%) and Ca (1.4 mol- %). This result demonstrates the
formation of calcium phosphate on the surface of the hybrid [84, 85].
In another approach using synthetic wollastonite powders as the active phase and modifying the
inorganic TEOS gel matrix with MTES (methyl triethoxysilane), the bioactive behaviour was
improved. The formation of a layer of HA crystals on the aerogel surface is apparent in the
micrograph of Figure 23; roughly spherical shaped particles, formed by fine lamellar crystals, can
be seen. A quantitative yield analysis on the EDX spectrum indicates a Ca/P ratio of 1.7, almost the
same composition as that of the HA.
72
Progress in Sol-Gel Production
Figure 23: SEM micrograph and EDX pattern of SiO2 /CaO co mposite aerogel after soaking in SBF during 25 days
[132b is].
4.3.2. CO2 sequestration
Among the gases contributing to the „„green house effect‟‟, the influence of CO 2 has been
estimated at 40–50% of the total. Currently, the policy proposed for dealing with this problem is the
reduction of gas emissions from power generation processes, transport and especiallycontaminating industries (such as cement factories). Any viable strategy with this objective requires
technologies that incorporate CO 2 elimination processes, i.e., separation, capture (sequestration),
disposal or storage and, finally, elimination. Everything must be carried out at costs that are feasible
for the gas-emitting industries, organisations and countries to bear.
Figure 24: a) Diffract ion profiles of the TEOS/wo llastonite composite before and after exposure to 30 min of CO2 flow
b) SEM of the byproducts of the same composite after exposure to 30 min of CO2 flow. There can be observed the
individual monocrystals (rho mbohedric-like) o f carbonates together with unreacted silica gel [80]
It is known that some natural silicates form carbonate when in contact with CO 2 . In particular,
wollastonite (CaSiO 3 ) reacts to produce calcite and silica. This is a natural reaction that has
operated over millions of years to reduce the CO 2 in the Earth‟s atmosphere. Wu et al. (2001) [133]
and Tai et al. (2006) [134] have published results on the rate of conversion of wollasto nite (CaSiO 3 )
into calcite (CaCO 3 ) for different experimental protocols. These values must be interpreted to
determine the efficiency of CO 2 fixation by the minerals. Thus, starting from a powdered
wollastonite sample in a reactor at atmospheric pressure a nd room temperature, Wu et al. obtained a
Key Engineering Materials Vol. 391
73
conversion rate of 14% of wollastonite into calcite after 22 days. Later, a comparative study has
been carried out [135] analysing different types of sample and experimental conditions, and has
given more encouraging results. In Figure 24 the carbonation reaction is illustrated by the X-Ray
diffraction peak and by the SEM micrograph. In this case, the wollastonite is almost entirely
dissolved and the calcite peak can be observed standing out from the other peaks. The reaction
efficiency can be estimated as relatively high: their analysis reveals that almost 85% of the
wollastonite is dissolved. The SEM micrograph shows the by-products of the aerogel composite
after exposure to 30 min of CO 2 flow, where the individual monocrystals (rhombohedric- like) of
carbonates together with unreacted silica gel can be observed. These results indicate advantages in
terms of fast reaction time and efficiency, which should open up new uses for composites of this
type [86, 87].
74
Progress in Sol-Gel Production
ACKNOWLEDGEMENTS
The authors thank to the Spanish Ministry of Science and Technology for financial support under
project MAT2005-1583. The authors are members of the PAI TEP-115 Research Group of the Junta
de Andalucía (Spain).
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
M.C. Rocco, J. Nanoparticle Res., 1, 1 (1999).
“Sol-gel technology for thin films, fibers, preforms, electronics and specialty shapes” Ed. L.
Klein, Noyes publications, New Jersey, (1988).
Y.H. Xu, CH. Cheng and J.D. Mackenzie, J. Non-Cryst. Solids 176, 1 (1994).
CH. Cheng, Y.H. Xu and J.D. Mackenzie, Ferroelectrics 152, 31 (1994).
H. Schmidt, J. Non-Cryst. Solids 73, 681 (1985)
J.D. Mackenzie, Q. Huang and T. Iwamoto, J. Sol-Gel Sci. and Tech., 7,151 (1996)
J.D. Mackenzie, J. Sol-Gel Sci. and Tech. 26, 23 (2003).
S.J. Kramer, F. Rubio-Alonso and J.D. Mackenzie, in “Better Ceramics through Chemistry
VII: Organic-Inorganic Hybrid Materials”, Eds. by B.K. Coltrain, C. Sanchez, D.W.
Schaefer and G.L. Wilkes, MRS, Pittsburgh (1996) p. 295.
S. Amberg-Schwab, M. Hoffmann, H. Bader and M. Gessler, J. Sol-Gel Sci. and Tech. 13,
141 (1998).
C. Sánchez, G.J.de A.A. Soler-Illia, F. Ribot, T. Lalot, C.R. Mayer and V. Cabuil, Chem.
Mater. 13, 3061 (2001)
P. Gómez-Romero and C. Sánchez, “Functional Hybrid Materials”, Willey-VCH, Germany,
(2004).
K.S. Suslick and G.J. Price, Annu. Rev. Mater. Sci. 29, 295 (1999).
K.S. Suslick, Science 247, 1439 (1990).
Lord Rayleigh, Phil. Mag. 34, 94 (1917).
T.G. Leighton, in “The Acoustic Bubble”, London Academic (1994).
E.B. Flint and K.S. Suslick, Science 253, 1397 (1991).
K.S. Suslick, Y. Didenko, M.M. Fang, T. Hyeon, K.J. Kolbeck, W.B. McNamara III, M.M.
Mdleleni and M. Wong, Phil. Trans. R. Soc. Lond. A 357, 335 (1999).
E. Blanco, L. Esquivias, R. Litran, M. Piñero, M. Ramírez-del-Solar, and N. de la Rosa-Fox,
Appl. Organomet. Chem. 13, 399 (1999).
J. Zarzycki, Heterogeneous Chem. Rev. 1, 243 (1994).
J. P. Lorimer and T. J. Mason, Chem. Soc. Rev. 16, 239 (1987).
E. A. Neppiras, Ultrasonics 22-1, 25 (1984).
J.L. Luche, in “Current TRENDS I Sonochemistry” (G.J. Price, Ed.) p.34 RSC, Cambridge
1192.
T. J. Mason, “Practical Sonochemistry.” Ellis Horwood, Chichester, 1991.
P. Kruus, Ultrasonics 21, 201 (1983).
K. Morita, Y. Hu, and J. D. Mackenzie, J. Sol–Gel Sci. Technol. 3, 109 (1994).
J. L. Luche, in “Current Trends in Sonochemistry” (G. J. Price, Ed.), p. 34. RSC,
Cambridge, 1992
P. Boudjouk, in “Science of Ceramics Chemical Processing”, L. L. Hench and D. R. Ulrich,
Eds., Wiley, New York (1986). p. 363.
Key Engineering Materials Vol. 391
[28]
[29]
[30]
[31]
75
G. J. Price, M. P. Hearn, E. Wallace, and A. M. Patel, Polymer 37, 2303 (1996).
M. Tarasevich, Ceram. Bull. 63, 500 (1984)(Abstract only).
L. Esquivias and J. Zarzycki, in “Current Topics on Non Crystalline Solids” (M. D. Baró
and N. Clavaguera, Eds.), p. 409. World Scientific, Singapore, 1986.
L. Esquivias and J. Zarzycki, in “Ultrastructure Processing of Ceramics, Glasses and
Composites” (J. D. Mackenzie and D. R. Ulrich, Eds.), p. 255. Wiley, New York, 1988.
N. de la Rosa-Fox, L. Esquivias, and J. Zarzycki, Diffusion Defect Data 53/54, 363 (1987).
D.A. Donatti and D.R. Vollet, J. Sol–Gel Sci. Technol. 4, 99 (1995).
D.A. Donatti and D.R. Vollet, J. Non-Cryst. Solids. 208, 99 (1996).
D.A. Donatti, A. Ibañez Ruiz and D.R. Vollet, Ultrasonics Sonochem. 9, 133 (2002).
D.A. Donatti, D.R. Vollet and A. Ibañez Ruiz, J. Sol-Gel Sci. Technol. 18, 5 (2000).
H.-J. Tiller, R. Gobel, and U. Hartung, J. Non-Cryst. Solids 105, 162 (1988).
H. Coudurier, R. Baudru, and J. B. Donnet, Bull. Soc. Chim. Fr. 9, 3147, 3154, 3161 (1971).
Pouxviel J C , Boilot J P, Beloeil J C . and Lallemand Y , J. Non-Cryst. Solids,1987, 89 :
[32]
[33]
[34]
[35]
[36]
[38]
[39]
[40]
345.
[41] Orcel G and Hench L , J. Non-Cryst. Solids, 1986, 79 : 177.
[42] Y. Hoshino and J. D. Mackenzie, J. Sol–Gel Sci. Technol. 5, 83 (1995).
[43] L.L. Hench and J.K. West, Chem. Rev. 90, 33 (1990)
[44] G. Orcel and L.L. Hench, J. Non-Cryst. Solids 79, 177 (1986).
[45] T. Adachi and S. Sakka, J. Mater. Sci. 22, 4407 (1987)
[46] L.L. Hench, in Science of Ceramic Chemical Processing, Eds. L.L. Hench and D.R. Ulrich,
Wiley, NY, (1986) p. 52
[47] T. Woigner, PhD Thesis, University of Montpellier, France (1984).
[48] P.H. Tewari, A.J. Hunt and K.D. Loffus, Mater. Lett., 3, 363 (1985).
[49] S. Bruin and K.Ch.A.M. Luyben in Advances in Drying, vol. 1, ed. A.S. Mujumdar,
Hemisphere, NY, (1980) p. 155.
[50] J. Rodriguez-Ortega and L. Esquivias, J. Sol-Gel Sci. Tech., 8, 117 (1997)
[51] J. Rodriguez-Ortega, PhD thesis, Universidad de Cadiz, Spain (1996)
[52] L. Esquivias, J. Rodríguez-Ortega, C. Barrera-Solano and N. de la Rosa-Fox, J. Non-Cryst.
Solids, 225,239 (1998)
[53] J.D. Bernal and J. Mason, Nature, 188, 910 (1960)
[54] G.D. Scott, Nature, 188,908 (1960)
[55] J.L. Finney, Proc. R. Soc (London) A319, 479 (1970)
[56] J. Zarzycki, J. Non-Cryst. Solids, 147&148, 176 (1992)
[57] L. Esquivias, N. de la Rosa-Fox, M. Bejarano and M. J. Mosquera (2004) Langmuir
20:3416.
[58] A. Emmerling, J. Fricke. J. Sol-Gel Sci. and Tech. 8 (1997) 781-788.
[59] M. Grzegorczyk, M. Rybaczuk, K. Maruszewski. Chaos, Solitons & Fractals 19 (2004)
1003-1011.
[60] H. Ma, J. H. Prevost, G. W. Scherer. International Journal of Solids and Structures 39
(2002) 4605-4614
[61] H. Ma, A. P. Roberts, J. H. Prevost, R. Jullien, G. W. Scherer. J. Non-Cryst. Solids 277
(2000) 127-141
[62] H. Ma, J. H. Prevost, R. Jullien, G. W. Scherer, J. Non-Cryst. Solids 285 (2001) 216-221.
[63] T. Woignier, J. Reynes, A. H. Alaoui, I. Beurroies, J. Phallipou, J. Non-Cryst. Solids 241
(1998) 45-52.
[64] J. Primera, A. Hasmy, T. Woignier, J. Sol-Gel Sci. and Tech. 26 (2003) 671-675.
76
[65]
Progress in Sol-Gel Production
T. Woignier, F. Despetis, A. Alaoui, P. Etienne, J. Phalippou, J.Sol-Gel Sci. and Tech. 19
(2000) 163-169.
[66] T. Woignier, J. Phalippou, J. Non-Cryst. Solids 100 (1988) 404-408.
[67] A. Rumpf, Chem. Ing. Tech. 30 (1958) 144.
[68] V. Morales-Florez, N. de la Rosa-Fox, M. Pin˜ero, L. Esquivias, J. Sol–Gel Sci. Technol. 35
203 (2005).
[69] Persistence Of Vision, www.povray.org. Last visit, 23/nov/2006.
[70] V. Morales-Flórez, M. Piñero, N. de la Rosa-Fox, L. Esquivias, J.A. Anta, J. Primera, J.
Non-Cryst. Solids, 354, 193-198 (2008)
[71] M.C. Barrera-Solano, N. de la Rosa-Fox and L. Esquivias. J. Non-Cryst. Solids 147/148
(1992) 194-200.
[72] Piñero M., Atik M. , Zarzycky J., J. Non-Cryst. Sol.. 147&148, 523 (1992)
[73] Piñero M., Zarzycki J., J. Sol-Gel Sci.Tech., 1, 275 (1994)
[74] M. Piñero, R. Litran, C. Fernández-Lorenzo, E. Blanco, M. Ramírez-del-Solar, N. De la
Rosa-Fox, L. Esquivias, A. Craievich, J. Zarzycki, J. Sol-Gel Sci.Tech., 2, 689 (1994)
[75] A. Craievich, N. De la Rosa-Fox, E. Blanco, M. Piñero, M. Ramírez del Solar, L. Esquivias,
Nanostruc. Mater., 5 (3) 363 (1995).
[76] R. Erce-Montilla, M. Piñero, N. de la Rosa-Fox, A. Santos, and L. Esquivias, J. Mater. Res.
16, 2572 (2001).
[77] L. Esquivias, V. Morales-Flórez, M. Piñero, N. de la Rosa- fox, J. Ramírez, J. GonzálezCalbet, A. Salinas, M. Vallet-Regí, Mater. Res. Soc. Symp. Proc. Vol.847 EE12.1.1 (2005)
[78] J. A. Toledo-Fernández, R. Mendoza-Serna, V.Morales-Flórez, N. de la Rosa-Fox, A.
Santos, M. Piñero, L. Esquivias, Bol. Soc. Exp. Ceram. V., 46 [3] 138-144 (2007)
[79] A. Santos, J. A. Toledo-Fernández, R. Mendoza-Serna, L. Gago-Duport, N. de la Rosa-Fox,
M.Piñero, L. Esquivias, Ind. Eng. Chem. Res, 46, 103-107 (2007)
[80] A. Santos, M. Ajbary, A. Kherbeche, M. Piñero, N. de la Rosa-Fox, L. Esquivias, J. Sol-Gel
Sci. Tech. DOI: 10.1007/s10971-007-1672-1 (2008)
[81] Jain, R.K., Lind, R.C., J. Opt.Soc. Am. 73, 647 (1983)
[82] J. Butty,Y.Z. Hu, N. Peyghambarian, Y.H. Kao and J.D. Mackenzie, Appl. Phys. Lett. 67,
2672 (1996).
[82bis] N. de la Rosa-Fox, M. Piñero, R. Litrán, L. Esquivias. J. Sol-Gel Science and Technology
26, 947-951, (2003)
[83] J. Butty, N. Peyghambarian, Y.H. Kao and J.D. Mackenzie, Appl. Phys. Lett. 69, 3224
(1996).
[84] K.L. Shaklee, R.E. Nahory and R.F.Leheny, J. of Lumin. 7, 284 (1973).
[85] N. Peyghambarian, S. Koch and A. Mysyrowicz, in “Introduction to Semiconductor Optics”.
(Prentice-Hall, New Jersey, 1993). Pp. 245-253.
[86] A. Lipovskii, E. Kolobkova, V. Petrikov, I.Kang, A. Olkhovets, T. Krauss, M. Thomas, J.
Silcox, F. Wise, Q. Shen, S. Kycia, Appl. Phys. Lett. 71(23), 3406,(1997).
[87] N. F. Borrelli, D. W. Smith, J. Non-Cryst. Solids., 180, 25 (1994).
[88] M. Mukherjee, A. Datta, D. Chakravorty, J. Mater. Res., 12(10), 2507, (1997)
[89] F del Monte, Y. Xu, J. D. Mckenzie, J. Sol-Gel Sci.&Tech., 17, 37 (2000)
[90] S. Gallardo, M. Gutiérrez, A. Henglein, E. Janata, Ver. Bunsenges. Phys. Chem. 93,1080,
(1989)
[91] M. Guglielmi, A. Martucci, J. Fick, G. Vitrant, J. Sol-Gel Sci.&Tech., 11, 229 (1998)
[92] A. Martucci, P. Innocenzi, J. Fick, J.D. Mackenzie, J. Non-Cryst. Solids., 244, 55, (1999)
[93] L. Spanhel, E. Arpac, and H. Schmidt, J. Non-Cryst. Solids., 147&148, 657, (1992)
[94] N.N. Parvathy, G.M. Pajonk, and A. Venkateswara, J. Cryst. Growth, 179, 249, (1997).
Key Engineering Materials Vol. 391
77
[95] N.N. Parvathy, G.M. Pajonk, and A. Venkateswara, Mat. Res. Bull., 32(4), 397, (1997)
[96] A. Martucci, M. Guglielmi, and K. Urabe, J. Sol-Gel Sci. &Tech., 11, 105 (1998).
[97] A. Martucci, J. Fick, J. Scell, G. Battaglin and M. Guglielmi, J. Appl. Phys. 86(1), 79
(1999).
[98] M. Guglielmi, A. Martucci, E. Menegazzo, G.C. Righini, S. Pelli, J. Fick, G. Vitrant, J. SolGel Sci. & Tech., 8, 1017, (1997)
[99] H. Lin, E. Bescher, J.D. Mackenzie, H.Dai, and O.M. Stafsudd, J. Mat. Sci., 27, 5523,
(1992).
[100] P. T. Guerreiro, S. Ten, N. F. Borrelli, J. Butty, G. E. Jabbour, N. Peyghambariam, Appl.
Phys. Lett. 71(12), 1595, (1997)
[101] A. Martucci, J. Fick, J. Scell, G. Battaglin and M. Guglielmi, J. Appl. Phys. 86(1), 79
(1999).
[102] M. Piñero, N. de la Rosa-Fox, R. Erce-Montilla and L. Esquivias, J. Sol-Gel Sci. Tech., 26,
527 (2003)
[103] N. de la Rosa-Fox, R. Erce-Montilla, M. Piñero and L. Esquivias, Opt. Mater., 22, 1 (2003)
[104] G.Banfi, V. Degiorgio and B. Sepeit, J. Appl. Phys. 74, 6925 (1993).
[105] P. Debye and M. Bueche, J. Appl. Phys. 20, 518 (1949).
[106] P. Debye, H.R. Anderson Jr. and H. Brumberger, J. Appl. Phys. 28, 179 (1957).
[107] N. Ashcroft, J. Lekner, Phys. Rev, 145, 83 (1966).
[108] E. Geissler, F. Horkay, A.M. Hecht and C. Rochas, Polymer 38 (1), 15 (1997).
[109] K. Morita, Y. Hu and J.D. Mackenzie, Mat. Res. Soc. Symp. Proc., Vol 271, 693 (1992).
[110] E. Blanco, L. Esquivias, R. Litran, M. Piñero, M. Ramírez-del-Solar and N. de la Rosa-Fox,
Appl. Organomet. Chem. 13, 399 (1999).
[111] N. de la Rosa-Fox, L. Esquivias and M. Piñero in Handbook Of Organic-Inorganic Hybrid
Materials and Nanocomposites, Vol. 1, Ed. H.S. Nalwa, American Scientific Publishers,
USA (2003) Ch. X (in press)
[112] J.D. Mackenzie, J. Ceram. Soc. Japan 101, 1 (1993).
[113] B. Dunn, J.D. Mackenzie, J.I. Zink and O.M. Stafsudd, SPIE Proc. 1328, 174 (1990).
[114] H. Zang and C. Pantano, J. Am. Ceram. Soc. 73, 958 (1990).
[115] F. Babonneau, L. Bois and J. Livage, J. Non-Cryst. Solids 147&148, 280 (1992).
[115bis] N. de la Rosa-Fox, V. Morales-Flórez, J.A Toledo-Fernandez, M. Piñero, R. MendozaSerna, L. Esquivias. “Journal of the European Ceramic Society” 27, 3311-3316 (2007)
[116] J.D. Ferry, “Viscoelastic properties of Polymers”, 3rd Ed. (Wiley, NY, 1980).
[117] I.M. Ward, J. Sweeney. “An introduction to the mechanical properties of solid polymers”.
(2nd Ed.) John Wiley & Sons, (2004).
[118] Schuh, C.A., Mater. Today, 2006, 9, 32-40.
[119] Marshall, G.W., Habelitz, S., Gallagher, R., Balooch, M., Balooch, G. and Marshall, S.J., J.
Dent. Res. 2000, 80, 1768.
[120] Ebenstein, D.M. and Pruitt, L.A., J. Biomed. Mater. Res. A, 2004, 69A, 222.
[121] Carrillo, F., Gupta, S., Balooch, M., Marshall, S. J., Marshall, G.W., Pruitt, L. and Puttlitz,
C.M., J. Mater. Res., 2005, 20, 2820-2830.
[122] Moner-Girona, M., Roig, A., Molins, E., Martínez, E. and Esteve, J. Appl. Phys. Lett., 1999,
75, 653-655
[123] Oliver, W. C. and Pharr, G.M., J. Mater. Res., 1992, 7, 1564-1583.
[124] Pharr, G.M., Oliver, W.C. and Brotzen, F.R., J. Mater. Res., 1992, 7, 613-617.
[125] Piñero, M., Morales-Flórez, V., de la Rosa-Fox, N. and Esquivias, L., Mechanical properties
of hybrid aerogels, Bol. Soc. Esp. Ceram. V., 2005, 44(5), 291-293.
[126] Ferry, J.D., “Viscoelastic Properties of Polymers”, Wiley, N. Y., 3rd. ed., 1980.
78
Progress in Sol-Gel Production
[127] Callister Jr., W. D. “Introducción a la Ciencia e Ingeniería de los Materiales”, Ed. Reverté,
Barcelona, 1995.
[128] U. Keiderling, Physica B, 234–236, (1997), 1111.
[128bis] N. de Rosa-Fox, V. Morales-Flórez , J. A. Toledo-Fernández M. Piñero, L. Esquivias y
U. Keiderling. J Sol-Gel Sci. Technol , 45, 245-250.(2008)
[129] Kirste, R.G. and Oberthur, R.C., in “Small Angle X-ray Scattering”, Ed. Glatter, O. and
Kratky, O., Academic Press, (1982), Chap. 12, 387-431.
[130] Pedersen, J.S., J. Appl. Cryst., 33, (2000), 637-640.
[131] M. Kamitahara, M. Kawashita, N. Miyata, T. Kokubo and T. Nakamura, J. Mat. Sci.: Mat.
In Med. 13 (2002) 1015-1020
[132] Hench L. L. J. Am. Ceram. Soc. 81 (1998) 1705.
[133] Wu JC-S, Sheen J-D, Chen S-Y, Fan Y-Ch. Ind Eng Chem Res 40, 3902 (2001)
[134] Tai CY, Chen W-R, Shih S-M AIChE J 52, 292 (2006)
[135] Zevenhoven R, Eloneva S, Teir S Catal Today 115, 73 (2006)
View publication stats