Accepted Manuscript

Title: Nanodomains within bicontinuous silicone/water

microemulsions retard TiO2 nanoparticle aggregation

Author: Marlena Whinton Michael A. Brook

PII: S0927-7757(16)30825-1
DOI: http://dx.doi.org/doi:10.1016/j.colsurfa.2016.09.075
Reference: COLSUA 21053

To appear in: Colloids and Surfaces A: Physicochem. Eng. Aspects

Received date: 15-5-2016

Revised date: 6-9-2016
Accepted date: 22-9-2016

Please cite this article as: Marlena Whinton, Michael A.Brook, Nanodomains
within bicontinuous silicone/water microemulsions retard TiO2 nanoparticle
aggregation, Colloids and Surfaces A: Physicochemical and Engineering Aspects
http://dx.doi.org/10.1016/j.colsurfa.2016.09.075

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Nanodomains within bicontinuous silicone/water microemulsions retard TiO 2 nanoparticle

aggregation

Marlena Whinton and Michael A. Brook*

Department of Chemistry and Chemical Biology, McMaster University, 1280 Main St. W.,

Hamilton, Ontario, Canada L8S 4M1

mabrook@mcmaster.ca, +1 905 525 9140 ext 23483

1
Graphical abstract

TiO2 particle growth/aggregation is retarded by the surfactant-reinforced interface (left) in

bicontinuous silicone/water microemulsions.

Highlights

 Sol gel processes at neutral pH normally lead to rapid formation of TiO2 particles
 Sol gel processes in W/O and bicontinuous silicone microemulsions are orders of
magnitude slower
 Titania particles growth and aggregation are initially restricted by the microemulsion
domains
 Bicontinuous domains are fluid, so eventually particle aggregation will occur at a rate
affected by the domain viscosity and anchoring of surfactants at the interface

Abstract

Nanoparticles of TiO2 are used in a variety of applications because of their high refractive index

and photocatalytic behaviour. We report that both the interactions between the surfactant and

titanium alkoxide and templating provided by a microemulsion significantly retard the formation

of visible TiO2 aggregates. TiO2 nanoparticles (< 30 nm) were prepared under mild, neutral

conditions within water-in-oil and bicontinuous silicone microemulsions. The particles

2
aggregated only slowly, as shown by UV absorption spectroscopy transmission electron

microscopy, and particle sizing. Retarded particle growth is ascribed primarily to constricted

domains provided by self-organization within a water/surfactant/silicone microemulsion.

Keywords: Bicontinuous microemulsion; silicone/water microemulsion; nanodomains; stabilized

TiO2 (titania) particle synthesis.

3
Introduction

Titanium dioxide is extensively used in coatings, water treatment, cosmetics, and as a pigment in

paint because of its photocatalytic properties and UV blocking capabilities [1]. Many

applications, including sunscreens, thin films, transparent coatings, and photovoltaics use titania

in the form of nano-sized particles because they can reflect UVA/UVB radiation while being

transparent to visible light and, depending on structure, can have enhanced photocatalytic

properties [2]. Amorphous nano-titania, in particular, is desirable for cosmetic and coating

applications owing to its optical transparency while exhibiting decreased photocatalytic

properties when compared to crystalline TiO2 [3].

TiO2 nanoparticles are commonly prepared by a sol-gel method where alkoxytitanates are

hydrolyzed under a variety of conditions [4]. The reaction is extremely rapid (typically <1

second) and followed by extremely fast condensation, then particle formation and

aggregation/growth; these processes are difficult to control. This high reactivity often leads to

highly polydisperse and inhomogeneous precipitates, which can be undesirable, particularly when

used in coatings or films [5].

Decreasing the rate of hydrolysis and condensation of titania precursors can be achieved, to a

degree, by altering the structure of the starting titanium alkoxide. For example, derivatives of

polyols, such as diglyceryltitanium, hydrolyze and condense to TiO2 much more slowly than

titanium isopropoxide (TTIP, Ti(O-iPr)4) [6]. Constraining TiO2 particle growth also requires

colloidal stabilization to prevent particle aggregation, which can be provided by steric or

electrostatic stabilization. For example, the use of the bidentate ligand acetylacetone at low pH

leads to transparent solutions containing 1-10 nm titania particles. The dispersions remained
4
transparent for up to a year under acidic conditions owing to the stability of surface hydroxyl

groups against further condensation [4] and electrostatic stabilization provided by cationic

species on the particle surface [7-10]. However, that stabilization was lost as the pH was

increased.

During our exploration of the use of titanium dioxide as a reinforcing agent in a previously

reported silicone microemulsion system [11] we were surprised to learn that the formation of

visible titania particles was very slow at neutral pH within the microemulsion. In this report, we

examine the factors that affect the rates of hydrolysis and condensation of alkoxytitanium

species, and show that, at neutral pH, TiO2 particle formation and aggregation is much slower

within W/O or bicontinuous microemulsions as a consequence of the microemulsion structure

itself.

Experimental Section

Materials

The surfactant Silsurf A008-UP (632 g/mol, Figure 1) was donated by Siltech Corporation

(Toronto, Canada). Titanium tetraisopropoxide (TTIP) (284.22 g/mol) was purchased from

Sigma Aldrich (Oakville, ON) and silanol-terminated polydimethylsiloxane DMS-S14 (35-45

cSt, 700-1500 g/mol) was obtained from Gelest (Morrisville, Pennsylvania).

Methods

UV Absorption Spectroscopy

UV absorption measurements were taken at different time points after the reaction was initiated

(see below) using a Cary 50 UV spectrophotometer at a scan rate of 300 nm/min. A baseline was

5
obtained using an analogous microemulsion differing only in that it did not contain a titanium

alkoxide (). Formulations were directly measured by depositing ~50 mg of microemulsion

between two quartz slides.

Transmission Electron Microscopy

Titania particles were isolated by centrifugation (Eppendorf Centrifuge 5424) at 15000 rpm for

~1.5 h after diluting the microemulsion (~0.25 g) with isopropanol (IPA) (1 mL). The supernatant

(a mixture of microemulsion and IPA) was removed using a pipette and the isolated titania was

then dispersed in a fresh aliquot of 1 mL isopropanol. The titania dispersion was deposited onto a

formvar-coated copper grid. The solvent was evaporated from the grid and images were

immediately obtained with a JEOL JEM 1200 EX TEMSCAN scanning transmission electron

microscope (JEOL, Peabody, MA, USA) operated in transmission mode. Images were obtained

after 4, 24, and 48 h, respectively, of initial microemulsion preparation and subsequent in-situ

particle formation (no particles could be isolated before 4 h).

Nanoparticle Tracking

Nanoparticle Tracking Analysis (NTA) was used to evaluate the size and distribution of titanium

dioxide nanoparticles formed in-situ from silicone microemulsions. Particle size data was

collected using a NanoSight LM10 instrument (Nanosight, Amesbury, UK), equipped with a

sample chamber and analyzed with NanoSight Software 3.1. Particle solutions were prepared by

diluting each microemulsion formulation in isopropanol to an optimized titania particle

concentration of between 107-109 particles/mL. Samples were directly injected into the

NanoSight chamber using a syringe and exposed to a monochromatic beam (green laser, 633

nm). In order to ensure a random sampling of particles for each measurement, an automated
6
syringe pump (20 L/min) was used to inject the particle solution into the chamber. Particle

movement was tracked frame-by-frame in a video of 30 s duration by the NanoSight 3.1

software, which applies the Stokes-Einstein equation to calculate the particle size based on each

particle’s individual velocity.

Microemulsion Formulations

Microemulsions were prepared using the formulations shown in

. The order of addition was important to obtain transparent microemulsions (otherwise rapid

formation of opaque dispersions of TiO2 occurred). The surfactant, Silsurf A008-UP, was first

placed in a ~7 mL (2-dram) vial, followed by the slow addition of titanium isopropoxide while

mixing with a magnetic stir bar. Silanol-terminated polydimethylsiloxane (PDMS-OH, DMS-

S14, Gelest) was added to the surfactant/TTIP solution while mixing, and led to a transparent,

yellow mixture. Water, in large molar excess, was then slowly added into the formulation drop-

wise to form a pale yellow, transparent microemulsion.

Results

The formation of large visible titania particles is usually quite rapid – essentially instantaneous –

when titanium alkoxides are exposed to even small quantities of water at neutral pH. Control

experiments demonstrated it was necessary to prepare formulations by first mixing the surfactant

Silsurf A008-UP with TTIP, followed by the addition of silicone oil (hydroxy-terminated

polydimethylsiloxane, PDMS, Figure 1). Water must be added last in order to obtain optically

transparent materials in which TiO2 particle formation and aggregation was delayed. Note that

7
mixtures also immediately became opaque upon the addition of water if either the silicone oil or

surfactant were missing from the formulation, due to the instantaneous formation of TiO2.

One possible explanation for slower TiO2 formation involves ligand exchange reactions between

TTIP and Silsurf A008-UP, to eventually give Ti(Silsurf)4, prior to hydrolysis (Figure 1). This

hypothesis was supported by the UV absorption spectra of surfactant Silsurf A008-UP/TTIP

mixtures in isopropanol that exhibited a decrease in absorbance and slight change in the

absorbance curve (Figure S6, Supporting Information), but not TiO2 formation.

Four microemulsions containing different concentrations of water were prepared, which

represented examples from all three different microemulsion structure types: oil/water (Figure

2A); bicontinuous (Figure 2B,C); and water/oil (Figure 2D). All the formulations contained a

large molar excess of water when compared to the molar quantity of Ti-OR groups present.

The microemulsion formulation containing the largest excess of water (224:1 mole ratio

[H2O]:[TTIP) represents an oil-in-water structure. Immediately after addition of water to form the

microemulsion there was, unsurprisingly, the formation of visible titania – the dispersion became

opaque – which resulted in low initial absorbance readings due to light scattering and a decrease

in ligand-metal charge transfer absorbance of the isolated Ti4+ tetrahedrally-coordinated

precursor Ti(Silsurf)4 (Figure 2A).

The outcomes with bicontinuous microemulsions were quite different. Two water-loading levels

were examined within the bicontinuous region. At water concentration levels of ~18wt% and

~13wt%, respectively, visible titania formation could only be observed after 24 hours; the

materials became translucent and then increasingly opaque slowly over time. A decrease in UV

8
absorbance at 215 nm was observed after 24 hours and continual decrease in UV absorbance

thereafter was observed over the course of 166 hours (Figure 2B,C). This absorption decrease

was accompanied by the growth over time of a small shoulder between 280-310 nm.

At lower concentrations of water (7.67wt%), representing a water-in-oil microemulsion structure

that still contains a huge molar excess when compared to TTIP (35:1 molar ratio [H2O]:[TTIP]),

no change in transparency of the microemulsion was observed after 4 days. The UV absorbance

for this low water content microemulsion decreased very gradually only after the first 24 hours

(Figure 2D).

Particle Size Distribution

Titania particle sizes and distribution were initially determined by the analysis of TEM images of

dried particles from the formulation containing ~13wt% water at 4, 24 and 48 hours, respectively,

using Image J software (see Supporting Information): the particles were too small to harvest

before 4 hours. The TEM images of the particles did not show well-defined structures, likely due

to high silicone content as shown by EDX of the particles formed in situ (Figure 3, Figure S7,

Supporting Information). Particle analysis at 4 hours revealed more small particles (<40 nm) than

at later time points. After 24 hours, fewer isolated small particles (<20 nm) were observed, and a

larger fraction of the particles were present as compact aggregates up to sizes of between 900-

2000 nm. That is, the major activity over time within the microemulsion was particle

aggregation, although particle growth did continue.

Particle concentrations and size distribution within two titania-containing microemulsion

formulations, which differed in the fraction of water added, were directly determined using

9
Nanoparticle Tracking Analysis (NTA) after dilution of the microemulsion. The W/O

microemulsion (~7wt% water) produced relatively few observable particles over the first 24

hours with the greatest concentration of particles seen at approximately 200 nm (Figure 4, see

Supporting Information). The bicontinuous microemulsion formed from ~13wt% water was

found to contain a much higher concentration of particles that grew to much larger sizes over

time, such that they began to sediment out. The particles formed were highly polydisperse, such

that the histogram at each time point was highly dependent on the particles observed at that

particular moment.

Data for the particle sizes from nanoparticle tracking are consistent with those from TEM. While

aggregation of particles during drying cannot be avoided with the TEM, the technique is able to

measure the presence of small particles (Figure 3). By contrast, NTA is better able to provide

particle size distributions, but only of particles that scatter light sufficiently for detection: it does

not provide consistent data on small (<30 nm) particles. Thus, the TEM data is more reliable for

small nanoparticles, while the NTA data is more useful for larger particles.

Discussion

Titanium dioxide has a refractive index ranging between 2.5-2.8 depending on its structure, and

amorphous v crystalline state [12]. However, titania nanoparticles do not scatter visible light and

are ideal for formulations and applications requiring UV absorption yet high optical transparency

in the visible region [13-15]. Titania monomers and nanoparticles can be characterized using UV-

Vis absorption spectroscopy. UV absorption bands are attributed to ligand-metal charge transfer

from the oxygen to titanium(IV) centre. The position of the absorption band is affected by the

geometry of the titanium centre and, eventually, the size of titanium dioxide particle.
10
Tetracoordinate monomers show absorption maxima in the region of 210-240 nm, while particle

sizes below 5 nm exhibit absorption maxima between 280-300 nm, and particles larger than 5 nm

will absorb between 350-400 nm [16].

Controlling titania particle size using a sol-gel process with alkoxytitanium derivatives is difficult

to achieve under neutral conditions due, in part, to the high reactivity of the metal center towards

nucleophilic attack, which normally leads to rapid rates of hydrolysis [17]: the nature of the

alkoxide ligand [18], the concentration of TiOR, and concentration of water in the sol-gel system

also affect hydrolysis and gel times [17, 19].

We have shown the partial or complete transesterification of O-iPr groups on TTIP with the

surfactant Silsurf A008-UP (e.g., to give Ti(Silsurf)4) occurred when the two materials were

mixed (Supporting Information). At water contents of ~18% or ~13% water, respectively (94:1 or

66:1 molar ratio [H2O]:[TTIP), the microemulsions were bicontinuous with domain sizes

between 5-10 nm [20] (Supporting Information). Hydrolysis of the TTIP/Silsurf compounds

should normally be very efficient because the titanium alkoxide species in bicontinuous

microemulsions would be readily exposed to water due to the very high interfacial area. UV

spectra show, however, that hydrolysis was significantly less efficient than in water, (Figure 2)

except in the O/W emulsion (Figure 2A). The Ti(OR)4 absorption peak (~210 nm) was initially

observed with bicontinuous or W/O microemulsions (Figure 2B-D), but decreased over 24-48

hours and additional absorption bands could been seen between 280-310 nm, indicating the

presence of octahedrally coordinated titanium dioxide [21]. The onset of opacity in the

microemulsions was longer than these changes in UV absorption. Thus, the attachment of the

11
longer alkyl chains in Ti(Silsurf)4, with their terminal hydrophobic trisiloxane residues, was only

partly responsible for the slower rate of titania formation within microemulsions.

With the exception of O/W microemulsions the key factor retarding large titania particle

formation was the presence of the microemulsion structure itself. The one W/O and two

bicontinuous microemulsions containing TTIP were initially transparent, and only slow increases

in opacity resulting from particle formation were observed over the course of 72 hours at a rate

that depended on the water content and microemulsion structure type: particles could not be

harvested before 4 hours. We propose that the reinforcement of interfaces by Silsurf A008-UP at

domain intersections within the microemulsion retards the rate of particle formation and growth

once particle sizes approach the domain sizes.

The lower water content TTIP microemulsion (35:1 molar ratio [H2O]:[TTIP]) had a water-in-oil

structure. The reaction process begins with hydrolysis that can take place at the droplet interface,

but the possibility of swollen micellar structures of partly hydrolyzed Ti(Silsurf)4 and of

sterically stabilized TiO2 particles cannot be excluded. Particle growth and aggregation may also

be inhibited by the presence of sterically stabilized TiO2 particles at the aqueous interface to form

a sort of Pickering emulsion (Figure 1H,I).

The bicontinuous microemulsions similarly showed very slow TiO2 particle growth. In this case,

the nanodomains are estimated to be 5-10 nm based on SAXS data [20] (Supporting

Information). We propose that particle growth and aggregation will be hindered once the particles

grow to the size of the domain (Figure 1G). TEM micrographs and ImageJ analysis obtained of

TiO2 formed within the bicontinuous structure type revealed that at the four-hour time point there

was a much greater number of small particles (20 nm) when compared to later time points:
12
particles could not be harvested prior to this. Only after 24 hours did the smaller particles begin

to aggregate to form a smaller number of larger aggregates (Figure 2), data that was reinforced

with the nanoparticle tracker (Figure 4, see Supporting Information). That is, there is significantly

more resistance to particle aggregation in a bicontinuous emulsion than in solution, due to the

presence of the interface.

These observations support the contention that both hydrolysis/condensation and particle

aggregation are retarded within a bicontinuous microemulsion. We attribute the difference in the

rate of formation of large TiO2 particles to reinforcement of the domain walls by the surfactant.

The first formed particles (< 20 nm) would be optically transparent (Figure 5A→B), as was

initially observed, but further growth would be inhibited once the particles had filled the available

space within the domain, that is, once they reach the domain walls/interface. The highly

viscoelastic interface will restrict the ability of particles to grow and aggregate. For further

particle growth or aggregation to occur the interface needs to distort and/or reform to facilitate

particle clustering and then aggregation (Figure 5B→C). It took over a day before this occurred

to a significant degree.

Water-in-oil microemulsions have previously been shown to serve as a template for titanium

dioxide particle synthesis from titanium chloride [22]. However, for the reasons noted above,

only under acidic conditions was it possible to form titania particles that did not aggregate. In

addition, the reactants (aqueous ammonia solution and titanium chloride in HCl) were

incorporated into separate W/O microemulsions that were then mixed together to initiate particle

formation. Such a system, in which visible titania had already formed within three hours [22], is

quite different from the one under consideration: the system described using Silsurf A008-UP

13
avoids visible particle formation for up to three days (depending on the formulation) under

neutral conditions.

The growth of TiO2 particles has previously been limited by using constricting domains of a

physical template. Szymanski et al. gelled a polymerizable microemulsion to form a monolithic

titania structure comprised of aggregated particles [23]. Exchanging water into the porous

template triggered the hydrolysis of the precursor leading to the formation of titania particles

ranging from 200-250 nm. The template did not delay hydrolysis or condensation processes, but

restricted the dimensions that the particles could ultimately achieve in analogy to the proposition

above (Figure 5).

In experiments reported here, the interactions between TTIP and Silsurf A008-UP slow the rate at

which TiO2 particles forms by hydrolysis and condensation. More importantly, although the

formation of visible titania particles is not arrested completely following the hydrolysis of a

tetraalkoxysilane, it is significantly retarded in a W/O or bicontinuous microemulsion. Despite

the fluidity of the microemulsion ingredients, and the use of neutral pH, it appears that surfactant

structuring at the water/oil interface particularly in bicontinuous microemulsions is able to retard

TiO2 particle growth and aggregation.

Conclusion

The rate of formation of titanium dioxide particles from TTIP can be significantly retarded in

W/O or bicontinuous microemulsions: the use of Silsurf A008-UP in O/W microemulsions did

not significantly retard titania formation. Water-in-oil type microemulsions exhibited the slowest

titania formation (opacity begins at 72 hours), while the bicontinuous structure type resulted in an

14
increased, but still very slow, rate of titania formation compared to the reaction with neutral water

in the absence of a microemulsion structure (the systems became translucent only after 24 hours).

In the latter two cases, even though the silicone fluids used in the microemulsion were of low

viscosity, the nanostructuring provided by water-in-oil and bicontinuous type microemulsions

retarded the growth and aggregation of titanium dioxide nanoparticles. This is ascribed to

reinforcement of the nanodomains by the surfactant at the oil/surfactant/water interface,

movement of which is needed for particle aggregation to occur.

Supporting Information

General procedure for the synthesis of titanium-siloxane complexes (iPrO)nTi(Silsurf)4-n, 1H

NMR spectra, UV-Vis absorption spectra of Ti(O-iPr)4 and Ti(Silsurf)4 compounds, Table giving

formulations in g, TEM micrographs and EDX spectra for particles derived from a

microemulsion containing 13% water at 4, 24 and 48 hours, ImageJ analysis of the micrographs.

Acknowledgements

We thank the Natural Sciences and Engineering Research Council of Canada and 20/20:

NSERC Ophthalmic Materials Network for financial support of this research. We are very

15
grateful to Siltech Corp. for providing surfactants. We would also like to express our gratitude to

Marcia Reid for assistance in obtaining TEM data.

References

[1] D.P. Macwan, P.N. Dave, S. Chaturvedi, A review on nano-TiO2 sol-gel type syntheses and
its applications, J. Mater. Sci., 46 (2011) 3669-3686.
[2] B. Ohtani, Y. Ogawa, S. Nishimoto, Photocatalytic activity of amorphous-anatase mixture of
titanium(IV) oxide particles suspended in aqueous solutions, J. Phys. Chem. B, 101 (1997) 3746-
3752.
[3] M.E. Carlotti, E. Ugazio, S. Sapino, I. Fenoglio, G. Greco, B. Fubini, Role of particle coating
in controlling skin damage photoinduced by titania nanoparticles, Free Radical Res., 43 (2009)
312-322.
[4] S. Okunaka, H. Tokudome, Y. Hitomi, R. Abe, Facile preparation of stable aqueous titania
sols for fabrication of highly active TiO2 photocatalyst films, J. Mater. Chem. A, 3 (2015) 1688-
1695.
[5] D.D. Dunuwila, C.D. Gagliardi, K.A. Berglund, Application of Controlled Hydrolysis of
Titanium(IV) Isopropoxide to Produce Sol Gel-Derived Thin-Films, Chem. Mater., 6 (1994)
1556-1562.
[6] Y. Chen, Y.Y. Yi, J.D. Brennan, M.A. Brook, Development of macroporous Titania
monoliths using a biocompatible method. Part 1: Material fabrication and characterization, Chem.
Mater., 18 (2006) 5326-5335.
[7] T. Sugimoto, X.P. Zhou, A. Muramatsu, Synthesis of uniform anatase TiO2 nanoparticles by
gel-sol method - 1. Solution chemistry of Ti(OH)(n)((4-n)+) complexes, J. Colloid Interface Sci.,
252 (2002) 339-346.
[8] T. Sugimoto, X.P. Zhou, A. Muramatsu, Synthesis of uniform anatase TiO2 nanoparticles by
gel-sol method 3. Formation process and size control, J. Colloid Interface Sci., 259 (2003) 43-52.
[9] S. Mahshid, M. Askari, M.S. Ghamsari, Synthesis of TiO2 nanoparticles by hydrolysis and
peptization of titanium isopropoxide solution, J. Mater. Proc. Tech., 189 (2007) 296-300.
[10] M. Pacia, P. Warszynski, W. Macyk, UV and visible light active aqueous titanium dioxide
colloids stabilized by surfactants, Dalton Trans., 43 (2014) 12480-12485.
[11] M.A. Brook, M. Whinton, F. Gonzaga, N. Li, Elastomeric hydrogels by polymerizing
silicone microemulsions, Chem. Commun., 47 (2011) 8874-8876.
[12] M. Pelaez, N.T. Nolan, S.C. Pillai, M.K. Seery, P. Falaras, A.G. Kontos, P.S.M. Dunlop,
J.W.J. Hamilton, J.A. Byrne, K. O'Shea, M.H. Entezari, D.D. Dionysiou, A review on the visible
light active titanium dioxide photocatalysts for environmental applications, Appl. Catal., B, 125
(2012) 331-349.
[13] J.L.H. Chau, C.T. Tung, Y.M. Lin, A.K. Li, Preparation and optical properties of
titania/epoxy nanocomposite coatings, Mater. Lett., 62 (2008) 3416-3418.

16
[14] J. Kockler, M. Oelgemoller, S. Robertson, B.D. Glass, Photostability of sunscreens, J
Photoch Photobio C, 13 (2012) 91-110.
[15] Z.A. Lewicka, W.W. Yu, B.L. Oliva, E.Q. Contreras, V.L. Colvin, Photochemical behavior
of nanoscale TiO2 and ZnO sunscreen ingredients, J. Photochem. Photobio. A: Chem., 263
(2013) 24-33.
[16] Y.Z. Li, S.J. Kim, Synthesis and characterization of nano titania particles embedded in
mesoporous silica with both high photocatalytic activity and adsorption capability, J. Phys.
Chem. B, 109 (2005) 12309-12315.
[17] C. Sanchez, J. Livage, M. Henry, F. Babonneau, Chemical Modification of Alkoxide
Precursors, J. Non-Cryst. Solids, 100 (1988) 65-76.
[18] C.J. Brinker, Scherer, G. W., Sol-Gel Science: The Physics and Chemistry of Sol-Gel
Processing, Academic Press, San Diego, CA, 1990.
[19] N.V. Golubko, M.I. Yanovskaya, I.P. Romm, A.N. Ozerin, Hydrolysis of titanium
alkoxides: Thermochemical, electron microscopy, SAXS studies, J. Sol-Gel Sci. Technol., 20
(2001) 245-262.
[20] M.E. Whinton, T.C. Hughes, S. Peng, M.A. Brook, Silicone Microemulsion Structures Are
Maintained During Polymerization Using Reactive Surfactants, in preparation.
[21] Y. Li, S.J. Kim, Synthesis and Characterization of Nano titania Particles Embedded in
Mesoporous Silica with Both High Photocatalytic Activity and Adsorption Capability, J. Phys.
Chem. B, 109 (2005) 12309-12315.
[22] P. Lin, C.Y. Xian, Z.P. Mao, Q. Zhu, X.L. Chen, L.M. Zheng, Synthesis and
characterization of titania nanoparticles by microemulsion process, Indian J. Chem. A, 45 (2006)
2017-2020.
[23] L. Szymanski, P. Surolia, O. Byrne, K.R. Thampi, C. Stubenrauch, Porous “sponge-like”
anatase TiO2 via polymer templates: synthesis, characterization, and performance as a light-
scattering material, Colloid Polym. Sci., 291 (2013) 805–815.

17
Figure 1: Proposed formation of TiO2 particles in microemulsion: A,B: transesterification of

TTIP with the surfactant Silsurf A008 UP giving Ti(Silsurf)4. C/D/E/F: hydrolysis and

condensation/micellization/conversion to TiO2 particles that adsorb onto the aqueous surface of

G: bicontinuous microemulsion, or H: a W/O microemulsion that can be stabilized like a

Pickering emulsion I.

18
Figure 2: TTIP microemulsions. A: Oil-in-water with a total water content 34.73 wt% (224:1
mole ratio [H2O]:[TTIP]). B-C: Bicontinuous with a total water content 18.25 wt% (94:1 mole
ratio [H2O]:[TTIP]) and 12.93 wt% (66:1 mole ratio [H2O]:[TTIP]), respectively. D: Water-in-oil
with a total water content of 7.67wt% (35:1 mole ratio [H2O]:[TTIP]).

2
2
Absorbance (a.u.)

Absorbance (a.u.)
1.5 1.5 Initial
Initial
24h 24h
48h 1 48h
1
168h
168h
0.5 0.5

0 0
190 290 390 190 290 390
Wavelength (nm) Wavelength (nm)
A B

2 2
Absorbance (a.u.)

Absorbance (a.u.)

Initial Initial
1.5 1.5
24h 24h
48h 48h
1 168h 1
168h

0.5 0.5

0 0
190 290 390 190 290 390
Wavelength (nm) Wavelength (nm)
C D

19
Figure 3: TEM analysis of particles formed within a silicone microemulsion containing ~13wt%
water (scale bar 2 m) at A: 4 hours showing most particles to be < 20 nm in size; B: 24 hours
showing many small particles have aggregated, and C: 48 hours. D: Particle size distribution of
titania formed over time in a silicone microemulsion with ~13wt% water.

A B C

700
600
Number of Particles

500
400
300 4hrs 24hrs 48hrs
200
100
0

Particle Size Range (nm)

D

20
Figure 4: A: Average particle size for titanium dioxide formed. B: Mode particle size for
titanium dioxide formed. C: Particles formed over time within microemulsions with different
water content and internal structuring.
Microemulsion 7wt% Water 6.0E+07
Microemulsion 13wt% Water

Concentration (particles/ug)
5.0E+07
Average Particle Size (nm)

500

Mode Particle Size (nm)

500
400 4.0E+07
400
300 3.0E+07
300

200 2.0E+07
200

100 100 1.0E+07

0 0 0.0E+00
0 100 0 100 0 100
Time (hours) Time (hours) Time (hours)
A B C

21
Figure 5: Model showing bicontinuous microemulsion domains constraining TiO2 particle
aggregation initially to the size of the aqueous domains. A: hydrolysis and condensation leading
to micelles and TiO2 particles. B: Expansion of domains over time, which requires reorganization
of the domains, allows particle aggregation.

22
Table 1: Microemulsion Formulationsa

Formulation Titanium Silsurf PDMS-OH Water

Isopropoxide A008-UP (DMS S14) (wt%)
(wt%) (wt%) (wt%)

1 3.51 49.18 39.64 7.67

2 51.12 41.57 7.31

3 3.31 46.32 37.44 12.93

4 47.80 38.84 13.36

5 3.04 43.36 35.35 18.25

6 44.69 36.50 18.81

7 2.47 34.60 28.20 34.73

8 35.55 28.91 35.54

a
All data on the table represents weight%. For formulations in grams, see Supporting

Information.

23