Spontaneous water release inducing nucleation during the nonaqueous synthesis of TiO2 nanoparticles

Article abstract of DOI:10.1039/c2ce25934f

The formation of anatase nanoparticles by reaction of titanium(iv) isopropoxide in benzyl alcohol was studied. In contrast to previous reports on the nonaqueous synthesis, in this system the particle formation occurs within a very limited time span in the course of the synthesis, concurrently to a fast step-type pressure increase within the closed reaction system. By Karl Fischer titration and ¹H NMR spectroscopy of both the liquid and the gaseous phase at different stages of the reaction, it is shown that water formation occurs during the pressure increase due to catalytic ether formation from benzyl alcohol. The generated water leads to instant nucleation and fast growth of crystalline nanoparticles, which is traced by powder X-ray diffraction as well as small-angle X-ray scattering and thereby shown to play a crucial role in the particle formation process. The Royal Society of Chemistry.

Full text of DOI:10.1039/c2ce25934f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
CrystEngComm
Cite this: CrystEngComm, 2012, 14, 8562–8568
www.rsc.org/crystengcomm
PAPER
Spontaneous water release inducing nucleation during the nonaqueous
synthesis of TiO2 nanoparticles{
Mandy Zimmermann and Georg Garnweitner*
Received 11th June 2012, Accepted 9th October 2012
DOI: 10.1039/c2ce25934f
The formation of anatase nanoparticles by reaction of titanium(IV) isopropoxide in benzyl alcohol
was studied. In contrast to previous reports on the nonaqueous synthesis, in this system the particle
formation occurs within a very limited time span in the course of the synthesis, concurrently to a fast
step-type pressure increase within the closed reaction system. By Karl Fischer titration and 1H NMR
spectroscopy of both the liquid and the gaseous phase at different stages of the reaction, it is shown
that water formation occurs during the pressure increase due to catalytic ether formation from benzyl
alcohol. The generated water leads to instant nucleation and fast growth of crystalline nanoparticles,
which is traced by powder X-ray diffraction as well as small-angle X-ray scattering and thereby
shown to play a crucial role in the particle formation process.
nanoparticles via thermal decomposition of iron oleate in an
Introduction
inert organic solvent, Kwon et al. have presented significant
insights into the particle formation kinetics.20,21 For that system,
Highly crystalline metal oxide nanoparticles of defined size are in
great demand for a variety of applications. Titanium dioxide
it was shown that the decomposition of the precursor results in
nanoparticles in particular have been of high interest due to their
the formation of monomers that lead to a sudden nucleation of
excellent photocatalytic properties1–4 but also their antibacterial
nanoparticles (nucleation burst), which could be described by the
performance5 and potential use in dye-sensitized solar cells6–9 as
well-known LaMer model22,23 and is analogous to models of
well as for cancer treatment.10,11 For their synthesis, there are
semiconductor nanocrystal nucleation and growth.24,25 In
different techniques available, from gas phase methods produ-
contrast to this method, the benzyl alcohol route is carried out
cing metal oxide nanoparticles in large quantities at relatively
at substantially lower temperatures, but also requires signifi-
low cost12,13 to liquid-based methods that generally are more
cantly longer reaction times of several hours to days. The
controllable as well as flexible with regard to the particle
molecular reaction mechanisms have been studied for the
characteristics.14 In the conventional aqueous sol–gel synthesis,
formation of TiO2 nanoparticles via alkyl halide elimination,
however, metal oxides are mostly obtained in high crystallinity
revealing the steady formation of organic side products via first-
only after an additional calcination step. In the last decade,
order kinetics.26 Therefore, the formation of nanoparticles in a
various nonaqueous techniques were established, such as the
nucleation burst event appears unlikely, as the generated
benzyl alcohol route being broadly applicable not only for TiO2
supersaturation would not be as high as for thermal decomposi-
but for a multitude of metal oxides and leading directly to highly
tion processes. Similarly, a pseudo-first order kinetics was
crystalline nanoparticles.15–18 This route includes the reaction of
observed for the formation of ZnO via ester elimination.27,28
a molecular precursor, such as a metal alkoxide, with the organic
For the microwave-assisted synthesis of ZnO nanoparticles, it
solvent benzyl alcohol under relatively mild conditions, resulting
was shown that the crystal growth can be described by means of
in well-defined nanoparticles.
the Lifshitz–Slyozov–Wagner (LSW) theory, which points to a
However, the detailed reaction mechanisms and the formation
diffusion limited growth behavior.29 It remains unclear however
mechanism of the resulting nanoparticles are not yet fully
as to why the thermally induced nonaqueous synthesis of metal
understood for the nonaqueous synthesis. We have reported that
oxide nanoparticles in contrast to the microwave-assisted
the molecular reaction mechanisms preceding the metal oxide
synthesis for many systems requires days to obtain crystalline
formation are surprisingly complex for a number of systems,19
nanoparticles despite reaction temperatures of 175–250 uC. We
preventing the application of simple nucleation and growth
performed a study on the formation of iron oxide nanoparticles
models. For the apparently similar synthesis of iron oxide
to elucidate this issue, and showed that the particle crystal-
lization can be separated from nucleation, with crystallization
Institute for Particle Technology, Technische Universita¨t Braunschweig,
kinetics even being dependent on the nature of the utilized
reaction medium.30 The nonaqueous synthesis of metal oxides
hence has turned out to be a highly complex process, with the
Germany. E-mail: g.garnweitner@tu-braunschweig.de;
Fax: +49 (0)531-3919633; Tel: +49 (0)531-3919615
{ Electronic supplementary information (ESI) available: GC measure-
ment, LSW plot. See DOI: 10.1039/c2ce25934f
8562 | CrystEngComm, 2012, 14, 8562–8568
This journal is ß The Royal Society of Chemistry 2012

View Article Online
series 2, PANalytical, PIXcel-3D detector). For thermo-gravi-
metric analysis (TGA), 15–20 mg of the dried powder was
analyzed with a Mettler Toledo TGA/SDTA851 instrument by
heating up to 600 uC with a heating rate of 20 uC min21 under
oxygen atmosphere. The reaction mixture was analyzed by 1H
NMR (400 MHz, CDCl3 or DMSO, TMS, Bruker DRX-400)
and GC/MS (dichloromethane (HPLC grade), JEOL AccuTOF
GC/MS, Agilent 7820A GC). Furthermore, the content of water
in the reaction mixture was determined by Karl Fischer titration
(Aqua 40.00, Analytik Jena). Transmission electron microscopy
(TEM) images were received from a JEOL FEM-2100 instru-
ment at 100 kV; for sample preparation the washed particles
were redispersed in a solution of 0.6 M oleic acid in chloroform
and dropped on a Formvar-coated copper grid (Plano). Small-
angle X-ray scattering (SAXS) measurements of the untreated
reaction mixture were carried out on a Nanostar instrument
(Bruker AXS GmbH) using Cu Ka radiation and a position
sensitive area detector (HiStar). The obtained 2D patterns were
azimuthally averaged and corrected for transmission and
instrument background.
Scheme 1 Reaction mechanism according to the ether elimination
mechanism.
precise mechanisms involved in particle nucleation still being
largely unknown.
In order to shed further light on these aspects, we present
investigations on the formation of TiO2 nanoparticles by
reaction of titanium isopropoxide in benzyl alcohol as a rather
simple model system. The reaction of metal alkoxides was
reported by us earlier to proceed in two steps via an ether
elimination mechanism.31,32 The first step is a partial exchange of
the isopropoxide ligand against benzyl alcohol. Subsequently,
the formation of Ti–O–Ti bonds happens via the elimination of
organic ethers analogous to the mechanism presented in
Scheme 1.31,33 For the detailed investigation of the synthesis,
we used a 1.5 L rector system enabling the monitoring of the
process temperature and pressure and moreover the withdrawal
of samples at any point during the course of the synthesis.
Results and discussion
From the reaction of titanium isopropoxide in benzyl alcohol we
obtained highly crystalline TiO2 nanoparticles, with the PXRD
patterns clearly assignable to the anatase modification, as shown
in Fig. 1(a). The experimental data are in good agreement with
the reference data for anatase (ICSD database, no. 98-000-9852).
Experimental
Synthesis of TiO2 nanoparticles
Titanium(IV) isopropoxide (TIP, 97%) and benzyl alcohol
(BnOH, 99%) (both obtained from Sigma Aldrich) were mixed
in a molar ratio of 1 : 50 and transferred into a stainless steel
double walled reactor of 1.5 L capacity (Polyclave type 3/1,
Bu¨chi Glas Uster) which was heated via an external thermostat
(Huber Tango HT) at 175 uC for 48 h. The reactor was equipped
with a blade agitator operated at 250 rpm, a temperature probe
and a manometer enabling the recording of the temperature and
the development of the pressure of the system during the whole
reaction. Additionally a sampling system was installed, allowing
the withdrawal of samples at different reaction times. The
reaction was quenched by fast cooling of the samples in a water
bath. To isolate the formed nanoparticles, the reaction suspen-
sion was centrifuged at 6500g for 10 min, washed twice with
ethanol followed by centrifuging again and dried under reduced
pressure at room temperature for at least 24 h. The fraction of
particles that could not be separated by centrifugation was
precipitated by addition of 1 mL reaction mixture to 5 mL
methanol and then centrifuged, washed and dried as described
above.
Characterization
The crystallite size was determined by powder X-ray diffraction
(PXRD) from the dried samples with Cu Ka radiation
(Empyrean Cu LEF HR goniometer) on a Si sample holder in
a range of 2h from 20 to 90u and a step size of 0.05u (Empyrean
Fig. 1 (a) PXRD pattern of the obtained TiO2 nanoparticles compared
to the anatase reference (ICSD database, no. 98-000-9852), (b)
representative TEM images of obtained anatase nanoparticles with a
magnification of 180 000 and 1 000 000 (inset; crystallite size y15 nm).
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 8562–8568 | 8563

View Article Online
Fig. 2 Development of temperature and pressure within the reactor
during the particle synthesis; the green crosses indicate points in time
where samples were withdrawn for further analysis.
The sample consists of small crystallites, as recognizable by the
peak broadening; applying the Scherrer equation, a crystallite
size of 14.5 nm in the [011] direction was calculated with an
accuracy of 0.5 nm. From TEM images of typical samples
(Fig. 1(b)) we infer that particle sizes and crystallite sizes are
equal, meaning that the primary particles are single crystals.
Interestingly, in contrast to the synthesis from TiCl4, the
nanoparticles exhibit
larger.12,13
a cubic shape and are significantly
By monitoring the reaction process, we observed a sudden and
pronounced increase in pressure after a certain reaction time.
Fig. 2 shows the development of reaction temperature and
pressure within the reactor over time. Upon heating, which was
completed within 20 min, a first rise in pressure to 1 bar
overpressure was observed, most likely caused by the evapora-
tion of isopropyl alcohol from the reaction mixture due to the
ligand exchange reaction. Afterwards, the reaction temperature
was kept constant at 175 uC and the pressure remained at this
level for about 24 h before suddenly increasing to ca. 2 bar
overpressure.
Fig. 3 1H NMR spectra of (a) the reaction mixture in CDCl3 and (b)
the reaction mixture (blue) and a sample from the gaseous headspace
phase (red) in DMSO-d6; the peaks are assigned to the components as
follows: 1 isopropyl alcohol, 2 benzyl alcohol, 3 dibenzyl ether, 4 benzyl
isopropyl ether, 5 water, 6 toluene, 7 DMSO.
and close to it at 4.42 ppm another peak which is assigned to the
mixed ether (4). The signals at 3.91 and 3.60 ppm are attributed
to the CH groups and at 1.10 and 1.13 ppm to the CH3 groups of
isopropyl alcohol (1) and the mixed ether, respectively.
Additionally, a small amount of toluene (6) was detected,
stemming from decomposition of benzyl alcohol through
heating.34 The molar ratio of benzyl alcohol : dibenzyl ether :
benzyl isopropyl ether is calculated from the integrals of the CH2
groups to about 20 : 4 : 1, whereby the integral of the mixed
ether was normalized to 1. Remarkably, the amount of dibenzyl
ether is considerably higher than expected from a stoichiometric
reaction according to Scheme 1. Therefore we infer that
additionally ether is formed via a catalytic ether condensation
reaction at the titanium centers. The formation of the organic
side products was also confirmed by GC/MS measurements (see
ESI{). Notably, an additional peak is visible in the 1H NMR
spectrum at a chemical shift of 1.49 ppm, which is assigned to
water (5). This is somewhat surprising, as the nonaqueous
synthesis usually does not result in the formation of significant
amounts of free water in the reaction system; only in special
cases such as for metal niobates, the formation of excess
amounts of water has been reported.35 The presence of water
is more visible when using DMSO-d6 as solvent for the NMR
measurement, as shown in Fig. 3(b). Here, the peak at 3.53 ppm
To the best of our knowledge, such an instantaneous step-type
intrinsic effect has not been described for the nonaqueous
synthesis of metal oxide nanoparticles before. For example, in
the synthesis of TiO2 nanoparticles from titanium tetrachloride,
which we performed in an open system, only continuous
processes were observed,26 and also during the microwave
synthesis of ZnO nanoparticles in closed vessels, no pronounced
increase in pressure was noted.29 The step-type effect rather
resembles the kinetics of the thermal decomposition synthesis.20
Hence, we decided to study the reaction kinetics in detail. First
we analyzed the final reaction mixture by 1H NMR spectro-
scopy, as illustrated in Fig. 3(a), to determine the organic
components formed during the reaction. As elucidated above,
the formation of various ethers as organic side products was
expected,31,33 which could be confirmed by our measurements,
detecting predominantly dibenzyl ether but also small amounts
of benzyl isopropyl ether and traces of diisopropyl ether. The 1H
NMR spectrum shows, next to the chemical shift of the CH2
group of benzyl alcohol (2) at 4.57 ppm, clearly a signal of the
CH2 group of dibenzyl ether (3) at a chemical shift of 4.47 ppm,
8564 | CrystEngComm, 2012, 14, 8562–8568
This journal is ß The Royal Society of Chemistry 2012

View Article Online
(blue line) can be clearly assigned to water in the reaction
mixture. To gain insight into the pressure increase observed
during the reaction, we further analyzed samples from the
gaseous phase above the reaction mixture by withdrawing
gaseous samples through an outlet valve while the mixture was
held at the reaction temperature. After cooling to room
temperature, a clear liquid was obtained and analyzed by 1H
NMR spectroscopy (Fig. 3(b), red line). In this spectrum, a large
quantity of water at 3.42 ppm in a molar ratio of 2 : 1 to
isopropyl alcohol (CH3: 1.05 ppm, CH: 3.79 ppm, OH: 4.40
ppm) is visible; the slight differences in chemical shift to the
solution sample are attributed to the great differences in
concentration. Benzyl alcohol is existent only in small amounts
(0.28 mol/mol (isopropyl alcohol)); CH2: 4.51 ppm. The
existence of water in the reaction mixture was also confirmed
by Karl Fischer titration (see below for quantification). These
findings explain the detection of ethers in the final reaction
mixture in much higher amounts than expected for the
stoichiometric reaction of the alkoxide. Evidently, via a catalytic
ether formation at the titanium centers, water is formed through
a condensation reaction of the benzyl alcohol solvent (Scheme 2).
Hence, the pressure increase can be clearly correlated to the
formation of water, which is of high importance for the
molecular reaction mechanism. The high reactivity of titanium
isopropoxide as well as other metal alkoxides towards water is
well known and utilized in aqueous sol–gel synthesis,36 but also
in nonaqueous sol–gel methods there are studies about the
influence of limited amounts of water on the particle forma-
tion.37,38 The hydrolysis accelerates the reaction, but in contrast
to previously described reaction types water is not added nor
formed constantly by organic condensation reactions but is
formed in a spontaneous and fast process during the reaction.
The pressure increase also plays an important role in the
particle formation, as prior to this step, the reaction mixture
appears completely transparent but is turbid afterwards.
Therefore, we investigated the processes occurring during the
time span of the pressure increase in detail both with regard to
the molecular reaction kinetics as well as the particle formation
and growth. The formation kinetics of dibenzyl ether, being
representative for the sum of ethers, and for water in the reaction
mixture during the pressure increase is presented in Fig. 4(a). For
better visibility, the time scale is set to 0 at the beginning of the
pressure increase. The molar ratio of dibenzyl ether to the total
amount of aromatic compounds was calculated by comparing
the integrals of the CH2 group of dibenzyl ether with the
aromatic signals from the 1H NMR spectra of the individual
samples. The relative molar amount of the ether after completion
of the reaction was determined as nearly 10%, whilst at the
beginning of the pressure rise it is rather low with just 1.7%. The
subsequent increase for the ether is very similar to the
development of pressure and follows first order kinetics, as
shown in Fig. 4(b). The data for the kinetics, obtained by fitting
the concentration development by an exponential function, is
given in Table 1. The molar ratio of water was determined by
combining the amount of water determined in the liquid and
gaseous phases. The water content in the liquid phase was
thereby obtained by Karl Fischer titration; in the gaseous phase
it was calculated using the equation of state for real gases
estimating the compressibility factor as 0.993.39 The total values
were then related to the total amount of aromatic compounds,
equivalent to the initial amount of benzyl alcohol. Similar first-
order kinetics as in the case of the ether is apparent. We have
reported first-order kinetics also before for the formation of side
products during the reaction of TiCl4 in benzyl alcohol,26 but in
contrast to our earlier study, the actual molecular mechanisms
leading to particle formation do not start after the reaction
mixture has reached the target temperature, but at a much later
stage at the beginning of the pressure increase and occur within a
very limited timescale as compared to the total reaction time.
Fig. 4(c) presents the PXRD patterns of samples taken during
and after the pressure rise. An increase of intensity as well as a
substantial sharpening of the reflections can be observed. The
diminution of the full width at half maximum (FWHM) of
reflections with time corresponds to a growth of the particles. As
mentioned above, the particle size could be calculated from the
PXRD data using the Scherrer equation for the (011) reflection.
The particles isolated 10 minutes after the beginning of the
pressure increase showed a size of 6.9 nm and are of high
crystallinity. During the next 2 hours, the particle size increased
to 13.6 nm and within the following 20 hours it increased only
slightly further to 14.5 nm. The kinetics of particle growth is
illustrated in Fig. 4(d), appearing similar to the formation
kinetics of the byproducts and again following first-order
kinetics (Table 1). For comparison, the development of pressure
is given as well. Initially the particle growth proceeds analo-
gously to the pressure rise, but while the pressure converges to a
limiting level, the crystallite growth exhibits a small further
increase. The LSW analysis of the development of particle size
over time29 (see ESI{) reveals that particle growth occurs via two
distinct mechanisms, and only the later growth stage might be
attributable to Ostwald ripening (according to the modified LSW
theory40). Our investigations also indicate that with the increase
of pressure, corresponding to the formation of water, the
formation of the nanoparticles is initiated as well. No particles
could be isolated or observed before this step. Moreover, after
cessation of the pressure increase, the molecular reactions also
do not proceed further, with the amount of ethers remaining
constant. Hence, the formation and growth of particles happen
concurrently to the formation of the byproducts, most notably
water, within the well detectable increase of pressure. With the
beginning of the water formation, hydrolysis is initiated and the
nanoparticles are formed.
On the other hand, it cannot be excluded in principle that
highly stabilized nanoparticles are present before the pressure
rise, even though the addition of methanol did not lead to any
precipitation. In order to elucidate this issue, SAXS measure-
ments of samples of the reaction mixture withdrawn from the
system at different stages of reaction were performed. In Fig. 5
the scattering curves measured for the pure solvent benzyl
alcohol, a freshly prepared mixture of the reactants TIP and
BnOH as well as the reaction mixture at different stages of the
Scheme 2 Condensation of benzyl alcohol to dibenzyl ether at the
titanium centers.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 8562–8568 | 8565

View Article Online
Fig. 4 (a) Concentration development over reaction time for dibenzyl ether and water in comparison to the increase of pressure, (b) first-order kinetics
for the formation of dibenzyl ether and water, (c) PXRD data showing the particle growth and the development of crystallinity during reaction time
and (d) crystallite size over time, showing the formation kinetics of anatase nanoparticles related to the pressure rise. The little decrease in pressure is
caused by sampling. Note: the reaction time is set to 0 at the beginning of the pressure increase.
Table 1 Kinetic data of dibenzyl ether, water and particle formation processes, with ce = final concentration and de = final size (reaction time in h)
Kinetics
ce [mol L21] or de [nm]
A
k [h21
]
Dibenzyl ether
Water
Particle size
c = ce [1 2 A exp(2kt)]
c = ce [1 2 A exp(2kt)]
d = de [1 2 A exp(2kt)]
9.98
5.18
13.52
0.83
0.95
0.92
0.10
0.06
0.04
pressure rise are displayed. Due to the addition of Ti scattering
centers, the curve for the mixture of reactants shows higher
intensity at larger q values because of higher contrast as
compared to the pure solvent. For low scattering vectors,
corresponding to the contribution of bigger structures, the curve
however shows the same intensity as the solvent, proving the
absence of larger clusters or nanoparticles in the sample. The
scattering curve of the sample immediately before the pressure
increase shows no significant differences to the precursor
solution, verifying that no particles are formed up to this stage
of the reaction. With the beginning of the pressure rise, the
scattering intensity is strongly enhanced at low q values,
evidencing the formation of larger structures. During the
pressure rise, the scattering intensity continuously increases,
corresponding to particle formation and growth. The increase
occurs within a substantially shorter time than reported earlier
Fig. 5 SAXS measurements of benzyl alcohol, the mixed reactants and
the reaction mixture before (0 min), at the beginning of (5 min) and
during (10 min) the pressure increase.
8566 | CrystEngComm, 2012, 14, 8562–8568
This journal is ß The Royal Society of Chemistry 2012

View Article Online
Conclusions
We have studied the formation and growth mechanism of TiO2
nanoparticles for the nonaqueous synthesis by reaction of
titanium(IV) isopropoxide in benzyl alcohol. In contrast to the
reaction of TiCl4 via alkyl halide elimination, cubic shaped
anatase nanoparticles are obtained only after an induction
period of almost 24 h. Subsequently, a sudden and pronounced
pressure increase was found to occur, being attributable to the
formation of water. Concurrently to the pressure increase, fast
nucleation and growth of the nanoparticles were observed,
presumably being caused by a hydrolysis of the precursor
species. At the same time the rate of formation of organic ethers
increases rapidly, with the formed quantities clearly exceeding
the stoichiometric amount expected from the nonaqueous
reaction mechanism. Hence, the spontaneous water formation
is explained by a catalytic ether condensation of benzyl alcohol
at the titanium centers. Although the formation of excess
amounts of water has been identified before for the synthesis
of metal niobates, it has not been reported for the simple
reaction of a metal alkoxide in benzyl alcohol nor for titanium as
the metal species. Moreover, the particle formation and growth
are initiated at the beginning of the pressure increase and show
analogous kinetics as the water formation and ether formation
reactions, except for subsequent Ostwald ripening processes.
SAXS measurements proved the absence of any nanoparticles
before the pressure surge, evidencing the crucial influence of the
water formation on the particle formation kinetics. Therefore,
the observed mechanism determines the formation and also the
properties of the formed nanoparticles and possibly is the
reason for the strong differences in morphology as compared
to the TiCl4–benzyl alcohol system. Because the reaction of a
metal alkoxide in an alcohol is a widely employed approach for
the synthesis of metal oxide nanostructures, the mechanisms
reported here might occur in a variety of other systems, also
especially for the metal niobates. Moreover, the nonaqueous
synthesis is known to lead to nanoparticles with very different
sizes and morphologies for different systems, which yet can only
partially be explained. As the formation of well-faceted
nanocrystals is shown here to be greatly facilitated by
catalytic water formation, the occurrence of this mechanism,
possibly only to a small extent in other systems, might be a
general explanation for the highly different product particle
morphologies.
Fig. 6 Development of (a) yield and (b) number of TiO2 particles over
reaction time as determined by TGA. Note: the reaction time is set to 0 at
the beginning of the pressure increase.
for the formation of nanoparticles by alkyl halide elimination.41
A calculation of the size distribution of the scattering structures
was not possible with the given measurement setup as the sizes
exceeded the resolution of the instrument due to the fast
agglomeration of the particles.
To complement the kinetic studies of particle growth, the
development of the yield of TiO2 nanoparticles was determined
by TGA measurements. The dried precipitates were heated to
600 uC, leading to decomposition of all organics bound to the
particle surface so the actual amount of titanium dioxide was
determined. Fig. 6(a) shows the development of the calculated
yield over time in comparison to the kinetics of particle size
growth. The lower yields at some points are caused by losses
during the washing process, so that we assume that the solid
content in the reaction mixture increases continuously to almost
100% at the end of the experiment, with analogous kinetics as
compared to the particle size and the formation of byproducts.
Supposing that the particles possess uniform size, the number of
particles at each stage of the reaction can be calculated from the
yield as illustrated in Fig. 6(b). The density of TiO2 is assumed
equal to the bulk material as 3.9 g cm21. The slight decrease in
the number of particles indicates that only particle growth and
possibly some Ostwald ripening happen during the pressure
surge but no further nucleation. Hence, the nucleation process is
confirmed to occur within a very limited timespan.
Acknowledgements
This work was financially supported by the Deutsche
Forschungsgemeinschaft (DFG), grant GA 1492/3-1. The
authors would like to thank Mr. Bilal Temel from the Institute
for Particle Technology, TU Braunschweig, for the XRD
measurements, Dr. Till Beuerle, Institute of Organic
Chemistry, TU Braunschweig, for the GC/MS measurements
and Mrs. Ingrid Zenke from the Department of Biomaterials,
Max Planck Institute of Colloids and Interfaces, Potsdam, for
the SAXS measurements. Dr. Kerstin Ibrom, Institute of
Organic Chemistry, TU Braunschweig, is acknowledged for
helpful discussions on the NMR analyses.
This journal is ß The Royal Society of Chemistry 2012
CrystEngComm, 2012, 14, 8562–8568 | 8567

View Article Online
22 V. K. LaMer and P. H. Dinegar, J. Am. Chem. Soc., 1950, 70,
4847–4854.
References
23 V. K. LaMer, Ind. Eng. Chem., 1952, 44, 1270–1277.
24 C. B. Murray, D. J. Norris and M. G. Brawendi, J. Am. Chem. Soc.,
1993, 115, 8706–8715.
25 J. Y. Rempel, M. G. Brawendi and L. F. Jensen, J. Am. Chem. Soc.,
2009, 131, 4479–4489.
26 G. Garnweitner and C. Grote, Phys. Chem. Chem. Phys., 2009, 11,
3767–3774.
27 B. Ludi, M. J. Su¨ess, I. A. Werner and M. Niederberger, Nanoscale,
2012, 4, 1982–1995.
28 I. Bilecka and M. Niederberger, Electrochim. Acta, 2010, 55,
7717–7725.
29 I. Bilecka, P. Elser and M. Niederberger, ACS Nano, 2009, 3,
467–477.
30 I.-M. Grabs, C. Bradtmo¨ller, D. Menzel and G. Garnweitner, Cryst.
Growth Des., 2012, 12, 1469–1475.
31 N. Pinna, G. Garnweitner, M. Antonietti and M. Niederberger, Adv.
Mater., 2004, 16, 2196–2200.
32 G. Garnweitner and M. Niederberger, J. Mater. Chem., 2008, 18,
1171–1182.
33 M. Niederberger, G. Garnweitner, N. Pinna and M. Antonietti, J.
Am. Chem. Soc., 2004, 126, 9120–9126.
34 A. R. Katritzky and M. Balasubramanian, Energy Fuels, 1990, 4,
499–505.
35 L. Zhang, G. Garnweitner, I. Djerdj, M. Antonietti and M.
Niederberger, Chem.–Asian J., 2008, 3, 746–752.
36 J. Livage, M. Henry and C. Sanchez, Prog. Solid State Chem., 1988,
18, 259–341.
37 M. Inoue, H. Kominami and T. Inui, Appl. Catal., A, 1995, 121,
L1–L6.
38 H. Kominami, Y. Takada, H. Yamagiwa and Y. Kera, J. Mater. Sci.
Lett., 1996, 15, 197–200.
39 R. H. Perry and D. Green, Perry’s Chemical Enigneers’ Handbook,
McGraw-Hill, New York, 1984.
40 I. M. Lifshitz and V. V. Slyozov, J. Phys. Chem. Solids, 1961, 19,
35–50.
1 A. Mills, R. H. Davies and D. Worsley, Chem. Soc. Rev., 1993, 22,
417–425.
2 A. L. Linsebigler, G. Lu and J. T. Yates, Chem. Rev., 1995, 95,
735–758.
3 J. Peral, X. Domenech and D. F. Ollis, J. Chem. Technol. Biotechnol.,
1997, 70, 117–140.
4 E. Stathatos, D. Tsiourvas and P. Lianos, Colloids Surf., A, 1999,
149, 49–56.
5 Y. Kikuchi, K. Sunada, T. Iyoda, K. Hashimoto and A. Fujishima, J.
Photochem. Photobiol., A, 1997, 106, 51–56.
6 B. O’Regan and M. Gra¨tzel, Nature, 1991, 353, 737–740.
7 M. Gra¨tzel, Acc. Chem. Res., 2009, 42, 1788–1798.
8 I. Mora-Sero´ and J. Bisquert, J. Phys. Chem. Lett., 2010, 1,
3046–3052.
9 T. Miyasaka, J. Phys. Chem. Lett., 2011, 2, 262–269.
10 D. M. Blake, P. C. Maness, Z. Huang, E. J. Wolfrum, J. Huang and
W. A. Jacoby, Sep. Purif. Methods, 1999, 28, 1–50.
11 S. Ivankovic, M. Gotic, M. Jurin and S. Music, J. Sol-Gel Sci.
Technol., 2003, 27, 225–233.
12 A. S. Gurav, Z. Duan, L. Wang, M. J. Hampden-Smith and T. T.
Kodas, Chem. Mater., 1993, 5, 214–216.
13 F. E. Kruis, H. Fissan and A. Peled, J. Aerosol Sci., 1998, 29,
511–535.
14 B. L. Cushing, V. L. Kolesnichenko and C. J. O’Connor, Chem. Rev.,
2004, 104, 3893–3946.
15 M. Niederberger, M. H. Bartl and G. D. Stucky, J. Am. Chem. Soc.,
2002, 124, 13642–13643.
16 M. Niederberger, M. H. Bartl and G. D. Stucky, Chem. Mater., 2002,
14, 4364–4370.
17 G. Garnweitner and M. Niederberger, J. Am. Ceram. Soc., 2006, 89,
1801–1808.
18 N. Pinna and M. Niederberger, Angew. Chem., Int. Ed., 2008, 47,
5292–5304.
19 M. Niederberger and G. Garnweitner, Chem.–Eur. J., 2006, 12,
7282–7302.
20 S. G. Kwon, Y. Piao, J. Park, S. Angappane, Y. Jo, N.-M. Hwang,
J.-G. Park and T. Hyeon, J. Am. Chem. Soc., 2007, 129,
12571–12584.
41 G. V. Jensen, M. Bremholm, N. Lock, G. R. Deen, T. R. Jensen, B.
B. Iversen, M. Niederberger, J. S. Pedersen and H. Birkedal, Chem.
Mater., 2010, 22, 6044–6055.
21 S. G. Kwon and T. Hyeon, Acc. Chem. Res., 2008, 41, 1696–1709.
8568 | CrystEngComm, 2012, 14, 8562–8568
This journal is ß The Royal Society of Chemistry 2012