Journal of Colloid and Interface Science 348 (2010) 342–347

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

www.elsevier.com/locate/jcis

Effect of the agglomeration of TiO2 nanoparticles on their photocatalytic

performance in the aqueous phase
Gang Li a, Lu Lv a, Haitao Fan b, Junyan Ma b, Yanqiang Li b, Yong Wan b, X.S. Zhao a,b,*
a
Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, 117576, Singapore
b
Institute of Multifunctional Materials, Laboratory of New Fiber Materials and Modern Textile, Growing Basis for State Key Laboratory, Qingdao University, Qingdao 266071, China

a r t i c l e i n f o a b s t r a c t

Article history: TiO2 nanoparticles have been widely explored as photocatalysts in the degradation of organic matters
Received 4 February 2010 present in water. However, spontaneous agglomeration of TiO2 nanoparticles in a suspension is a crucial
Accepted 20 April 2010 issue that must be addressed before the photocatalyst can be used for water treatment. In the present
Available online 24 April 2010
work, the nature of the agglomeration of TiO2 nanoparticles in aqueous suspension was investigated.
Two approaches to minimize the agglomeration of colloidal TiO2 particles were investigated. A careful
Keywords: control over the pH of the system was found to be an effective method for stabilizing colloidal TiO2 par-
Colloidal TiO2
ticles and to significantly enhance the adsorption of orange II. As a result, the overall photocatalytic deg-
Agglomeration
Photocatalysis
radation rate was greatly accelerated. In addition to pH control, modification of TiO2 particles using
Zeta potential polyelectrolyte poly allylamine hydrochloride (PAH) was observed to be an effective approach for pre-
venting colloidal TiO2 particles from agglomeration.
Ó 2010 Elsevier Inc. All rights reserved.

1. Introduction strategies such as surface modification [7,13–18] and adjustment

Nanosized titanium dioxide (TiO2) is a promising material for
water purification because of its high specific surface area [1–6].
But practical applications of TiO2 nanomaterials are hindered due
to a number of problems, one of which is the spontaneous agglom-
eration of TiO2 nanoparticles when dispersed in aqueous media,
resulting in a rapid decrease in specific surface area, thus photocat-
alytic activity. The agglomeration behavior of TiO2 particles in the
aqueous phase has been studied previously [6–9]. Unfortunately,
the potentially important effects of agglomeration on subsequent
adsorption/photoreaction have not been considered in experi-
ments despite mounting evidence that agglomeration is the rule
rather than the exception [6]. A common concept is that such par-
ticle agglomeration can rapidly decrease the effective surface area
of the photocatalyst, thus resulting in a fast loss in photocatalytic
activity. However, there is no report on the agglomeration behavior
of TiO2 nanoparticles at different particle concentrations and how
such agglomeration can affect the photocatalytic performance.
Recent studies [10,11] have shown that special means like pro-
tracted sonication cannot prevent the suspended nanoparticles
from agglomeration. The agglomeration process is strongly depen-
dent on parameters, such as ionic strength and pH, of the suspen-
sion [6,12]. To minimize the agglomeration of TiO2 particles,
* Corresponding author at: Department of Chemical and Biomolecular Engineer-
ing, National University of Singapore, 4 Engineering Drive 4, 117576, Singapore.
Fax: +65 67791936.
E-mail address: chezxs@nus.edu.sg (X.S. Zhao).
of the system pH [19–21] have been reported. The idea is to stabi-
lize colloidal TiO2 particles by enabling their surfaces to carry elec-
tric charge to create electrostatic repulsive forces to repel each
other. It has been observed that the adsorption affinity of the mod-
ified TiO2 particles toward organic matters can be greatly enhanced
if the organic matter carries a charge opposite to the TiO2 particle
surface charge [21]. This in turn can dramatically improve the pho-
tocatalytic reaction rate.
In the present work, the nature of the agglomeration of colloidal
TiO2 particles at different concentrations was examined. The influ-
ence of particle agglomeration on photocatalytic properties was
investigated. Two strategies, namely modification of TiO2 particles
using polyelectrolyte and varying the system pH, were used to min-
imize the agglomeration. New insights into the nature of TiO2 parti-
cle agglomeration were gained.
2. Material and methods
2.1. Materials
A TiO2 powder sample (P25) with a specific surface area of
50 m2/g was obtained from Degussa and used as received. Poly-
electrolyte poly allylamine hydrochloride (PAH, Mw  70,000)
was obtained from Aldrich. Orange II purchased from Aldrich was
used as a model organic pollutant in the measurement of photocat-
alytic properties. Orange II is a nonbiodegradable synthetic dye
with a molecular formula of C16H11N2NaO4S, being widely used
in the textile industry.

0021-9797/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcis.2010.04.045
 G. Li et al. / Journal of Colloid and Interface Science 348 (2010) 342–347
2.2. Modification of P25 TiO2 nanoparticles by PAH
First, a given amount of P25 powder was dispersed in deionized
(DI) water under ultrasonication for 10 min. Second, the suspen-
sion pH was adjusted to 9.3 using a 1 M NaOH solution under mag-
netic stirring. Then, various amounts of PAH were subsequently
added to the suspension to allow deposition of PAH on the surface
of the TiO2 nanoparticles.
2.3. Preparation of a fibrous TiO2 sample
A TiO2 sample with a fibreous morphology was prepared by
using P25 powder as the titanium source according to the method
described in Ref. [22]. One gram of P25 powder was combined with
50 mL of 10 M NaOH aqueous solution under stirring. The suspen-
sion was then hydrothermally treated at 200 °C for 24 h in an auto-
clave. The white solid was separated from the liquid phase by
centrifugation, washed with DI water, and dried at 70 °C. The resul-
tant solid was then added to a HNO3 solution (0.1 M), hydrother-
mally heated at 180 °C for 24 h, to form layered hydrogen
titanate (H-titanate) nanofibers.
2.4. Characterization
Zeta potential profiles of P25 TiO2 particles before and after
modification were measured using a Zetaplus 100. A solid sample
was suspended in a 0.001 M KCl solution and the pH of the suspen-
sion was adjusted using a 1 M NaOH or HCl solution. Particle size
distribution curves were measured using the dynamic laser light
scattering technique (Brookhaven Instruments, 90 plus). A field-
emission scanning electron microscope (FESEM) (JEOL JSM-
6700F) was used to observe the morphology of the TiO2 particles.
2.5. Evaluation of photocatalytic activity
The photocatalytic activities of the photocatalysts were evalu-
ated by measuring the degradation rate of orange II under UV light
illumination. The photocatalytic measurements were carried out in
a semibatch swirl flow reactor [23]. An aqueous suspension con-
taining orange II and TiO2 was pumped tangentially into the reac-
tor from a reservoir using a peristaltic pump. The reservoir was
jacketed and maintained at a temperature of 25 °C throughout
the experiment. A high-pressure mercury vapor lamp (Phillips
HPR 125 W) with a wavelength of 365 nm was used as the UV-light
source. Before the UV lamp was switched on for photodegradation
there is a dark adsorption period in which the P25 particles were
mixed with orange II solution to reach the adsorption equilibrium.
This dark adsorption was studied to evaluate the adsorption capac-
ity of P25 particles based on the concentration variance of orange II
before and after mixing with P25. Samples were taken from the
reservoir at regular intervals and filtered with Millex Millipore fil-
ter (0.1 lm) to remove the catalyst before analysis. The orange II
concentrations before and after adsorption/reaction were mea-
sured using a UV–Vis spectrophotometer (UV-1601, Shimadzu).
The pH of the orange II solution was adjusted using either a 1 M
NaOH or 1 M HCl solution.
3. Results and discussion
Experiments were carried out to evaluate the photocatalytic
activities of commercially available P25 nanoparticles at various
catalyst dosages in order to investigate the effect of particle con-
centration on the agglomeration nature of the P25 nanoparticles
and how this agglomeration would influence photocatalytic prop-
erties. A previous study [21] showed that the degradation of
343
orange II catalyzed by P25 follows the first-order kinetics model
with respect to orange II. Table 1 summarizes the first-order pho-
toreaction rate constant, K, at different catalyst dosages.
As can be seen in Table 1, the photoreaction rate increased with
the catalyst dosage. However, the increment was not linear in the
whole range of TiO2 dosage. For example, the photodegradation
rate was only enhanced slightly when the catalyst dosage was
high. The photoreaction rates were then normalized to the catalyst
dosages (K0 ) and the results are presented in Table 1 as well. It is
seen that the normalized photoreaction rate decreased with the in-
crease in catalyst dosage. According to our experimental results,
the most efficient catalyst dosage for photodegradation of orange
II was about 0.15 g/L, which was the lowest catalyst dosage studied
in this work. This is an indication of agglomeration of the nano-
sized P25 particles, which resulted in a decrease in the effective
surface area, thus the photocatalytic activity.
To confirm the effect of particle agglomeration on the photocat-
alytic activity of P25, a highly agglomerated P25 fiber sample was
prepared [22]. The morphologies of both P25 and TiO2 fiber sam-
ples were confirmed by the FESEM images shown in Fig. 1. The
photocatalytic activity of the TiO2 fiber sample was evaluated
and is also reported in Table 1. It can be seen that at the same cat-
alyst dosage, the TiO2 fiber showed a lower photoactivity than P25,
about a 20% decrease in comparison with P25. These comparative
results strongly suggest that with increase in solid dosage in the
suspension, spontaneous agglomeration occurred, resulting in the
formation of large TiO2 particles (like the TiO2 fibers), leading to
a loss in the effective surface area, and consequently a decrease
in photocatalytic activity.
The P25 particle sizes in suspension at different dosages were
monitored using the dynamic laser light scattering technique and
the results are shown in Fig. 2. As can be seen, the mean particle
sizes of P25 in suspensions at concentrations of 0.15 and 0.5 g/L
were about 407 and 1220 nm, respectively. It is obvious that the
particle size increased greatly with the increase in catalyst dosage.
This observation further confirmed that P25 nanoparticles agglom-
erated spontaneously in suspension, especially at high dosages.
In this work, two methods were attempted to minimize the
agglomeration of the P25 TiO2 nanoparticles. One was to change
the pH of the dispersion and the other one was to use polyelectro-
lyte to modify the surface of TiO2 nanoparticles. Depending on pH,
TiO2 particles can be rendered to carry either positive or negative
charges. If pH is below its isoelectric point (IEP), P25 particles will
be ionized to carry a positive charge while if pH is above its IEP P25
will be ionized to carry a negative charge as illustrated below:
þ
pH < IEP; TiOH þ Hþ ! TiOH2 ; ð1Þ

pH > IEP; TiOH þ OH ! TiO þ H2 O: ð2Þ
With a surface charge, P25 nanoparticles can be colloidally sta-
bilized because of the presence of electrostatic repulsive forces be-
tween particles. Hence, particle agglomeration can be minimized
Table 1
Photocatalytic reaction rate for decomposition of orange II with various catalyst
dosages.
Catalyst Photoreaction Normalized photoreaction
rate constant, rate constant,
K (min1) K0 (min1 g1 of TiO2)
P25 (0.15 g/L) 0.022 0.49
P25 (0.30 g/L) 0.03 0.33
P25 (0.50 g/L) 0.035 0.23
P25 (0.70 g/L) 0.04 0.19
P25 (1.0 g/L) 0.041 0.14
P25 fiber (0.5 g/L) 0.029 0.19
Experimental conditions: initial orange II concentration = 30 mg/L, UV light inten-
sity = 260 w/m2, neutral pH (7.1).
344 G. Li et al. / Journal of Colloid and Interface Science 348 (2010) 342–347
Fig. 1. FESEM images of P25 nanoparticles (A) and TiO2 nanofibers (B).
Fig. 2. Particle size distribution curves of P25 suspended under different condi-
tions: (a) at pH 7.1 with a concentration of 0.15 g/L; (b) at pH 7.1 with a
concentration of 0.5 g/L, (c) at pH 2.6 with a concentration of 0.15 g/L; (d) P25
(0.15 g/L) modified by PAH (6.67 mg/L).
or avoided. With consideration of the fact that orange II will disso-
ciate into negatively charged sulfonate ions in aqueous phase [21],
an acidic P25 suspension system with various pH values was used
to enable strong electrostatic interactions between positively
charged TiO2 particles and negatively charged sulfonate ions.
Fig. 2 shows the particle size distribution curves of P25 dis-
persed in water under different conditions. It can be seen that
the particle size is centered at about 407 nm at pH 7.1 (about neu-
tral pH conditions), more than 10 times larger than that of the par-
ticles in powder form (about 30–40 nm). This result confirmed the
agglomeration of TiO2 nanoparticles in aqueous suspension at
neutral pH. By adjusting the system pH to be about 2.6, the mean
particle size became about 79 nm, significantly smaller than that
under neutral pH conditions. Therefore, particle agglomeration
can be minimized by altering the system pH.
Fig. 3 compares the photocatalytic reaction rate constant at pH
7.1 (neutral conditions) with that at pH 2.6 (acidic conditions) in
the presence of different catalyst dosages. It is clearly seen that
TiO2 displayed a better performance under acid conditions than
it did under neutral conditions, indicating that altering the system
pH is an effective means for improving the photocatalytic activity
of TiO2 nanoparticles. To elucidate this observation, experiments
were conducted to study the dependence of dye adsorption and
photoactivity on pH.
Fig. 3. Effect of system pH on photocatalytic reaction rate at different catalyst
dosages. Experimental conditions: initial orange II concentration = 30 mg/L, UV
light intensity = 260 w/m2.
Fig. 4 depicts the adsorption amount of orange II on P25 at var-
ious pH values. It can be seen that the adsorption increased with
decreasing pH. This may be attributed to the electrostatic interac-
tions between the TiO2 particles and the orange II. As seen in Fig. 5
(curve b), the TiO2 particles are positively charged at system pH be-
low pHpzc and zeta potential increased with the decrease of system
pH. Therefore, as the pH was lowered the affinity between posi-
tively charged TiO2 particles and negatively charged sulfonate ions
of orange II increased, resulting in strong adsorption of orange II
over TiO2 particles. The curves c–g in Fig. 5 represents zeta poten-
tial trends of PAH-modified TiO2 particles and will be discussed in
the subsequent section.
The dependence of photodegradation rate on pH is shown in
Fig. 6. It is seen that the dependence followed the same trend as
the dependence of the adsorption of orange II on pH as shown in
Fig. 4. As the pH was increased from 2.29 to 6.15, the reaction rate
linearly decreased. The strong adsorption of orange II at low pH fa-
vored the mass transfer of orange II from the bulk to the catalyst
surface, leading to the enhancement of the photocatalytic activity.
Therefore, it can be concluded that the strong adsorption on the
TiO2 surface due to the strong electrostatic interactions at low sys-
tem pH significantly facilitated the photocatalytic reaction.
According to the theory of classical heterogeneous catalysis,
this conclusion implies that mass transfer from the bulk to the
catalyst surface is a rate-limiting step. However, it has been
reported that for photocatalytic reactions in a liquid phase, the
 G. Li et al. / Journal of Colloid and Interface Science 348 (2010) 342–347
Fig. 4. Dependence of adsorption of orange II on system pH. Experimental
conditions: initial orange II concentration = 50 mg/L, catalyst dosage = 0.5 g/L.
Fig. 5. Zeta potential profiles of P25 + PAH (6.67 mg/L) after 95 min UV illumination
(a), pure P25 (b), P25 + PAH (1 mg/L) (c), P25 + PAH (2 mg/L) (d), P25 + PAH
(3.33 mg/L) (e), P25 + PAH (5 mg/L) (f), and P25 + PAH (6.67 mg/L) (g).
photo-induced radicals may diffuse into the bulk liquid phase to
initiate the reaction [24,25], implying that the adsorption of orange
II on the P25 surface was not a necessary step in the photocatalytic
reaction process. To ascertain this, photodegradation of orange II
was conducted at pH 9.3. Under this pH condition, the surface of
P25 particles is negatively charged, thus adsorption of negatively
charged orange II ions would be strongly hindered because of the
Fig. 6. Dependence of photocatalytic rate constant on system pH. Experimental
conditions: initial orange II concentration = 50 mg/L, catalyst dosage = 0.5 g/L, UV
light intensity = 180 w/m2.
345
Fig. 7. Photodegradation of orange II over P25. Experimental conditions: initial
orange II concentration = 30 mg/L, catalyst dosage = 0.5 g/L, system pH 9.30, UV
light intensity = 260 w/m2.
strong electrostatic repulsion between the solid surface and the
dye ions. However, it is seen from Fig. 7 that at pH 9.3 photodegra-
dation of orange II still occurred despite at a lower reaction rate
than that under pH 2.6. According to this experimental observa-
tion, it can be hypothesized that photo-induced radicals indeed dif-
fused in the bulk liquid from the catalyst surface to initiate the
decomposition of orange II molecules [24,25]. However, based on
the fact that these radicals have very short life spans and are too
reactive to move far from surface it is reasonable to believe this dif-
fusion could only occur within a few nanometers away from cata-
lyst surface, which greatly reduces the reaction rate. In addition,
such low photoreaction rate can also be ascribed to the fact that
a Coulombic repulsion between the negatively charged surface of
TiO2 and the hydroxide anions could prevent the formation of
OH radical and thus decreased the photooxidation rate [26].
Based on observation from Figs. 6 and 7, two plausible reaction
pathways for photocatalytic decomposition of orange II over TiO2
dispersion are proposed, namely surface reaction over TiO2 parti-
cles and reaction in suspension due to radicals diffusion. It should
be noted that the former dominates the heterogeneous reaction
when the enrichment of reactants over the surface occurs whereas
the later could only happen at a lower rate within a limited dis-
tance from solid catalyst surface.
In addition to varying system pH, another method, namely
polyelectrolyte modification, was also applied to minimize sponta-
neous agglomeration and improve adsorption of orange II on TiO2
nanoparticles. As is seen from Fig. 5 (curves c–g), the surface of
Fig. 8. Adsorption of orange II on P25 at various PAH concentrations. Experimental
conditions: initial orange II concentration = 30 mg/L, solid dosage = 0.15 g/L, pH
6.5–8.5.
346 G. Li et al. / Journal of Colloid and Interface Science 348 (2010) 342–347
Fig. 9. Photocatalytic degradation of orange II over PAH-modified P25 photocata-
lyst. Experimental conditions: initial orange II concentration = 30 mg/L, catalyst
dosage = 0.15 mg/L, system pH 6.5–8.5, UV light intensity = 260 w/m2.
modified P25 was positively charged in the pH range studied. It is
also seen that zeta potential of P25 increased with the PAH concen-
tration under a given pH value, indicating that the portion of P25
surface covered by PAH increases with addition of PAH. Fig. 2
(curve d) showed that the high zeta potential after modification
with PAH can greatly reduce the particle agglomeration (about
115 nm). It should be noted that the minimizing of particle aggre-
gation was also due to steric stabilization derived from applied
polymer (PAH) [27]. Therefore, by coating P25 with PAH, two syn-
ergetic effects, namely surface charging and steric stabilization,
work together to prevent TiO2 particle aggregation.
Fig. 8 shows the dependence of the orange II adsorption over
PAH-modified P25 on the dosage of PAH. As expected, the modified
P25 showed strong adsorption capacity in comparison with the
bare P25 due to the electrostatic attraction. With the increase of
PAH concentration, there was a smooth improvement of the
adsorption. Therefore, it can be concluded that both polyelectro-
lyte modification of TiO2 and system pH variation are effective
ways to minimize the particle agglomeration and enhance the
adsorption.
The photocatalytic activity of PAH-modified P25 was also inves-
tigated and shown in Fig. 9. Surprisingly, the photoreaction rate
Fig. 10. Schematic representation of the dispersion of TiO2 nanoparticles under various experimental conditions.
decreased with the increase of PAH concentration. In addition, all
PAH-modified P25 showed a poorer photocatalytic activity in com-
parison with bare P25, although the former possessed a stronger
adsorption capacity than the latter. This result is inconsistent with
the one derived from the study of varying system pH that the
strong adsorption could accelerate the photoreaction rate. One
plausible reason is that for the PAH-modified P25 the polymer
layer attached on the P25 surface might impede the migration of
photo-induced electrons to target pollutants (orange II). The
photo-induced electrons were consumed by the decomposition of
PAH layer before they reached the organic dye molecule. As a re-
sult, the polymer layer was decomposed and the amount of elec-
trons which participated in the photodegradation of organic
pollutants decreased. Subsequently, the overall photoreaction effi-
ciency was lowered. This hypothesis was further confirmed by the
zeta potential results shown in Fig. 5 (see curve a). After 95-min UV
illumination of PAH-modified P25 (without orange II), the zeta po-
tential profile of modified P25 became similar to that of bare P25,
indicating that the PAH layer was completely decomposed during
UV illumination.
It should be highlighted that at low PAH dosage, the P25 surface
was not completely covered by PAH as shown in Fig. 5; therefore,
another possible path for degradation of orange II by PAH-modified
P25 is that the photo-induced electrons might directly diffuse to
the adsorbed orange II molecules from the area which was not cov-
ered by PAH. However, in the case of high PAH dosage, the surface
was almost fully modified by PAH, thus, the electrons need to cross
through the PAH layer and then reach orange II molecules, result-
ing in the low photoreaction rate. As a result, the photoreaction
rate for degradation of orange II over PAH-modified P25 decreased
with the increase of PAH concentration. Fig. 10 summarizes the
proposed pathways for photo-induced electron transfer over
PAH-modified P25 and dispersion status of P25 nanoparticles un-
der various experiment conditions.
4. Conclusions
Spontaneous agglomeration of nano-sized TiO2 nanoparticles of
different concentrations was evaluated, to investigate the effect of
particle on photoactivity. By controlling the system pH, the zeta
potential of the TiO2 suspension was varied and particle agglomer-
 G. Li et al. / Journal of Colloid and Interface Science 348 (2010) 342–347
ation was significantly minimized because of the strong electro-
static repulsions between colloidal TiO2 particles. Adsorption of or-
ange II on the surface of TiO2 nanoparticles was found to be
strengthened at low pH values because of the electrostatic attrac-
tion between the dye and the TiO2 particles. This in turn acceler-
ated the photoreaction rate. It was also experimentally observed
that photo-induced electrons can diffuse into the suspension, thus
participating in photoreactions. Modification of TiO2 nanoparticles
by polyelectrolyte PAH can greatly abate particle agglomeration,
thus improving the adsorption of orange II dye. However, the over-
all photocatalytic reaction efficiency decreased in comparison with
the TiO2 nanoparticles before modification because the polyelec-
trolyte layer retarded the migration of electrons from the bulk to
the surface.
Acknowledgment
Financial support from Environment and Water Industry Devel-
opment Council (EWI) of Singapore under project number MEWR
651/06/161 is acknowledged.
References
[1] Y. Gao, R. Wahi, A.T. Kan, J.C. Falkner, V.L. Colvin, A.B. Tomson, Langmuir 20
(2004) 9585.
[2] D.E. Giammar, C.J. Maus, L.Y. Xie, Environ. Eng. Sci. 24 (2007) 85.
[3] H.Z. Zhang, R.L. Penn, R.J. Hamers, J.F. Banfield, J. Phys. Chem. B 103 (1999)
4656.
347
[4] U.I. Gaya, A.H. Abdullah, J. Photochem. Photobiol. C 9 (2008) 1.
[5] M. Pena, X.G. Meng, G.P. Korfiatis, C.Y. Jing, Environ. Sci. Technol. 40 (2006)
1257.
[6] R.A. French, A.R. Jacobson, B. Kim, S.L. Isley, R. Lee Penn, P.C. Baveye, Environ.
Sci. Technol. 43 (2009) 1354.
[7] Z.M. Yaremko, N.H. Tkachenko, C. Bellmann, A. Pich, J. Colloid Interface Sci. 296
(2006) 565.
[8] Z.M. Yaremko, D.M. Nikipanchuk, L.B. Fedushinskaya, I.G. Uspenskaya, Colloid
J. 63 (2001) 253.
[9] Z.M. Yaremko, L.B. Fedushinska, D.M. Nikipanchuk, Colloid J. 57 (1995) 249.
[10] B.M. Rothen-Rutishauser, S. Schurch, B. Haenni, N. Kapp, P. Gehr, Environ. Sci.
Technol. 40 (2006) 4353.
[11] P.C. Baveye, M. Laba, Environ. Health Perspect. 116 (2008) A152.
[12] K.A.D. Guzman, M.P. Finnegan, J.F. Banfield, Environ. Sci. Technol. 40 (2006)
7688.
[13] N.G. Hoogeveen, M.A.C. Stuart, G.J. Fleer, J. Colloid Interface Sci. 182 (1996)
133.
[14] B.J. Ninness, D.W. Bousfield, C.P. Tripp, Colloid Surf. A 203 (2002) 21.
[15] H.Y. Li, C.P. Tripp, Langmuir 20 (2004) 10526.
[16] T. Sato, S. Kohnosu, J. Colloid Interface Sci. 143 (1991) 434.
[17] S. Lebrette, C. Pagnoux, P. Abélard, J. Colloid Interface Sci. 280 (2004) 400.
[18] Q.Y. Li, Y.F. Chen, D.D. Zeng, W.M. Gao, Z.J. Wu, J. Nanopart. Res. 7 (2005)
295.
[19] F. Mange, P. Couchot, A. Foissy, A. Pierre, J. Colloid Interface Sci. 159
(1993) 58.
[20] I.N. Martyanov, E.N. Savinov, K.J. Klabunde, J. Colloid Interface Sci. 267 (2003)
111.
[21] G. Li, X.S. Zhao, M.B. Ray, Sep. Purif. Technol. 55 (2007) 91.
[22] Y.X. Yu, D.S. Xu, Appl. Catal. B 73 (2007) 166.
[23] A. Bhattacharyya, S. Kawi, M.B. Ray, Catal. Today 98 (2004) 431.
[24] C.S. Turchi, D.F. Ollis, J. Catal. 122 (1990) 178.
[25] M.W. Peterson, J.A. Turner, A.J. Nozik, J. Phys. Chem. 95 (1991) 221.
[26] I.K. Konstantinou, T.A. Albanis, Appl. Catal. B 49 (2004) 1.
[27] J.B. Hooper, K.S. Schweizer, Macromolecules 38 (2005) 8858.