Surface modification of titania nanoparticles using ultrathin ceramic films

Article abstract of DOI:10.1111/j.1551-2916.2006.01216.x

Surface modification of titania nanoparticles was achieved by coating them with ultrathin alumina films using atomic layer deposition. The coating process was performed in a fluidized bed reactor at low pressure and under mechanical vibration. Films were deposited using self-limiting, sequential surface reactions of trimethylaluminum and water. The composition of aluminacoated particles was verified using infrared spectroscopy. The deposited films had an average growth rate of 0.2 nm/coating cycle and were highly uniform and conformal as demonstrated by transmission electron microscopy and X-ray spectroscopy. Deposited alumina films were amorphous as verified through X-ray diffraction and high-resolution electron microscopy. The coating process did not promote particle sintering as validated via particle size and surface area analysis.

Full text of DOI:10.1111/j.1551-2916.2006.01216.x

J. Am. Ceram. Soc., 89 [10] 3070–3075 (2006)
DOI: 10.1111/j.1551-2916.2006.01216.x
r 2006 The American Ceramic Society
ournal
J
Surface Modification of Titania Nanoparticles Using Ultrathin
Ceramic Films
,w
Luis F. Hakim,* Jarod A. McCormick, Guo-Dong Zhan,* and Alan W. Weimer*
Department of Chemical and Biological Engineering, University of Colorado, Boulder, Colorado 80309
Peng Li*
Department of Earth and Planetary Sciences, University of New Mexico, Albuquerque, New Mexico 87131
Steven M. George
Department of Chemistry and Biochemistry, University of Colorado, Boulder, Colorado 80309
Surface modification of titania nanoparticles was achieved by
coating them with ultrathin alumina films using atomic layer
deposition. The coating process was performed in a fluidized bed
reactor at low pressure and under mechanical vibration. Films
were deposited using self-limiting, sequential surface reactions
of trimethylaluminum and water. The composition of alumina-
coated particles was verified using infrared spectroscopy. The
deposited films had an average growth rate of 0.2 nm/coating
cycle and were highly uniform and conformal as demonstrated
by transmission electron microscopy and X-ray spectroscopy.
Deposited alumina films were amorphous as verified through
X-ray diffraction and high-resolution electron microscopy. The
coating process did not promote particle sintering as validated
via particle size and surface area analysis.
There is therefore a need to modify the surface of individual
nanoparticles while keeping their properties unchanged. This
goal can be achieved by conformally coating particles with ul-
trathin ceramic films using atomic layer deposition (ALD) in a
fluidized bed reactor.
The use of coating techniques such as chemical vapor depo-
sition (CVD) and wet chemistry to deposit ultrathin films on
primary particles is difficult. In a CVD reaction, the chemical
6
species are fed simultaneously in the gas phase where they may
homogeneously react and form particulates that can deposit
onto the substrate particles. A CVD reaction is dependent upon
pressure, temperature, reaction time, and exposure, and occurs
7
,8
continuously with no self-control. Different CVD technolo-
However,
9
–14
gies have been successfully applied to particles.
truly uniform and conformal films on individual particles have
not been achieved mainly due to particle aggregation and due to
a lack of control over the film thickness.
I. Introduction
ALD provides an ideal method for depositing ultrathin films
on high aspect ratio surfaces as it is independent of line of
ITANIA is an excellent absorber of ultraviolet radiation and is
commonly used in sunscreen materials. Owing to their high
photocatalytic activity, titania particles may react with the skin
1
T
in the presence of UV light and cause severe damage. Organic
15–17
sight.
ALD uses sequential surface chemical reactions to de-
2
posit highly conformal films with a precise control at the atomic
scale. ALD reactions take place only after precursors have been
adsorbed on the substrate. These reactions are also self-limiting
and self-terminating as adsorption of gas phase species occurs
only while active sites are available on the particle surface.
coatings are commonly used as barriers to prevent direct contact
with the skin. However, these organic materials have a limited
lifetime as they can be decomposed by the photoactive titania
substrate. It has been proposed that conformal, nano-thick ce-
ramic films can be used to control the photocatalytic activity of
titania particles. For this application, it has to be assured that
the deposited films do not adversely affect the optical properties
The use of a progression of surface reactions to deposit
18–23
alumina films has been demonstrated.
methylaluminum [Al(CH ) ] (TMA) are used alternately in an
Water vapor and tri-
3 3
3
ABAB y binary reaction sequence. Film deposition is based on
the following binary CVD reaction
of the titania substrate.
Titania particles are also one of the most important fillers for
polymers in rubbers and thermoplastics applications. In many
cases, surfactant treatment of titania particles is necessary in or-
2
AlðCH3Þ þ 3H2O ! Al2O3 þ 6CH4
(1)
3
4
der to increase dispersability into the polymeric matrix. Never-
Alumina ALD is achieved by splitting this reaction into two
separate half-reactions
theless, the interaction between the particle and these surfactants
is only physical in nature and has a potentially low stability. It has
been shown that an ultrathin film of alumina deposited on ce-
ramic particles can improve their rheological behavior by mod-
ifying interactions between the particles and an epoxy resin. The
deposited alumina films should be extremely thin in order to as-
20
ꢀ
2AlOH þ 2AlðCH3Þ
3
ðAÞ
ðBÞ
(2)
(3)
ꢀ
!
2½AlOAlðCH Þ ꢁ þ 2CH4
3
2
5
sure that the thermal conductivity of the matrix is not reduced.
ꢀ
2
½AlOAlðCH3Þ ꢁ þ 3H2O
! Al2O3 þ 2AlOH þ 4CH4
2
ꢀ
W. Mullins—contributing editor
where asterisks represent the surface species. The sequential ex-
posures of TMA and water represent one coating cycle that de-
posits a complete monolayer of alumina. The thickness of this
Manuscript No. 21206. Received December 1, 2005; approved February 13, 2006.
Supported by the National Science Foundation, under Grant No. DMI-0400-292 and
the GAANN Program in Functional Materials, U.S. Department of Education.
2
4
monolayer is approximately 0.11 nm at 450 K.
Particles have been successfully coated with ultrathin films at
a small scale for several ALD reactions. Bulk quantities of
w
Author to whom correspondence should be addressed. e-mail: alan.weimer@colorado.
edu
2
5
*Member, American Ceramic Society.
3
070

October 2006
Surface Modification of Titania Nanoparticles
3071
To
pump
1
2
7
N2
supply
6
5
3
4
8
8
Fig. 1. Schematic of atomic layer deposition fluidized bed reactor apparatus: (1) fluidization column, (2) furnace, (3) vibro-motors, (4) spring supports,
5) pressure transducers, (6) pneumatic valves, (7) mass flow controller, (8) reactant containers.
(
s
metal and ceramic micron-sized particles have been coated with
alumina using ALD in a fluidized bed reactor.
As the use of nanoparticles increases, the need for large-scale
bulk processing also increases. The present study is the first
successful attempt to coat individual titania nanoparticles con-
formally at a large scale using ALD in a fluidized bed reactor.
¨
Titania particles (Aeroxide P-25, Degussa AG, Dusseldorf,
Germany) with a primary particle size of 21 nm and a BET sur-
5
,26
2
face area of 50715 m /g were used for this study. The bulk den-
sity of the powder was about 130 kg/m , and the skeletal density
3
3
of the primary particles was approximately 4200 kg/m . High ag-
gregation tendency and some sintering were observed in the un-
coated sample using a transmission electron microscope (TEM)
(Model CM10, Philips, Eindhoven, the Netherlands) as shown in
Fig. 2. Before loading, powders were sieved using a 100 mesh (150
mm) screen. Particles were then dried at 400 K and 70 Pa for 2 h
inside the reactor and under a continuous flow of nitrogen.
The FBR was encased by a clamshell-type furnace (Therm-
craft, Winston-Salem, NC). The reaction temperature was 450
K. The feeding lines were also heated to approximately 350 K.
Trimethylaluminum (Sigma Aldrich, St. Louis, MO) and deion-
ized water were used as reactants. These precursors were fed via
their vapor pressures and the system was maintained at low
pressure (B130 Pa).
During each coating cycle, the reactants were fed for sufficient
time so saturation of all active sites occurred for every dose. This
time was calculated based upon the surface area of the particles
and the sample mass utilized for each batch. After each reactant
dose, the system was flushed with inert gas to eliminate unreacted
species as well as any methane formed during the reaction.
II. Experimental Procedure
Titania nanoparticles were coated using an ALD fluidized bed
reactor (ALD-FBR). A schematic of the experimental apparatus
is shown in Fig. 1. The reactor is a stainless-steel column with a
2
0 mm pore size porous metal disc as the gas distributor. A po-
rous stainless-steel filter element (0.5 mm pore size) was used at
the inside top of the reactor column to retain the particles in the
reactor at all times.
The system was capable of operating at reduced pressures and
under mechanical vibration. Reduction of pressure in the reac-
tor was achieved by controlling the opening of a diaphragm
valve placed at the inlet of a vacuum pump (Model 2063, Al-
catel, Paris, France). Two synchronized electric vibrators (Mod-
el CD36210, Martin Engineering, Marine City, MI) were
utilized to vibrate the column. A vibration frequency of 20 Hz
and an amplitude of 3.3 mm were used. Four spring supports on
the platform imparted homogeneous vertical vibration to the
entire system.
III. Results
The fluidizing gas was high-purity nitrogen. Its flow rate was
controlled by a mass flow controller (Model 1179, MKS, Boul-
der, CO). Piezoelectric transducers (Model 902, MKS) were lo-
cated below the distributor plate and at the outlet of the
fluidization column to measure the pressure drop across the
bed of powder and determine minimum fluidization velocities.
(
1) Fluidization and Aggregation Characteristics
Understanding the fluidization of titania nanoparticles is a key
step in perfecting the coating process. In general, nanoparticles
fluidize as aggregates of several hundreds of microns in size due
2
7–31
to high London–Van der Waals interactions.

3
072
Journal of the American Ceramic Society—Hakim et al.
Vol. 89, No. 10
2
2
1
1
0
0
.5
.1
.7
.3
.9
.5
1
400
1300
1200
1100
1000
900
800
700
Frequency [cm 1
−
]
Fig. 4. Fourier transform infrared spectra of uncoated, reference alu-
mina powder, and alumina-coated titania nanoparticles after 50 and 70
cycles.
inspection, the growth rate is calculated to be in the range of
0
films may vary with the size and geometry of the substrate. For
surfaces with a high aspect ratio, less steric effects occur and the
available active sites are more easily exposed to the gas precur-
sors during the adsorption step of the reaction.
.19–0.21 nm per coating cycle. The growth rate of the ALD
Fig. 2. Transmission electron microscopy image of uncoated titania
nanoparticles.
The fluidization behavior of nanoparticles is determined by
the properties of aggregates, rather than by those of primary
particles. Fluidization properties such as bed expansion and
minimum fluidization velocity are dictated by aggregate prop-
erties such as size, density, sphericity, and surface roughness.
A fluidized bed offers great advantages such as a constant
recirculation of solids and an excellent fluid–solid contacting. A
fluidized bed also has high heat and mass transfer coefficients as
well as a rapid solids mixing. In an ALD system, this charac-
teristic helps to achieve fast saturation of active sites on the
particle surface. Therefore, a fluidized state must be assured at
all times during processing. During fluidization experiments, the
pressure drop across the bed of particles always reached a con-
stant value when the superficial gas velocity was increased,
The TEM image shows extremely conformal and uniform
films around primary particles. The contrast between the film
and the particle substrate is given by the difference in density
between alumina and titania. For alumina ALD films, the den-
3
24
sity has been determined to be approximately 3.5 g/cm ,
whereas the skeletal density of the primary titania particles is
3
about 4.2 g/cm . Therefore, the lighter alumina gives a brighter
contrast in the TEM image. The distinction between films de-
posited on different particles is also clear as no sintering between
particles is observed. Even though aggregation of particles may
occur, the interactions would be physical in nature and not due
to the coating process.
The composition of the particles after the coating process was
analyzed using Fourier transform infrared (FTIR) spectroscopy.
An FTIR spectrometer (Model 750 Magna-IR, Thermo Nicolet,
Waltham, MA) was used for this analysis. The titania particles
were supported by a tungsten photoetched grid (Buckbee Mears,
St. Paul, MN). The tungsten grid (50.8 mm thick with 3.9 lines/
mm) had dimensions of approximately 2 cm ꢂ 2 cm. The uncoated
and alumina-coated titania particles were pressed separately into
3
2
which is a typical indication of fluidization. This shows that
titania nanoparticles fluidize smoothly despite their small size
and high aggregation tendency and that a fluidized bed is very
suitable for ALD processing.
(2) Characterization of Coated Nanoparticles
25
TEM was used to analyze the conformality of alumina films as
well as to measure their thickness approximately. An image of
coated particles after 70 cycles is shown in Fig. 3. From visual
the grid using a manual press and polished stainless-steel dies.
The vibrational mode of amorphous alumina films has been
ꢃ1 33
As shown
located at a frequency of approximately 930 cm
.
in Fig. 4, no features appear at this frequency for the uncoated
titania sample. Absorption bands appear at this frequency for
coated particles after 50 and 70 cycles. This is a direct confir-
mation of the composition of the alumina films. As an addi-
tional reference, the vibrational mode of alumina powder (A-16
SG, Alcoa Chemicals, Point Comfort, TX) was included in
Fig. 4. Absorption bands of alumina particles have been ob-
ꢃ1
34
served in the 850–1050 cm range, which overlaps with the
1
0
0
0
0
0
.0
.8
.6
.4
.2
.0
1
000
800
600
400
200
0
Binding Energy [eV]
Fig. 3. Transmission electron microscopy image of alumina-coated ti-
tania nanoparticles after 70 coating cycles.
Fig. 5. X-ray photoelectron spectroscopy spectrum of alumina-coated
titania nanoparticles after 50 cycles.

October 2006
Surface Modification of Titania Nanoparticles
3073
region where amorphous alumina films have been observed. The
alumina spectrum for coated particles matches the one of the
reference alumina powder and no features from the titania spec-
trum appear. This behavior is expected for uniformly coated
particles.
The uniformity and composition of the deposited alumina
films were further tested by X-ray photoelectron spectroscopy
Coated (70 cycles)
2
5
(
XPS). An XPS system (Model PHI 5600, Physical Electronics,
Uncoated Titania
Chanhassen, MN) with a high-energy resolution analyzer
was used for this study. The analysis was performed using an
aluminum source, a pass energy of 187.8 eV, and an energy
step of 0.2 eV.
2
0
25
30
35
40
45
50
55
60
65
70
2
θ [degree]
As shown in Fig. 5, the highest intensity peaks for TiO
2
(2p,
58.5 eV) and Ti (2p, 453.8 eV) have been completely atten-
3
5
4
Fig. 7. X-ray diffraction spectra of uncoated and alumina-coated ti-
tania nanoparticles after 70 cycles ( , anatase; m, rutile).
uated as expected for a uniform film. Oxygen and aluminum
features derive from the alumina film deposited on the particles.
Carbon features correspond to background surface carbon.
The atomic structure of the deposited alumina films on titania
nanoparticles was analyzed using a 200 keV field emission TEM
ꢄ
As shown in Fig. 7, the relative intensity of some peaks ap-
pears slightly attenuated. However, the spectrum of uncoated
titania particles remains fairly unchanged after the coating proc-
ess, which suggests that the deposited alumina films have an
amorphous structure. Previous studies have also confirmed the
(
Model 2010F, JEOL, Tokyo, Japan). For this analysis, the
coated particles were dispersed and embedded in a polymeric
resin. Once solidified, cross-section TEM samples were made by
cutting across the polymeric matrix to expose the surface of in-
dividually coated particles. Figure 6 shows a representative high-
resolution electron microscopy (HREM) image of the cross-sec-
tion sample. The atomic fringes (indicated in the image by a
white arrow) demonstrate the crystalline character of an indi-
vidual titania nanoparticle. A very conformal alumina film en-
capsulates this nanoparticle. No atomic fringe is observed inside
the alumina film, indicating an amorphous structure.
1
8
amorphous nature of alumina ALD films.
The XRD spectra also show that the titania nanoparticles
have a mixture of anatase and rutile phases. From product
specifications, the ratio of anatase to rutile is approximately
80:20. These kinds of titania particles have been used commonly
as it has been shown that they yield better photocatalytic activity
3
6
than samples with just one crystalline state.
Another important remark is that the coated titania particle
shown in Fig. 6 has an approximate size of only 6 nm. Addi-
tionally, the thickness of the deposited alumina film is very uni-
form throughout the particle surface. This is further evidence
that ALD can coat extremely small structures with precise con-
trol over the film thickness.
Further testing of the structure of alumina films was per-
formed using X-ray diffraction (XRD). Figure 7 shows XRD
spectra of titania nanoparticles before and after 70 coating cy-
cles. The weight concentration of alumina on the coated parti-
cles is several times the minimum detection limit of the analytical
system. This means that if crystalline alumina is present, it
would be easily detected.
(
3) Particle Size and Surface Area Analysis
Owing to strong London–Van der Waals interactions, nanopar-
ticles fluidize as aggregates of several hundreds of microns in
size.
2
7–31
A drastic decrease in the specific surface area of the
nanopowders would occur if entire aggregates were coated, as
narrow interstices inside the aggregates would be filled. Never-
theless, even when particles are individually coated, their surface
area is expected to vary as the thickness of the film increases.
This change is due to the slight increase in the particle diameter
but most importantly, due to the change in the effective particle
density. The prediction in the change of specific surface area for
three different sizes of titania particles as the thickness of the
alumina film increases is shown in Fig. 8.
The surface area of titania particles was measured before and
after 50 coating cycles using a gas physisorption analyzer (Mod-
s
el Autosorb -1, Quantachrome, Boynton Beach, FL). The ex-
perimental surface areas for uncoated and coated samples are
49.071.1 and 30.570.4, respectively. The prediction presented
before shows an expected change of 43.0% in the specific surface
area after 50 cycles. The experimental change was 37.7%, which
is lower that the predicted value, showing that nanoparticles are
not being bonded together during the coating process.
The prediction shown in Fig. 8 assumes spherical particles.
Conversely, the TEM image shown in Fig. 2 demonstrates that
8
7
6
5
4
3
2
0
0
0
0
0
0
0
1
0
Particle Size
2
4
1
1 nm
0 nm
00 nm
0
4
8
12
16
20
Film Thickness [nm]
Fig. 6. Cross-section high-resolution electron microscopy image of an
alumina-coated titania nanoparticle.
Fig. 8. Prediction of change in specific surface area with increasing film
thickness for different primary particle sizes.

3
074
Journal of the American Ceramic Society—Hakim et al.
Vol. 89, No. 10
allowed the processing of bulk quantities of nanopowders.
The composition of the alumina films was confirmed by infra-
red spectroscopy and XPS, and their amorphous structure was
verified by HREM imaging and XRD. Alumina films had an
average growth rate of 0.2 nm/coating cycle. Highly conformal
and uniform films were observed via TEM. The coating process
did not promote particle sintering due to the dynamic aggrega-
tion behavior of nanoparticles and the self-limiting characteris-
tics of ALD.
Coated (50 cycles)
Uncoated Titania
0
0
0
0
0
.08
.06
.04
.02
.00
0
3
6
9
12
15
18
Size [microns]
References
Fig. 9. Particle size distribution of uncoated and alumina-coated par-
ticles after 50 cycles.
1
W. E. Vargas, P. Greenwood, J. E. Otterstedt, and G. A. Niklasson, ‘‘Light
Scattering in Pigmented Coatings: Experiments and Theory,’’ Sol. Energy, 68, 553
(
2000).
2
titania nanoparticles are not perfectly spherical. This is explained
by the sintering process that may occur during the aerosol man-
ufacture of particles. Deviations from the predicted change in
surface area can be attributed to this precoating condition.
In addition to the surface area, the particle size distribution of
titania samples before and after coating was measured. Figure 9
N. Serpone, A. Salinaro, S. Horikoshi, and H. Hidaka, ‘‘Beneficial Effects of
Photo-Inactive Titanium Dioxide Specimens on Plasmid DNA, Human Cells and
Yeast Cells Exposed to UVA/UVB Simulated Sunlight,’’ J. Photochem. Photobiol.
A: Chem, 179, 200 (2006).
3
D. M. King, J. A. McCormick, L. F. Hakim, G. D. Zhan, S. M. George, and
A. W. Weimer, ‘‘Benign Nano-Thin Film Composite Particles for Protection from
UVA–UVB—Rays’’; Presented at 2005 AIChE Annual Meeting. Cincinnati, OH,
October 30–Novembber 4, 2005.
shows the results obtained using a particle size analyzer (Model
225 Aerosizer , TSI, Shoreview, MN). As shown, the particle
4
s
M. Arellano, I. Manas-Zloczower, and D. L. Feke, ‘‘Effect of Surfactant
Treatment on the Formation of Bound Polymer on Titanium Dioxide Powders,’’
3
size volume distribution remains fairly unchanged after 50 coat-
ing cycles. It is important to mention that this system only de-
tects particles in the micron-sized range and above. Therefore,
individual nanoparticles do not appear in this distribution. The
measured aggregates of particles form due to high interparticle
forces for both uncoated and coated samples. This aggregation
is then physical in nature and not due to the coating. If sintering
was promoted during the coating process, a large shift to the
right for the distribution of coated particles would be expected.
Powder Technol., 84, 117 (1995).
J. R. Wank, S. M. George, and A. W. Weimer, ‘‘Nanocoating Individual Co-
5
hesive Boron Nitride Particles in a Fluidized Bed by ALD,’’ Powder Technol., 142,
5
9–69 (2004).
J. P. Dan, H. J. Boving, and H. E. Hintermann, ‘‘Hard Coatings,’’ J. Phys. IV,
6
3, 933–41 (1993).
B. Hong, J. Lee, R. W. Collins, Y. Kuang, W. Drawl, R. Messier, T. T. Tsong,
7
and Y. E. Strausser, ‘‘Effects of Processing Conditions on the Growth of Nano-
crystalline Diamond Thin Films: Real Time Spectroscopic Ellipsometry Studies,’’
Diamond Relat. Mater., 6, 55–80 (1997).
8
B. K. Kellerman, E. Chason, D. P. Adams, T. M. Mayer, and J. M. White,
‘Indashsitu X-Ray Reflectivity Investigation of Growth and Surface Morphology
‘
Evolution During Fe Chemical Vapor Deposition on Si(001),’’ Surf. Sci., 375,
3
31–9 (1997).
M. Karches, M. Morstein, and P. Rudolf von Rohr, ‘‘Encapsulation of
9
IV. Discussion
Abrasive Particles by Plasma CVD,’’ Surf. Coatings Technol., 169–170, 544–8
2003).
J. Kim, G. Y. Han, and C.-H. Chung, ‘‘Encapsulating the Electroluminescent
One of the main advantages that ALD has over other coating
techniques is its minimum film thickness. As shown before, ce-
ramic films of only a few Angstroms in thickness can be depos-
ited using this technique. When more traditional coating
methods such as wet chemistry and CVD are used, this mini-
mum film thickness is several orders of magnitude higher. When
nanoparticles aggregate, the separation distance between them is
very small. If ALD is used, complete monolayers can be depos-
ited without promoting particle sintering. When traditional
methods are utilized, the amount of material deposited is too
large, relative to the interparticle separation distance. This proc-
ess creates a bridging effect between particles and promotes ir-
reversible changes to the particle size distribution.
(
10
Phosphor Micro-Particles Using a Pulsed Metal-Organic Chemical Vapor Depo-
sition Process in a Fluidized Bed,’’ Thin Solid Films, 409, 58–65 (2002).
11
I. Sanchez, G. Flamant, D. Gauthier, R. Flamand, J. M. Badie, and G. Ma-
zza, ‘‘Plasma-Enhanced Chemical Vapor Deposition of Nitrides on Fluidized Par-
ticles,’’ Powder Technol., 120, 134–40 (2001).
12
H. S. Shin and D. G. Goodwin, ‘‘Deposition of Diamond Coatings on Par-
ticles in a Microwave Plasma-Enhanced Fluidized Bed Reactor,’’ Mater. Lett., 19,
1
19–22 (1994).
V. V. Srdic, M. Winterer, G. Miehe, and H. Hahn, ‘‘Different Zirconia–Alu-
13
mina Nanopowders by Modifications of Chemical Vapor Synthesis,’’ Nanostruct.
Mater., 12, 95 (1999).
D. Vollath and D. V. Szabo, ‘‘Nanocoated Particles: A Special Type of Ce-
14
ramic Powder,’’ Nanostruct. Mater., 4, 927 (1994).
S. M. George, A. W. Ott, and J. W. Klaus, ‘‘Surface Chemistry for Atomic
Layer Growth,’’ J. Phys. Chem., 100, 13121–31 (1996).
15
Aggregation of cohesive particles is dependent on flow con-
ditions as well as on the external energy that is transferred to
particles during processing. Therefore, nanoparticles will show
diverse aggregation patterns in different reactor configurations.
Owing to the high level of solids mixing in a fluidized bed, a
behavior called dynamic aggregation is observed during nano-
16
C. Goodman and M. Pessa, ‘‘Atomic Layer Epitaxy,’’ J. Appl. Phys., 60, R65–
1 (1986).
T. Suntola, ‘‘Atomic Layer Epitaxy,’’ Thin Solid Films, 216, 84–9 (1992).
8
17
18
A. C. Dillon, A. W. Ott, J. D. Way, and S. M. George, ‘‘Surface Chemistry of
Al O Deposition Using Al(CH ) and H O in a Binary Reaction Sequence,’’ Surf.
Sci., 322, 230–42 (1995).
2
3
3 3
2
19
2
7
J. F. Fan, K. Sugioka, and K. Toyoda, ‘‘Low-Temperature Growth of Thin-
Films of Al by Sequential Surface Chemical-Reaction of Trimethylaluminum
and H ,’’ Jpn. J. Appl. Phys. Part 2—Lett., 30, 1139–41 (1991).
G. S. Higashi and C. G. Fleming, ‘‘Sequential Surface Chemical-Reaction
Limited Growth Of High-Quality Al Dielectrics,’’ Appl. Phys. Lett., 55, 1963–5
1989).
particle fluidization. In this process, outer edges of fluidizing
aggregates are shed and picked up by other aggregates. Even
though this breakage is partial, due to frequent collisions and
continuous recirculation of solids, the entire surface of all the
particles will be exposed to the gaseous precursors. This mech-
anism allows for coating individual nanoparticles.
2 3
O
2 2
O
20
2 3
O
(
21
H. Kumagai and K. Toyoda, ‘‘In Situ Ellipsometric Diagnostics for Control-
led Growth of Metal Oxides with Surface Chemical Reactions,’’ Appl. Surf. Sci.,
82–83, 481–6 (1994).
ALD is the only known method capable of coating individual
particles with carefully controlled conformal films. Conformal,
ultrathin ceramic films allow for modifying the particle surface
properties while keeping their bulk characteristics unchanged.
Other methods have intrinsic limitations to achieve this objective.
22
M. Ritala, H. Saloniemi, M. Leskela, T. Prohaska, G. Friedbacher, and M.
Grasserbauer, ‘‘Studies on the Morphology of Al Thin Films Grown by Atom-
ic Layer Epitaxy,’’ Thin Solid Films, 286, 54–8 (1996).
2 3
O
23
C. Soto and W. T. Tysoe, ‘‘The Reaction Pathway for the Growth of
Alumina on High Surface-Area Alumina and in Ultrahigh-Vacuum by
Reaction Between Trimethyl Aluminum and Water,’’ J. Vac. Sci. Technol. A, 9,
686–95 (1991).
a
2
24
A. W. Ott, J. W. Klaus, J. M. Johnson, and S. M. George, ‘‘Al
2 3
O Thin Film
V. Conclusions
Growth on Si(100) Using Binary Reaction Sequency Chemistry,’’ Thin Solid Films,
2
92, 135–44 (1997).
J. D. Ferguson, A. W. Weimer, and S. M. George, ‘‘Atomic Layer Deposition
This work represents the first successful attempt to coat indi-
vidual titania nanoparticles conformally at a large scale using
ALD in a fluidized bed reactor. The proposed technology
25
of Al O and SiO2 on BN Particles Using Sequential Surface Reactions,’’ Appl.
2
3
Surf. Sci., 162–163, 280 (2000).

October 2006
Surface Modification of Titania Nanoparticles
3075
2
6
31
J. R. Wank, S. M. George, and A. W. Weimer, ‘‘Coating Fine Nickel Particles
with Al Utilizing an Atomic Layer Deposition-Fluidized Bed Reactor (ALD-
FBR),’’ J. Am. Ceram. Soc., 87, 762–5 (2004).
W. Yao, G. Guangsheng, W. Fei, and W. Jun, ‘‘Fluidization and Agglomerate
Nanoparticles,’’ Powder Technol., 124, 152–9 (2002).
D. Kunii and O. Levenspiel, Fluidization Engineering. John Wiley and Sons,
New York, 1969.
O
2 3
Structure of SiO
2
32
27
L. F. Hakim, J. L. Portman, M. D. Casper, and A. W. Weimer, ‘‘Aggregation
Behavior of Nanoparticles in Fluidized Beds,’’ Powder Technol., 160, 149 (2005).
33
J. D. Ferguson, A. W. Weimer, and S. M. George, ‘‘Atomic Layer Deposition
28
S. Matsuda, H. Hatano, and A. Tsutsumi, ‘‘Ultrafine Particle Fluidization and
its Application to Photocatalytic NO Treatment,’’ Chem. Eng. J., 82, 183–8 (2001).
C. H. Nam, R. Pfeffer, and R. N. Dave, ‘‘Fluidization of Nanoparticles’’;
2 3
of Ultrathin and Conformal Al O Films on BN Particles,’’ Thin Solid Films, 371,
x
95–104 (2000).
29
34
L. Yang and D. J. Watts, ‘‘Particle Surface Characteristics May Play an Impor-
tant Role in Phytotoxicity of Alumina Nanoparticles,’’ Toxicol. Lett., 158, 122 (2005).
pp. 1–8 in Fluidization XI, 11th International Conference on Fluidization. 2004
Engineering Conferences International, Naples, Italy, Edited by U. Arena, R.
Chirone, M. Miccio, and P. Salatino. Engineering Conferences International,
Brooklyn, NY, 2004.
35
C. D. Wagner, W. M. Riggs, L. E. Davis, J. F. Moulder, and G. E. Muilenb-
erg, Handbook of X-Ray Photoelectron Spectroscopy. Perkin-Elmer Corporation,
Eden Prairie, MN, 1979.
J. Sun, L. Gao, and Q. H. Zhang, ‘‘Synthesizing and Comparing the Photo-
30
36
J. R. Wank, A. W. Weimer, and S. M. George, ‘‘Vibro-Fluidization of
Fine Boron Nitride Powder at Low Pressure,’’ Powder Technol., 121, 195–204
2001).
catalytic Properties of High Surface Area Rutile and Anatase Titania Nanopar-
ticles,’’ J. Am. Ceram. Soc., 86, 1677–82 (2003).
(
&